Nearly OT: VF16 harddisk source

[Les Cargill]

Turns out that Micro Center ( at least the one in Houston ) has a
"vintage rack" and I found an EIDE 40GB drive that seems so far to
work perfectly on my old Fostex VF16.

it has a sticker that says it's a refurb.

--
Les Cargill

[Don Y]

Hi Les,

Sorry, I'm not aware of much of this kit. What's the "problem"
that the 40G drive seems to have solved for you? Does the device
only work with "40G PATA drives"?

[Mike Rivers]

On 5/28/2012 2:58 PM, Les Cargill wrote:
>
> Turns out that Micro Center ( at least the one in Houston )
> has a "vintage rack" and I found an EIDE 40GB drive

> it has a sticker that says it's a refurb.

I've bought a few of those to have on hand to use with my
Mackie hard disk recorder (which will actually use up to a
120 GB drive with the updated BIOS). I haven't worked them
very hard, but they format and record fine at least one pass
to fill up the whole drive.

I'm not sure what they do when they refurbish a disk drive.
Probably just test it to see if it works, and then put a
"refurbished" sticker on it. I doubt that they put in new
electronics, new motors, or new platters.



--
"Today's production equipment is IT based and cannot be
operated without a passing knowledge of computing, although
it seems that it can be operated without a passing knowledge
of audio." - John Watkinson

http://mikeriversaudio.wordpress.com - useful and
interesting audio stuff

[Les Cargill]

Don Y wrote:
> Hi Les,
>
> On 5/28/2012 11:58 AM, Les Cargill wrote:
>>
>> Turns out that Micro Center ( at least the one in Houston ) has a
>> "vintage rack" and I found an EIDE 40GB drive that seems so far to
>> work perfectly on my old Fostex VF16.
>>
>> it has a sticker that says it's a refurb.
>
> Sorry, I'm not aware of much of this kit. What's the "problem" that
> the 40G drive seems to have solved for you?

The Fostex VF16 is an old ( 2000 vintage ) all in one harddisk
recorder. The problem it solves for me is availability of
replacement drives, at least drives where the primary
failure was electronic rather than mechanical.

Prior to finding them in stock in a brick and mortar, you
took your chances on an online "retailer", and those
were a lot more risky.

My reason for posting is that others may still use something
like a Fostex or Roland all in one, and this may be a good
resource for them.

I still have it because it's about the only 16 track recorder in
its size. You have to use a Behringer ADA8000 equivalent 8 channel
lightpipe device to get to 16 tracks.

The VF16 and VF-160 both also chase SMPTE sync at 30nd ( with a
box to convert SMPTE to MTC - code comes into the box via
a MIDI port). It tolerates up to 5 or ten percent speed variation
before it loses sync.

Nobody cares much about remote multitrack recording any more,
but this is a darn good platform for it.

> Does the device
> only work with "40G PATA drives"?

It only works with PATA drives of 40 GB or less. There's a
firmware upgrade to get to 80GB, but for reasons I won't bore
you with, it's not possible where I sit.

--
Les Cargill

[Les Cargill]

Mike Rivers wrote:
> On 5/28/2012 2:58 PM, Les Cargill wrote:
>>
>> Turns out that Micro Center ( at least the one in Houston )
>> has a "vintage rack" and I found an EIDE 40GB drive
>
>> it has a sticker that says it's a refurb.
>
>
> I've bought a few of those to have on hand to use with my Mackie hard
> disk recorder (which will actually use up to a 120 GB drive with the
> updated BIOS). I haven't worked them very hard, but they format and
> record fine at least one pass to fill up the whole drive.
>
> I'm not sure what they do when they refurbish a disk drive. Probably
> just test it to see if it works, and then put a "refurbished" sticker on
> it. I doubt that they put in new electronics, new motors, or new platters.
>
>
>

I doubt that, too - but some people will replace
the circuit boards on dead drives. Motors and
platters sound like they would be too hard to do.

--
Les Cargill

[John Williamson]

On a refurbished drive, they *should* have done a low level format and
platter check, mapping any bad or marginal sectors, this will also
realign the tracks with the heads. They should also replace the drive
electronics.

Whether they do this or not is your guess, and you'll never be able to
tell if all they've done is pull it, format it and run a disk checking
program over it. It may last an hour or years, and you'll not find out
which until it fails.

--
Tciao for Now!

John.

[Don Y]

On 5/28/2012 1:06 PM, Les Cargill wrote:
> Don Y wrote:
>> Hi Les,
>>
>> On 5/28/2012 11:58 AM, Les Cargill wrote:
>>>
>>> Turns out that Micro Center ( at least the one in Houston ) has a
>>> "vintage rack" and I found an EIDE 40GB drive that seems so far to
>>> work perfectly on my old Fostex VF16.
>>>
>>> it has a sticker that says it's a refurb.
>>
>> Sorry, I'm not aware of much of this kit. What's the "problem" that
>> the 40G drive seems to have solved for you?
>
> The Fostex VF16 is an old ( 2000 vintage ) all in one harddisk
> recorder. The problem it solves for me is availability of
> replacement drives, at least drives where the primary
> failure was electronic rather than mechanical.

And, those replacements need to be small-ish. Understood.
I keep several 300MB (that's M, not G!) drives for odd bits
of kit I have around here...

> Prior to finding them in stock in a brick and mortar, you
> took your chances on an online "retailer", and those
> were a lot more risky.
>
> My reason for posting is that others may still use something
> like a Fostex or Roland all in one, and this may be a good
> resource for them.
>
> I still have it because it's about the only 16 track recorder in
> its size. You have to use a Behringer ADA8000 equivalent 8 channel
> lightpipe device to get to 16 tracks.
>
> The VF16 and VF-160 both also chase SMPTE sync at 30nd ( with a
> box to convert SMPTE to MTC - code comes into the box via
> a MIDI port). It tolerates up to 5 or ten percent speed variation
> before it loses sync.
>
> Nobody cares much about remote multitrack recording any more,
> but this is a darn good platform for it.
>
> > Does the device
>> only work with "40G PATA drives"?
>
> It only works with PATA drives of 40 GB or less. There's a
> firmware upgrade to get to 80GB, but for reasons I won't bore
> you with, it's not possible where I sit.

Have you tried a > 40G drive in it to see what happens?
E.g., 60G? It might just treat the drive as a 40G drive
which, to someone wanting 60G of storage, would seem like
a bummer. But, if all it *can* use is 40, this might
give you an option that you might not have had, otherwise.

Also, note that some larger drives have strapping options that
let you limit their apparent size to ~32G. Again, just another
potential option!

A good source of smaller IDE drives are neighbors with "old"
computers that they are upgrading. I now only save ~160G+
drives pulled from these sorts of things.

If your community has a hazardous waste recycling program,
you might enquire if you can cherry pick through the items
dropped off for that program.

[John Williamson]

There are also programs available which will tell the HD to lie about
its size to the BIOS of whatever it's attached to, and the operating
system will then often see the full size of the drive.

[Don Y]

Hi John,

On 5/28/2012 2:10 PM, John Williamson wrote:

>> Also, note that some larger drives have strapping options that
>> let you limit their apparent size to ~32G. Again, just another
>> potential option!
>>
> There are also programs available which will tell the HD to lie about
> its size to the BIOS of whatever it's attached to, and the operating
> system will then often see the full size of the drive.

The amusing thing is that IDE drives originally needed to be
*told* what their sizes/geometries were.

Anything more recent than that undoubtedly expects the
drive to do the "telling" :-/

You could also look into solid state media (40G isn't
that big!) but I don't know how "durable" it would be
as I don't know what the usage pattern of the device in
question is likely to be (and SSD's tend to be tuned to
particular types of usage)

Finally, you could also look for laptop drives (I'm assuming
the OP was speaking of a 3" drive) as they are trivial to
interface to a conventional IDE cable (OTOH, laptop drives
tend to be a lot slower than desktop drives)

[Les Cargill]

Yeah. Some work, some don't.

> E.g., 60G? It might just treat the drive as a 40G drive
> which, to someone wanting 60G of storage, would seem like
> a bummer. But, if all it *can* use is 40, this might
> give you an option that you might not have had, otherwise.
>

Aye. Yes, there are a lot of 80GB drives out there, and
some work, and some don't.

> Also, note that some larger drives have strapping options that
> let you limit their apparent size to ~32G. Again, just another
> potential option!
>

Yessir. That's failed every time I've tried it...

> A good source of smaller IDE drives are neighbors with "old"
> computers that they are upgrading. I now only save ~160G+
> drives pulled from these sorts of things.
>
> If your community has a hazardous waste recycling program,
> you might enquire if you can cherry pick through the items
> dropped off for that program.

Or you can go to MicroCenter and pay them < $20 for somebody to do it
for you...

--
Les Cargill

[Don Y]

Hi Les,

On 5/28/2012 2:43 PM, Les Cargill wrote:

>>>> Does the device
>>>> only work with "40G PATA drives"?
>>>
>>> It only works with PATA drives of 40 GB or less. There's a
>>> firmware upgrade to get to 80GB, but for reasons I won't bore
>>> you with, it's not possible where I sit.
>>
>> Have you tried a > 40G drive in it to see what happens?
>
> Yeah. Some work, some don't.

Have you looked at the "good" drives and "bad" drives to
see if there is anything they have in common? E.g.,
a certain number of cylinders, sectors per track, etc.?

Of course, this depends on just how curious you are.
You may prefer to just "move on" each time you stumble across
a problem drive in the hope that a "good" drive is waiting in
the queue. I tend to see too many things as puzzles and
want to "solve" them -- if only to give me some insight into
how something was designed or intended to behave.

>> E.g., 60G? It might just treat the drive as a 40G drive
>> which, to someone wanting 60G of storage, would seem like
>> a bummer. But, if all it *can* use is 40, this might
>> give you an option that you might not have had, otherwise.
>
> Aye. Yes, there are a lot of 80GB drives out there, and
> some work, and some don't.
>
>> Also, note that some larger drives have strapping options that
>> let you limit their apparent size to ~32G. Again, just another
>> potential option!
>
> Yessir. That's failed every time I've tried it...

Worth knowing. I've never tried strapping a drive for a smaller
size. Seemed wasteful :-/

>> A good source of smaller IDE drives are neighbors with "old"
>> computers that they are upgrading. I now only save ~160G+
>> drives pulled from these sorts of things.
>>
>> If your community has a hazardous waste recycling program,
>> you might enquire if you can cherry pick through the items
>> dropped off for that program.
>
> Or you can go to MicroCenter and pay them < $20 for somebody to do it
> for you...

Yup. Though you might find that they've just "pulled"
the drive and done little more to prove its functionality
than apply power and see if it is recognized as a drive!
(for $20, you don't get a helluvalot of someone's *time*!)

E.g., one of the places where I volunteer just pulls drives
(CD/DVD readers/writers, etc.) and piles them up in the store
for sale by the pound. If it doesn't work, bring it back for
an equivalent "weight" drive. It's just not worth testing
them at the quantities we see.

Of course, if it *fails* a month later, it's *yours*! :>
(we don't guarantee reliability -- just that it *should*
work when you buy it -- but we don't want to bother
*testing* it... which is why the drives are so cheap!
*You* take on the role of providing that testing)

We can't use the same approach with recordable media as
donors would scream if the drives weren't "wiped" before
being reused. (amazing the amount of cruft people leave
on their disks! :< ) So, small drives (< 200GB) tend
to just be destroyed as the time to test them doesn't
justify their value (in refurbished systems).

[Les Cargill]

Don Y wrote:
> Hi Les,
>
> On 5/28/2012 2:43 PM, Les Cargill wrote:
>
>>>>> Does the device
>>>>> only work with "40G PATA drives"?
>>>>
>>>> It only works with PATA drives of 40 GB or less. There's a
>>>> firmware upgrade to get to 80GB, but for reasons I won't bore
>>>> you with, it's not possible where I sit.
>>>
>>> Have you tried a > 40G drive in it to see what happens?
>>
>> Yeah. Some work, some don't.
>
> Have you looked at the "good" drives and "bad" drives to
> see if there is anything they have in common? E.g.,
> a certain number of cylinders, sectors per track, etc.?
>
> Of course, this depends on just how curious you are.

it depends more on how many drives I can lay hands on. I've
had... a half dozen? The first 80GB drive I tried was an old
castoff from a machine we otherwise gave away. I happened to try it,
and it worked. I bought two more, and neither of those did.

> You may prefer to just "move on" each time you stumble across
> a problem drive in the hope that a "good" drive is waiting in
> the queue. I tend to see too many things as puzzles and
> want to "solve" them -- if only to give me some insight into
> how something was designed or intended to behave.
>

Sure. It's a mystery to me why the Fostex folks had any
dependency on drive size at all. I suppose it was memory
limitations, but it's not that hard to pack rectangles
into a single rectangle, then stripe that out as a
linear structure.

It's more like they built it to a drive they knew they
could get, then tested others to see if they worked.

>>> E.g., 60G? It might just treat the drive as a 40G drive
>>> which, to someone wanting 60G of storage, would seem like
>>> a bummer. But, if all it *can* use is 40, this might
>>> give you an option that you might not have had, otherwise.
>>
>> Aye. Yes, there are a lot of 80GB drives out there, and
>> some work, and some don't.
>>
>>> Also, note that some larger drives have strapping options that
>>> let you limit their apparent size to ~32G. Again, just another
>>> potential option!
>>
>> Yessir. That's failed every time I've tried it...
>
> Worth knowing. I've never tried strapping a drive for a smaller
> size. Seemed wasteful :-/
>

I haven't tried it many times.

>>> A good source of smaller IDE drives are neighbors with "old"
>>> computers that they are upgrading. I now only save ~160G+
>>> drives pulled from these sorts of things.
>>>
>>> If your community has a hazardous waste recycling program,
>>> you might enquire if you can cherry pick through the items
>>> dropped off for that program.
>>
>> Or you can go to MicroCenter and pay them < $20 for somebody to do it
>> for you...
>
> Yup. Though you might find that they've just "pulled"
> the drive and done little more to prove its functionality
> than apply power and see if it is recognized as a drive!
> (for $20, you don't get a helluvalot of someone's *time*!)
>

I wouldn't be surprised if that were the case. Don't see
how they can recover the cost of keeping the thing on the shelf,
really. The value here is aggregating a bunch of drives for sale
that you can pick through.

> E.g., one of the places where I volunteer just pulls drives
> (CD/DVD readers/writers, etc.) and piles them up in the store
> for sale by the pound. If it doesn't work, bring it back for
> an equivalent "weight" drive. It's just not worth testing
> them at the quantities we see.
>

Right.

> Of course, if it *fails* a month later, it's *yours*! :>
> (we don't guarantee reliability -- just that it *should*
> work when you buy it -- but we don't want to bother
> *testing* it... which is why the drives are so cheap!
> *You* take on the role of providing that testing)
>
> We can't use the same approach with recordable media as
> donors would scream if the drives weren't "wiped" before
> being reused. (amazing the amount of cruft people leave
> on their disks! :< ) So, small drives (< 200GB) tend
> to just be destroyed as the time to test them doesn't
> justify their value (in refurbished systems).

Right. Although in this case, I will be formatting it as a Fostex
proprietary F/S "FDMS" drive, so there'll be no reading of what's
on there before.


I kind of expected there to be a replacement for the Fostex on
offer, but there really hasn't been. Fostex' own offering is
interesting, but overkill ( and has a price that matches ). If
I could make my money back, I'd invest in a JoeCo Blackbox and
the other stuff needed, but people make do with cameraphones,
small HD movie cams or H4 type devices.

--
Les Cargill

[Don Y]

Hi Les,

On 5/28/2012 3:51 PM, Les Cargill wrote:

>> Have you looked at the "good" drives and "bad" drives to
>> see if there is anything they have in common? E.g.,
>> a certain number of cylinders, sectors per track, etc.?
>>
>> Of course, this depends on just how curious you are.
> it depends more on how many drives I can lay hands on. I've
> had... a half dozen? The first 80GB drive I tried was an old
> castoff from a machine we otherwise gave away. I happened to try it,
> and it worked. I bought two more, and neither of those did.

But you didn't "talk to the drives" to see what they
said. I.e., they could have had different geometries
(assuming they actually *worked* at all).

I.e., pretend you are the product and see what it
sees -- and figure out what it likes/dislikes based
on those observations.

>> You may prefer to just "move on" each time you stumble across
>> a problem drive in the hope that a "good" drive is waiting in
>> the queue. I tend to see too many things as puzzles and
>> want to "solve" them -- if only to give me some insight into
>> how something was designed or intended to behave.
>
> Sure. It's a mystery to me why the Fostex folks had any
> dependency on drive size at all. I suppose it was memory
> limitations, but it's not that hard to pack rectangles
> into a single rectangle, then stripe that out as a
> linear structure.

They might have opted for a simplistic mapping of samples
and tracks onto the disk's structure. I am always amazed at
how "uninspired" many solutions actually are! E.g.,
all the contortions over FAT12, FAT16, FAT32, 137GB limits,
Y2K, Y2038, etc. Sheesh! Talk about saving micropennies
and wasting megadollars!! :<

> It's more like they built it to a drive they knew they
> could get, then tested others to see if they worked.

Is the device tolerant of a wider variety of < 40G drives?
Or, is it "picky" there, too?

>>> Or you can go to MicroCenter and pay them < $20 for somebody to do it
>>> for you...
>>
>> Yup. Though you might find that they've just "pulled"
>> the drive and done little more to prove its functionality
>> than apply power and see if it is recognized as a drive!
>> (for $20, you don't get a helluvalot of someone's *time*!)
>
> I wouldn't be surprised if that were the case. Don't see
> how they can recover the cost of keeping the thing on the shelf,
> really. The value here is aggregating a bunch of drives for sale
> that you can pick through.

Yup. We treat cables and wall warts similarly. Just toss
them in a big "bin" and let folks sort through them. Some
folks can get great deals -- if you want to be patient
and sort through the mess. Other folks with less patience
pay higher prices from "real vendors".

<shrug>

I learned early on that cables will eat your lunch (budget)
so I tend to stockpile all different sorts -- even if I
haven't a clue as to their intended uses! :> (sooner or
later, I'll stumble on just such a "need" and I'll be
tickled to remember that I have "just the right cable"
squirreled away)

>> We can't use the same approach with recordable media as
>> donors would scream if the drives weren't "wiped" before
>> being reused. (amazing the amount of cruft people leave
>> on their disks! :< ) So, small drives (< 200GB) tend
>> to just be destroyed as the time to test them doesn't
>> justify their value (in refurbished systems).
>
> Right. Although in this case, I will be formatting it as a Fostex
> proprietary F/S "FDMS" drive, so there'll be no reading of what's
> on there before.

Sure. I'm just commenting that you might be surprised at
what you'll find on the drive if you put it into a more
conventional environment! :> (taking care as to how you
*access* those "foreign" files!)

One donor bulk erases all drives before disposing of them.
I.e., even the servo tracks are wiped so the drives are
essentially paperweights.

Another drills 3/8" holes through the platters (effective for
the sort of casual "snoop")

> I kind of expected there to be a replacement for the Fostex on
> offer, but there really hasn't been. Fostex' own offering is
> interesting, but overkill ( and has a price that matches ). If
> I could make my money back, I'd invest in a JoeCo Blackbox and
> the other stuff needed, but people make do with cameraphones,
> small HD movie cams or H4 type devices.

Have you tried laptop drives? Or even a CF device? Both
should be wins in terms of weight and power...

[Mike Rivers]

On 5/28/2012 4:53 PM, Don Y wrote:

> Have you tried a > 40G drive in it to see what happens?
> E.g., 60G? It might just treat the drive as a 40G drive
> which, to someone wanting 60G of storage, would seem like
> a bummer. But, if all it *can* use is 40, this might
> give you an option that you might not have had, otherwise.

My experience with the original Mackie hard disk recorder,
which originally shipped with a 20 GB ATA-100 drive, had a
BIOS that simply wouldn't recognize a drive greater than 32
GB as a disk drive. You could use a 40 GB drive with a
reduced-capacity jumper installed, but depending on the
drive's firmware, the BIOS would recognize it as 30, 8, or 2
GB.

Several years after the HDR24/96's introduction, Mackie came
up with a replacement BIOS for the motherboard that would
recognize a larger drive, however the operating system
(although it was a PC motherboard, it didn't run under
Windows or DOS) pooped out at 130 GB. Back when 160 GB was
the smallest IDE drive that could be bought over the
counter, I used a few of those as 130 GB drives.

The simple solution would have been to use the flash BIOS
update that the motherboard manufacturer offered, but Mackie
tested that and found that there were some instabilities
with their software that weren't easy to fix.

Welcome to the world of obsolete software-based equipment.
The Ampex AG-440 will probably live forever.

Oh, and did you notice that we never had to say "PATA" until
someone came up with "SATA." Couldn't they come up with a
new name so we wouldn't have to change the old name? ;)

> A good source of smaller IDE drives are neighbors with "old"
> computers that they are upgrading. I now only save ~160G+
> drives pulled from these sorts of things.

I thought that Mackie's original design with a 20 GB drive
was good. It was enough "tape" for a single project recorded
with a certain amount of common sense and in the style of
multitrack tape - not keeping every bad take or useless idea
because there's pleny of disk space. A removable drive
could be put on the shelf, backed up however you want,
handed to the client at the end of the job, or re-formatted
and used for another project. I never felt very good about
putting a bunch of (usually unrelated) projects on a single
disk drive.

[Mike Rivers]

On 5/28/2012 6:08 PM, Don Y wrote:

> You may prefer to just "move on" each time you stumble across
> a problem drive in the hope that a "good" drive is waiting in
> the queue. I tend to see too many things as puzzles and
> want to "solve" them -- if only to give me some insight into
> how something was designed or intended to behave.

The nice thing about buying stuff at local retailers is that
you can return it if it doesn't work. When I was buying
drives for the Mackie recorder on a fairly regular basis, I
watched out for sales. If I bought one that didn't work, I'd
just take it back and either get a refund or exchange it for
another brand. I'm curious about a lot of things, but not
when it comes to computer stuff.

>> Or you can go to MicroCenter and pay them < $20 for
>> somebody to do it
>> for you...
>
> Yup. Though you might find that they've just "pulled"
> the drive and done little more to prove its functionality
> than apply power and see if it is recognized as a drive!
> (for $20, you don't get a helluvalot of someone's *time*!)

That may very well be enough. I've had a lot of hard drives
over the last 25+ years and only two actual failures. They
get too small or too slow for our bloating applications, but
rarely break. Your (the editorial "you") drives, on the
other hand, DO fail in ways that none of your data is
recoverable, and it's happened enough times so you now back
up everything every day. Nearly everyone has a story (except
me).

> We can't use the same approach with recordable media as
> donors would scream if the drives weren't "wiped" before
> being reused.

It may also not be a good idea to use a "non-certified"
drive as music recording media lest it fail with thousands
of hours of studio time and never-to-be-repeated takes
stored on it.

[Mike Rivers]

On 5/28/2012 6:51 PM, Les Cargill wrote:

> Sure. It's a mystery to me why the Fostex folks had any
> dependency on drive size at all. I suppose it was memory
> limitations, but it's not that hard to pack rectangles
> into a single rectangle, then stripe that out as a
> linear structure.

Back when that recorder was designed, 40 GB seemed like more
space than any sensible person could use. Operating systems
could only count so high so there was a limit to the number
of bytes that could be addressed. In the early days of big
drives, we'd update the BIOS so that the hardware could
recognize a larger drive than could originally be
accommodated, and then we partitioned it so that it looked
like several drives small enough for the OS to use.

In the case of the Mackie recorder, the operating system
didn't recognize DOS partitions beyond the primary
partition, and there were no provisions in that OS for
partitioning in a way that it could recognize.

As far as drives go, Mackie torture tested all of the drives
on the market that were in the price range that made sense
for the price of the recorder. They found some that simply
didn't work. Either they failed to format or failed the real
time read/write test that was part of the drive
initialization procedure. This was why they resisted, for a
long time, making it easy to "roll your own" removable
drive. They settled on two models, one from Maxtor and one
from whoever was making drives under the IBM name at the
time (probably Hitachi) and got contracts with both
companies to supply that model, unchanged, for, I think, six
years. By that time, just about any drive you could buy
would work reliably in the recorder.

I don't know if Fostex was that methodical, but the
certainly (should have) had the same concerns as Mackie

I still have an uneasy feeling about the recorders I'm using
now that record to postage stamp sized memory cards. I've
never backed up so much in my life! And I've never really
been unable to find a reel of 2" tape that I knew I had
SOMEWHERE. Not so with SD memory cards. ')

[Mike Rivers]

On 5/28/2012 4:26 PM, John Williamson wrote:

> On a refurbished drive, they *should* have done a low level
> format and platter check, mapping any bad or marginal
> sectors, this will also realign the tracks with the heads.
> They should also replace the drive electronics.
>
> Whether they do this or not is your guess, and you'll never
> be able to tell if all they've done is pull it, format it
> and run a disk checking program over it. It may last an hour
> or years, and you'll not find out which until it fails.

And the only difference between that and a brand new drive
is that the new drive carries a 2 or 3 or 5 year warranty.
If it fails within the warranty period, they'll send you a
new one, but they won't send it with all your data nor
compensate you for your data loss.

[John Williamson]

Mike Rivers wrote:
> On 5/28/2012 4:26 PM, John Williamson wrote:
>
>> On a refurbished drive, they *should* have done a low level
>> format and platter check, mapping any bad or marginal
>> sectors, this will also realign the tracks with the heads.
>> They should also replace the drive electronics.
>>
>> Whether they do this or not is your guess, and you'll never
>> be able to tell if all they've done is pull it, format it
>> and run a disk checking program over it. It may last an hour
>> or years, and you'll not find out which until it fails.
>
> And the only difference between that and a brand new drive is that the
> new drive carries a 2 or 3 or 5 year warranty. If it fails within the
> warranty period, they'll send you a new one, but they won't send it with
> all your data nor compensate you for your data loss.
>

Which is why I run a backup regime that some people think is excessive.

[Don Y]

Hi Mike,

On 5/29/2012 4:10 AM, Mike Rivers wrote:
> On 5/28/2012 4:53 PM, Don Y wrote:
>
>> Have you tried a > 40G drive in it to see what happens?
>> E.g., 60G? It might just treat the drive as a 40G drive
>> which, to someone wanting 60G of storage, would seem like
>> a bummer. But, if all it *can* use is 40, this might
>> give you an option that you might not have had, otherwise.
>
> My experience with the original Mackie hard disk recorder, which
> originally shipped with a 20 GB ATA-100 drive, had a BIOS that simply
> wouldn't recognize a drive greater than 32 GB as a disk drive. You could
> use a 40 GB drive with a reduced-capacity jumper installed, but
> depending on the drive's firmware, the BIOS would recognize it as 30, 8,
> or 2 GB.

OK. So, someone was short-sighted and/or wrote buggy code.

> Several years after the HDR24/96's introduction, Mackie came up with a
> replacement BIOS for the motherboard that would recognize a larger
> drive, however the operating system (although it was a PC motherboard,
> it didn't run under Windows or DOS) pooped out at 130 GB. Back when 160
> GB was the smallest IDE drive that could be bought over the counter, I
> used a few of those as 130 GB drives.

It is (somewhat) understandable for the BIOS to be short-sighted
and not comprehend LBA's exceeding 48 bits. After all, they
are dealing with "real hardware" and that hardware didn't
even know what larger (wider) LBA's would be!

For the operating system to carry this limitation *up* through
it's internal structure is just plain stupid. Akin to only being
able to print dates in which the first digit of the year is a '1'!

> The simple solution would have been to use the flash BIOS update that
> the motherboard manufacturer offered, but Mackie tested that and found
> that there were some instabilities with their software that weren't easy
> to fix.

Yeah. They didn't think about anything bigger than 128GiB (137GB).
Abd, since this oversight has ramifications throughout the OS,
it's not trivial to fix.

OTOH, had the interface from the hardware abstraction layer
implemented, for example, a 64b LBA -- or even a 64b medium
size (64b - 9b/block = 55b LBA... for a ~15TB media limitation
IN THE OS!) -- then the BIOS could later easily be updated
and that extra capacity made available throughout the OS.

> Welcome to the world of obsolete software-based equipment. The Ampex
> AG-440 will probably live forever.

Yeah, I keep a flint for the day we forget how to make fire! :>

> Oh, and did you notice that we never had to say "PATA" until someone
> came up with "SATA." Couldn't they come up with a new name so we
> wouldn't have to change the old name? ;)

IDE -> EIDE -> PATA

At least it's not as bad as the alphabet soup surrounding SCSI drives!

>> A good source of smaller IDE drives are neighbors with "old"
>> computers that they are upgrading. I now only save ~160G+
>> drives pulled from these sorts of things.
>
> I thought that Mackie's original design with a 20 GB drive was good. It
> was enough "tape" for a single project recorded with a certain amount of
> common sense and in the style of multitrack tape - not keeping every bad
> take or useless idea because there's pleny of disk space. A removable
> drive could be put on the shelf, backed up however you want, handed to
> the client at the end of the job, or re-formatted and used for another
> project. I never felt very good about putting a bunch of (usually
> unrelated) projects on a single disk drive.

Good point. With "computers", we tend to view the drive as
a closet or desk drawer into which we can shove all our
(unrelated) crap.

OTOH, I suspect that if you used this device for post processing
work, as well, your attitude might change. I.e., having to pull
projects (which may be only partially completed) off of it to
make room for new projects would tend to cause you to look for
a larger medium -- meaning cruft would tend to accumulate. :>

("Files grow and multiply to fill all space available")

[Don Y]

Hi Mike,

On 5/29/2012 4:21 AM, Mike Rivers wrote:
> On 5/28/2012 6:08 PM, Don Y wrote:
>
>> You may prefer to just "move on" each time you stumble across
>> a problem drive in the hope that a "good" drive is waiting in
>> the queue. I tend to see too many things as puzzles and
>> want to "solve" them -- if only to give me some insight into
>> how something was designed or intended to behave.
>
> The nice thing about buying stuff at local retailers is that you can
> return it if it doesn't work.

Sure -- if the "doesn't work" is related to an incompatibility
in the design. If you can suss out what aspect of the design
upsets your device, then you can make smarter choices without
investing the time to buy-and-try.

OTOH, if "doesn't work" means *breaks* -- especially after you've
started using it -- then that's a non-starter.

> When I was buying drives for the Mackie
> recorder on a fairly regular basis, I watched out for sales. If I bought
> one that didn't work, I'd just take it back and either get a refund or
> exchange it for another brand. I'm curious about a lot of things, but
> not when it comes to computer stuff.

Ah, I design devices for a living so I am always interesteed in
seeing how things are designed, fabricated, flaws they might
have, etc. Learn from other peoples' efforts and mistakes
in the hope of reducing the number that *I* make! :>

>>> Or you can go to MicroCenter and pay them < $20 for
>>> somebody to do it
>>> for you...
>>
>> Yup. Though you might find that they've just "pulled"
>> the drive and done little more to prove its functionality
>> than apply power and see if it is recognized as a drive!
>> (for $20, you don't get a helluvalot of someone's *time*!)
>
> That may very well be enough. I've had a lot of hard drives over the
> last 25+ years and only two actual failures.

I've had one over the past 30+ years -- and that was only
recently (a laptop drive that probably got just a wee bit
too warm for too long -- but, I expect less from laptops
so I wasn't disappointed. I also have a stack of laptop
drives in case of this sort of thing)

> They get too small or too
> slow for our bloating applications, but rarely break.

Exactly. Hence my comment re: pulling drives out of your
friends' and neighbors' old machines as they upgrade
(that's where all my laptop drives came from. Though,
for the life of me, I can't see the need for 100G drives
in laptops -- no doubt because I wouldn't use a laptop
as a "work machine")

> Your (the
> editorial "you") drives, on the other hand, DO fail in ways that none of
> your data is recoverable, and it's happened enough times so you now back
> up everything every day. Nearly everyone has a story (except me).

I've only (almost!) lost data from a non-hardware failure once.
That was caused by a bug introduced into the disk "driver" in
an OS upgrade. Apparently tickled a problem in this particular
make/model of disk drive which became apparent when I replaced
the scrambled disk with the cold backup -- same make/model and
watched it suffer the same fate as quickly as the first! "Hmmm...
there's something screwwy going on, here! Let's not mount the
M.O. backup and put *it* at risk until we understand what
trashed these two drives!"

>> We can't use the same approach with recordable media as
>> donors would scream if the drives weren't "wiped" before
>> being reused.
>
> It may also not be a good idea to use a "non-certified" drive as music
> recording media lest it fail with thousands of hours of studio time and
> never-to-be-repeated takes stored on it.

Do you really think you're going to pick up a "used, obsolete"
drive for $20 and have the supplier give you any assurances other
than, "If it dies in the first 30 days, bring it back and we'll
give you another one?" :> (Heck, they won't do much more than
that -- witht he exception of the timeframe -- for *new* drives!)

[Don Y]

Hi John,

How often do you *test* your backups? I've always been amused
at folks who buy oddball devices for backup (e.g., I can recall
*businesses* using repurposed VCR's with special external
processors to encode the data as video signal!) and/or don't
religiously use them. Or, don't practice recovery operations
to keep their skills honed and verify the backups are of high
integrity.

I panic, now, as my archives have grown so large that I
have to keep a lot on far fewer spindles. I.e., one goes
down and I there's a lot at stake. Even restoring the data
on a single spindle can take the better part of a day. :<

[Les Cargill]

Mike Rivers wrote:
> On 5/28/2012 6:51 PM, Les Cargill wrote:
>
>> Sure. It's a mystery to me why the Fostex folks had any
>> dependency on drive size at all. I suppose it was memory
>> limitations, but it's not that hard to pack rectangles
>> into a single rectangle, then stripe that out as a
>> linear structure.
>
> Back when that recorder was designed, 40 GB seemed like more space than
> any sensible person could use. Operating systems could only count so
> high

They can count to 2^32. It's in hardware. Like I said, there's
no compelling reason for it.

> so there was a limit to the number of bytes that could be
> addressed. In the early days of big drives, we'd update the BIOS so that
> the hardware could recognize a larger drive than could originally be
> accommodated, and then we partitioned it so that it looked like several
> drives small enough for the OS to use.
>

... or we'd load drivers. We were Doing It Wrong. That's because
the PC was a toy with delusions of grandeur.

> In the case of the Mackie recorder, the operating system didn't
> recognize DOS partitions beyond the primary partition, and there were no
> provisions in that OS for partitioning in a way that it could recognize.
>
> As far as drives go, Mackie torture tested all of the drives on the
> market that were in the price range that made sense for the price of the
> recorder. They found some that simply didn't work.

Yep. Part of that is because there wasn't a large number of filesystems
about they felt they could use, or they had NIH syndrome, or it was
simply a cultural blind spot.

> Either they failed to
> format or failed the real time read/write test that was part of the
> drive initialization procedure. This was why they resisted, for a long
> time, making it easy to "roll your own" removable drive.

No, the "why" is "we want to sell drives" or "we want to ship now,
rather than hold release for finishing that part of it."


Either decision has advantages or disadvantages.

> They settled on
> two models, one from Maxtor and one from whoever was making drives under
> the IBM name at the time (probably Hitachi)

Probably - those figure prominently in the Approved
Media List for Fostex, too.

> and got contracts with both
> companies to supply that model, unchanged, for, I think, six years. By
> that time, just about any drive you could buy would work reliably in the
> recorder.
>
> I don't know if Fostex was that methodical, but the certainly (should
> have) had the same concerns as Mackie
>
> I still have an uneasy feeling about the recorders I'm using now that
> record to postage stamp sized memory cards. I've never backed up so much
> in my life! And I've never really been unable to find a reel of 2" tape
> that I knew I had SOMEWHERE. Not so with SD memory cards. ')
>
>
>

Yep.

--
Les Cargill

[John Williamson]

On 29/05/2012 17:56, Don Y wrote:
> Hi John,
>
> On 5/29/2012 4:50 AM, John Williamson wrote:
>> Mike Rivers wrote:
>>> And the only difference between that and a brand new drive is that the
>>> new drive carries a 2 or 3 or 5 year warranty. If it fails within the
>>> warranty period, they'll send you a new one, but they won't send it
>>> with all your data nor compensate you for your data loss.
>>
>> Which is why I run a backup regime that some people think is excessive.
>
> How often do you *test* your backups? I've always been amused
> at folks who buy oddball devices for backup (e.g., I can recall
> *businesses* using repurposed VCR's with special external
> processors to encode the data as video signal!) and/or don't
> religiously use them. Or, don't practice recovery operations
> to keep their skills honed and verify the backups are of high
> integrity.
>

The backups get restored from every few months, normally when I
re-install Windows or buy a new, larger, HD for one of the computers.
The files are also individually verified on copy.

Now watch the next restore fall over in a comprehensive manner...

> I panic, now, as my archives have grown so large that I
> have to keep a lot on far fewer spindles. I.e., one goes
> down and I there's a lot at stake. Even restoring the data
> on a single spindle can take the better part of a day. :<

My record so far has been a backup that took a couple of days to run
across what turned out to be a USB 1.0 link. Which crashed during the
backup operation, corrupting the FAT.:-/

The scariest was when I was told at work many years ago to "Run this
backup program, which copies everything onto that Zip disc." When it
only took a few seconds to "back up" the whole of the company database,
I checked. It hadn't. Most of the files weren't even on the backup disc
as old versions, so it hadn't been doing anything since the backup
program had been installed.

[Don Y]

Hi John,

On 5/29/2012 10:20 AM, John Williamson wrote:

>>>> And the only difference between that and a brand new drive is that the
>>>> new drive carries a 2 or 3 or 5 year warranty. If it fails within the
>>>> warranty period, they'll send you a new one, but they won't send it
>>>> with all your data nor compensate you for your data loss.
>>>
>>> Which is why I run a backup regime that some people think is excessive.
>>
>> How often do you *test* your backups? I've always been amused
>> at folks who buy oddball devices for backup (e.g., I can recall
>> *businesses* using repurposed VCR's with special external
>> processors to encode the data as video signal!) and/or don't
>> religiously use them. Or, don't practice recovery operations
>> to keep their skills honed and verify the backups are of high
>> integrity.
>
> The backups get restored from every few months, normally when I

You get a gold star! You'd be amazed at how many people MAKE
backups but are clueless as to how (or if) they would actually
restore them when confronted with that (real!) possibility!

I don't worry about things that have inherent backups that
I could recover "the hard way" (i.e., with lots of time and
effort). For example, photos that have been scanned, ripped
music, etc. That's just "time" if a backup fails -- "How
much do you *really* want to recover this in this particular
form?"

I'm considerably more nervous about "original works" which
might represent thousands of hours of time -- and be almost
impossible to recreate in exactly the same form (imagine
recreating a painting from memory -- it might resemble
the original but is hardly IDENTICAL to the original).

I can push copies of "works for hire" back onto my clients
and *hope* they preserve them as they should (of course, I
have had many come back to me having expected *me* to have
preserved the work I did for them previously :< ).

But, I can't always push the 3rd party tools that were used
(and now, no longer available -- at least in those specific
versions) back into their hands. At least, not legally.
So, if I take a hit in one of those spots, I'm really up
the creek... (this is why I backup important stuff in lots
of different places and on different media -- the cost of
loss is just too high!)

> re-install Windows or buy a new, larger, HD for one of the computers.
> The files are also individually verified on copy.
>
> Now watch the next restore fall over in a comprehensive manner...

<grin> Touch wood.

>> I panic, now, as my archives have grown so large that I
>> have to keep a lot on far fewer spindles. I.e., one goes
>> down and I there's a lot at stake. Even restoring the data
>> on a single spindle can take the better part of a day. :<
>
> My record so far has been a backup that took a couple of days to run
> across what turned out to be a USB 1.0 link. Which crashed during the
> backup operation, corrupting the FAT.:-/

Ah. My "slow" media links are now "over the network" (which
tends to max out at 10MB/s -- unless I am talking to one
of the servers over GB fabric). I prefer to bring the media
to the machine, where possible, as it usually results in
higher bandwidth.

I've also discovered that MS network stacks get confused
easily (or at least the SMB redirector -- think: network
neighborhood) which can cut performance down to *bytes*
per second. <frown>

> The scariest was when I was told at work many years ago to "Run this
> backup program, which copies everything onto that Zip disc." When it
> only took a few seconds to "back up" the whole of the company database,

Ha! Small database, eh? :>

> I checked. It hadn't. Most of the files weren't even on the backup disc
> as old versions, so it hadn't been doing anything since the backup
> program had been installed.

But, it probably made everyone BEFORE YOU feel good about themselves
for being *diligent* in their backups!! :-/

I used to use a lot of tape. There, a backup can look like it
is progressing correctly (even with read-after-write technology)
and still not be recoverable, at a later date (or on a different
transport).

Ages ago, with 9-track tape, I used to have to retension the
tapes periodically (essentially, fast forward to the end, then
rewind to the beginning, then unmount) just to minimize the
effects of print-through. Sure happy NOT having to do that
with the current size of my archives! :( I'd be spinning tape
every day -- just to restart the process each time I got the
last reel finished!

[Mike Rivers]

On 5/29/2012 12:37 PM, Don Y wrote:

> OK. So, someone was short-sighted and/or wrote buggy code.

Hey, it was fine for at least 5 years after the product's
introduction. That's end-of-life for most computer-based
products. And as I pointed out, a 20 or 30 GB drive is
practical for a replacement for 2" tape. That's what the
Mackie HDR24/96 was designed to do, and that's what it did,
darn well. I'm still using mine in preference to a
computer-based DAW. The code wasn't buggy, and in an era
when a 20 GB drive was about as large as you could buy, it
isn't fair to call the design "short sighted."

> For the operating system to carry this limitation *up* through
> it's internal structure is just plain stupid. Akin to only
> being
> able to print dates in which the first digit of the year is
> a '1'!

Well, it works for its intended purpose. Mackie purchased a
real time kernel from someone else and wrote the application
around it. Probably the most short-sighted thing they did
(and I know this isn't the perfect machine) was to compile
drivers into the operating system. It seems to work OK with
just about any mouse, but in the HDR, the version with the
GUI, it restricts the graphics board to an ATI Radeon 7000
series, which are getting difficult to find should a
replacement be needed. Worse (and this isn't really Mackie's
problem) with so many "Radeon" graphics cards having been
made, the average for finding the right one that you buy via
an eBay auction seems to be running about 20% these days.
There really isn't a good way to identify the correct one
(too many different numbers) except by trying. And the
motherboard has been recently discontinued (not bad after 12
years) and apparently a modern one isn't a workable
replacement. Obsolete computers are like that.

> Good point. With "computers", we tend to view the drive as
> a closet or desk drawer into which we can shove all our
> (unrelated) crap.
>
> OTOH, I suspect that if you used this device for post
> processing
> work, as well, your attitude might change. I.e., having to pull
> projects (which may be only partially completed) off of it to
> make room for new projects would tend to cause you to look for
> a larger medium -- meaning cruft would tend to accumulate. :>

It has two disk drives, an internal one that contains the
software (only a couple of megabytes worth) and the rest is
available for data. Then there's a removable drive. When I
was recording fairly regularly, my workflow was to dedicated
a removable drive to a project. I'd record on that, and then
back up to the internal drive. I didn't work in an
environment where I had several projects to juggle at one
time, but I'd use the internal drive as backup if I had
projects that involved just a couple of songs.

After the project is mixed, the removable drive goes on the
shelf or given to the artist and the material is deleted
from the internal drive.

> ("Files grow and multiply to fill all space available")

Only if you let them.

[Mike Rivers]

On 5/29/2012 12:51 PM, Don Y wrote:

> > for the life of me, I can't see the need for 100G drives
> in laptops -- no doubt because I wouldn't use a laptop
> as a "work machine")

Get with the times, man! Today, the laptop, as a "desktop
replacement" is the most popular configuration with the
largest customer base. That's not the corporate users, it's
the Facebookers and movie watchers. You know, the people who
have cable TV so they can get high speed internet, then
download movies and television programs so they can watch
them on their computers.

> Do you really think you're going to pick up a "used, obsolete"
> drive for $20 and have the supplier give you any assurances
> other
> than, "If it dies in the first 30 days, bring it back and we'll
> give you another one?" :>

Of course not. And if you buy a brand new one and it fails
in two months, they'll tell you to send it back to the
manufacturer. The working in the warranty section of the
Mackie HDR manual was pretty clear that they'll replace
failed hard drives but will not compensate the user for lost
time or money for a lost recording. I was in a meeting there
where there was a lot of discussion about "What if [insert
the name of a famous artist at the time] lost his disk drive
in the middle of a session?" While nobody actually said it,
I think there would be special cases where they'd helpl out
with some cash in exchange for being able to say that
[Famous Artist] used the recorder.

I don't think the issue ever came up, but then the people
who work for Famous Artists back up their work, and Famous
Artists also get used to the occasional studio goof where
something is lost. Reasonable people are usually pretty
reasonable. It's only the ones with unreasonable
expectations that can be a pain in the butt. Suppose Joe
Songster who had been working on his CD for four years had a
disk crash and of course he never backed it up. He'd be
screaming "What if this happened to Michael Jackson?"

[hank alrich]

Don Y <th...@isnotme.com> wrote:

> Though,
> for the life of me, I can't see the need for 100G drives
> in laptops -- no doubt because I wouldn't use a laptop
> as a "work machine")

Meanwhile, I just pulled the old 329GB drive from my MacBook Pro, and
replaced it with a 1TB drive. I can track 10 x 24/96 now without
worrying about storage capacity, while running a mirror drive via USB or
FIrewire, copy over the ubiquitous videos people are phone.camera
shooting of my/our shows, and have slop left over. This _is_ my work
machine, just as the TiBook was for nearly seven years.

--
shut up and play your guitar * http://hankalrich.com/
http://www.youtube.com/walkinaymusic
http://www.sonicbids.com/HankandShaidri

[Don Y]

Hi Mike,

On 5/29/2012 12:42 PM, Mike Rivers wrote:
> On 5/29/2012 12:51 PM, Don Y wrote:
>
>> > for the life of me, I can't see the need for 100G drives
>> in laptops -- no doubt because I wouldn't use a laptop
>> as a "work machine")
>
> Get with the times, man!

<grin> Not quite a "luddite" but *really* hesitant to change
just for the sake of change! :>

> Today, the laptop, as a "desktop replacement"
> is the most popular configuration with the largest customer base. That's
> not the corporate users, it's the Facebookers and movie watchers. You
> know, the people who have cable TV so they can get high speed internet,
> then download movies and television programs so they can watch them on
> their computers.

Exactly. I have a TV for watching movies. And, a HiFi for listening
to music. Just because you *can* use a device for a purpose doesn't
mean you *want* to! :>

I find laptop keyboards "icky". The disks are slow. And, the
processors are often stripped down for less demanding applications
(e.g., want to do some 3D CAD rendering?). The screens are too
small (I use a pair of 20" monitors on each of my machines) -- and
too close to the keyboard. They also are less reliable and require
lots of "custom" parts (aside from disks and memory -- and *some*
optical drives -- there is very little from Machine A that can be
used to repair or upgrade Machine B. Couple that with a short
product lifetime and you are almost guaranteed to be replacing the
machine whenever something breaks.

So, for me, a laptop is little more than a portable typewriter, mail
reader/web browser on which I can also watch DVDs while traveling.
Sort of like *carrying* music on an iPod...

>> Do you really think you're going to pick up a "used, obsolete"
>> drive for $20 and have the supplier give you any assurances
>> other
>> than, "If it dies in the first 30 days, bring it back and we'll
>> give you another one?" :>
>
> Of course not. And if you buy a brand new one and it fails in two
> months, they'll tell you to send it back to the manufacturer. The
> working in the warranty section of the Mackie HDR manual was pretty
> clear that they'll replace failed hard drives but will not compensate
> the user for lost time or money for a lost recording.

That's true of damn near everything, these days. "If our software
screws up, tough luck!" etc.

> I was in a meeting
> there where there was a lot of discussion about "What if [insert the
> name of a famous artist at the time] lost his disk drive in the middle
> of a session?" While nobody actually said it, I think there would be
> special cases where they'd helpl out with some cash in exchange for
> being able to say that [Famous Artist] used the recorder.

I recall watching a pseudo-documentary about one of the first films
made using a new "all digital" camera ("Red" being the manufacturer)
and how they had a guy who's sole job was to produce the *two*
backups of the footage shot each day (on PCMCIA/CF cards). Note
that they could, technically, still recover from a loss there -- but
at considerable expense (reshoot those scenes).

> I don't think the issue ever came up, but then the people who work for
> Famous Artists back up their work, and Famous Artists also get used to
> the occasional studio goof where something is lost. Reasonable people
> are usually pretty reasonable.

*If* the mistake was a "reasonable" one. OTOH, if you incur a
loss due to incompetence or negligence, attitudes tend to be a
bit different!

> It's only the ones with unreasonable
> expectations that can be a pain in the butt. Suppose Joe Songster who
> had been working on his CD for four years had a disk crash and of course
> he never backed it up. He'd be screaming "What if this happened to
> Michael Jackson?"

"Michael Jackson probably wouldn't have put himself in that
situation!" (Or, MJ *may* have been in a similar situation,
got bit, and LEARNED from it!)

[Don Y]

Hi Mike,

On 5/29/2012 12:30 PM, Mike Rivers wrote:
>> OK. So, someone was short-sighted and/or wrote buggy code.
>
> Hey, it was fine for at least 5 years after the product's introduction.
> That's end-of-life for most computer-based products.

Well, that would depend on the market and what users'
expectations are. I tend to design with very long
product lifetimes in mind -- 15 - 30 years. Often,
the corners that you "cut" for the short term just
don't amount to real savings in the long run.

E.g., I'll live through the Y2038 mess (assuming average
mortality). Wanna bet folks are designing systems *today*
that will fall apart, *then*?!

<frown>

> And as I pointed
> out, a 20 or 30 GB drive is practical for a replacement for 2" tape.
> That's what the Mackie HDR24/96 was designed to do, and that's what it
> did, darn well. I'm still using mine in preference to a computer-based
> DAW. The code wasn't buggy, and in an era when a 20 GB drive was about
> as large as you could buy, it isn't fair to call the design "short
> sighted."

When 20G drives were common, it wouldn't take a rocket scientist
to look *back* and see how drive sizes had evolved in a very
short timespan. And, to see that something produced a year
or two ago would become "unobtanium" a year or two hence. With
that in mind, I'd call it short-sighted.

>> For the operating system to carry this limitation *up* through
>> it's internal structure is just plain stupid. Akin to only
>> being
>> able to print dates in which the first digit of the year is
>> a '1'!
>
> Well, it works for its intended purpose. Mackie purchased a real time
> kernel from someone else and wrote the application around it. Probably
> the most short-sighted thing they did (and I know this isn't the perfect
> machine) was to compile drivers into the operating system. It seems to
> work OK with just about any mouse, but in the HDR, the version with the
> GUI, it restricts the graphics board to an ATI Radeon 7000 series, which

It's a separate *board* and not part of the "system board"?

> are getting difficult to find should a replacement be needed. Worse (and
> this isn't really Mackie's problem) with so many "Radeon" graphics cards
> having been made, the average for finding the right one that you buy via
> an eBay auction seems to be running about 20% these days. There really
> isn't a good way to identify the correct one (too many different
> numbers) except by trying.

Yup. Even purchasing something that *looks* like it should
work can leave you with problems. Wrong BIOS onboard, etc.
Correct BIOS no longer available from manufacturer's web site,
etc.

Nowadays, when you buy something, grab *everything* you can
and archive it somewhere. Never know when you will need it
only to discover it's not even available on the Way Back Machine...

> And the motherboard has been recently
> discontinued (not bad after 12 years) and apparently a modern one isn't
> a workable replacement. Obsolete computers are like that.
>
>> Good point. With "computers", we tend to view the drive as
>> a closet or desk drawer into which we can shove all our
>> (unrelated) crap.
>>
>> OTOH, I suspect that if you used this device for post
>> processing
>> work, as well, your attitude might change. I.e., having to pull
>> projects (which may be only partially completed) off of it to
>> make room for new projects would tend to cause you to look for
>> a larger medium -- meaning cruft would tend to accumulate. :>
>
> It has two disk drives, an internal one that contains the software (only
> a couple of megabytes worth) and the rest is available for data. Then
> there's a removable drive.

OK. So, if you intend to handle multiple projects, you would
prefer to use the removable drive for the medium (?)

Are both subject to the same size limitation? Can they be
different sizes, etc.?

> When I was recording fairly regularly, my
> workflow was to dedicated a removable drive to a project. I'd record on
> that, and then back up to the internal drive. I didn't work in an
> environment where I had several projects to juggle at one time, but I'd
> use the internal drive as backup if I had projects that involved just a
> couple of songs.

OK, makes sense. And saves you the hassle of having to deal with
yet another piece of kit.

OTOH, if the machine goes screwwy, there's a change your backup
and original can get trashed at the same time.

> After the project is mixed, the removable drive goes on the shelf or
> given to the artist and the material is deleted from the internal drive.
>
>> ("Files grow and multiply to fill all space available")
>
> Only if you let them.

Yeah, which means perpetually being vigilant and beating them
back with a big stick!

I think when I had a *real* (wooden) desk, it was a lot easier to
keep it organized...

[Don Y]

Hi Hank,

On 5/29/2012 1:17 PM, hank alrich wrote:
> Don Y<th...@isnotme.com> wrote:
>
>> Though,
>> for the life of me, I can't see the need for 100G drives
>> in laptops -- no doubt because I wouldn't use a laptop
>> as a "work machine")
>
> Meanwhile, I just pulled the old 329GB drive from my MacBook Pro, and
> replaced it with a 1TB drive. I can track 10 x 24/96 now without
> worrying about storage capacity, while running a mirror drive via USB or
> FIrewire, copy over the ubiquitous videos people are phone.camera
> shooting of my/our shows, and have slop left over. This _is_ my work
> machine, just as the TiBook was for nearly seven years.

Different type of work. :> The prototype of <whatever> that I'm
tethered to is likely to be bigger than my actual workstation.

E.g., when developing the network speakers, I had two legacy
PC's acting as the "network speakers", a pair of generic amplified
speakers connected to them, another machine acting as media server,
a logic analyzer to let me measure instantaneous phase differences
between the outputs of the two "speakers" and my actual workstation
letting me probe all of the above (as well as snoop the network).

More like a small *room* full of assorted kit.

I have other prototypes that are the size of washing machines, etc.
(And I've worked with others that were literally the size of a
small room!)

So, aside from just "writing" (software, specifications, test
suites, etc.), a laptop doesn't buy me much in terms of "work".

[hank alrich]

Don Y <th...@isnotme.com> wrote:

> So, for me, a laptop is little more than a portable typewriter, mail
> reader/web browser on which I can also watch DVDs while traveling.

My laptop is my extremely portable pro audio workstation, as well as all
that other typical laptop stuff. <g> A Mac laptop and a Metric Halo
interface is the replacement for my Studer 1" machine, and lots easier
to move around, even though the A80 was the "transportable" version,
with electronics and deck interconnected by a fat multicable - i.e.,
transportable in sections by two strong humans.

[Don Y]

Hi Hank,

On 5/29/2012 5:11 PM, hank alrich wrote:
> Don Y<th...@isnotme.com> wrote:
>
>> So, for me, a laptop is little more than a portable typewriter, mail
>> reader/web browser on which I can also watch DVDs while traveling.
>
> My laptop is my extremely portable pro audio workstation, as well as all
> that other typical laptop stuff.<g> A Mac laptop and a Metric Halo
> interface is the replacement for my Studer 1" machine, and lots easier

The "Metric Halo" is an external bit of kit (A/DC's and D/AC's)
*controlled* by the Mac? Is it a true "black box" (with all
controls on the Mac) or does it have a few knobs and displays?

Besides control/user interface, is the Mac involved in the
signal processing loop? Or, just act as bulk storage?

> to move around, even though the A80 was the "transportable" version,
> with electronics and deck interconnected by a fat multicable - i.e.,
> transportable in sections by two strong humans.

Yeah, my workstation would be more like that -- a couple
of extra hands and/or a flat bed cart! :< Anything *but*
"portable" :-/

[hank alrich]

Don Y <th...@isnotme.com> wrote:

> Hi Hank,
>
> On 5/29/2012 5:11 PM, hank alrich wrote:
> > Don Y<th...@isnotme.com> wrote:
> >
> >> So, for me, a laptop is little more than a portable typewriter, mail
> >> reader/web browser on which I can also watch DVDs while traveling.
> >
> > My laptop is my extremely portable pro audio workstation, as well as all
> > that other typical laptop stuff.<g> A Mac laptop and a Metric Halo
> > interface is the replacement for my Studer 1" machine, and lots easier
>
> The "Metric Halo" is an external bit of kit (A/DC's and D/AC's)
> *controlled* by the Mac? Is it a true "black box" (with all
> controls on the Mac) or does it have a few knobs and displays?

The model I have, which was their first offering, and which they have
managed to keep completely workable through a decade of Apple "upgrades"
is the 2882, orginally +DSP model with an extra SHARC chip for
andditional processing power. I have recently had that upgraded to what
they're calling a 2d card, which is much more powerful than the original
extra processing chip. It's essentially a black box with elementary
metering and various I/O.

http://mhsecure.com/metric_halo/products/hardware/2882.html

They also have other models now that include some external control.

http://mhsecure.com/metric_halo/products.html

This is good kit, from a company that provides stellar service, too.

> Besides control/user interface, is the Mac involved in the
> signal processing loop? Or, just act as bulk storage?

Absent the optional +DSP or 2d card the Mac handles much of the
processing. With the additional card power is greatly expanded, with a
lot of work being done in the interface.

While other companies now offer similar features, MH was the first of
which I am aware to provide for direct streaming of audio to disk via
the control app, which they call the Console, without an intermediary
DAW app.

> > to move around, even though the A80 was the "transportable" version,
> > with electronics and deck interconnected by a fat multicable - i.e.,
> > transportable in sections by two strong humans.
>
> Yeah, my workstation would be more like that -- a couple
> of extra hands and/or a flat bed cart! :< Anything *but*
> "portable" :-/

I get to carry it with me, and I am grateful for that.

[Mike Rivers]

On 5/29/2012 4:28 PM, Don Y wrote:

> I find laptop keyboards "icky". The disks are slow. And, the
> processors are often stripped down for less demanding
> applications (e.g., want to do some 3D CAD rendering?).

Or mixing your latest music project while flying across the
ocean to the mastering lab? <g>

> They also are less reliable and
> require
> lots of "custom" parts (aside from disks and memory -- and
> *some*
> optical drives -- there is very little from Machine A that
> can be
> used to repair or upgrade Machine B.

The parts that you can get to are pretty much
interchangeable, not with full sized computers, but with
other laptop sized computers. It's a different industrial
design and has its own set of parts. It's true that few
people upgrade a laptop any other way than to replace it,
but I've replaced memory (needed more to work smoother),
hard drives (needed more space) and CD drive (the only thing
that ever stopped working on one of my laptops) with parts
that I could buy over the counter at Micro Center.
Mechanical parts like screens, cases, and keyboards,
however, usually require original parts.

> Couple that with a short
> product lifetime and you are almost guaranteed to be
> replacing the
> machine whenever something breaks.

I think most get replaced when they run out of disk space or
decide that an iPhone is a better toy, rather than failure.
But like so many things nowadays, repair costs are high,
purchase costs are low, so when there's a problem (sometimes
as simple as a corrupted program or dead battery) the
average laptop user, who doesn't have much "legacy"
software, will replace the computer as long as the Best Buy
Geek Squad can transfer their pictures from the old computer
to the new one.

> So, for me, a laptop is little more than a portable
> typewriter, mail
> reader/web browser on which I can also watch DVDs while
> traveling.

Ah, so you DO get it. But for many people in our small
community here, a laptop computer is not only all that, but
also their portable recorder for capturing live performances
for later mixing, and serve as their compulete music
production system.

>> Famous Artists back up their work, and Famous Artists also
>> get used to
>> the occasional studio goof where something is lost.
>> Reasonable people
>> are usually pretty reasonable.
>
> *If* the mistake was a "reasonable" one. OTOH, if you incur a
> loss due to incompetence or negligence, attitudes tend to be a
> bit different!

I assume you mean incompetence or negligence on the part of
the equipment manufacturer, not the user. But some would
complain about that, too. ("Your stupid recorder let me
erase the tape!") Most manufacturers do a pretty good job
testing and debugging but sometimes problems slip through.
But they can't predict the future and accommodate what might
be ten years later as part of their original design,
particularly when they're adapting existing technology to a
new product.

[John Williamson]

On 30/05/2012 12:49, Mike Rivers wrote:
> On 5/29/2012 4:28 PM, Don Y wrote:

<Laptops>

Some of them don't even bother asking the Geeks to transfer their
pictures now, as they're all kept on the camera memory card or in The Cloud.

I've picked up a couple of used laptops at bargain prices becuase
they're not shiny and new, too. They work fine, thank you very much.

>> So, for me, a laptop is little more than a portable
>> typewriter, mail
>> reader/web browser on which I can also watch DVDs while
>> traveling.
>
> Ah, so you DO get it. But for many people in our small community here, a
> laptop computer is not only all that, but also their portable recorder
> for capturing live performances for later mixing, and serve as their
> compulete music production system.
>

<Holds hand up> That's me, that is.

[Mike Rivers]

On 5/29/2012 6:28 PM, Don Y wrote:

> Well, that would depend on the market and what users'
> expectations are. I tend to design with very long
> product lifetimes in mind -- 15 - 30 years. Often,
> the corners that you "cut" for the short term just
> don't amount to real savings in the long run.

This is why there are companies like Mackie and companies
like yours. But then I don't believe I've heard of any of
your products. They probably aren't marketed to people like
me. Market research tells them what they can sell to whom
and at what price, then they decide what to build. So, yes,
they builid products that are close to the users'
expectations, hopefully wowing them enough to get their
business.

If you're buying something that costs as much or more to
install and operate as it does to buy, of course you'll want
it to last for a long time. If you're buying something that
you can plot down on the desk and go to work doing what you
want to do, then it's not all that important that it may
become obsolete in a few years. If I cared about recording
more than 24 tracks or at higher sample rates, then my
Mackie recorder would be obsolete. But - specifically
because it's hardly changed in the dozen years I've been
using it (more years, actually, than my Ampex MM-1100,
though not more years than my Soundcraft 600 console) - it's
still doing the job I need it to do and more conveniently
than a kit of parts with a computer at its heart.

When I was working for the FAA, buying ground based landing
systems for airports, it became impossible to buy something
completely designed from the ground up for the budgets that
Congress gave us. So the days of keeping a 35 year old
localizer operating and landing planes safely were over.
When I retired at the end of 1999, they were already
planning for replacement of ILS equipment designed and built
in the 1992-1995 time frame. And it still cost too much money!

> When 20G drives were common, it wouldn't take a rocket
> scientist
> to look *back* and see how drive sizes had evolved in a very
> short timespan. And, to see that something produced a year
> or two ago would become "unobtanium" a year or two hence. With
> that in mind, I'd call it short-sighted.

But it's not trivial to modify an operating system or BIOS
to accommodate what you think might happen. I know it can be
done, but not at the price where the product would sell
enough to make money. For certain products (for example the
Mackie hard disk recorder) resourceful users have been able
to keep them fed and healthy for longer than was originally
anticipated.


>> in the [Mackie] HDR, the

>> version with the
>> GUI, it restricts the graphics board to an ATI Radeon 7000
>> series

> > It's a separate *board* and not part of the "system board"?

Yes. The motherboard is from the era where graphics weren't
expected to be built in. This board has no monitor port.

> Nowadays, when you buy something, grab *everything* you can
> and archive it somewhere. Never know when you will need it
> only to discover it's not even available on the Way Back
> Machine...

When the Alesis ADAT first came out, I suggested that anyone
who wanted to keep their recordings buy a recorder and spare
mechanical parts, box it up, and store it along with their
tapes. I don't know if anyone did that. I doubt it. But
still, today, people are buying ADATs through eBay to
transfer their old tapes to disk.

> OK. So, if you intend to handle multiple projects, you would
> prefer to use the removable drive for the medium (?)
>
> Are both subject to the same size limitation? Can they be
> different sizes, etc.?

Yes, just like I use a reel of tape. Both drives talk to the
same operating system and BIOS so therefore are subject to
the same size limitation, though they can be of different
sizes. My internal drive is now 120 GB so I have lots of
room for little temporary things. I have external drives
ranging from 20 GB to 160 GB depending on what I can get
when the getting gets good. I just bought two 80 GB used
drives at a hamfest last weekend for $5 each. They seem to
work fine.

This design was from the USB 1.1 days, so Mackie built an
FTP server into the machine so that both the internal and
external drives can be accessed from a network. Files can be
dragged and dropped to drives on a network computer. While
not recommended, it can even be put out on the Internet.

I have the external part of a mobile rack mounted in a
USB2/Firewire case so I can plug in a removable drive from
the recorder and back it up to another drive easily. Backing
up the internal drive is a little more complicated. Either
it can be done via FTP or it can be copied to an external
drive and that can be stored (or backed up to something that
you'll store). It's really pretty flexible.

> OTOH, if the machine goes screwwy, there's a change your backup
> and original can get trashed at the same time.

This is true with any computer based device. And if my tape
deck goes screwy, it can ruin a section of tape, too.

>>> ("Files grow and multiply to fill all space available")

> Yeah, which means perpetually being vigilant and beating them
> back with a big stick!

I delete far more than I save, and I hardly ever listen to
what I save.

[hank alrich]

Mike Rivers <mri...@d-and-d.com> wrote:

> But some would complain about that, too. ("Your stupid recorder let me
> erase the tape!")

Nothing about the Studer was able to prevent that. <g>

[Don Y]

Hi Hank,

On 5/29/2012 7:22 PM, hank alrich wrote:

>>> My laptop is my extremely portable pro audio workstation, as well as all
>>> that other typical laptop stuff.<g> A Mac laptop and a Metric Halo
>>> interface is the replacement for my Studer 1" machine, and lots easier
>>
>> The "Metric Halo" is an external bit of kit (A/DC's and D/AC's)
>> *controlled* by the Mac? Is it a true "black box" (with all
>> controls on the Mac) or does it have a few knobs and displays?
>
> The model I have, which was their first offering, and which they have
> managed to keep completely workable through a decade of Apple "upgrades"
> is the 2882, orginally +DSP model with an extra SHARC chip for
> andditional processing power. I have recently had that upgraded to what
> they're calling a 2d card, which is much more powerful than the original
> extra processing chip. It's essentially a black box with elementary
> metering and various I/O.
>
> http://mhsecure.com/metric_halo/products/hardware/2882.html
>
> They also have other models now that include some external control.
>
> http://mhsecure.com/metric_halo/products.html
>
> This is good kit, from a company that provides stellar service, too.
>
>> Besides control/user interface, is the Mac involved in the
>> signal processing loop? Or, just act as bulk storage?
>
> Absent the optional +DSP or 2d card the Mac handles much of the
> processing.

I'm confused -- and admit to only skimming the user manual
(which doesn't seem to cleanly separate the "with +DSP"
and the "without +DSP" capabilities).

It seems that all the "audio" (even in digital form) is
handled inside the interface (?). I.e., the Mac can
sink (and source?) "audio data" (blech but you know what
I mean)but it doesn't actively *process* it "in real time".

In other words, the Mac doesn't "*implement* effects" like
reverb, filtering, mixing, etc. But, rather, *controls*
mechanisms in the "interface" (2882) based on settings that
the user manipulates in a GUI (?)

And, that the +DSP option is what you add to allow these
sorts of "effects" to happen IN REAL TIME in the interface (?)

In other words, "audio" doesn't come into the Mac from the
interface, get "number crunched" inside the Mac and then
routed back out to the interface for delivery to studio
monitors, etc. "live".

It seems that any non-optional DSP (SHARC) built into the
device would be preoccupied just moving digital streams
around with some mixing functions, etc. (?) So, the +DSP
adds horsepower to actually implement effects in real time (?)

> With the additional card power is greatly expanded, with a
> lot of work being done in the interface.
>
> While other companies now offer similar features, MH was the first of
> which I am aware to provide for direct streaming of audio to disk via
> the control app, which they call the Console, without an intermediary
> DAW app.

I'm surprised that the Mac provides a data channel that can
make those guarantees! I.e., the Mac can't "get distracted"
and delay processing that incoming (or outgoing) data as it
would result in data being *lost* ("Sorry guys, we lost the
47th bar of that piece. Can we try it again, starting at the
30th?")

>>> to move around, even though the A80 was the "transportable" version,
>>> with electronics and deck interconnected by a fat multicable - i.e.,
>>> transportable in sections by two strong humans.
>>
>> Yeah, my workstation would be more like that -- a couple
>> of extra hands and/or a flat bed cart! :< Anything *but*
>> "portable" :-/
>
> I get to carry it with me, and I am grateful for that.

Understood. And, you're more likely (than me) to need to
"go to the mountain" (instead of the mountain coming to you!)

[Don Y]

Hi Mike,

On 5/30/2012 4:49 AM, Mike Rivers wrote:
> On 5/29/2012 4:28 PM, Don Y wrote:
>
>> I find laptop keyboards "icky". The disks are slow. And, the
>> processors are often stripped down for less demanding
>> applications (e.g., want to do some 3D CAD rendering?).
>
> Or mixing your latest music project while flying across the ocean to the
> mastering lab? <g>

Can you actually *do* that (practically)? I can't hear myself
*think* on most planes, let alone try to come up with a
bit of audio that I might be able to appreciate in the absence
of all that background noise.

>> They also are less reliable and require
>> lots of "custom" parts (aside from disks and memory -- and *some*
>> optical drives -- there is very little from Machine A that
>> can be used to repair or upgrade Machine B.
>
> The parts that you can get to are pretty much interchangeable, not with
> full sized computers, but with other laptop sized computers. It's a

IME, there seems to be more variety in the appearance (form factor)
and implementation. E.g., some models support readily removable
"drive bays" (swap out a CD for a DVD or floppy); or cosmetic
cruft like the front of the optical disk "tray" being sculpted
to blend in with the shape of the laptop's edge; etc.

For example, I have some tiny machines (about the size of a
bible you might find in a hotel room) that largely use
laptop parts (memory, disk, optical drive) -- *but* use a
funky, non-standard connector on the back of the optical
drive. And, of course, there's no extra room in which you
could install a cable or any kind of adapter board. So,
I'm stuck trying to locate an oddball "standard" CD/DVD
writer :-/

> different industrial design and has its own set of parts. It's true that
> few people upgrade a laptop any other way than to replace it, but I've
> replaced memory (needed more to work smoother), hard drives (needed more
> space) and CD drive (the only thing that ever stopped working on one of
> my laptops) with parts that I could buy over the counter at Micro
> Center. Mechanical parts like screens, cases, and keyboards, however,
> usually require original parts.

I guess I don't consider things like screens to be mechanical :>

> > Couple that with a short
>> product lifetime and you are almost guaranteed to be
>> replacing the
>> machine whenever something breaks.
>
> I think most get replaced when they run out of disk space or decide that
> an iPhone is a better toy, rather than failure.

In my experience, the two biggest reasons *claimed* are:
- battery won't hold a charge (why spend $80 for a battery?)
- a "virus" (i.e., performance has deteriorated enough that
I am now bored with this model and need an excuse for a new one)

As an aside, single biggest reason I've heard for upgrading a
*phone*:
- it fell in the toilet ("Jeez, guys, do you carry them around
wedged between your BUTT CHEEKS???")

> But like so many things
> nowadays, repair costs are high, purchase costs are low, so when there's
> a problem (sometimes as simple as a corrupted program or dead battery)
> the average laptop user, who doesn't have much "legacy" software, will
> replace the computer as long as the Best Buy Geek Squad can transfer
> their pictures from the old computer to the new one.

Exactly. Though I find people don't even care about salvaging their
files. I think most folks forget the sort of cruft that they've
accumulated on the machine and put little value on it.

Sort of like old snapshots or slides -- you toss them in a box
and rarely recall what the heck is in there!

I've inherited two laptops from neighbors who asked me to "see
if I could fix them". Usually, something funky in software.
A day or, at most, *two* later, I'll bring it back to them
and be met with, "Oh, we decided to just go out and buy a new
one. (Huh? 24-48 hours and $0 wasn't good enough service for
you?) If you want,m you can just keep it!"

<shrug> I guess I must be the only one that doesn't have a money
tree growing in his back yard. (I *told* her it was a mistake to
plant all those ORANGE trees!! :> )

>> So, for me, a laptop is little more than a portable typewriter, mail
>> reader/web browser on which I can also watch DVDs while
>> traveling.
>
> Ah, so you DO get it.

But that's not "work". I.e., the machine I am currently typing
on has exactly the same sort of software that I put on my laptops:
firefox (web browser), thunderbird (email, news), telnet client,
emacs (text editor), adobe reader and a handful of system tools
to manipulate the registry, examine archive files, etc.

If I want to do *work*, I get up and walk down the hallway...

> But for many people in our small community here, a
> laptop computer is not only all that, but also their portable recorder
> for capturing live performances for later mixing, and serve as their
> compulete music production system.

Understood.

>>> Famous Artists back up their work, and Famous Artists also
>>> get used to
>>> the occasional studio goof where something is lost.
>>> Reasonable people
>>> are usually pretty reasonable.
>>
>> *If* the mistake was a "reasonable" one. OTOH, if you incur a
>> loss due to incompetence or negligence, attitudes tend to be a
>> bit different!
>
> I assume you mean incompetence or negligence on the part of the
> equipment manufacturer, not the user.

Both -- if the user is an employee, etc. (I am always surprised
as how intolerant some folks are of mistakes made by others -- yet
expecting of forgiveness or understanding of mistakes made by
themselves!)

> But some would complain about
> that, too. ("Your stupid recorder let me erase the tape!") Most

We had a discussion about this, recently. How much should the
device insist on "confirmations" for actions? When does your
"hand holding" become *nagging*? When do you just *trust* the
user?

Imagine you started recording a gig only to realize you had
something set improperly. Perhaps want to save the audio to
a different drive as you just realized you can use the remaining
free space on the OTHER drive so that you can preserve as much
of the space on the CURRENTLY SELECTED drive to hold the program
for the headliner that will follow this warmup act.

You rush to make the changes before the performers play
their first few notes... first, stop the current recording:
"Are you sure you want to stop this recording?"
"Yes, dammit!"
Then you go to change the destination drive for the first group:
"Are you sure you want to switch to the second drive? There
is only enough free space to store 15 minutes of audio?"
"YES, DAMMIT!"

> manufacturers do a pretty good job testing and debugging but sometimes
> problems slip through. But they can't predict the future and accommodate
> what might be ten years later as part of their original design,
> particularly when they're adapting existing technology to a new product.

They can't predict ALL (or even MOST!) future developments.
But, if you couldn't foresee laptops, tablets, smart phones,
etc. in the 90's, you weren't *thinking* about it! :>

And, if you couldn't imagine TB drives 10 years ago (when GB
drives were common -- a decade after MB drives!) you aren't
thinking about that, either! :<

Think about it: 5 years represents a tenfold increase in
computing power, etc. Even if the products that you offer
don't have the ability to exploit that performance curve, you,
as a company, should be designing with that in mind! Otherwise,
you will find yourself having to reinvent the wheel every five
years! You may not worry about your customers having to "buy
new" that often, but do you also want to subject *yourself* to
having to "buy new" (re-designs from scratch!) as well?
(i.e., you didn't design the OS to be able to handle those
inevitable advances right around the corner)

Kinda like being "surprised" that Christmas falls on the 25th
this year! ;-)

<shrug> I don't like doing things over just because someone
failed to think ahead. "Let's sort out where you want the
sofa BEFORE we start moving furniture..." :>

[Mike Rivers]

On 5/30/2012 7:06 PM, Don Y wrote:

>> Or mixing your latest music project while flying across
>> the ocean to the mastering lab? <g>
>
> Can you actually *do* that (practically)?

I can't imagine doing anything like that seriously but it's
one of those things that "they" tell you that you can do
with your laptop computer. I expect that you might be able
to play around with the sequence of songs on a CD, or maybe
do some basic editing, maybe even compose a loop-based
melody. Me, though, I put on my noise canceling headphones,
wind up my MP3 player, and listen to some radio shows that
I'd recorded previously and loaded on there. Makes for a
good sound track for watching The Food Channel on the
on-board free satellite TV.

> I guess I don't consider things like screens to be
> mechanical :>

On a laptop, the screen things that "break" are the hinges,
latch, or somehow the screen gets cracked. You can lose a
few pixels and not be bothered too much by it.

>> I think most get replaced when they run out of disk space
>> or decide that
>> an iPhone is a better toy, rather than failure.

> In my experience, the two biggest reasons *claimed* are:
> - battery won't hold a charge (why spend $80 for a battery?)
> - a "virus" (i.e., performance has deteriorated enough that
> I am now bored with this model and need an excuse for a new
> one)

Yes, those are two common reasons to replace a laptop
computer, though for what seems to be the statistical
majority of laptop users today, they wouldn't know if the
battery no longer held a charge until they had a power
failure. Most of them are plugged in and stay in the same
place all the time. But Grandma likes them because it's all
one piece and takes up less space than a tower case,
keyboard, mouse, and monitor. Doesn't require any hookup
either, other than plugging in the power, if you're using a
wireless Internet connection.

> "YES, DAMMIT!"

Yes, dammit, I've said that to my computers many times.

> And, if you couldn't imagine TB drives 10 years ago (when GB
> drives were common -- a decade after MB drives!) you aren't
> thinking about that, either! :<

These things tend to go in leaps. Once they realized that
people will be buying 20 GB drives, they made computers
accommodate up to, I dunno, maybe 500 GB. But the first
computer I upgraded with a 20 GB drive previously had a 540
MB drive in it, and I thought THAT was big. So I suppose
that a few terabytes isn't a barrier any longer.

> Think about it: 5 years represents a tenfold increase in
> computing power, etc. Even if the products that you offer
> don't have the ability to exploit that performance curve, you,
> as a company, should be designing with that in mind!

I think that what most of them have in mind when designing
their system is that in five years, they don't want you to
still be using the same old system, they want you to buy a
new one. When it comes to things like laptop computers, they
don't have to be underpowered to run software that comes
along 5 years from now in order to be obsolete, they can be
obsolete as soon as a part that fails is no longer available
from any source, or it's too difficult to obtain because
(like the hard drives that we started this discussion with)
you may have to buy them from a questionable source or in
questionable condition rather than brand new with a factory
warranty if it's DOA or dies an early death.

> You may not worry about your customers having to "buy
> new" that often, but do you also want to subject *yourself* to
> having to "buy new" (re-designs from scratch!) as well?

It's what keeps engineers employed.

[Don Y]

Hi Mike,

On 5/30/2012 5:36 AM, Mike Rivers wrote:
> On 5/29/2012 6:28 PM, Don Y wrote:
>
>> Well, that would depend on the market and what users'
>> expectations are. I tend to design with very long
>> product lifetimes in mind -- 15 - 30 years. Often,
>> the corners that you "cut" for the short term just
>> don't amount to real savings in the long run.
>
> This is why there are companies like Mackie and companies like yours.
> But then I don't believe I've heard of any of your products. They
> probably aren't marketed to people like me.

No, you'll never find anything with my name on it. :>
I tend to make for others to sell as "their own". Even
the network speakers/displays and home automation products
(the designs of which I own, outright) will never be "sold"
as products. Rather, just published so others can reproduce
them (as hobbyists or commercially) as well as to shift the
practical state of the art (and force commercial suppliers
to step it up, a notch! :> )

> Market research tells them
> what they can sell to whom and at what price, then they decide what to
> build. So, yes, they builid products that are close to the users'
> expectations, hopefully wowing them enough to get their business.
>
> If you're buying something that costs as much or more to install and
> operate as it does to buy, of course you'll want it to last for a long

You also want things that cost a lot to buy (even if they cost
only a little to install/operate!) to last for a long time.
You wouldn't be keen on buying a new car just because it
"went out of style", etc.

> time. If you're buying something that you can plot down on the desk and
> go to work doing what you want to do, then it's not all that important
> that it may become obsolete in a few years.

What if the recordings you've made with it weren't reproducible
(portable) on newer kit? Or, required a 3rd party device or service
to "port" them? My point is that there are lots of costs to
"premature obsolescence". I'm sure you'd prefer to be able to
install a 200GB drive in the device. But can't because someone
was shortsighted.

I.e., that's the boat I'm in with many of the designs for hire
I've done over the past few decades. Tools that version X+1 will
refuse (i.e., be completely incapable of) to process files that
version X created last year! Leaving me with three choices:
- stick to version X and hope I never need any of the fixes or
added functionality provided by version X+1
- move to version X+1 and abandon all work done with version X
- move to X+1 and personally carry the cost of maintaining a
"live" copy of version X at the same time
As a consumer (of that tool), none of these are acceptable options.

And, of course, I'm going to pass those costs on to the next
consumer downstream from me...

> If I cared about recording
> more than 24 tracks or at higher sample rates, then my Mackie recorder
> would be obsolete. But - specifically because it's hardly changed in the
> dozen years I've been using it (more years, actually, than my Ampex
> MM-1100, though not more years than my Soundcraft 600 console) - it's
> still doing the job I need it to do and more conveniently than a kit of
> parts with a computer at its heart.
>
> When I was working for the FAA, buying ground based landing systems for
> airports, it became impossible to buy something completely designed from
> the ground up for the budgets that Congress gave us. So the days of
> keeping a 35 year old localizer operating and landing planes safely were
> over. When I retired at the end of 1999, they were already planning for
> replacement of ILS equipment designed and built in the 1992-1995 time
> frame. And it still cost too much money!

Possibly a bad example as every overly regulated and bureaucratic
organization tends to be inherently inefficient. :<

>> When 20G drives were common, it wouldn't take a rocket
>> scientist
>> to look *back* and see how drive sizes had evolved in a very
>> short timespan. And, to see that something produced a year
>> or two ago would become "unobtanium" a year or two hence. With
>> that in mind, I'd call it short-sighted.
>
> But it's not trivial to modify an operating system or BIOS to
> accommodate what you think might happen.

Exactly! It's not trivial to MODIFY an operating system to
do these things. So, *when* they eventually have to do it
for their FUTURE products, they will incur that cost. And,
not be able to benefit from having that "design" in place
getting hundreds of man-years of testing, "for free" from
their (legacy) user base. Instead, it will be as if they
just invented the product anew and have no real preformance
or reliability data on it.

On the other hand, of you design this *into* the operating system
to begin with, then when the future arrives (or, becomes more
affordable), you -- and your customers -- can step into it
without a big reinvestment. You and they aren't rediscovering
bugs that should have been fixed previously, etc.

It's like building a house as a newly wed with just one bedroom.
Then, discovering you're going to have a kid and having to build
an addition for the extra bedroom. But, *only* building an
addition large enough for a single bedroom! A few years later,
"Surprise!". And now you realize that your lot size and general
floorplan makes adding yet another bedroom on the ground floor,
impractical. So, you think about building *up*, instead. But,
the first floor framing wasn't designed with that sort of
a load in mind. So you reinforce the interior and exterior
walls in preparation for "lifting the roof". etc.

Sure, all of these things are possible -- you could also just
get up and relocate to a larger house! But, each represents
additional risk that can cause problems for your family as
a "going concern" (i.e., draw the analogy to a company that
is already producing product and now faced with the NEED to
release a new product to keep up with rivals that threaten
to steal market share and, perhaps, put the company's
existence in jeopardy).

One can argue that putting these things "up front" burdens
the "start up" too much. Sure. But, if you are successful,
you're going to have to do this anyway! Are you hedging
your bet just in case you AREN'T successful? Planning for
failure?? :-/

> I know it can be done, but not
> at the price where the product would sell enough to make money. For
> certain products (for example the Mackie hard disk recorder) resourceful
> users have been able to keep them fed and healthy for longer than was
> originally anticipated.

So the users have born the cost regardless! If the manufacturer
had increased the selling price, the users would have born the
cost up front. And, in the future, would have benefitted from
the *next* model being a more efficient upgrade (from the
manufacturer's point of view -- less development effort and
risk reinventing what they already *had* prior to adding
new features).

Of course, this is a matter of personal opinion. I just abhor
having to do things over again when I *know* what's likely
to be coming :<

>>> in the [Mackie] HDR, the
>>> version with the
>>> GUI, it restricts the graphics board to an ATI Radeon 7000
>>> series
>
>> > It's a separate *board* and not part of the "system board"?
>
> Yes. The motherboard is from the era where graphics weren't expected to
> be built in. This board has no monitor port.

Oh, OK. Is the motherboard of a "traditional" size/shape?
I think the 7000 series were used on some motherboards so
this could be a possible option in the future (though you'd
probably have to depopulate the video connector and "remote"
it on a ribbon cable, etc. so that the physical connector
could end up in the right place.

>> Nowadays, when you buy something, grab *everything* you can
>> and archive it somewhere. Never know when you will need it
>> only to discover it's not even available on the Way Back
>> Machine...
>
> When the Alesis ADAT first came out, I suggested that anyone who wanted
> to keep their recordings buy a recorder and spare mechanical parts, box
> it up, and store it along with their tapes. I don't know if anyone did
> that. I doubt it. But still, today, people are buying ADATs through eBay
> to transfer their old tapes to disk.

That's what I did with my DLT (computer) tape! Several "drives"
just so I wouldn't have to worry that *the* drive wasn't up to
snuff when I needed it!

> This design was from the USB 1.1 days, so Mackie built an FTP server
> into the machine so that both the internal and external drives can be
> accessed from a network. Files can be dragged and dropped to drives on a
> network computer. While not recommended, it can even be put out on the
> Internet.

Ah, OK. And, presumably, the USB i/f is what makes the external
drive possible? Have you tried adding *multiple* external drives
off the USB port?

> I have the external part of a mobile rack mounted in a USB2/Firewire
> case so I can plug in a removable drive from the recorder and back it up
> to another drive easily. Backing up the internal drive is a little more
> complicated. Either it can be done via FTP or it can be copied to an
> external drive and that can be stored (or backed up to something that
> you'll store). It's really pretty flexible.

What happens when/if the internal drive dies? Is there any
way to recover or "reinitialize" a new internal drive so
that any SOFTWARE on the drive is preserved?

E.g., I have several NAS (Network Attached Storage -- basically
a disk drive with an ethernet connection) drives here that have
their software (firmware) residing *on* the disk drive itself.
So, when the drive dies, the software is gone.

This hasn't been a problem -- yet. Rather, it has made upgrading
the drives a bit more challenging as I have to remove the "old"
drive, install it in a PC along with the "new" drive and then
use 3rd party tools to move the files onto the new drive so
that it will behave like the new drive -- only larger!

>> OTOH, if the machine goes screwwy, there's a change your backup
>> and original can get trashed at the same time.
>
> This is true with any computer based device. And if my tape deck goes
> screwy, it can ruin a section of tape, too.

Yes. Point I was making is that you have your "original" and
"backup" both on the same machine at the same time. It can
damage *both* in the time it decides to damage *one*.

This is one of the reasons I like to keep my backups "offline".
Also, a strong incentive for keeping them on different types
of media.

>>>> ("Files grow and multiply to fill all space available")
>> Yeah, which means perpetually being vigilant and beating them
>> back with a big stick!
>
> I delete far more than I save, and I hardly ever listen to what I save.

I tend to err on the other side and have lots of cruft lying
around that can become overwhelming to keep track of.

E.g., I have saved all of the "test utterances" from a speech
synthesizer that I developed. From each *version* of that
synthesizer! So, I can go back and decide if the pronunciation
embodied in version X was better or worse than version Y. Or,
if just certain aspects of it were better (maybe it handles the
"hard ch" sound as in the american pronunciation of "school"
but the fundamental frequency of the "long a" might be wrong...)

[Don Y]

Hi Mike,

On 5/30/2012 6:42 PM, Mike Rivers wrote:
> On 5/30/2012 7:06 PM, Don Y wrote:
>
>>> Or mixing your latest music project while flying across
>>> the ocean to the mastering lab? <g>
>>
>> Can you actually *do* that (practically)?
>
> I can't imagine doing anything like that seriously but it's one of those
> things that "they" tell you that you can do with your laptop computer. I

Ah!

> expect that you might be able to play around with the sequence of songs
> on a CD, or maybe do some basic editing, maybe even compose a loop-based
> melody. Me, though, I put on my noise canceling headphones, wind up my
> MP3 player, and listen to some radio shows that I'd recorded previously
> and loaded on there. Makes for a good sound track for watching The Food
> Channel on the on-board free satellite TV.

I read. Since most flights are ~2hrs (for me) the time in the
air plus the time at the gate is just enough to finish many
books. So, bring two on each trip -- one for outbound and the
other for inbound!

>> In my experience, the two biggest reasons *claimed* are:
>> - battery won't hold a charge (why spend $80 for a battery?)
>> - a "virus" (i.e., performance has deteriorated enough that
>> I am now bored with this model and need an excuse for a new
>> one)
>
> Yes, those are two common reasons to replace a laptop computer, though
> for what seems to be the statistical majority of laptop users today,
> they wouldn't know if the battery no longer held a charge until they had
> a power failure. Most of them are plugged in and stay in the same place
> all the time.

Exactly how *we* use them! (though only when on-the-road)

> But Grandma likes them because it's all one piece and
> takes up less space than a tower case, keyboard, mouse, and monitor.
> Doesn't require any hookup either, other than plugging in the power, if
> you're using a wireless Internet connection.

Except it tends to mean you have to put it away when you're
done with it -- since it doesn't have a genuine "home".
(My other half would probably *like* that! Get me to put some
of this kit away when I'm "done" with it! Of course, I'm NEVER
really "done" with it, so...)

>> And, if you couldn't imagine TB drives 10 years ago (when GB
>> drives were common -- a decade after MB drives!) you aren't
>> thinking about that, either! :<
>
> These things tend to go in leaps. Once they realized that people will be
> buying 20 GB drives, they made computers accommodate up to, I dunno,
> maybe 500 GB. But the first computer I upgraded with a 20 GB drive
> previously had a 540 MB drive in it, and I thought THAT was big. So I
> suppose that a few terabytes isn't a barrier any longer.

For me, 1.4MB (i.e., a second 8" floppy!) -> 60MB winnie ->
2 x 600MB -> 4 * 4GB -> now I *limit* machines to 1TB (though
I probably have 10-20TB "offline").

>> Think about it: 5 years represents a tenfold increase in
>> computing power, etc. Even if the products that you offer
>> don't have the ability to exploit that performance curve, you,
>> as a company, should be designing with that in mind!
>
> I think that what most of them have in mind when designing their system
> is that in five years, they don't want you to still be using the same
> old system, they want you to buy a new one.

Yet that's exactly contrary to what Hank's kit manufacturer
claims in their sales literature.

I think a lot depends on the user, vendor and market served.

For example, I think most folks wouldn't be caught *dead*
with a 2yr old cell phone. OTOH, a carpenter probably has
the original hammer that he used on his first job site!
("Can't you AFFORD a new hammer?" :> "What's wrong with the
one that I *have*??")

> When it comes to things like
> laptop computers, they don't have to be underpowered to run software
> that comes along 5 years from now in order to be obsolete, they can be
> obsolete as soon as a part that fails is no longer available from any
> source, or it's too difficult to obtain because (like the hard drives
> that we started this discussion with) you may have to buy them from a
> questionable source or in questionable condition rather than brand new
> with a factory warranty if it's DOA or dies an early death.
>
>> You may not worry about your customers having to "buy
>> new" that often, but do you also want to subject *yourself* to
>> having to "buy new" (re-designs from scratch!) as well?
>
> It's what keeps engineers employed.

I wonder how many of the engineers that designed that piece of 5 year
old kit are still employed at the same firm *and* in the same
capacity whereby the firm could benefit from their past experiences?
Or, do they have to watch a new crop of engineers make the same
mistakes as their predecessors??

<frown>

As I said elsewhere, it will be HILARIOUS to watch the scurrying
as Y2038 approaches. And, all the comments from folks saying,
"Hey, didn't we just go through this in Y2K? Why are we doing
it AGAIN? Should we start planning for Y2076? Or, Y2.1K??"

[Neil Gould]

Don Y wrote:
> Hi Mike,
>
> On 5/30/2012 5:36 AM, Mike Rivers wrote:
>> If you're buying something that costs as much or more to install and
>> operate as it does to buy, of course you'll want it to last for a

>> long [...]

>> time. If you're buying something that you can plot down on the desk
>> and go to work doing what you want to do, then it's not all that
>> important that it may become obsolete in a few years.
>
> What if the recordings you've made with it weren't reproducible
> (portable) on newer kit? Or, required a 3rd party device or service
> to "port" them? My point is that there are lots of costs to
> "premature obsolescence". I'm sure you'd prefer to be able to
> install a 200GB drive in the device. But can't because someone
> was shortsighted.
>

I don't believe it's reasonable to conclude that someone was shortsighted
because specialized devices such as audio recorders that were made for a
20GB drive can't handle a 200GB drive. There's much more to it when it comes
to streaming media devices than there would be for a word processor that
runs on a generic motherboard and OS combination. These devices have a
custom, embedded OS designed and optimized for the particular hardware
available at the time. The interface and control ROMs of newer hardware may
not be (and in many cases *were* not) predictable.

Some devices utilized SCSI interfaces, which meant that the particular
requirements of newer hardware would be handled by their conformance to the
SCSI standard. In my experience, this has mitigated issues such as the size,
or even the operational technology of HDs. But the cost of licensing made
this option unattractive in the marketplace for low-volume devices, so
having the "foresight" to provide future upgradability came with a cost that
could affect the success of a device in the market. If it doesn't sell, who
cares that it can't be upgraded?

>> When I was working for the FAA, buying ground based landing systems
>> for airports, it became impossible to buy something completely
>> designed from the ground up for the budgets that Congress gave us.
>> So the days of keeping a 35 year old localizer operating and landing
>> planes safely were over. When I retired at the end of 1999, they
>> were already planning for replacement of ILS equipment designed and
>> built in the 1992-1995 time frame. And it still cost too much money!
>
> Possibly a bad example as every overly regulated and bureaucratic
> organization tends to be inherently inefficient. :<
>

This has nothing to do with over regulation or bureaucratic inefficiency.
This problem is one of screwed up priorities, where we're willing to wage
wars on credit cards but neglect our infrastructure because it "costs too
much" to fix.

--
best regards,

Neil

[Mike Rivers]

On 5/30/2012 10:33 PM, Don Y wrote:

> No, you'll never find anything with my name on it. :>
> I tend to make for others to sell as "their own".

That's the best. If you have to support your designs, at
least you're not dealing with the #%*!@ end users. ;)

> You also want things that cost a lot to buy (even if they cost
> only a little to install/operate!) to last for a long time.
> You wouldn't be keen on buying a new car just because it
> "went out of style", etc.

Cars are a good example. When I was growing up, people
replaced their car every 3 years or so, but that wasn't
because it was going out of style, it was just plain worn
out or (in many places in the country) rusted out. The last
four cars I've owned, however, I've kept for 10 years or
more because they just don't wear out.

But now, cars are getting to be more and more like computers
and phones, and they DO go out of style (or leave you
feeling that you're missing out on some great stuff) in a
couple of years. People want the GPS to be integrated with
the entertainment system,, they want a rear-facing video
camera, a DVD player in the back to entertain the kids, and
of course integration with their smart phone for hands-free
operation (required in many states now for
phoning-while-driving. Leasing is popular with people who
are like that since when the lease term is up, you can
exchange it for one with the latest gadgets and just keep
making the payments.

> What if the recordings you've made with it weren't reproducible
> (portable) on newer kit? Or, required a 3rd party device or
> service to "port" them?

Life's like that.

> My point is that there are lots of costs to
> "premature obsolescence". I'm sure you'd prefer to be able to
> install a 200GB drive in the device. But can't because someone
> was shortsighted.

I get your point, but I believe that to get over what you
insist in calling "shortsighted" in consumer products would
push the cost for every purchaser up to the point where
there would be a significant loss of business. Paying up
front for the design of a computer that could be updated
indefinitely wouldn't suit the average computer buyer.

It might be a smart investment for people buying integrated
audio workstations (like the VF16 - remember what started
this discussion) but have you noticed the lack of products
like this in the market any more? What you find is $300
Portastudios that use SD flash memory which will become
obsolete as soon as you can no longer buy cards in that
format, or, like hard drives, you can no longer buy one
that's small enough to fit the operating system. But someone
who graduates from a system like this usually goes the
computer route where he's accustomed to dealing with a
pattern of replacement rather than upgrading.

> I.e., that's the boat I'm in with many of the designs for hire
> I've done over the past few decades. Tools that version X+1
> will
> refuse (i.e., be completely incapable of) to process files that
> version X created last year!

This isn't being shortsighted, it's making a decision not to
provide backward compatibility. I don't like that one bit,
but I really haven't encountered it very much these days.
Sonic came up with an upgrade that they didn't tell users
wouldn't open their old files and now the company that made
the best audio mastering software is off making something
else. I occasionally get sent .docx files (whatever that
is?) that my copy of Word 2000 won't open, but I found a
converter (from Microsoft, I think) that makes them work at
least to the extent that I need.

But one has to question how far back it's necessary to go. I
talk about this a lot when discussions come around to
archiving (or rather, long term storage) of audio
information. I suspect that software designers don't change
file formats just for the heck of it, they need to do that
to incorporate features that simply didn't exist when the
original program was developed. I keep a copy of Lotus 1-2-3
running for my business bookkeeping because the macros don't
translate to Excel and I haven't found an equivalent Excel
application (and I'm not clever enough to write my own). And
I think that Quicken is too complicated, and too expensive
to use in a very simple manner. I may need to keep a WinXP
computer going for the next 10 years (which isn't going to
be too hard to do) just to keep my books.

>> When I retired at the end of 1999, they were already
>> planning for
>> replacement of ILS equipment designed and built in the
>> 1992-1995 time
>> frame. And it still cost too much money!
>
> Possibly a bad example as every overly regulated and
> bureaucratic
> organization tends to be inherently inefficient. :<

No, a great example. In this case, it's more important for
safety to be able to have consistent maintenance and
verification procedures across the board. When there's an
accident (and thankfully there's never been an accident
that's been attributed to misleading information transmitted
by an ILS) the first question that the lawyers ask is "Are
you sure the system was working properly?" When equipment
lasts for a long time and isn't subject to changing
requirements (or consumer whims) you can develop a reliable
maintenance and test program that will hold up in court.
This is worth paying for.

However, I can say from experience that the process by whch
the Government buys things like this involves a lot of
bickering, approvals, revisions, and sometimes agreements to
change internal procedures before the specification ever
goes out to bid. And there's very room left in there for
creativity. The government, for better or worse, knows
exactly what they want, and that's what they buy. This
doesn't work for the common consumer, however.

> Exactly! It's not trivial to MODIFY an operating system to

> do these things.So, *when* they eventually have to do it

> for their FUTURE products, they will incur that cost. And,
> not be able to benefit from having that "design" in place
> getting hundreds of man-years of testing, "for free" from
> their (legacy) user base. Instead, it will be as if they
> just invented the product anew and have no real preformance
> or reliability data on it.
>
> On the other hand, of you design this *into* the operating
> system
> to begin with, then when the future arrives (or, becomes more
> affordable), you -- and your customers -- can step into it
> without a big reinvestment. You and they aren't rediscovering
> bugs that should have been fixed previously, etc.

Rant and rave all you want, but I maintain that you can only
see so far into the future, and the future is changing
rapidly so it's really hard to keep up. You can incorporate
a lot of "what ifs" in your design that will never be needed
because that "if" gets bypassed by something you don't
foresee. And every "what if" costs money.

To take myself as an example, I've been wanting to set up a
Pro Tools 10 system for a while now. I don't need it to do
my day-to-day work, I'd just like to learn something about
it and have it available when I'm reviewing a new piece of
computer audio hardware so I can verify that it works
properly with the program that's rapidly becoming the
industry standard. I don't want to update any of my existing
computers with Windows 7 even though it will run on at least
one or two of them, because I'd have to spend a lot of time
updating applications that I use every day on those
computers. So my plan is to keep them going under WinXP
until the wheels fall off. I'll probably be dead in 20 years
so if I run out of IDE drives and DIMMs by then, I won't care.

When I absolutely have to get Pro Tools 10 up and running, I
can buy a computer for $300 (less than half the cost of the
program) that I can install it on and it'll work adequately
for my purposes.

What's keeping me from doing that right now? The real
problem is space. Computers don't take up much room, but
they need a keyboard, mouse, and monitor. I share the UI
setup now between my working studio computer and Mackie
recorder using a 2-port KVM switch. I could (planning for
the future) get a 4-port KVM switch, but there's a problem,
and here's where your obsolescence argument kicks in.

The Mackie recorder has a PS/2 mouse port, an AT keyboard
port, and a VGA monitor port. Adapting a PS/2 keyboard to
the AT socket is no problem, though I've never tried a USB
keyboard (with yet another adapter in the string). Most new
computers use USB to connect the mouse and keyboard
(fortunately most still include a VGA port) and I'd have to
get one a few years old to find a PS/2 port.

What I can't find is a KVM switch that has both USB and PS/2
outputs. I could probably make this work using a manual
switch, but when I tried a similar setup using my existing
switch (which has no buttons on it, I ran into a problem.
The switch detects the computer that's started first and
connects to that one. Hitting the Scroll Lock key on the
keyboard switches to the other computer. But when going from
a USB keyboard port on the computer through a PS/2 adapter
to the switch, apparently something doesn't get through. If
I start that computer first, the switch will detect it. If I
start the recorder first, the switch will detect it. I don't
remember all the details now, but there was some combination
that didn't work.

Computers with no "normal" mouse and keyboard ports have
made my system obsolete.

> It's like building a house as a newly wed with just one
> bedroom.
> Then, discovering you're going to have a kid and having to
> build
> an addition for the extra bedroom. But, *only* building an
> addition large enough for a single bedroom! A few years later,
> "Surprise!". And now you realize that your lot size and general
> floorplan makes adding yet another bedroom on the ground floor,
> impractical. So, you think about building *up*, instead. But,
> the first floor framing wasn't designed with that sort of
> a load in mind. So you reinforce the interior and exterior
> walls in preparation for "lifting the roof". etc.

So the answer is to buy your first house large enough to
contain your whole family whether you planned on one or not.
But a lot of newlyweds can't afford a house large enough for
a family of six. Some will choose to expand the present
house but most just move and buy a bigger house.

> One can argue that putting these things "up front" burdens
> the "start up" too much. Sure. But, if you are successful,
> you're going to have to do this anyway! Are you hedging
> your bet just in case you AREN'T successful? Planning for
> failure?? :-/

What? Build a device that will never need to be replaced?
How can you stay in business that way? Once everyone who
needs one has one, your sales dry up.

> So the users have born the cost regardless!

Of course. But by replacing rather than modifying or
upgrading, you often get more than the single benefit of
having a larger disk drive, or more memory, or a new
program. And usually at lower cost.

> If the manufacturer
> had increased the selling price, the users would have born the
> cost up front. And, in the future, would have benefitted from
> the *next* model being a more efficient upgrade (from the
> manufacturer's point of view -- less development effort and
> risk reinventing what they already *had* prior to adding
> new features).

But he might have priced himself out of the market from the
get-go.

> Oh, OK. Is the motherboard of a "traditional" size/shape?
> I think the 7000 series were used on some motherboards so
> this could be a possible option in the future (though you'd
> probably have to depopulate the video connector and "remote"
> it on a ribbon cable, etc. so that the physical connector
> could end up in the right place.

The problem is that the Radeon 7000 series just grew and
grew, and the drivers that are fixed in the now orphaned
software won't work with the newer chip sets. Mackie
(smartly, I think) made the decision half a dozen or so
years ago to get out of the hard disk recorder business in
favor of supporting computer-based recording which is where
the customers are. Sustaining engineering is very expensive
for a company to maintain, so they just don't do it.

> That's what I did with my DLT (computer) tape! Several "drives"
> just so I wouldn't have to worry that *the* drive wasn't up to
> snuff when I needed it!

Do you test your spare tape drives periodically? Analog tape
decks can last forever but they have rubber parts (pinch
rollers and belts) that deteriorate over time, and
eventually replacements (with good rubber anyway) become
unavailable. One of the best roller refurbishers that people
used to send pinch rollers to just closed up shop. I think
their primary business was for refurbishing rollers for
printing presses so it's not surprising that they found
their business dwindling.

> Ah, OK. And, presumably, the USB i/f is what makes the external
> drive possible? Have you tried adding *multiple* external
> drives
> off the USB port?

I don't remember if this motherboard has USB at all. There's
no standard connector on the back of the board so if it's
there, it's a ribbon connector on the board. No, the
external drive is IDE in a mobile rack connected to the
second IDE port on the motherboard. There's a SCSI driver
built into the operating system since at one point they
contemplated offering an external CD-R drive (most were SCSI
at the time) for backup connected to a card in an accessory
slot on the motherboard but never went ahead with it.
There's a 3-1/2" floppy drive for loading the software. To
many users, their point of obsolescence is that they don't
have a computer with a floppy drive so they can't make
installation disks from the self-extracting installer file
which is still available for download from the Mackie web
site (bless their heart). I tell 'em to buy a USB floppy
drive for about $20 and then they'll have one.

> What happens when/if the internal drive dies? Is there any
> way to recover or "reinitialize" a new internal drive so
> that any SOFTWARE on the drive is preserved?

Simple. Install a new drive (sized for the BIOS, of course),
put in the #1 installation floppy, power up, and it boots
from the floppy. From there, you can install the software on
the new drive. You need to know the secret to edit a file on
the floppy which will enable formatting for the Mackie OS
(which you need to do - it also installs a special boot
loader) but restoring the software is really a
straightforward process and will continue to be one as long
as floppy disks and drives are avaialble (which won't be
forever).

The disks are readable from a standard computer, so if, for
example, the boot sector gets corrupted and the only problem
is that it won't boot from the internal drive, you can pull
it, connect it to another computer (I use one of those
IDE-USB adapters) and usually recover project data.

> E.g., I have several NAS (Network Attached Storage -- basically
> a disk drive with an ethernet connection) drives here that have
> their software (firmware) residing *on* the disk drive itself.
> So, when the drive dies, the software is gone.

See, if they were as far sighted as Mackie they'd provide a
way to reload the drives. ;)

[Mike Rivers]

On 5/30/2012 11:02 PM, Don Y wrote:

> I wonder how many of the engineers that designed that piece
> of 5 year
> old kit are still employed at the same firm *and* in the same
> capacity whereby the firm could benefit from their past
> experiences?
> Or, do they have to watch a new crop of engineers make the same
> mistakes as their predecessors??

I don't know what it's like in the computer business, but in
the audio business, everybody I know moves around every few
years. I hardly know anyone at Mackie any more. Engineers
from there are now working at Avid, the guy who led the
design team for the d8b console and HDR24/96 recorder is now
with PreSonus (as is the former chief marketing guy), This
is what makes sustaining an old product impossible. A lot of
knowledge moves out of the company and the newcomers,
because the product has been discontinued, don't have an
opportunity to learn it.

> As I said elsewhere, it will be HILARIOUS to watch the
> scurrying
> as Y2038 approaches.

What's wrong with 2038? I hadn't heard about that one.

[John Williamson]

On 31/05/2012 14:57, Mike Rivers wrote:
> On 5/30/2012 11:02 PM, Don Y wrote:
>
>> As I said elsewhere, it will be HILARIOUS to watch the
>> scurrying
>> as Y2038 approaches.
>
> What's wrong with 2038? I hadn't heard about that one.
>

Nor had I, but apparently it's when the 32 bit Unix clock/ calendar runs
out of bits and wraps to a negative number. The deadline is 03:14:07 UTC
on Tuesday, 19 January 2038. Programmes that work on long lead times
like mortage and pension plans are getting worryingly close to their
deadlines.

Originally known as the AOL bug, accoring to Wikipedia, as it was first
noted on the AOL servers.

[david gourley]

John Williamson <johnwil...@btinternet.com> said...news:a2pdaaF5kaU1
@mid.individual.net:

And I thought it was more prone to affect embedded systems.

Either way, how does it affect pro audio or a VF16 ?

david

[John Williamson]

Thread drift. HTH.

That and the way that a lot of embedded systems in things like sound
recorders are based on the Linux kernel, which uses the Unix clock. It'd
be a right so and so if your latest epic was wiped because the file
datestamp was in 1901 and the library couldn't believe it was so old, or
couldn't be included in a future project because it hadn't been recorded
yet according to the operating system.

[Les Cargill]

So if it really doesn't need a clock. don't put one on it. Most stuff
doesn't.

--
Les Cargill

[Les Cargill]

Neil Gould wrote:
> Don Y wrote:
>> Hi Mike,
>>
>> On 5/30/2012 5:36 AM, Mike Rivers wrote:
>>> If you're buying something that costs as much or more to install and
>>> operate as it does to buy, of course you'll want it to last for a
>>> long [...]
>>> time. If you're buying something that you can plot down on the desk
>>> and go to work doing what you want to do, then it's not all that
>>> important that it may become obsolete in a few years.
>>
>> What if the recordings you've made with it weren't reproducible
>> (portable) on newer kit? Or, required a 3rd party device or service
>> to "port" them? My point is that there are lots of costs to
>> "premature obsolescence". I'm sure you'd prefer to be able to
>> install a 200GB drive in the device. But can't because someone
>> was shortsighted.
>>
> I don't believe it's reasonable to conclude that someone was shortsighted
> because specialized devices such as audio recorders that were made for a
> 20GB drive can't handle a 200GB drive.

it's quite reasonable to conclude that - at least with hard drive
"drivers". I don't think you can get a CS degree without having seen
that math at least once.

And if all else failed, you coulda stole the source code to "dd" plus
a BIOS and used it as a template - or hired a contractor to do that
using an in camera development process to maintain plausible deniability...

> There's much more to it when it comes
> to streaming media devices than there would be for a word processor that
> runs on a generic motherboard and OS combination.

??? Whut??? Word processors are so complicated that
there hasn't really been a new one since the late 70s/early 80s.

Sure people have built them, but the center of gravity
is wildly with the two top contenders. You don't think
Microsoft wrote Word, do you???

> These devices have a
> custom, embedded OS designed and optimized for the particular hardware
> available at the time. The interface and control ROMs of newer hardware may
> not be (and in many cases *were* not) predictable.
>

For IDE harddisks??? They're like blocks of cheese - a purely commodity
thing and have been for a long time. Er, when they were IDE - now
they're SATA blocks of cheese.

> Some devices utilized SCSI interfaces, which meant that the particular
> requirements of newer hardware would be handled by their conformance to the
> SCSI standard. In my experience, this has mitigated issues such as the size,
> or even the operational technology of HDs. But the cost of licensing made
> this option unattractive in the marketplace for low-volume devices, so
> having the "foresight" to provide future upgradability came with a cost that
> could affect the success of a device in the market. If it doesn't sell, who
> cares that it can't be upgraded?
>

Hello? In 2000, IDE hard drives were firmly established. SCSI was
ostensibly obsolete by then. Sure, people used them, but there was
a very small actual market.

We've all dropped "nice to have" things over the side
to make the balloon rise faster, but that's what it is.

>>> When I was working for the FAA, buying ground based landing systems
>>> for airports, it became impossible to buy something completely
>>> designed from the ground up for the budgets that Congress gave us.
>>> So the days of keeping a 35 year old localizer operating and landing
>>> planes safely were over. When I retired at the end of 1999, they
>>> were already planning for replacement of ILS equipment designed and
>>> built in the 1992-1995 time frame. And it still cost too much money!
>>
>> Possibly a bad example as every overly regulated and bureaucratic
>> organization tends to be inherently inefficient. :<
>>
> This has nothing to do with over regulation or bureaucratic inefficiency.
> This problem is one of screwed up priorities, where we're willing to wage
> wars on credit cards but neglect our infrastructure because it "costs too
> much" to fix.
>

the big driver *there* is having air travel be so grandma can
see the grandkids in Minnesota twice a year. If we went back to
a 1970s paradigm of air travel, it wouldn't be a problem.

the sort of thing Mike is talking about is actually probably worthy of
a book-length treatise. Or more than one book.

--
Les Cargill

[Don Y]

Hi Mike,

On 5/31/2012 6:57 AM, Mike Rivers wrote:
> On 5/30/2012 11:02 PM, Don Y wrote:
>
>> I wonder how many of the engineers that designed that piece
>> of 5 year
>> old kit are still employed at the same firm *and* in the same
>> capacity whereby the firm could benefit from their past
>> experiences?
>> Or, do they have to watch a new crop of engineers make the same
>> mistakes as their predecessors??
>
> I don't know what it's like in the computer business, but in the audio
> business, everybody I know moves around every few years. I hardly know
> anyone at Mackie any more. Engineers from there are now working at Avid,
> the guy who led the design team for the d8b console and HDR24/96
> recorder is now with PreSonus (as is the former chief marketing guy),
> This is what makes sustaining an old product impossible.

Exactly. *If* the old product(s) and the new product(s) don't
share a common implementation base -- because someone opted to
cut a corner in "old product" thereby making that aspect of
the design non-portable to the "new product".

"This OS doesn't support volumes larger than XXX bytes. I guess
we'll have to find *another* OS for the GizmaCorder 2" (which
means the newer folks working on GizmaCorder 2 will tend not
to ever see any of the internals of GizmaCorder 1 -- at least
never with an eye towards supporting the G1!

OTOH, if G2 was largely G1 with new and improved features
(higher sampling rates, wider samples, more channels plus
a new set of "effects") then the G2 designers would have
been intimately familiar with the G1 *as* they adapted it
to the G2 and could, then, support the G1 as needed. I.e.,
then support becomes a marketing/political issue instead of
a technical consequence of design!

> A lot of
> knowledge moves out of the company and the newcomers, because the
> product has been discontinued, don't have an opportunity to learn it.
>
>> As I said elsewhere, it will be HILARIOUS to watch the
>> scurrying
>> as Y2038 approaches.
>
> What's wrong with 2038? I hadn't heard about that one.

The wheels fall off the bus. :>

The "clock" by which many computer systems track time "wraps around"
("overflows"). It's sort of analagous to the "Mayan" calendar
"ending" in 2012 (no, it doesn't herald the "end of time" but,
rather, the end of a particular system of keeping track of time...
just like the way computers keep track of time :> )

It's interesting to note that the Maya concocted a scheme for
tracking time over THOUSANDS of years. Yet, modern techies
cant seem to imagine a future of more than ~70 years! Progress?

<grin>

[Neil Gould]

Les Cargill wrote:
> Neil Gould wrote:
>> Don Y wrote:
>>> Hi Mike,
>>>
>>> On 5/30/2012 5:36 AM, Mike Rivers wrote:
>>>> If you're buying something that costs as much or more to install
>>>> and operate as it does to buy, of course you'll want it to last
>>>> for a long [...]
>>>> time. If you're buying something that you can plot down on the desk
>>>> and go to work doing what you want to do, then it's not all that
>>>> important that it may become obsolete in a few years.
>>>
>>> What if the recordings you've made with it weren't reproducible
>>> (portable) on newer kit? Or, required a 3rd party device or service
>>> to "port" them? My point is that there are lots of costs to
>>> "premature obsolescence". I'm sure you'd prefer to be able to
>>> install a 200GB drive in the device. But can't because someone
>>> was shortsighted.
>>>
>> I don't believe it's reasonable to conclude that someone was
>> shortsighted because specialized devices such as audio recorders
>> that were made for a 20GB drive can't handle a 200GB drive.
>
> it's quite reasonable to conclude that - at least with hard drive
> "drivers". I don't think you can get a CS degree without having seen
> that math at least once.
>

Is the math the only issue? What about drive formats? What about the
on-board logic? It sounds as if your perspective is based on the issues that
weren't necessarily settled at the time.

>> There's much more to it when it comes
>> to streaming media devices than there would be for a word processor
>> that runs on a generic motherboard and OS combination.
>
> ??? Whut??? Word processors are so complicated that
> there hasn't really been a new one since the late 70s/early 80s.
>
> Sure people have built them, but the center of gravity
> is wildly with the two top contenders. You don't think
> Microsoft wrote Word, do you???
>

I was comparing the system requirements of streaming media devices vs. word
processing, not talking about *dedicated* word processors a la from the
'70s.

>> These devices have a
>> custom, embedded OS designed and optimized for the particular
>> hardware available at the time. The interface and control ROMs of
>> newer hardware may not be (and in many cases *were* not) predictable.
>>
>
> For IDE harddisks??? They're like blocks of cheese - a purely
> commodity thing and have been for a long time. Er, when they were IDE
> - now they're SATA blocks of cheese.
>

Oh? Why, for example, would a 20GB or 40GB IDE drive work while a 30GB drive
would not with some computer BIOSes of that era? Things are different TODAY,
but that is due to the commodity that HDs have become. Consider the vintage
of gear that intiated this thread, since the issues of that time are
relevant to the design options that were exercised.

>> Some devices utilized SCSI interfaces, which meant that the
>> particular requirements of newer hardware would be handled by their
>> conformance to the SCSI standard. In my experience, this has
>> mitigated issues such as the size, or even the operational
>> technology of HDs. But the cost of licensing made this option
>> unattractive in the marketplace for low-volume devices, so having
>> the "foresight" to provide future upgradability came with a cost
>> that could affect the success of a device in the market. If it
>> doesn't sell, who cares that it can't be upgraded?
>>
>
> Hello? In 2000, IDE hard drives were firmly established. SCSI was
> ostensibly obsolete by then. Sure, people used them, but there was
> a very small actual market.
>

SCSI was *always* a very small "actual market", however the interface was
common in specialized, pro hardware, such as audio/video recorders,
scanners, control interfaces, etc., and the reason was that it *was* a
standard.

>>>> When I was working for the FAA, buying ground based landing systems
>>>> for airports, it became impossible to buy something completely
>>>> designed from the ground up for the budgets that Congress gave us.
>>>> So the days of keeping a 35 year old localizer operating and
>>>> landing planes safely were over. When I retired at the end of
>>>> 1999, they were already planning for replacement of ILS equipment
>>>> designed and built in the 1992-1995 time frame. And it still cost
>>>> too much money!
>>>
>>> Possibly a bad example as every overly regulated and bureaucratic
>>> organization tends to be inherently inefficient. :<
>>>
>> This has nothing to do with over regulation or bureaucratic
>> inefficiency. This problem is one of screwed up priorities, where
>> we're willing to wage wars on credit cards but neglect our
>> infrastructure because it "costs too much" to fix.
>>
>
> the big driver *there* is having air travel be so grandma can
> see the grandkids in Minnesota twice a year. If we went back to
> a 1970s paradigm of air travel, it wouldn't be a problem.
>

Those of us that are pilots have a very different take on what "the big
driver" behind ILS maintenance is, and it hasn't changed for at least 50
years, and probably won't for some time to come.

--
best regards,

Neil

[Don Y]

Hi David,

Audio is temporal in nature. Somewhere inside your little
DIGITAL black box is a gizmo that is dealing with time.
Even if it thinks of time in terms of "sample clocks", it
deals with time.

If that device ever has to communicate time to a *user* in
units that we deal with -- seconds, minutes, hours -- then
there is a mapping, somewhere, between the two. Once
you're thinking in terms of human units (in a product),
it's usually not a big step to thinking in terms of
calendar time.

Do files get time stamped? Chances are, that would be
something more than just "8:23:00" (or even "8:23:00AM")
as it would be impractical to tell if this file was
"newer" (a more recent take) than one stamped "7:45:00"
(since this latter could have been a PM recording)
And, from there to add a real date! Silly to add a
date but not a *year*.

You can see where this goes...

So, if you have a product that makes reference to "human
time units", due diligence suggests you review all uses
of those time entities to verify that they are not
prone to some (bizarre) future screwup.

People design hardware (yes, hardware can also have this
problem -- since it has to choose *some* means of representing
time) and software inconsistently. So, within a given product
(possibly by "the same designer"), these things can be
referenced in incompatible ways.

E.g., I saw the year 2000 displayed as "11000" (think of this as
1 10 00 -- the "1" being a fixed character, the "10" being the
next larger number than '9' and the "00" being the successor
to "99"), "1A00" ('A' being the digit after '9' when counting
in hexadecimal), "1000" (obvious) and "2000" (hooray!).

If you had "set an alarm" (i.e., do <whatever> at <sometime>)
for 1 Jan 2000, there is no guarantee that *any* of those
representations would work -- or NOT work!

If, for example, you chose to compare the string representations
(i.e., characters that a user would see -- like I have typed,
above), then only the last (hooray) example stands a chance
of working.

On the other hand, if you chose some internal representation
(of 1 Jan 2000) to compare against some other (possibly slightly
incompatible) internal representation, then you might get lucky or
fail miserably (e.g., maybe one of those internal representations
only looks at the rightmost two digits of the year -- so, 1900
and 2000 are the same, as far as it is concerned! "Heck, no
one would ever set the clock to 1900 so that's a safe test!")

Even if you don't deal with dates, chances are, the software
is using standard functions ("subroutines") that were written
to handle "times". So, you have to look at how those are
actually implemented to assure yourself that nothing could
go astray. I.e., your Xray machine measures radiation
exposure times in "human units". Best be safe to verify
that it has no dependence on this -- even though it might not
be aware of the *date*!

OTOH, had you sat down and said, in 1970, "Gee, I'm a vital
20-year-old. There is a real probability that *I* will be
alive 68 years hence. Why shouldn't I craft my software so
that it can REPRESENT those times (even if it isn't actively
running 68 years hence).

I.e., the Maya had no concept of 2013 in their calendar.
"We'll worry about that when the time comes..." :>

The year 2100 -- something kids born today are likely to
experience -- will NOT be a leap year. How much software
written today will need to be (wastefully) re-examined
to verify the proper handling of this? "Crap! It never
occurred to me that 2100 would follow 2099!" <grin>

Since the early 80's, my designs have been immune from these
sorts of issues. They'll be vulnerable in the 9999 -> 10,000
New Year :>

More recently, I've implemented time as a "monotonically
increasing value" that bears no relationship to calendar
time. If the user wants a notion of "now" that coincides
with the time displayed on the clock/calendar hanging on
the wall, he can tell me "treat NOW as 2012 May 31 19:04:23".
Hereafter, I'll *normalize* my internal notion of time to
correlate to that (or whatever NEW notion he has specified,
since). But, my device will always have a consistent
notion of time -- regardless of how often the user moves
his notion forwards OR BACKWARDS (!)

"Hmmm... this file has a May 31st timestamp but this other
one is May 30th. Which one *really* is oldest??"

[Mike Rivers]

On 5/31/2012 2:20 PM, Don Y wrote:

> Exactly. *If* the old product(s) and the new product(s) don't
> share a common implementation base -- because someone opted to
> cut a corner in "old product" thereby making that aspect of
> the design non-portable to the "new product".

If you're going to design products for the mass market, in
order to make them price-competitive you have to adopt
someone else's technology. Mackie couldn't make the MDR24/96
and sell it for $2,000 if they had to design their own
motherboard rather than using an Intel style one. They chose
well, I thought, buying an industrial grade board for about
$400 rather than an Asus or whatever was around at the time
for $100, and failures have been very rare, and replacements
were available for a long time. But still, that board was
based on standard technology of the day. It was only better
than a run-of-the-mill one because they tested better and
used better components.

Contrast this with a commercial computer that might be the
heart of a studio's recorder and mixer. You can't buy a Dell
one month and another one a couple of months later and
expect them to have the same motherboard components. Sure,
both of them will run Windows, but they may have different
Firewire chips (or no Firewire chip at all) and your
Firewire audio interface won't work the newer one.

Our $100 Zoom recorders are practical because the
photography business buys enough flash memory cards that
they're dirt cheap. Our hard disk recorders are practical
because for many years, IDE drives could be bought for well
under $100 because there were so many general purpose
computers that used them.

The RADAR 24 track hard disk recorder runs under Linux and
prices start at about $7500. Some people are still buying
them. The Alesis HD24 (recently discontinued) used a disk
format that nobody can read. You want WAV files? You have to
go through the recorder or an accessory dock (discontinued
way too soon). But it allowed them to make a recorder that
used cheaper parts than the Mackie but still cost about the
same.

So, yeah, there are people who are being somewhat
innovative, but it just doesn't pay in this particular
market. And when something based on unique technology dies,
you're really sunk if you bought one. But at least it
didn't cost much, and if you were diligent about backing up
and migrating your data, you could use the data from one
system on another system - because the companies were
forward thinking enough to accommodate an audio file format
specified by Microsoft that will probably be around as long
as cockroaches (but maybe not for 1,000 years).

> "This OS doesn't support volumes larger than XXX bytes. I guess
> we'll have to find *another* OS for the GizmaCorder 2" (which
> means the newer folks working on GizmaCorder 2 will tend not
> to ever see any of the internals of GizmaCorder 1 -- at least
> never with an eye towards supporting the G1!
>
> OTOH, if G2 was largely G1 with new and improved features
> (higher sampling rates, wider samples, more channels plus
> a new set of "effects") then the G2 designers would have
> been intimately familiar with the G1 *as* they adapted it
> to the G2 and could, then, support the G1 as needed. I.e.,
> then support becomes a marketing/political issue instead of
> a technical consequence of design!

Easy to say, but it doesn't always work like that in
practice. You have to be able to adopt to the technology
that's available when you design the product unless you want
to take responsibility for sustaining and extending the
technology. Commercial products that are based on Linux, for
instance the Harrison digital consoles, might be supportable
for a long time because they control the operating system.
But they may need to make extensive changes to it in order
to support a motherboard that they can buy 10 years after
the initial design. They're in the console business, not the
motherboard business.

[Mike Rivers]

On 5/31/2012 1:30 PM, Les Cargill wrote:

> the big driver *there* is having air travel be so grandma can
> see the grandkids in Minnesota twice a year. If we went back to
> a 1970s paradigm of air travel, it wouldn't be a problem.
>
> the sort of thing Mike is talking about is actually probably
> worthy of
> a book-length treatise. Or more than one book.

I was going to write that book, but Dilbert beat me to it. ;)

[Les Cargill]

Perhaps I forget - but I recall that drives of that era were mainly
about the geometry. IOW, a 40 GB was different from an 80 GB ( of the
same rough line of disks ) in that the 80 GB had more
platters. Of course, density was advancing too.

>>> There's much more to it when it comes
>>> to streaming media devices than there would be for a word processor
>>> that runs on a generic motherboard and OS combination.
>>
>> ??? Whut??? Word processors are so complicated that
>> there hasn't really been a new one since the late 70s/early 80s.
>>
>> Sure people have built them, but the center of gravity
>> is wildly with the two top contenders. You don't think
>> Microsoft wrote Word, do you???
>>
> I was comparing the system requirements of streaming media devices vs. word
> processing, not talking about *dedicated* word processors a la from the
> '70s.
>

No, I mean there are two word processors still - Wordperfect and Word.

>>> These devices have a
>>> custom, embedded OS designed and optimized for the particular
>>> hardware available at the time. The interface and control ROMs of
>>> newer hardware may not be (and in many cases *were* not) predictable.
>>>
>>
>> For IDE harddisks??? They're like blocks of cheese - a purely
>> commodity thing and have been for a long time. Er, when they were IDE
>> - now they're SATA blocks of cheese.
>>
> Oh? Why, for example, would a 20GB or 40GB IDE drive work while a 30GB drive
> would not with some computer BIOSes of that era?

Because one of the geometry parameters of the drive exceeded a
limitation of the model in the BIOS. :(

The real "why" is that they did it wrong....

> Things are different TODAY,
> but that is due to the commodity that HDs have become. Consider the vintage
> of gear that intiated this thread, since the issues of that time are
> relevant to the design options that were exercised.
>

To be sure, I might be making a bad assumption about 2000 era drives,
but I recall that they were mostly all one thing.

>>> Some devices utilized SCSI interfaces, which meant that the
>>> particular requirements of newer hardware would be handled by their
>>> conformance to the SCSI standard. In my experience, this has
>>> mitigated issues such as the size, or even the operational
>>> technology of HDs. But the cost of licensing made this option
>>> unattractive in the marketplace for low-volume devices, so having
>>> the "foresight" to provide future upgradability came with a cost
>>> that could affect the success of a device in the market. If it
>>> doesn't sell, who cares that it can't be upgraded?
>>>
>>
>> Hello? In 2000, IDE hard drives were firmly established. SCSI was
>> ostensibly obsolete by then. Sure, people used them, but there was
>> a very small actual market.
>>
> SCSI was *always* a very small "actual market", however the interface was
> common in specialized, pro hardware, such as audio/video recorders,
> scanners, control interfaces, etc., and the reason was that it *was* a
> standard.
>

Sure! They'd made more sense a few years before.

>>>>> When I was working for the FAA, buying ground based landing systems
>>>>> for airports, it became impossible to buy something completely
>>>>> designed from the ground up for the budgets that Congress gave us.
>>>>> So the days of keeping a 35 year old localizer operating and
>>>>> landing planes safely were over. When I retired at the end of
>>>>> 1999, they were already planning for replacement of ILS equipment
>>>>> designed and built in the 1992-1995 time frame. And it still cost
>>>>> too much money!
>>>>
>>>> Possibly a bad example as every overly regulated and bureaucratic
>>>> organization tends to be inherently inefficient. :<
>>>>
>>> This has nothing to do with over regulation or bureaucratic
>>> inefficiency. This problem is one of screwed up priorities, where
>>> we're willing to wage wars on credit cards but neglect our
>>> infrastructure because it "costs too much" to fix.
>>>
>>
>> the big driver *there* is having air travel be so grandma can
>> see the grandkids in Minnesota twice a year. If we went back to
>> a 1970s paradigm of air travel, it wouldn't be a problem.
>>
> Those of us that are pilots have a very different take on what "the big
> driver" behind ILS maintenance is, and it hasn't changed for at least 50
> years, and probably won't for some time to come.
>

--

Les Cargill

[Neil Gould]

There were other issues, too, as the desire to increase HD capacity lead to
some poorly supported "solutions".

> No, I mean there are two word processors still - Wordperfect and Word.
>

Well, I suspect that OpenOffice and its spin-off would shift that a bit, but
my point was that the demands of a word processor are not remotely similar
to a real-time processing app, such as is typical of audio. Surely, we don't
disagree about that?

>>>> These devices have a
>>>> custom, embedded OS designed and optimized for the particular
>>>> hardware available at the time. The interface and control ROMs of
>>>> newer hardware may not be (and in many cases *were* not)
>>>> predictable.
>>>>
>>>
>>> For IDE harddisks??? They're like blocks of cheese - a purely
>>> commodity thing and have been for a long time. Er, when they were
>>> IDE - now they're SATA blocks of cheese.
>>>
>> Oh? Why, for example, would a 20GB or 40GB IDE drive work while a
>> 30GB drive would not with some computer BIOSes of that era?
>
> Because one of the geometry parameters of the drive exceeded a
> limitation of the model in the BIOS. :(
>
> The real "why" is that they did it wrong....
>

"Wrong" is hard to claim when there wasn't a common standard, as there is
today. Also, one real need was for backward compatibility. So, there were
contending solutions, even from a single manufacturer.

>> Things are different TODAY,
>> but that is due to the commodity that HDs have become. Consider the
>> vintage of gear that intiated this thread, since the issues of that
>> time are relevant to the design options that were exercised.
>>
>
> To be sure, I might be making a bad assumption about 2000 era drives,
> but I recall that they were mostly all one thing.
>

In the commodity market things were starting to gel by Y2000, but, this ng
is NOT about that market, even with the higher-volume products. How many
VF16s do you think were sold all together? How about ProTools? I'd bet there
are more iPads than ProTools rigs in the general marketplace.

>>> Hello? In 2000, IDE hard drives were firmly established. SCSI was
>>> ostensibly obsolete by then. Sure, people used them, but there was
>>> a very small actual market.
>>>
>> SCSI was *always* a very small "actual market", however the
>> interface was common in specialized, pro hardware, such as
>> audio/video recorders, scanners, control interfaces, etc., and the
>> reason was that it *was* a standard.
>>
>
> Sure! They'd made more sense a few years before.
>

SCSI still makes sense for some applications, but it is more expensive to
implement than firewire or USB, so it isn't practical for the mass market
where any advantages it might offer are useless to the majority of users.

--
best regards,

Neil

[Les Cargill]

A bit. It's good stuff.

> but
> my point was that the demands of a word processor are not remotely similar
> to a real-time processing app, such as is typical of audio. Surely, we don't
> disagree about that?
>

No, not at all. Although I don't think harddisk performance has been
a consideration for a very long time.

>>>>> These devices have a
>>>>> custom, embedded OS designed and optimized for the particular
>>>>> hardware available at the time. The interface and control ROMs of
>>>>> newer hardware may not be (and in many cases *were* not)
>>>>> predictable.
>>>>>
>>>>
>>>> For IDE harddisks??? They're like blocks of cheese - a purely
>>>> commodity thing and have been for a long time. Er, when they were
>>>> IDE - now they're SATA blocks of cheese.
>>>>
>>> Oh? Why, for example, would a 20GB or 40GB IDE drive work while a
>>> 30GB drive would not with some computer BIOSes of that era?
>>
>> Because one of the geometry parameters of the drive exceeded a
>> limitation of the model in the BIOS. :(
>>
>> The real "why" is that they did it wrong....
>>
> "Wrong" is hard to claim when there wasn't a common standard, as there is
> today. Also, one real need was for backward compatibility. So, there were
> contending solutions, even from a single manufacturer.
>

I am desperately trying to remember the last time I had to load a
driver for a disk drive, and it's probably back when you had to do
things like HIMEM and such, when things tended to be a lot Real Mode.

>>> Things are different TODAY,
>>> but that is due to the commodity that HDs have become. Consider the
>>> vintage of gear that intiated this thread, since the issues of that
>>> time are relevant to the design options that were exercised.
>>>
>>
>> To be sure, I might be making a bad assumption about 2000 era drives,
>> but I recall that they were mostly all one thing.
>>
> In the commodity market things were starting to gel by Y2000, but, this ng
> is NOT about that market, even with the higher-volume products. How many
> VF16s do you think were sold all together? How about ProTools? I'd bet there
> are more iPads than ProTools rigs in the general marketplace.
>

Oh, absolutely.

>>>> Hello? In 2000, IDE hard drives were firmly established. SCSI was
>>>> ostensibly obsolete by then. Sure, people used them, but there was
>>>> a very small actual market.
>>>>
>>> SCSI was *always* a very small "actual market", however the
>>> interface was common in specialized, pro hardware, such as
>>> audio/video recorders, scanners, control interfaces, etc., and the
>>> reason was that it *was* a standard.
>>>
>>
>> Sure! They'd made more sense a few years before.
>>
> SCSI still makes sense for some applications, but it is more expensive to
> implement than firewire or USB, so it isn't practical for the mass market
> where any advantages it might offer are useless to the majority of users.
>


--

Les Cargill

[hank alrich]

Don Y <th...@isnotme.com> wrote:

> Hi Mike,
>
> On 5/30/2012 4:49 AM, Mike Rivers wrote:
> > On 5/29/2012 4:28 PM, Don Y wrote:
> >
> >> I find laptop keyboards "icky". The disks are slow. And, the
> >> processors are often stripped down for less demanding
> >> applications (e.g., want to do some 3D CAD rendering?).
> >
> > Or mixing your latest music project while flying across the ocean to the
> > mastering lab? <g>
>
> Can you actually *do* that (practically)? I can't hear myself
> *think* on most planes, let alone try to come up with a
> bit of audio that I might be able to appreciate in the absence
> of all that background noise.

Mixing, not so much, but editing in headphones, absolutely. Some years
ago I read about the production of a Bela Fleck album, and at several
stages heading to and from different parts of the country to meet with
other musicians and record in their locale he'd be editing parts on the
plane.

[Neil Gould]

Les Cargill wrote:
> Neil Gould wrote:
>> my point was that the demands of a word processor are not remotely
>> similar to a real-time processing app, such as is typical of audio.
>> Surely, we don't disagree about that?
>>
>
> No, not at all. Although I don't think harddisk performance has been
> a consideration for a very long time.
>

Perhaps not for two-track recording, but those pushing their systems to
record 16+ simultaneous tracks have always had to take HD performance into
consideration, and that hasn't changed because some of those folks went from
44.1/16 to 192/xxx. Performance matters when you have to flush that much
data reliably.

>> "Wrong" is hard to claim when there wasn't a common standard, as
>> there is today. Also, one real need was for backward compatibility.
>> So, there were contending solutions, even from a single manufacturer.
>>
>
> I am desperately trying to remember the last time I had to load a
> driver for a disk drive, and it's probably back when you had to do
> things like HIMEM and such, when things tended to be a lot Real Mode.
>

Whether one has to load a driver for a disk drive depends on whether the
generic driver supplied with the OS works witht the particular HD that you
bought. That is still true today, which is probably why drivers are
available for, if not sold with HDs on the market.

--
best regards,

Neil

[Les Cargill]

Neil Gould wrote:
> Les Cargill wrote:
>> Neil Gould wrote:
>>> my point was that the demands of a word processor are not remotely
>>> similar to a real-time processing app, such as is typical of audio.
>>> Surely, we don't disagree about that?
>>>
>>
>> No, not at all. Although I don't think harddisk performance has been
>> a consideration for a very long time.
>>
> Perhaps not for two-track recording, but those pushing their systems to
> record 16+ simultaneous tracks have always had to take HD performance into
> consideration, and that hasn't changed because some of those folks went from
> 44.1/16 to 192/xxx. Performance matters when you have to flush that much
> data reliably.
>

16 tracks of 192/24 is < 80 kBit/sec. Not very demanding.

>>> "Wrong" is hard to claim when there wasn't a common standard, as
>>> there is today. Also, one real need was for backward compatibility.
>>> So, there were contending solutions, even from a single manufacturer.
>>>
>>
>> I am desperately trying to remember the last time I had to load a
>> driver for a disk drive, and it's probably back when you had to do
>> things like HIMEM and such, when things tended to be a lot Real Mode.
>>
> Whether one has to load a driver for a disk drive depends on whether the
> generic driver supplied with the OS works witht the particular HD that you
> bought. That is still true today, which is probably why drivers are
> available for, if not sold with HDs on the market.
>

So do it like Linux does it. No drivers required... I realize Winboxen
had legacy issues, but a standalone doesn't.

--
Les Cargill

[John Williamson]

Les Cargill wrote:

>
> 16 tracks of 192/24 is < 80 kBit/sec. Not very demanding.
>

Pardon? 16 tracks X 192,000 samples/ second x 3 (possibly 4, allowing
for the fact that writing samples as four bytes is normally faster than
writing three bytes per sample) bytes per sample is a *lot* more than
80kbits/ second. At least 9,216,000 *bytes*/ second. or 73,728,000 bits,
plus overheads. That's all got to be got into the buffer and back out
onto the platter in the right order, while simultaneously writing to at
least 16 files.

[Don Y]

Hi Mike,

>> No, you'll never find anything with my name on it. :>
>> I tend to make for others to sell as "their own".
>
> That's the best. If you have to support your designs, at least you're
> not dealing with the #%*!@ end users. ;)

Exactly. Customers are tricky and the more customers, the
more "issues" come up. When all of those customers are hidden
behind a *single* customer (or a small number of different
customers for different products), you only have to interact
and reason with that *one*.

Customers also never know what they WANT. They *think* they
do but actually only know what they DON'T want -- after
they've seen it! So, the best approach is to listen to
customers, figure out what they REALLY mean, then ignore
what they've said :>

I provide lifetime "fixes" for no charge. When dealing with
another businessman, you're less likely to find him trying
to con you EXPECTING to get away with it! ("Honest, it
just stopped working one day. I have no idea what's wrong..."
"Really? Why are there bits of toilet paper stuck to it
and a general feeling of *damp*?")

>> You also want things that cost a lot to buy (even if they cost
>> only a little to install/operate!) to last for a long time.
>> You wouldn't be keen on buying a new car just because it
>> "went out of style", etc.
>
> Cars are a good example. When I was growing up, people replaced their
> car every 3 years or so, but that wasn't because it was going out of
> style, it was just plain worn out or (in many places in the country)
> rusted out. The last four cars I've owned, however, I've kept for 10
> years or more because they just don't wear out.

I drive cars into the ground. My current ride is 25 years old.
But, with only 60K original miles on it and no "salt", here,
I figure it's good for another 25! :>

> But now, cars are getting to be more and more like computers and phones,
> and they DO go out of style (or leave you feeling that you're missing
> out on some great stuff) in a couple of years. People want the GPS to be
> integrated with the entertainment system,, they want a rear-facing video
> camera, a DVD player in the back to entertain the kids, and of course
> integration with their smart phone for hands-free operation (required in
> many states now for phoning-while-driving. Leasing is popular with
> people who are like that since when the lease term is up, you can
> exchange it for one with the latest gadgets and just keep making the
> payments.

And those people PAY for that "fickleness".

Even if their new purchase is "fair value for money", they spend
more time shopping for cars, arranging financing, getting used
to new features, etc.

>> What if the recordings you've made with it weren't reproducible
>> (portable) on newer kit? Or, required a 3rd party device or
>> service to "port" them?
>
> Life's like that.

Only if you let it be. There are hidden costs with these
sorts of decisions. If you assign no/little value to being
able to recover those recordings, then that aspect of the
cost is small/nonexistent.

OTOH, if you later discover these costs, you can quickly
regret what had previously seemed like a good purchase
decision.

E.g., Richard and I were discussing portable speech synthesizers
elsewhere. On the one hand, portable is a big win. And, a
*rechargeable* battery seems like an even *bigger* win -- cuts
down on the cost of batteries! (if you are using this sort
of device for several HOURS daily, you go through batteries
surprisingly fast!) But, when you discover, belatedly, that
the rechargeable battery in question is a "special", costs
$20 and can only be mail-ordered from the original vendor
(assuming he is still in business and still supporting the
product), suddenly there are costs that offset a lot of those
originally perceived benefits/gains.

E.g., I own no "cordless (hand) tools". I'm not a "contractor"
who is out on job sites daily using these sorts of tools. As
such, any battery pack would *always* be flat when I needed
the tool. And, quite possibly in need of replacement at having
run down too far, too long and in too inhospitable an environment.

OTOH, I have many *powered* hand tools. I just insist on
an AC power cord instead of the "(IN)convenience" of a battery
pack. The cost/benefit of the cord outweighs the cost/benefit
of the cordLESS (both of which outweighing the cost of the
non-powered options, in most cases)

>> My point is that there are lots of costs to
>> "premature obsolescence". I'm sure you'd prefer to be able to
>> install a 200GB drive in the device. But can't because someone
>> was shortsighted.
>
> I get your point, but I believe that to get over what you insist in
> calling "shortsighted" in consumer products would push the cost for
> every purchaser up to the point where there would be a significant loss
> of business. Paying up front for the design of a computer that could be
> updated indefinitely wouldn't suit the average computer buyer.

But we're not (just) talking about computers. We're talking about
special kit intended to be used by special consumers (don't let
it go to your head! :> ). Computers (esp laptops) are treated
as disposable products. From the comments I've heard here, it
doesn't seem like you folks are upgrading kit every two or three
years (the normal "business cycle" for computers).

> It might be a smart investment for people buying integrated audio
> workstations (like the VF16 - remember what started this discussion) but
> have you noticed the lack of products like this in the market any more?
> What you find is $300 Portastudios that use SD flash memory which will
> become obsolete as soon as you can no longer buy cards in that format,
> or, like hard drives, you can no longer buy one that's small enough to
> fit the operating system. But someone who graduates from a system like
> this usually goes the computer route where he's accustomed to dealing
> with a pattern of replacement rather than upgrading.

But that would be an equally short-sighted design decision.
Why does it have to be a "5 pound sack" for "5 pounds of sh*t"?
Is there something wrong with a 50 pound sack??

Again, you *know* that things will only be getting smaller,
faster, more capacious, cheaper, etc. Why deliberately
choose to ignore those facts?

If you deliberately want to artificially limit your customers to
purchasing "approved parts", you can always add a line of code
that says "if part != approved then complain and refuse to operate".

If you want them to be coerced into upgrading to take advantage
of different features (at a different price point!), you can
artificially limit the availability of those features.

But, to impose the same constraints on YOURSELF as you go
about designing the successor product seems, um, "shortsighted" :>

>> I.e., that's the boat I'm in with many of the designs for hire
>> I've done over the past few decades. Tools that version X+1
>> will
>> refuse (i.e., be completely incapable of) to process files that
>> version X created last year!
>
> This isn't being shortsighted, it's making a decision not to provide
> backward compatibility.

It may not have been a conscious "decision". They may have rushed
to get a new product out before a competitor -- hoping/planning to
back-fill these capabilities later. Or, may have made some
"little" changes to the implementation without realizing the
ramifications and how they would affect old file formats. Or,
they may simply have contempt for their customers and figure
they can get away with these sorts of affronts.

<shrug> Never know. But, when you spend a few kilobucks on
a piece of software only to find yourself spending even *more*
AND having to maintain *both*, it doesn't inspire much loyalty!

> I don't like that one bit, but I really haven't
> encountered it very much these days. Sonic came up with an upgrade that
> they didn't tell users wouldn't open their old files and now the company
> that made the best audio mastering software is off making something
> else. I occasionally get sent .docx files (whatever that is?) that my
> copy of Word 2000 won't open, but I found a converter (from Microsoft, I
> think) that makes them work at least to the extent that I need.
>
> But one has to question how far back it's necessary to go.

How about "the release immediately preceeding this one"? :<

> I talk about
> this a lot when discussions come around to archiving (or rather, long
> term storage) of audio information. I suspect that software designers
> don't change file formats just for the heck of it, they need to do that
> to incorporate features that simply didn't exist when the original
> program was developed.

Actually, the results are all over the map when it comes to these
sorts of decisions! Some firms treat software as an "art" and
the individuals responsible for it as (pampered?!) artists. Some
treat it as BFM and almost *cringe* at the prospect of trying
to understand what "those guys with the robes and pointed hats"
are doing. Some have formal standards in place that dampen
"unnecessary" change (i.e., by making it to costly to implement).

Some firms have very loose (non existent?) testing procedures.
Other firms are very diligent in their testing methodologies.
(if you have to build a formal test suite -- a product in its
own right -- for each piece of software that you create, there
is an incentive to stick with the status quo as that allows
you to reuse parts of the "old" test suite... saving you that
labor and potential uncertainty).

Some firms cater to regulated industries (where "outsiders"
impose restrictions on what you do and how you do it) while
others have no formal standards at all.

I.e., it is very likely that a small number of minds (often 1)
were involved in making big decisions. For probably half of
the products I've worked on in my career, I was the sole
(or one of two or three) voice driving the technical and
marketing aspects of the design. (scary, eh? :> Now you
see why I think real hard about the needs of different
application domains)

I used Ventura Publisher for many years. One nice feature
was that it maintained files in a human readable ASCII format.
So, I could (carefully) wander through those files and make
changes that the program couldn't do easily for me.

Corel bought the product and changed the file format to
something proprietary effectively locking me out. They
added no functionality -- just rebranded the product as
"Corel Ventura". So, Corel lost my business and I cling
to an old "Ventura branded" release.

> I keep a copy of Lotus 1-2-3 running for my
> business bookkeeping because the macros don't translate to Excel and I
> haven't found an equivalent Excel application (and I'm not clever enough
> to write my own).

Nor do you necessarily *want* to! I don't want to have to
cast an iron head for a hammer and whittle a hickory handle for it.
Whether I can or can't is immaterial.

> And I think that Quicken is too complicated, and too
> expensive to use in a very simple manner. I may need to keep a WinXP
> computer going for the next 10 years (which isn't going to be too hard
> to do) just to keep my books.

Exactly. *You* want to determine the lifetime of these products,
not a vendor operating under a different set of constraints.

>>> When I retired at the end of 1999, they were already
>>> planning for
>>> replacement of ILS equipment designed and built in the
>>> 1992-1995 time
>>> frame. And it still cost too much money!
>>
>> Possibly a bad example as every overly regulated and
>> bureaucratic
>> organization tends to be inherently inefficient. :<
>
> No, a great example.

It's a bad example because its a single HUGE customer with
"personalities" making decisions instead of "ideas" making
them. The lack of competition (among customers) skews the
types of results that you get.

How many consumers "influence" their cell phone providers to
use a particular communication technology? :>

Small numbers of big customers distort decisions. Just like
huge numbers of tiny customers. In the first case, a smaller
customer has no say. In the second, a customer that doesn't
have the same herd mentality suffers.

> In this case, it's more important for safety to be
> able to have consistent maintenance and verification procedures across
> the board. When there's an accident (and thankfully there's never been
> an accident that's been attributed to misleading information transmitted
> by an ILS) the first question that the lawyers ask is "Are you sure the
> system was working properly?" When equipment lasts for a long time and
> isn't subject to changing requirements (or consumer whims) you can
> develop a reliable maintenance and test program that will hold up in
> court. This is worth paying for.

You mean like having an operating system that knows how to handle
"bigger files" and "bigger volumes" in place with years of
customer experience testifying to its stability? Instead of a
new OS designed as a result of needing to handle larger disk
sizes "a couple years hence"?? :>

>> On the other hand, of you design this *into* the operating
>> system
>> to begin with, then when the future arrives (or, becomes more
>> affordable), you -- and your customers -- can step into it
>> without a big reinvestment. You and they aren't rediscovering
>> bugs that should have been fixed previously, etc.
>
> Rant and rave all you want, but I maintain that you can only see so far
> into the future, and the future is changing rapidly so it's really hard
> to keep up.

Those changes might include different physical (and virtual) media
types. But, one thing is for sure is that capacities are going up.
Whether it is physical memory used within a device, processor
speed, bulk storage sizes, etc. the numbers never go *down*.
You also overestimate how much effort is required to accommodate
these things. And, how inexpensive (in parts, labor and development
costs) they are to implement.

> You can incorporate a lot of "what ifs" in your design that
> will never be needed because that "if" gets bypassed by something you
> don't foresee. And every "what if" costs money.

Every "what if" that becomes a *real* "if" costs money as well!
And, often considerably more as it now has to be retrofitted
to a design that might not have taken it into consideration
originally.

There are also consequences (costs) in terms of usability
and consistency. Here are two lists of files:
-9 files.txt 1 files.txt
-fred.txt 12 files.txt
1 files.txt 100 files.txt
100 files.txt -9 files.txt
12 files.txt fred.txt
fred.txt -fred.txt
Both lists refer to the same 6 files. The left column is
how a "DIR" command in a DOS box ("command prompt") shows
them. The right column is how they are listed in an explored
window.

There is no way to get either representation to agree
with the other! Their sort criteria are incompatible.

The same sort of thing happens when you view a file
locally vs. remotely: "Why is this file 12K in this
view and 13K in this other view? Aren't they the same
file? Do I have to physically compare them to be sure?
They're both 12,621 bytes in length..."

There's a cost to these inconsistencies. Much born by the
user. But, also by the vendor -- having to document
"exceptions", having to field support calls related to
them, etc. Because they didn't think ahead to address
these issues.

Adding something to an existing product after-the-fact
rarely comes without significant consequences. Look at
adding long filename support to MS's products (present
in other OS's long before!). Or real-time capabilities
to the Linux kernel.

> To take myself as an example, I've been wanting to set up a Pro Tools 10
> system for a while now. I don't need it to do my day-to-day work, I'd
> just like to learn something about it and have it available when I'm
> reviewing a new piece of computer audio hardware so I can verify that it
> works properly with the program that's rapidly becoming the industry
> standard. I don't want to update any of my existing computers with
> Windows 7 even though it will run on at least one or two of them,
> because I'd have to spend a lot of time updating applications that I use
> every day on those computers. So my plan is to keep them going under
> WinXP until the wheels fall off. I'll probably be dead in 20 years so if
> I run out of IDE drives and DIMMs by then, I won't care.

The bigger problem will be keeping the PC's themselves running.
PC's are just not designed for longevity -- even if you replace
the pseudo-mechanical parts. Capacitor plague being a common
ailment. I used to think this only decimated consumer products
but have seen it in "industrial grade" servers, as well (though
not as soon).

Buy a new PC after Xp has been EOL'd and you probably won't find
XP being compatible with whatever the hardware du jour is at that
time. Try reinstalling XP (from CD) on a SUPPORTED machine at that
time and you might find all the "updates" have disappeared. (This
is why I make images of all my installs)

> When I absolutely have to get Pro Tools 10 up and running, I can buy a
> computer for $300 (less than half the cost of the program) that I can
> install it on and it'll work adequately for my purposes.

That's the approach I've taken with things (though nearly every
Wintel machine is running XP, currently). The peripherals and
user I/O's that each type of application needs are just not
practical to try to get to coexist in a single machine. Scanners
and photoquality printers on machines for DTP; tablets and pen
plotters on machines for CAD, etc.

> What's keeping me from doing that right now? The real problem is space.
> Computers don't take up much room, but they need a keyboard, mouse, and
> monitor. I share the UI setup now between my working studio computer and
> Mackie recorder using a 2-port KVM switch. I could (planning for the
> future) get a 4-port KVM switch, but there's a problem, and here's where
> your obsolescence argument kicks in.
>
> The Mackie recorder has a PS/2 mouse port, an AT keyboard port, and a
> VGA monitor port.

> What I can't find is a KVM switch that has both USB and PS/2 outputs. I
> could probably make this work using a manual switch, but when I tried a
> similar setup using my existing switch (which has no buttons on it, I
> ran into a problem. The switch detects the computer that's started first
> and connects to that one. Hitting the Scroll Lock key on the keyboard
> switches to the other computer. But when going from a USB keyboard port
> on the computer through a PS/2 adapter to the switch, apparently
> something doesn't get through. If I start that computer first, the
> switch will detect it. If I start the recorder first, the switch will
> detect it. I don't remember all the details now, but there was some
> combination that didn't work.
>
> Computers with no "normal" mouse and keyboard ports have made my system
> obsolete.

Imagine what it's like when you have Sun keyboards and PC keyboards,
etc! (i.e., far more incompatible -- including the actual keys on
the keyboard -- than a USB vs PS2 vs. AT vs. XT) :>

I've never had any luck with KVM's -- and I've tried a LOT of them
(I have lots of machines in use, here). I had a pair of these:
<http://www.tomshardware.com/reviews/saving-space,784-9.html>
at one point (they may even be hiding in the garage for all I know)
but they didn't meet my needs. Again, it had to do with the
order that machines were powered up, etc. The box in question
has PS2 & USB connections to the individual machines but only
PS2 connections to the shared keyboard/mouse (see photo at URL).
No idea if it would work for you. Note that when you share
a keyboard, the KVM has to remember what the keyboard was
last *told* by the last computer it talked to (e.g., should
CAPS lock be enabled?) so that the keyboard can be returned
to the appropriate "state" when it is reconnected to that
particular computer (i.e., it's not just a "switch")

Rose makes some capable -- but pricey -- KVM's. Though no guarantees
there, either.

Instead, I use monitors with A/B (or A/B/C/D) switches built in
to eliminate the need for the video switch. And, a keyboard
drawer under each keyboard for the "extra" keyboard. A
workstation thus only deals with two computers. Each of those
have two video outputs feeding a *pair* of monitors. If I
am working on just one machine, this gives me a big desktop.

OTOH, if I happen to need to deal with those two machines at
the same time (for example, wanting to look at the schematic
for a design on machine A while writing software on machine B
to drive that design ), then I can toggle the A/B switch on
one of the monitors while leaving the other as is. This leaves
me with displays from two computers "side by side" where I can
consult or interact with whichever is necessary.

Of course, it is really easy to forget which keyboard talks
to which display :> But, there's no way around that when
you have to deal with two different machines or toolsets.

>> One can argue that putting these things "up front" burdens
>> the "start up" too much. Sure. But, if you are successful,
>> you're going to have to do this anyway! Are you hedging
>> your bet just in case you AREN'T successful? Planning for
>> failure?? :-/
>
> What? Build a device that will never need to be replaced? How can you
> stay in business that way? Once everyone who needs one has one, your
> sales dry up.

This is a red herring. Just because you "have one" doesn't mean
you never buy something else. And, that followup purchase is one
you look forward to -- because of good experiences with the
previous product.

I worked for a firm that manufactured a ~$30K piece of kit.
Probably a 10% margin. But, they only sold a handful a year.
I showed them how to redesign for a ~$500 product cost, sell
it for ~$5K (90% margin and more actual dollars profit!)
and INCREASE the number of units sold (at $5K, a customer could
easily afford to keep a spare or two on hand -- not so at $30K!)

And, you're more likely to generate "buzz" among customers!

If you decide you've genuinely outgrown a product, it's likely
that your old kit will end up resold or gifted to a friend
thereby turning them onto the product line. People rarely complain
that they CAN'T upgrade because their current device is "too good".

>> So the users have born the cost regardless!
>
> Of course. But by replacing rather than modifying or upgrading, you
> often get more than the single benefit of having a larger disk drive, or
> more memory, or a new program. And usually at lower cost.

Lower $$$ price not lower *total* cost. Each time you have to
learn something new, there is a cost that you bear even if you
don't put a dollar figure on it.

I've seen published statistics about the number of products
returned (lost sales + cost of absorbing the returned product!)
simply because they were too much "work" for the user to bear.
They'd prefer "doing without" or returning to their older
product.

> > If the manufacturer
>> had increased the selling price, the users would have born the
>> cost up front. And, in the future, would have benefitted from
>> the *next* model being a more efficient upgrade (from the
>> manufacturer's point of view -- less development effort and
>> risk reinventing what they already *had* prior to adding
>> new features).
>
> But he might have priced himself out of the market from the get-go.

But not for anticipating larger disk sizes! :> (trust me,
this sort of thing is easy to address "from the start")

However, you have to think clearly and precisely about your goal
before you start implementing. Otherwise, you "discover"
its limitations down the road -- when its more costly to
recover.

[One of my first employers told me that the hardest thing for
engineers to do was come up against a brick wall and start
*beating* on it instead of rethinking their approach and
starting over. The more you have "invested", the greater the
temptation to rationalize a solution that you would never have
accepted initially]

>> That's what I did with my DLT (computer) tape! Several "drives"
>> just so I wouldn't have to worry that *the* drive wasn't up to
>> snuff when I needed it!
>
> Do you test your spare tape drives periodically? Analog tape decks can
> last forever but they have rubber parts (pinch rollers and belts) that
> deteriorate over time, and eventually replacements (with good rubber
> anyway) become unavailable. One of the best roller refurbishers that
> people used to send pinch rollers to just closed up shop. I think their
> primary business was for refurbishing rollers for printing presses so
> it's not surprising that they found their business dwindling.

DLT's have no rubber parts, no capstan. But, yes, they all get "tested"
regularly as it's easiest to just attach one to each machine for its
backup device (instead of sharing a device or backing up over the
network). There are 5 in the office, currently, and 3 more (used) in
storage for "spare parts" (though the only part I've had fail has been
the leader which pulls the *tape* leader out of the tape cartridge and
into the tape drive; sometimes they snap)

>> Ah, OK. And, presumably, the USB i/f is what makes the external
>> drive possible? Have you tried adding *multiple* external
>> drives
>> off the USB port?
>
> I don't remember if this motherboard has USB at all.

I thought you said it had USB1.1? "FTP server"? etc.

> There's no standard
> connector on the back of the board so if it's there, it's a ribbon
> connector on the board. No, the external drive is IDE in a mobile rack
> connected to the second IDE port on the motherboard. There's a SCSI
> driver built into the operating system since at one point they
> contemplated offering an external CD-R drive (most were SCSI at the
> time) for backup connected to a card in an accessory slot on the
> motherboard but never went ahead with it. There's a 3-1/2" floppy drive
> for loading the software. To many users, their point of obsolescence is
> that they don't have a computer with a floppy drive so they can't make
> installation disks from the self-extracting installer file which is
> still available for download from the Mackie web site (bless their
> heart). I tell 'em to buy a USB floppy drive for about $20 and then
> they'll have one.

I'd also suggest making an image of the floppy so you aren't
tied to whatever OS the "self-extracting installer" expects to
run under. I.e., rely on the image format being more portable
than the executable! (My PROM programmer runs off floppies
so I go to great pains to keep the ability to burn floppies,
here. I actually still have a working 8" drive -- soft
sectored -- available :-/ )

>> What happens when/if the internal drive dies? Is there any
>> way to recover or "reinitialize" a new internal drive so
>> that any SOFTWARE on the drive is preserved?
>
> Simple. Install a new drive (sized for the BIOS, of course), put in the
> #1 installation floppy, power up, and it boots from the floppy. From
> there, you can install the software on the new drive.

So the system software is small-ish? Fits on one (or several?)
floppies?

> You need to know
> the secret to edit a file on the floppy which will enable formatting for
> the Mackie OS (which you need to do - it also installs a special boot

Why is it a "secret"? Do they not want you doing this? Or, is it
just "poorly documented" (i.e., not an intentional secret)?

> loader) but restoring the software is really a straightforward process
> and will continue to be one as long as floppy disks and drives are
> avaialble (which won't be forever).

> The disks are readable from a standard computer, so if, for example, the
> boot sector gets corrupted and the only problem is that it won't boot
> from the internal drive, you can pull it, connect it to another computer
> (I use one of those IDE-USB adapters) and usually recover project data.

So they probably use a FAT32 filesystem (more portable than NTFS).
Similar to many USB thumb/kangaroo drives

>> E.g., I have several NAS (Network Attached Storage -- basically
>> a disk drive with an ethernet connection) drives here that have
>> their software (firmware) residing *on* the disk drive itself.
>> So, when the drive dies, the software is gone.
>
> See, if they were as far sighted as Mackie they'd provide a way to
> reload the drives. ;)

I don't think they want you mucking with them. Much of their
"value added" lies in being able to sell you different size
models.

The attitude towards this seems to vary between manufacturers.
I.e., some you just install drives, connect to the box over
a web interface and click on "format drive(s)". Others require
you to boot the box into a recovery mode and make services
(BOOTP/TFTP/DHCP) available on a "local machine" through which
the box can fetch it's firmware and initialize itself. Others
expect you to use a RAID 1 configuration (mirroring) so a failure
in the "system" drive can be recovered by using the mirrored
copy (you then have to replace the failed drive and let the
box recreate the mirror-of-the-mirror to return to the
redundant configuration).

Personally, I find hindering the user to be counterproductive.
Design so the user *can* do these things. Do you want to be in
the "service" business or in the "manufacturing/sales" business??
Do you want to be selling finished products or (overpriced)
spare parts? etc.

[Neil Gould]

Les Cargill wrote:
> Neil Gould wrote:
>> Les Cargill wrote:
>>> Neil Gould wrote:
>>>> my point was that the demands of a word processor are not remotely
>>>> similar to a real-time processing app, such as is typical of audio.
>>>> Surely, we don't disagree about that?
>>>>
>>>
>>> No, not at all. Although I don't think harddisk performance has been
>>> a consideration for a very long time.
>>>
>> Perhaps not for two-track recording, but those pushing their systems
>> to record 16+ simultaneous tracks have always had to take HD
>> performance into consideration, and that hasn't changed because some
>> of those folks went from
>> 44.1/16 to 192/xxx. Performance matters when you have to flush that
>> much data reliably.
>>
>
> 16 tracks of 192/24 is < 80 kBit/sec. Not very demanding.
>

You may want to review that math... ;-)

>>>> "Wrong" is hard to claim when there wasn't a common standard, as
>>>> there is today. Also, one real need was for backward compatibility.
>>>> So, there were contending solutions, even from a single
>>>> manufacturer.
>>>>
>>>
>>> I am desperately trying to remember the last time I had to load a
>>> driver for a disk drive, and it's probably back when you had to do
>>> things like HIMEM and such, when things tended to be a lot Real
>>> Mode.
>>>
>> Whether one has to load a driver for a disk drive depends on whether
>> the generic driver supplied with the OS works witht the particular
>> HD that you bought. That is still true today, which is probably why
>> drivers are available for, if not sold with HDs on the market.
>>
>
>
> So do it like Linux does it. No drivers required... I realize Winboxen
> had legacy issues, but a standalone doesn't.
>

Linux??? There's a significant difference between "no drivers required" and
"no drivers available"... <rofl>

--
best regards,

Neil

[Don Y]

Hi Les,

On 6/1/2012 10:42 AM, Les Cargill wrote:

[attributions elided]

>>>> my point was that the demands of a word processor are not remotely
>>>> similar to a real-time processing app, such as is typical of audio.
>>>> Surely, we don't disagree about that?
>>>
>>> No, not at all. Although I don't think harddisk performance has been
>>> a consideration for a very long time.
>>>
>> Perhaps not for two-track recording, but those pushing their systems to
>> record 16+ simultaneous tracks have always had to take HD performance
>> into
>> consideration, and that hasn't changed because some of those folks
>> went from
>> 44.1/16 to 192/xxx. Performance matters when you have to flush that much
>> data reliably.

16 tracks * ~50K sample/second * 3 bytes/sample ~= 2.5MB/s
This is well within the capabilities of low end drives in the
1995 era with internal transfer rates in the 5MB/s ballpark.
(Note that you are concerned witht he *internal* rate and not
the "interface rate" as the latter can be substantially
higher -- though also has to include some tiny command
overhead).

The hard disk isn't the issue at those data rates. Rather, it
is a function of the GUARANTEES that your RTOS provides (if
you are using an OS that does not provide real-time guarantees
then 8KHz/8b audio could be an issue! :> ).

> 16 tracks of 192/24 is < 80 kBit/sec. Not very demanding.

Recheck your math, Les :> Closer to 8M *Byte* than 80K *bit*
(an annoying problem with slide rules... counting zeroes! :> )

>>>> "Wrong" is hard to claim when there wasn't a common standard, as
>>>> there is today. Also, one real need was for backward compatibility.
>>>> So, there were contending solutions, even from a single manufacturer.
>>>
>>> I am desperately trying to remember the last time I had to load a
>>> driver for a disk drive, and it's probably back when you had to do
>>> things like HIMEM and such, when things tended to be a lot Real Mode.
>>>
>> Whether one has to load a driver for a disk drive depends on whether the
>> generic driver supplied with the OS works witht the particular HD that
>> you
>> bought. That is still true today, which is probably why drivers are
>> available for, if not sold with HDs on the market.
>
> So do it like Linux does it. No drivers required... I realize Winboxen
> had legacy issues, but a standalone doesn't.

If you are designing an *appliance*, you can do whatever
makes sense for that appliance! And, can exploit patterns
in the operation of that appliance to effectively increase
performance.

E.g., you could losslessly encode data before writing it to
disk -- as long as you losslessly *decode* that same data before
replay (or transfer to other media). A factor of 2 is pretty
easy to attain -- especially with "music" (OTOH, if you were
recording 16 arbitrary waveforms, that sort of compression
might not be feasible).

You can write directly to the "raw" device instead of letting
an operating system / file system impose its overhead on your
transactions. (Indeed, a common hack is to write the data
to the raw device and build the file structure *afterwards*).

But, you have to understand the technology that you are using
and not just treat it as a "black box". E.g., a hard disk
performs periodic (thermal) recalibration cycles (AV drives
do this much less often to achieve their higher performance
figures). How often does *this* drive perform them? Are
they always the same duration -- regardless of which cylinder
might be accessed at the time the cycle is initiated? How
does the drive handle miswrites? How long will a retry
take before the drive reports it as an error? Will your
OS "silently" try to repeat the operation? How many times?
How often (statistically) does this happen? What happens to
the data queuing up behind this fault? When will that queue
overflow? What if the nonzero probability of an unrecoverable
error actually occurs -- will you simply "halt"? Will you
somehow mark the data stream "an error occurred HERE"?

What sort of guarantees do you convey to your *user*? How
can you come to those conclusions without fully understanding
the technology on which your product ultimately depends??

Or, do you just try to empirically derive these figures and
"hope" the device always operates "nominally"?

[Don Y]

Hi Mike,

On 5/31/2012 1:05 PM, Mike Rivers wrote:
> On 5/31/2012 2:20 PM, Don Y wrote:
>
>> Exactly. *If* the old product(s) and the new product(s) don't
>> share a common implementation base -- because someone opted to
>> cut a corner in "old product" thereby making that aspect of
>> the design non-portable to the "new product".
>
> If you're going to design products for the mass market, in order to make
> them price-competitive you have to adopt someone else's technology.

Actually, that isn't true. It depends a lot on what your product
wants to be and the "commodity" components on which you can draw.

E.g., if I had opted to build my network speakers using "someone
else's technology", my *cost* would be close to $100/device.
I'd need a microcomputer/microcontroller with enough memory to
store its program (plus a big enough buffer to hold the streaming
audio data), a network interface and two D/AC's. They would
have to be connected to a two channel audio amplifier capable of
digital control (volume, etc.). And, that amplifier would have to
be very efficient to make the best use of the limited power
available from a PoE distribution system -- which would have
to be accommodated by appropriate hardware in that package, as
well.

And, that assumes I care nothing about the physical form that the
final device assumes (hint: that sort of approach wouldn't yield
the "ice cubes" that my current design fits).

OTOH, building the proof-of-principle prototypes for these devices
was easy to do with off-the-shelf components! A pair of PC's
with sound cards and network cards running software that I
spoon fed to them over the network. Too big and too expensive
but provides the same "functionality"!

This is a common misconception that folks building custom kit
get distracted by. "Let's just use a PC!" As if that, magically,
solves their problems!

I've long ago stopped arguing with these folks. Instead, I "agree"
and go through the list of what the project will entail.

- "OK, so how are we going to interface to the motors? The PC has
no concept of 'motors'..."
- "We'll buy a motor driver board".
- "How many? We have four motors to control?"
- "Four boards?"
- "So we need at least 4 slots for these four boards..."
- "No, maybe we can buy a four-motor board!"
- "OK, and how do we interface to the ultrasonic range detectors?"
- "We'll use a sound card(s)!"
- "No, doesn't have an appropriate frequency response. And, we
don't have the computational horsepower to be analyzing echoes
in real time while doing these other things"
- "OK, so we'll have to design a board to do that."
- "With a separate processor? That communicates with the processor
in the PC? How?"

Etc. When you're all done, you pick up one of the many catalogs from
board vendors and start filling in some prices.

Then, you look at the designs you *still* have to do "by yourself"
(or subcontract).

Finally, you look at how "exposed" you now are to all these
other providers. What happens if the motor board manufacturer
goes out of business? Or, discontinues that model? Or, makes
a change to the implementation that somehow doesn't work with
*your* application?

[I know a company that regularly scours eBay and ham fests for
old kit on which they have based their products -- kit that is
no longer available in regular channels! <shudder>]

> Mackie couldn't make the MDR24/96 and sell it for $2,000 if they had to
> design their own motherboard rather than using an Intel style one. They
> chose well, I thought, buying an industrial grade board for about $400
> rather than an Asus or whatever was around at the time for $100, and
> failures have been very rare, and replacements were available for a long
> time. But still, that board was based on standard technology of the day.
> It was only better than a run-of-the-mill one because they tested better
> and used better components.

Again, it depends on what they needed from that board/design.
My network speakers don't need serial or parallel ports, nor
a display, keyboard, mouse. Nor a disk drive. Nor a PCI
expansion bus. Nor megabytes of RAM, etc.

What often happens is people "pick" something and then try
to rationalize additional uses that they might not have
originally intended: "We can hook up a printer and print song
lists!" "We can interface to a MIDI device!" "We can transfer
files over the network!" All might be desirable features
but you are paying for them even if you don't use them (and
paying MORE, if you *do* -- as they need to be implemented
and maintained).

> Contrast this with a commercial computer that might be the heart of a
> studio's recorder and mixer. You can't buy a Dell one month and another
> one a couple of months later and expect them to have the same
> motherboard components. Sure, both of them will run Windows, but they
> may have different Firewire chips (or no Firewire chip at all) and your
> Firewire audio interface won't work the newer one.

You face this every time you bring someone else's technology
into your product. They could use the *same* Firewire chips
only to discover that the chip manufacturer updated their
masks and the chips now behave slightly differently! (that's
one reason chips have date codes)

> Our $100 Zoom recorders are practical because the photography business
> buys enough flash memory cards that they're dirt cheap. Our hard disk
> recorders are practical because for many years, IDE drives could be
> bought for well under $100 because there were so many general purpose
> computers that used them.
>
> The RADAR 24 track hard disk recorder runs under Linux and prices start
> at about $7500. Some people are still buying them. The Alesis HD24
> (recently discontinued) used a disk format that nobody can read. You
> want WAV files? You have to go through the recorder or an accessory dock
> (discontinued way too soon). But it allowed them to make a recorder that
> used cheaper parts than the Mackie but still cost about the same.
>
> So, yeah, there are people who are being somewhat innovative, but it
> just doesn't pay in this particular market. And when something based on
> unique technology dies, you're really sunk if you bought one. But at
> least it didn't cost much, and if you were diligent about backing up and
> migrating your data, you could use the data from one system on another
> system - because the companies were forward thinking enough to
> accommodate an audio file format specified by Microsoft that will
> probably be around as long as cockroaches (but maybe not for 1,000 years).

WAV format limits the size of the "piece" you can record (about 1 hr
at 192/24/mono). Stuff 16 channels in that container and you're
down to ~5 minutes. I.e., you are forced to record multiple
channels in multiple files -- and still limited to ~1 hr.

There's absolutely nothing wrong with "special formats" -- as
long as they are documented as such! Nowadays, if you had
data on a medium in damn near *any* bizarre format, it would
be relatively trivial to convert that data to another form
AS LONG AS THE ORIGINAL FORMAT WAS DOCUMENTED. If this was
a common piece of kit, you can be *assured* someone would
have done it, already!

>> "This OS doesn't support volumes larger than XXX bytes. I guess
>> we'll have to find *another* OS for the GizmaCorder 2" (which
>> means the newer folks working on GizmaCorder 2 will tend not
>> to ever see any of the internals of GizmaCorder 1 -- at least
>> never with an eye towards supporting the G1!
>>
>> OTOH, if G2 was largely G1 with new and improved features
>> (higher sampling rates, wider samples, more channels plus
>> a new set of "effects") then the G2 designers would have
>> been intimately familiar with the G1 *as* they adapted it
>> to the G2 and could, then, support the G1 as needed. I.e.,
>> then support becomes a marketing/political issue instead of
>> a technical consequence of design!
>
> Easy to say, but it doesn't always work like that in practice. You have
> to be able to adopt to the technology that's available when you design
> the product unless you want to take responsibility for sustaining and
> extending the technology. Commercial products that are based on Linux,
> for instance the Harrison digital consoles, might be supportable for a
> long time because they control the operating system. But they may need
> to make extensive changes to it in order to support a motherboard that
> they can buy 10 years after the initial design. They're in the console
> business, not the motherboard business.

Look at their choices:
- avoid digital technology altogether (stick with an all-analog console)
- use third party technology (and be beholding to those third parties)
- develop your own technology (gives you free reign over its direction)
- embrace some other "public" technology and develop the skills
required to support it (in the event it fizzles out, heads off in
a different direction, or doesn't attend to your particular needs)

I have first hand experience with companies that relied on 3rd party
technology and then were forced out of certain markets (which they
held controlling interests!) simply because the third party decided
not to keep *supplying* the product in question. (In one case, the
"product" was a piece of software. All the 3rd party would have
had to do was grant additional licenses -- for $$$$ -- to allow the
product to continue to be available!)

I try to take the 3rd or 4th options (above) with products. I don't
want to be stuck in the past (option 1) with my hands tied, needlessly.
Nor do I want to invest lots of time and money in a design only to
have someone else act as "gatekeeper" for my exploiting that investment
(option 2).

E.g., the network speakers straddle the option 3 and 4 boundary as
much of the CODEC design I borrowed -- along with ideas for the
peer-based recovery protocol and the time synchronization protocol.
Yet, the mechanisms that I developed to "protect" the device in
a potentially "non-benign" environment are completely unique. As
is the overall design of the "system".

[Mike Rivers]

On 6/1/2012 8:59 AM, Neil Gould wrote:

> those pushing their systems to
> record 16+ simultaneous tracks have always had to take HD performance into
> consideration, and that hasn't changed because some of those folks went from
> 44.1/16 to 192/xxx. Performance matters when you have to flush that much
> data reliably.

The Mackie HDR24/96 was originally specified with a 5400 RPM
drive and worked fine that way. Of course that was before
192 kHz sample rates, but anyone running 24 tracks at 192
kHz sample rate either isn't recording conventional audio,
is just plain nuts, has more money than brains, or has
already specified a hot-rod system just assuming that he'd
need it.

> Whether one has to load a driver for a disk drive depends on whether the
> generic driver supplied with the OS works witht the particular HD that you
> bought. That is still true today, which is probably why drivers are
> available for, if not sold with HDs on the market.

What operating system uses someone else's driver for the
hard disk? If there's anything standard in the personal
computer world, it's hard drives, and all operating systems
have a driver built in that works fine with them.

[Mike Rivers]

On 6/1/2012 3:47 PM, Don Y wrote:
Pages and pages . .

You've worn me out. I won't continue to respond to you about
what manufacturers should do. Some things offer you choices,
others don't. You can buy what you like, and refuse to buy
what you don't like. But sometimes, in order to get some
function, you may have to buy a product that isn't designed
as you would design it. Or you could continue to live
without that funciton.

[Don Y]

Hi Mike,

On 6/1/2012 3:01 PM, Mike Rivers wrote:
> On 6/1/2012 3:47 PM, Don Y wrote:
> Pages and pages . .
>
> You've worn me out.

Sorry, wasn't intended to do so. Rather, showing you what
things look like from the other side of the "development wall".

> I won't continue to respond to you about what
> manufacturers should do. Some things offer you choices, others don't.
> You can buy what you like, and refuse to buy what you don't like. But
> sometimes, in order to get some function, you may have to buy a product
> that isn't designed as you would design it. Or you could continue to
> live without that funciton.

Exactly. My goal is to encourage *suppliers* to design rationally
(i.e. *think* about what their goals actually are before embarking
on a development project) and *consumers* to think about what their
needs are likely to *be* (now and in the future).

Consumers want to be happy with their purchases (and value for
dollar) and suppliers want to be happy with their *sales*!

[Les Cargill]

Oops! 80 *mega* bit. okay, a little more of a challenge then. Peopel
really use 192? Yeesh.

16 tracks at 48k is 18 Mbit.

--
Les Cargill

[Les Cargill]

Neil Gould wrote:
> Les Cargill wrote:
>> Neil Gould wrote:
>>> Les Cargill wrote:
>>>> Neil Gould wrote:
>>>>> my point was that the demands of a word processor are not remotely
>>>>> similar to a real-time processing app, such as is typical of audio.
>>>>> Surely, we don't disagree about that?
>>>>>
>>>>
>>>> No, not at all. Although I don't think harddisk performance has been
>>>> a consideration for a very long time.
>>>>
>>> Perhaps not for two-track recording, but those pushing their systems
>>> to record 16+ simultaneous tracks have always had to take HD
>>> performance into consideration, and that hasn't changed because some
>>> of those folks went from
>>> 44.1/16 to 192/xxx. Performance matters when you have to flush that
>>> much data reliably.
>>>
>>
>> 16 tracks of 192/24 is< 80 kBit/sec. Not very demanding.
>>
> You may want to review that math... ;-)
>

No kidding! Off by three orders of magnitude. That's what i
get for not using a calculator.

>>>>> "Wrong" is hard to claim when there wasn't a common standard, as
>>>>> there is today. Also, one real need was for backward compatibility.
>>>>> So, there were contending solutions, even from a single
>>>>> manufacturer.
>>>>>
>>>>
>>>> I am desperately trying to remember the last time I had to load a
>>>> driver for a disk drive, and it's probably back when you had to do
>>>> things like HIMEM and such, when things tended to be a lot Real
>>>> Mode.
>>>>
>>> Whether one has to load a driver for a disk drive depends on whether
>>> the generic driver supplied with the OS works witht the particular
>>> HD that you bought. That is still true today, which is probably why
>>> drivers are available for, if not sold with HDs on the market.
>>>
>>
>>
>> So do it like Linux does it. No drivers required... I realize Winboxen
>> had legacy issues, but a standalone doesn't.
>>
> Linux??? There's a significant difference between "no drivers required" and
> "no drivers available"...<rofl>
>

But for drives....

--
Les Cargill

[Don Y]

Hi Mike,

On 6/1/2012 3:01 PM, Mike Rivers wrote:

> On 6/1/2012 3:47 PM, Don Y wrote:
> Pages and pages . .
>
> You've worn me out. I won't continue to respond to you about what
> manufacturers should do. Some things offer you choices, others don't.
> You can buy what you like, and refuse to buy what you don't like. But
> sometimes, in order to get some function, you may have to buy a product
> that isn't designed as you would design it. Or you could continue to
> live without that funciton.

If you've not read my post, you might still (personally) be interested
in this excerpt:

---------

[Mike Rivers]

On 6/1/2012 7:12 PM, Don Y wrote:

> I've never had any luck with KVM's -- and I've tried a LOT
> of them

> but they didn't meet my needs. Again, it had to do with the
> order that machines were powered up, etc. The box in question
> has PS2 & USB connections to the individual machines but only
> PS2 connections to the shared keyboard/mouse (see photo at
> URL).

Mine is a Linksys, just a little lump with 9 cables dangling
from it. One of the reasons why I bought it was that I
wouldn't have to find more cables. The other was that it was
on sale for $20. I've had it for about 10 years now and it
works perfectly with the present setup (which has only
changed computers once in the switch's lifetime). But when I
tried a computer that didn't have PS/2 ports, there were
some snags.

I guess these things must have some logic in them that make
the computer that's off look like it has a keyboard attached
so it won't complain when you turn it on.

[Neil Gould]

Mike Rivers wrote:
> On 6/1/2012 8:59 AM, Neil Gould wrote:
>
>> those pushing their systems to
>> record 16+ simultaneous tracks have always had to take HD
>> performance into consideration, and that hasn't changed because some
>> of those folks went from
>> 44.1/16 to 192/xxx. Performance matters when you have to flush that
>> much data reliably.
>
> The Mackie HDR24/96 was originally specified with a 5400 RPM
> drive and worked fine that way. Of course that was before
> 192 kHz sample rates, but anyone running 24 tracks at 192
> kHz sample rate either isn't recording conventional audio,
> is just plain nuts, has more money than brains, or has
> already specified a hot-rod system just assuming that he'd
> need it.
>

What... you don't know at least one of those nuts? ;-)

>> Whether one has to load a driver for a disk drive depends on whether
>> the generic driver supplied with the OS works witht the particular
>> HD that you bought. That is still true today, which is probably why
>> drivers are available for, if not sold with HDs on the market.
>
> What operating system uses someone else's driver for the
> hard disk? If there's anything standard in the personal
> computer world, it's hard drives, and all operating systems
> have a driver built in that works fine with them.
>

Ever wonder why today's OSs are so bloated that they take a GB to just boot?
One reason is that they have *lots* of drivers built in so that it "works
fine" with a lot of stuff. Do a search for hardware in the Control Panel >
System and check out the list.

--
best regards,

Neil

[david gourley]

Don Y <th...@isnotme.com> said...news:jq8fi8$75v$1...@speranza.aioe.org:
> Hi David,
-snippage of a lot of stuff, some useful

Hi Don,

Thanks for taking the time for that explanation. That was generous of you to
nake that kind of effort.

As such, I doubt I'll be using much of what I have now by 2038 for any given
reason, should I even be alive then. I'm sure updates or replacements will
be there otherwise, so no worries.

YMMV,
david

[hank alrich]

Some folks are running multiple Avid HD192 interfaces, some are running
multiple Metric Halo ULN-8's, at 24/192. It adds up. My old 2882 maxes
at 24/96, and that's where I've been running it since I got the 2d card
upgrade.

[Don Y]

Hi David,

On 6/1/2012 7:05 PM, david gourley wrote:
> Don Y<th...@isnotme.com> said...news:jq8fi8$75v$1...@speranza.aioe.org:
>> Hi David,
> -snippage of a lot of stuff, some useful
>

> Thanks for taking the time for that explanation. That was generous of you to
> nake that kind of effort.
>
> As such, I doubt I'll be using much of what I have now by 2038 for any given
> reason, should I even be alive then. I'm sure updates or replacements will
> be there otherwise, so no worries.

Wanna bet folks will still be designing with these same inherent
defects ten or twenty years hence?? :-(

[Don Y]

Hi Mike,

On 6/1/2012 5:25 PM, Mike Rivers wrote:
> On 6/1/2012 7:12 PM, Don Y wrote:
>
>> I've never had any luck with KVM's -- and I've tried a LOT
>> of them
>> but they didn't meet my needs. Again, it had to do with the
>> order that machines were powered up, etc. The box in question
>> has PS2 & USB connections to the individual machines but only
>> PS2 connections to the shared keyboard/mouse (see photo at
>> URL).
>
> Mine is a Linksys, just a little lump with 9 cables dangling from it.

I had one of those (or something similar). Indescript "blob"
with cables coming out like a "tail" (though I recall only
2 sets of K+V+M -- the "console" was just connectors on the
blob). Controlled by hot keys from the console keyboard?

> One of the reasons why I bought it was that I wouldn't have to find more
> cables.

Yup. I might discard a KVM but I'll always keep its cables! :>
(they are usually interchangeable with other KVM's except for
the odd ones)

> The other was that it was on sale for $20. I've had it for about
> 10 years now and it works perfectly with the present setup (which has
> only changed computers once in the switch's lifetime). But when I tried
> a computer that didn't have PS/2 ports, there were some snags.

IME, there are *always* snags. :< As I said, it seemed that there was
some ideal sequence in which you had to power things up or down to
avoid the things "locking up".

> I guess these things must have some logic in them that make the computer
> that's off look like it has a keyboard attached so it won't complain
> when you turn it on.

Yes. For small KVM's, they can derive power from the keyboard
and mouse ports of the machines in question. For bigger KVMs,
you need an external power supply to handle all the various
buffering and "smarts" inside. E.g., the unit I referenced had
a significant wall wart. But, it also handled the audio I/O's,
too (no "phantom power" to steal from those).

[Les Cargill]

hank alrich wrote:
> Les Cargill<lcarg...@comcast.com> wrote:
>
>> John Williamson wrote:
>>> Les Cargill wrote:
>>>
>>>>
>>>> 16 tracks of 192/24 is< 80 kBit/sec. Not very demanding.
>>>>
>>> Pardon? 16 tracks X 192,000 samples/ second x 3 (possibly 4, allowing
>>> for the fact that writing samples as four bytes is normally faster than
>>> writing three bytes per sample) bytes per sample is a *lot* more than
>>> 80kbits/ second. At least 9,216,000 *bytes*/ second. or 73,728,000 bits,
>>> plus overheads. That's all got to be got into the buffer and back out
>>> onto the platter in the right order, while simultaneously writing to at
>>> least 16 files.
>>>
>>
>>
>> Oops! 80 *mega* bit. okay, a little more of a challenge then. Peopel
>> really use 192? Yeesh.
>>
>> 16 tracks at 48k is 18 Mbit.
>
> Some folks are running multiple Avid HD192 interfaces, some are running
> multiple Metric Halo ULN-8's, at 24/192.

So I guess the higher bit rates make large number of
tracks impossible, which is what *really* makes the higher rates sound
better :)

> It adds up. My old 2882 maxes
> at 24/96, and that's where I've been running it since I got the 2d card
> upgrade.
>

--

Les Cargill

[hank alrich]

Les Cargill <lcarg...@comcast.com> wrote:

> hank alrich wrote:
> > Les Cargill<lcarg...@comcast.com> wrote:
> >
> >> John Williamson wrote:
> >>> Les Cargill wrote:
> >>>
> >>>>
> >>>> 16 tracks of 192/24 is< 80 kBit/sec. Not very demanding.
> >>>>
> >>> Pardon? 16 tracks X 192,000 samples/ second x 3 (possibly 4, allowing
> >>> for the fact that writing samples as four bytes is normally faster than
> >>> writing three bytes per sample) bytes per sample is a *lot* more than
> >>> 80kbits/ second. At least 9,216,000 *bytes*/ second. or 73,728,000 bits,
> >>> plus overheads. That's all got to be got into the buffer and back out
> >>> onto the platter in the right order, while simultaneously writing to at
> >>> least 16 files.
> >>>
> >>
> >>
> >> Oops! 80 *mega* bit. okay, a little more of a challenge then. Peopel
> >> really use 192? Yeesh.
> >>
> >> 16 tracks at 48k is 18 Mbit.
> >
> > Some folks are running multiple Avid HD192 interfaces, some are running
> > multiple Metric Halo ULN-8's, at 24/192.
>
>
> So I guess the higher bit rates make large number of
> tracks impossible, which is what *really* makes the higher rates sound
> better :)

Heh. They're collecting those tracks by the dozens per swipe, as in 24
to 32 at a time.

I have no 192 experience. I will use 96 any time I can, at least with
the kit I have, and I don't see getting a ULN-8 or LIO-8 in this
lifetime.

I'm only collecting ten tracks at a time, and rarely more than that
total now.

[Les Cargill]

hank alrich wrote:
> Les Cargill<lcarg...@comcast.com> wrote:
>
>> hank alrich wrote:
>>> Les Cargill<lcarg...@comcast.com> wrote:
>>>
>>>> John Williamson wrote:
>>>>> Les Cargill wrote:
>>>>>
>>>>>>
>>>>>> 16 tracks of 192/24 is< 80 kBit/sec. Not very demanding.
>>>>>>
>>>>> Pardon? 16 tracks X 192,000 samples/ second x 3 (possibly 4, allowing
>>>>> for the fact that writing samples as four bytes is normally faster than
>>>>> writing three bytes per sample) bytes per sample is a *lot* more than
>>>>> 80kbits/ second. At least 9,216,000 *bytes*/ second. or 73,728,000 bits,
>>>>> plus overheads. That's all got to be got into the buffer and back out
>>>>> onto the platter in the right order, while simultaneously writing to at
>>>>> least 16 files.
>>>>>
>>>>
>>>>
>>>> Oops! 80 *mega* bit. okay, a little more of a challenge then. Peopel
>>>> really use 192? Yeesh.
>>>>
>>>> 16 tracks at 48k is 18 Mbit.
>>>
>>> Some folks are running multiple Avid HD192 interfaces, some are running
>>> multiple Metric Halo ULN-8's, at 24/192.
>>
>>
>> So I guess the higher bit rates make large number of
>> tracks impossible, which is what *really* makes the higher rates sound
>> better :)
>
> Heh. They're collecting those tracks by the dozens per swipe, as in 24
> to 32 at a time.
>

That takes some careful systems engineering.

> I have no 192 experience. I will use 96 any time I can, at least with
> the kit I have, and I don't see getting a ULN-8 or LIO-8 in this
> lifetime.
>
> I'm only collecting ten tracks at a time, and rarely more than that
> total now.
>

Anything over ten is pretty much a waste.

--
Les Cargill

[Mike Rivers]

On 6/1/2012 10:05 PM, david gourley wrote:

> As such, I doubt I'll be using much of what I have now by 2038 for any given
> reason, should I even be alive then.

Ah, but if Don was in charge, you could be. ;)

[Mike Rivers]

On 6/1/2012 11:39 PM, Don Y wrote:

>> Mine is a Linksys, just a little lump with 9 cables
>> dangling from it.
>
> I had one of those (or something similar). Indescript "blob"
> with cables coming out like a "tail" (though I recall only
> 2 sets of K+V+M -- the "console" was just connectors on the
> blob). Controlled by hot keys from the console keyboard?

Yeah, I guess you're right. I guess they assume your mouse
and keyboard already have tails, and you must have had a
cable to connect your monitor, whether it was a separate
cable or captive with the monitor. There's only one hot key
- double-punching the Scroll Lock toggles between the two
sets of inputs. And, when used as intended (PS/2 to PS/2) it
always switches to whatever announces its presence first.

> Yup. I might discard a KVM but I'll always keep its cables! :>

Only worth while if they're two-ended cables. Not much point
in cutting off the cables at the "lump" end with the one I
have, unless I'm planning to build more keyboard, monitor,
and mouse cables by attaching another connector.

> IME, there are *always* snags. :< As I said, it seemed that
> there was
> some ideal sequence in which you had to power things up or
> down to
> avoid the things "locking up".

This one works perfectly in my present setup. No snags
whatever. I've never had the luck, when powering up both the
HDR and computer, to have them both send the "Hey! Pick me!"
signal to the switch simultaneously. That might confuse it.
I don't know. The only snags were when I couldn't connect a
real PS/2 device to both keyboard ports. You can make almost
anything screw up if you use it incorrectly.

My problem was simply that I was unable to find an
off-the-shelf solution. Maybe the Belkin you (I think)
cited. But it's probably obsolete now.

[david gourley]

Mike Rivers <mri...@d-and-d.com> said...news:jqcuti$rqi$1...@dont-email.me:

> On 6/1/2012 10:05 PM, david gourley wrote:
>
>> As such, I doubt I'll be using much of what I have now by 2038 for any
given
>> reason, should I even be alive then.
>
> Ah, but if Don was in charge, you could be. ;)
>
>

Yes, just my luck.

david

[david gourley]

Don Y <th...@isnotme.com> said...news:jqc0jv$vfs$1...@speranza.aioe.org:

Nope.

No doubt 'they' will find new, higher-tech (or not) mistakes to build by
then, but I couldn't care less at this point.

david

[hank alrich]

Easy to get to that though, when playing live as a duo.

Voc 1
Voc 2
Red Eye 1
Red Eye 2
Schoeps 1
Schoeps 2
Board Mix
Ambient pair

[Don Y]

Hi Mike,

On 6/2/2012 4:54 AM, Mike Rivers wrote:
> On 6/1/2012 10:05 PM, david gourley wrote:
>
>> As such, I doubt I'll be using much of what I have now by 2038 for any
>> given
>> reason, should I even be alive then.
>
> Ah, but if Don was in charge, you could be. ;)

Naw, but I'd make _Final Exit_ mandatory reading! :> (but that's a
whole 'nother issue...)

[Don Y]

Hi Mike,

On 6/2/2012 5:04 AM, Mike Rivers wrote:

>> Yup. I might discard a KVM but I'll always keep its cables! :>
>
> Only worth while if they're two-ended cables.

Yes. Unfortunately, they're always male-to-male (whereas
male-to-female would have use as "extension cords"). But, keeping
old cables at least gives me a choice between different lengths!

> Not much point in cutting
> off the cables at the "lump" end with the one I have, unless I'm
> planning to build more keyboard, monitor, and mouse cables by attaching
> another connector.

I *really* dislike having to make (or modify) cables. Hence the
reason for squirreling away as many as I do. Too expensive to
buy new, too tedious to build from scratch. Stuffing them in
a box (actually, a collection of boxes) in the garage is "just
right"! :>

>> IME, there are *always* snags. :< As I said, it seemed that
>> there was
>> some ideal sequence in which you had to power things up or
>> down to
>> avoid the things "locking up".
>
> This one works perfectly in my present setup. No snags whatever. I've
> never had the luck, when powering up both the HDR and computer, to have
> them both send the "Hey! Pick me!" signal to the switch simultaneously.
> That might confuse it. I don't know. The only snags were when I couldn't
> connect a real PS/2 device to both keyboard ports. You can make almost
> anything screw up if you use it incorrectly.
>
> My problem was simply that I was unable to find an off-the-shelf
> solution. Maybe the Belkin you (I think) cited. But it's probably
> obsolete now.

Appears to still be available (at a variety of prices :> )
Google "omniview soho kvm".

I think they also make a newer version of the same thing (but
not as daffy looking!)

[Mike Rivers]

On 6/2/2012 9:53 AM, Jeff Henig wrote:

> I knew some guys in an a cappella boy band-type group about 15 years ago
> who had to bounce with a 24 track rtr.
>
> They were doubling each part and had a BUNCH of parts. Very intricate
> arrangements--basically imitating the entire chord structures of each
> instrument in a rock band.

Sounds like they needed more singers. But I guess that's
kind of like saying that Les Paul needed more guitar players.

[Scott Dorsey]

Mike Rivers <mri...@d-and-d.com> wrote:
>
>I guess these things must have some logic in them that make
>the computer that's off look like it has a keyboard attached
>so it won't complain when you turn it on.

Yes. And if you don't like the existing units you have tried, take a look
at the Gefen ones. They're still RoHS but they are built a lot more solidly
and seem better-debugged than the run of the mill junk.
--scott


--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Arny Krueger]

"John Williamson" <johnwil...@btinternet.com> wrote in message
news:a2sk4v...@mid.individual.net...

> Les Cargill wrote:
>
>>
>> 16 tracks of 192/24 is < 80 kBit/sec. Not very demanding.
>>
> Pardon? 16 tracks X 192,000 samples/ second x 3 (possibly 4, allowing for
> the fact that writing samples as four bytes is normally faster than
> writing three bytes per sample) bytes per sample is a *lot* more than
> 80kbits/ second. At least 9,216,000 *bytes*/ second. or 73,728,000 bits,
> plus overheads. That's all got to be got into the buffer and back out onto
> the platter in the right order, while simultaneously writing to at least
> 16 files.

If the VF16 is looking for an IDE or PATA drive, just plug in a SATA SSD + a
PATA-> SATA converter. If it swings with SATA, leave the converter out.

[Arny Krueger]

"Neil Gould" <ne...@myplaceofwork.com> wrote in message
news:jqbpkf$iv8$1...@speranza.aioe.org...

> Ever wonder why today's OSs are so bloated that they take a GB to just
> boot?
> One reason is that they have *lots* of drivers built in so that it "works
> fine" with a lot of stuff. Do a search for hardware in the Control Panel >
> System and check out the list.

The Apple OSs solve that problem the old fashioned way - the owner totally
controls the hardware configs it supports and brings them down to just about
nothing, in comparison with say Win 7.

[Neil Gould]

That's a perpetuation of the design philosophy that Apple started with the
Mac (they must have gotten their fill of support issues with the Apple ][
e). You can keep the OS footprint small if you don't support much. ;-)

--
best regards,

Neil

[Don Y]

Hi Neil,

On 6/1/2012 6:15 PM, Neil Gould wrote:

> Mike Rivers wrote:

>>> Whether one has to load a driver for a disk drive depends on whether
>>> the generic driver supplied with the OS works witht the particular
>>> HD that you bought. That is still true today, which is probably why
>>> drivers are available for, if not sold with HDs on the market.
>>
>> What operating system uses someone else's driver for the
>> hard disk? If there's anything standard in the personal
>> computer world, it's hard drives, and all operating systems
>> have a driver built in that works fine with them.
>
> Ever wonder why today's OSs are so bloated that they take a GB to just boot?

I think that's only true of MS's products. Much of the bloat in
modern OS's comes from the presence of extra services that aren't
really essential for the OS's operation. Or, that could be moved
to "user land" (out of the kernel proper) if the OS had been designed
more intelligently.

> One reason is that they have *lots* of drivers built in so that it "works
> fine" with a lot of stuff. Do a search for hardware in the Control Panel>
> System and check out the list.

I am actually surprised at how many different hardware configurations
are supported with as *few* drivers as might otherwise be suggested.
I guess it testifies to how *little* variation there is in hardware
designs, these days. And, how much "compatibility" appears to be
valued by manufacturers (even at the chip level)...

Each time I build a custom kernel, I'm amused by how few devices
actually need to change from machine to machine (even though I
strip out support for devices that are not present in a given
machine).

[William Sommerwerck]

>> Ever wonder why today's OSs are so bloated that they take
>> a GB to just boot?
>> One reason is that they have *lots* of drivers built in so that
>> it "works fine" with a lot of stuff. Do a search for hardware in
>> the Control Panel > System and check out the list.

> The Apple OSs solve that problem the old fashioned way -- the

> owner totally controls the hardware configs it supports and brings
> them down to just about nothing, in comparison with say Win 7.

The average Apple owner is incapable of performing such an analysis and
pruning.

Several weeks ago I spent quite some time going through the Device Manager
and removing unneeded drivers. There weren't many, because as far as I can
tell, Windows doesn't load what it doesn't need.

For example, if you remove a hard drive you no longer use, a grayed-out
reference to it remains, but the driver isn't installed unless the drive is
in place during boot-up.

This is particularly true for USB devices, whose drivers are loaded
on-the-fly.

When you have a few minutes, browse the Device Manager and note how many
drivers are needed for the computer's basic hardware. I doubt Macintosh
machines -- or any other personal computer of any degree of
sophistication -- is much different.

[Neil Gould]

Hi Don,

Don Y wrote:
> Hi Neil,
>
> On 6/1/2012 6:15 PM, Neil Gould wrote:
>> Mike Rivers wrote:
>
>>>> Whether one has to load a driver for a disk drive depends on
>>>> whether the generic driver supplied with the OS works witht the
>>>> particular HD that you bought. That is still true today, which is
>>>> probably why drivers are available for, if not sold with HDs on
>>>> the market.
>>>
>>> What operating system uses someone else's driver for the
>>> hard disk? If there's anything standard in the personal
>>> computer world, it's hard drives, and all operating systems
>>> have a driver built in that works fine with them.
>>
>> Ever wonder why today's OSs are so bloated that they take a GB to
>> just boot?
>
> I think that's only true of MS's products.
>

Oh?
http://www.mathworks.com/support/sysreq/current_release/linux.html

> Much of the bloat in
> modern OS's comes from the presence of extra services that aren't
> really essential for the OS's operation. Or, that could be moved
> to "user land" (out of the kernel proper) if the OS had been designed
> more intelligently.
>

I think it's a choice, made to make an OS more amenable to novice users.

>> One reason is that they have *lots* of drivers built in so that it
>> "works fine" with a lot of stuff. Do a search for hardware in the
>> Control Panel> System and check out the list.
>
> I am actually surprised at how many different hardware configurations
> are supported with as *few* drivers as might otherwise be suggested.
>

Perhaps we're talking about different "drivers", then. I'm talking about
device-specific drivers, not generic interface "drivers" such as those used
for the on-board USB and video chips. There may be one SATA interface driver
for the on-board chipset, but every SATA HD will have its own drivers.
Without both, you get nothin' ... ;-)

> Each time I build a custom kernel, I'm amused by how few devices
> actually need to change from machine to machine (even though I
> strip out support for devices that are not present in a given
> machine).
>

Aha... I think you are referring to the on-board chipset drivers, right?

--
best regards,

Neil

[Mike Rivers]

On 6/5/2012 10:08 AM, Scott Dorsey wrote:

> And if you don't like the existing units you have tried, take a look
> at the Gefen ones.

When I couldn't find a KVM switch with both USB and PS/2
ports at the usual places of ill repute, I asked Gefen at a
trade show if they made something like that and the guy
scratched his head, asked why I wanted one (I guess thinking
that he could talk me into something else) and finally
admitted that they didn't have one and didn't see any reason
to have ever made one.

[William Sommerwerck]

> When I couldn't find a KVM switch with both USB and PS/2
> ports at the usual places of ill repute, I asked Gefen at a
> trade show if they made something like that and the guy
> scratched his head, asked why I wanted one (I guess
> thinking that he could talk me into something else) and
> finally admitted that they didn't have one and didn't see
> any reason to have ever made one.

This switcher supports both USB and PS/2, though perhaps not in the way you
want.

http://www.trendnet.com/products/proddetail.asp?prod=195_TK-1603R&cat=108


The ATEN UC-100KMa converts a PS/2 keyboard and a PS/2 mouse to a single USB
port. I own one, and it works beautifully.

http://www.amazon.com/ATEN-Keyboard-Mouse-Converter-UC100KMA/dp/B00065XUQA

[Mike Rivers]

On 6/5/2012 5:44 PM, William Sommerwerck wrote:

> The ATEN UC-100KMa converts a PS/2 keyboard and a PS/2 mouse to a single USB
> port. I own one, and it works beautifully.

I have one of those that I use on a laptop computer when
it's serving desktop duties (which is is, most of the time
now). Doubt if it's the same brand. I've had it for years
and years.