VRPC and Win/Tel hardware

[Brian Bailey]

Hi All

I have a Win/Tel (actually it's Xp OS on AMD hardware) machine that is
used principally for running VRPC. It is also used for photographic image
manipulation, with Adobe Lightroom, and media play back with VLC. It is
networked to a RISC OS machine, a NAS drive and a router.

It is showing signs of stress, but as it is custom built so I really don't
want to replace it in its entirety. I propose to rebuild it.

I am advised that an Intel processor would be preferable (possibly an i3)
and that for several reasons Windows 7 Pro is now also preferable.

Trouble is, there are just so many options and this task is outside my
current experience and knowledge base regarding what is suitable and
currently available. The ASUS motherboard is unreliable and I don't want
to use ASUS ever again. It has been suggested that I use a Mil grade MSI
or Gigabyte motherboard as a replacement and an Intel processor.

I don't want to cheese pare but a sensible budget is preferable, that
still leaves head room with the new setup.

I should be glad to hear of any comments, suggestions and guidance,
please. But, more especially any caveats.

TIA

Regards

Brian

[Brian Bailey]

[Rick Murray]

On 12/01/2012 09:55, Brian Bailey wrote:

> It is showing signs of stress,

? What, it goes "grrrr" and red-faced when you ask it to do something?

> I propose to rebuild it.

Price that, then price getting a similar spec machine pre-built. Some of
the deals around can be quite nice. Okay, you get ultimate flexibility
with a self-build, but then you have the problem of "does X work well
with Y"?

> and that for several reasons Windows 7 Pro is now also preferable.

Until you discover how much that costs, and then you'll be looking to
see if there are alternatives. There are. :-)

> current experience and knowledge base regarding what is suitable and
> currently available.

I think the main criteria for a motherboard is the FSB rate. If you bolt
a fast processor to a piss-poor motherboard, it isn't going to fly.

I'll give you a real life example. I have a laptop (dead display, used
to run it off an SVGA screen) with a 466MHz Celeron and a 66MHz FSB. It
was designed for Windows 98, so we aren't talking modern.
Of the same era, I have a self-assembled box with a Pentium II clocking
450MHz and a 100MHz FSB.

The *slower* processor reliably benchmarks at 45-55% *FASTER* than the
laptop. You can say the Celeron is a mickey-mouse version of a Pentium,
and you can say knock off a third for the memory bus lethargy, but no,
this machine is, genuinely, twice as fast. I have this in a real-world
application two, where ripping DVDs to XviD (mostly to get around
Macrovision copy-protection screwing up my telly with flickering white
bars and crap) the Acer laptop would manage 3fps and the box would chug
along at 5-6fps. Now I use a 1.1GHz AMD box which manages 8fps (the
apparent slowness is because I can set better quality options and just
leave it running overnight).

So I'd say the speed at which the motherboard will let the processor
communicate with memory is surely The Single Most Important Thing.


Secondly, ensure there aren't odd cost-cutting quirks. My Presario could
run a PCI network card, or it could run my PCI TV capture card. It could
not do both. It seems that the entire PCI setup (it was shared PCI and
(E)ISA) ran with one interrupt, so anything that made heavy use of
interrupts (as TV capture and ethernet both would) would co-exist with
*nothing*. Useful.

> The ASUS motherboard is unreliable

How? Have you looked on-line to see if this is something that can be
fixed? I once came across a machine that stalled for around *five*
minutes on boot-up. I rummaged around the CMOS options (noting we'd not
made it as far as Windows yet) and found an option for boot device
order. The order was:
network
removable disk (usb)
removable disk (usb)
floppy disk
cd-rom
harddisc
I altered that to be simply:
cd-rom
harddisc
and then the boot progressed as expected after a quick check of the DVD
drive.

> It has been suggested that I use a Mil grade MSI

Mil grade? You mean military grade?!?

> I don't want to cheese pare but a sensible budget is preferable, that
> still leaves head room with the new setup.

A legit copy of Windows 7 pro will hit you for something in the order of
£156 (Amazon.co.uk price), plus the motherboard (around £100), plus
£80-£130 for a processor...
...then there's memory, PSU, a box to put it in... You may or may not be
able to recycle older parts.

> But, more especially any caveats.

Once you've found bits you think you'll be happy with, it is time to
Google. Type in stuff like:
gigabyte Z68MA-D2H-B3 faults
and:
gigabyte Z68MA-D2H-B3 problems

and see if anything interesting turns up. Mostly it seems to be
overclockers that screw up, and a guy trying to boot his machine on a
way too low spec PSU (needs to be 550W+ for that board).


Further questions are perhaps best directed to a more appropriate
newsgroup. I notice nobody has replied otherwise. This might be because
a fair few people around here don't know the latest x86 processors,
which ones are 'best', and what socket types they go into. Indeed, for
me, my main criteria would be "can it play HD H.264?". If it can, then
it's good enough. ;-) I enjoy my animé, you see. My eeePC 901 is plenty
quick enough for anything else (it can emulate a RiscPC faster than any
RiscPC ever built, OvationPro works a treat, I have email and web
and...). So seriously, do I actually have a need for a quad core
ass-kicking bit of kit? I think weighing up extreme speed versus
five-six hour battery life, it's going to be the latter that wins. You
may disagree, though I do urge you to sit down and consider exactly what
you are looking for and what you expect. If you don't need bleeding edge
capabilities, you will probably find some good deals on obsolete (as in,
six month old!) hardware. ;-)

My mobile phone (Motorola Defy) that cost me €50 when I bought it (on
contract)? It was €1 a few months back, now you have to look around for
it. I hope to pick up a Galaxy Ace S2 (or whatever it is everybody is
currently drooling over) in a year or two. It's €250 on (small) contract
right now, ought to be a much nicer price when the next batch of
sexy-shiny drops into the shops. Would I like that nifty phone? Yes. Do
I need that nifty phone? No. It's the same sort of deal with PCs too.
Don't be afraid to look at slightly older kit, and to be realistic about
what you really truly need.


Best wishes,

Rick.

[David Holden]

On 12-Jan-2012, Brian Bailey <bba...@argonet.co.uk> wrote:

> I am advised that an Intel processor would be preferable (possibly an i3)
> and that for several reasons Windows 7 Pro is now also preferable.

If you're intending to use VRPC then Intel processors are much to be
preferred. They run VRPC significantly faster than equivalent AMD types.

As for the OS, I believe 7 is better, but I just don' like it and am
sticking with XP Pro for now. Probably because XP does everything I need and
I've got used to it and see no reason to learn a whole new lot of stuff to
do exactly the same things.

--
David Holden - APDL - <http://www.apdl.co.uk>

[Dave Symes]

In article <9nff5f...@mid.individual.net>,

Yes indeedy...

It might also be worth mentioning at this point...

While some folks seem to have VRPC running okay on Win 7 natively, so to
speak... Other of us have encountered a problem or two that required
measures to sort it.

1) Switch off UAC (Not actually a problem for me as that Sh one tee is the
first thing to be offed on an install)

1) Run in compatibility mode for XP SP3

That sorted my VRPC and Win 7 Problems.

Dave

--

Dave Triffid

[M Harding]

In article <5251dee...@triffid.co.uk>,

Be careful though. W7 32-bit version seems to be rare nowadays and is
limited to just over 3 GB of memory (which is probably enough anyway).
But the now-normal W7 64-bit version of Home Premium, although it can
use more than 3GB of RAM, has problems about Compatability Mode to
emulate XP.

My 64-bit version is perfectly happy with SA-VRPC.

I still haven't (as yet) managed to get it to talk to my Epson RX620
printer's scanner. Nor does (?did) Nero work, though I've found a
clone that does. It's the W7 Pro version for which you get all the
pluses - but at great cost. But except for the Compatability Mode
issue I'm as happy with W7 as with XP/3.

Michael Harding
Rev. Preb. M.D. Harding ris...@mdharding.org.uk

[Alan Adams]

In message <5251ea1e...@mdharding.org.uk>

M Harding <ris...@mdharding.org.uk> wrote:

> In article <5251dee...@triffid.co.uk>,
> Dave Symes <da...@triffid.co.uk> wrote:
>> In article <9nff5f...@mid.individual.net>,
>> David Holden <Spa...@apdl.co.uk> wrote:

>>> On 12-Jan-2012, Brian Bailey <bba...@argonet.co.uk> wrote:

<snip>

> Be careful though. W7 32-bit version seems to be rare nowadays and is
> limited to just over 3 GB of memory (which is probably enough anyway).
> But the now-normal W7 64-bit version of Home Premium, although it can
> use more than 3GB of RAM, has problems about Compatability Mode to
> emulate XP.

That's a good point. There are issues too with printer drivers for
W7-65 - HP for example point you to the "Universal driver" for most of
their older printers, and it can be a case of "one size fits none".

Generally though I'd recommend looking at, for example, Misco.co.uk,
and make sure you select W7 Pro and pick a good deal. You will get a
reasonable machine including W7 pro that way for about the same money
as a mainboard and windows license. You stll have to buy a processor,
PSU, maybe a case etc... You will get a year's warranty too.

<snip>

> Michael Harding
> Rev. Preb. M.D. Harding ris...@mdharding.org.uk

--
Alan Adams, from Northamptonshire
al...@adamshome.org.uk
http://www.nckc.org.uk/

[Brian Bailey]

Hi Rick

> > It is showing signs of stress,

> ? What, it goes "grrrr" and red-faced when you ask it to do something?

Not quite, more silent stubborn recalcitrance, perhaps 'dumb insolence',
kind of thing, with occasional blue-face and a tendency to re-boot without
being asked.

> > I propose to rebuild it.

> Price that, then price getting a similar spec machine pre-built. Some of
> the deals around can be quite nice. Okay, you get ultimate flexibility
> with a self-build, but then you have the problem of "does X work well
> with Y"?

Too right!

> > and that for several reasons Windows 7 Pro is now also preferable.

> Until you discover how much that costs, and then you'll be looking to
> see if there are alternatives. There are. :-)

Agreed. 8-)

[snip very useful information]

> > The ASUS motherboard is unreliable

> How?

After the third ASUS board in the same machine I've had enough. It's very
picky over RAM and won't accept more than 2G - heavens knows why.

> Have you looked on-line to see if this is something that can be
> fixed? I once came across a machine that stalled for around *five*
> minutes on boot-up. I rummaged around the CMOS options (noting we'd not
> made it as far as Windows yet) and found an option for boot device
> order. The order was:
> network
> removable disk (usb)
> removable disk (usb)
> floppy disk
> cd-rom
> harddisc
> I altered that to be simply:
> cd-rom
> harddisc
> and then the boot progressed as expected after a quick check of the DVD
> drive.

Thanks for that.

> > It has been suggested that I use a Mil grade MSI

> Mil grade? You mean military grade?!?

Yes, Military grade. My fingers got tired after Mil ....

> > I don't want to cheese pare but a sensible budget is preferable, that
> > still leaves head room with the new setup.

> A legit copy of Windows 7 pro will hit you for something in the order of
> £156 (Amazon.co.uk price),

OEM version much cheaper.

> plus the motherboard (around £100), plus
> £80-£130 for a processor...
> ...then there's memory, PSU, a box to put it in... You may or may not be
> able to recycle older parts.

The box and hard drives are fine, plus sound card. Will replace graphics
card.

> > But, more especially any caveats.

> Once you've found bits you think you'll be happy with, it is time to
> Google. Type in stuff like:
> gigabyte Z68MA-D2H-B3 faults
> and:
> gigabyte Z68MA-D2H-B3 problems

> and see if anything interesting turns up. Mostly it seems to be
> overclockers that screw up, and a guy trying to boot his machine on a
> way too low spec PSU (needs to be 550W+ for that board).

Useful info. But I'm not into overclocking.

> Further questions are perhaps best directed to a more appropriate
> newsgroup.

Agreed. Starting here seemed to be the right place because of emphasis on
RISC OS - VRPC.

> I notice nobody has replied otherwise. This might be because
> a fair few people around here don't know the latest x86 processors,
> which ones are 'best', and what socket types they go into. Indeed, for
> me, my main criteria would be "can it play HD H.264?".

Games?? Don't recognise that. Not into games.

> If it can, then
> it's good enough. ;-) I enjoy my animé, you see. My eeePC 901 is plenty
> quick enough for anything else (it can emulate a RiscPC faster than any
> RiscPC ever built, OvationPro works a treat, I have email and web
> and...). So seriously, do I actually have a need for a quad core
> ass-kicking bit of kit? I think weighing up extreme speed versus
> five-six hour battery life, it's going to be the latter that wins. You
> may disagree, though I do urge you to sit down and consider exactly what
> you are looking for and what you expect. If you don't need bleeding edge
> capabilities, you will probably find some good deals on obsolete (as in,
> six month old!) hardware. ;-)

Interesting!!

> My mobile phone (Motorola Defy) that cost me €50 when I bought it (on
> contract)? It was €1 a few months back, now you have to look around for
> it. I hope to pick up a Galaxy Ace S2 (or whatever it is everybody is
> currently drooling over) in a year or two. It's €250 on (small) contract
> right now, ought to be a much nicer price when the next batch of
> sexy-shiny drops into the shops. Would I like that nifty phone? Yes. Do
> I need that nifty phone? No. It's the same sort of deal with PCs too.
> Don't be afraid to look at slightly older kit, and to be realistic about
> what you really truly need.

What ever turns you on, Rick. 8-)

My mobile was given to me and is only used for emergencies. First used in
mid-Wales - nothing happened - no signal.

Thanks for the help.

Take care.

Regards

Brian

[Rick Murray]

On 15/01/2012 14:29, Brian Bailey wrote:

> After the third ASUS board in the same machine I've had enough. It's very
> picky over RAM and won't accept more than 2G - heavens knows why.

Some memory controllers simply *cannot* address more. I think my eeePC
may be likewise limited (but I only have 1Gb fitted so it's moot...).

See http://ark.intel.com/products/chipsets/26558 (etc) and note the "Max
Memory" entry.

>> Mil grade? You mean military grade?!?
> Yes, Military grade. My fingers got tired after Mil ....

If the price compares with a standard board, it is probably *not* a true
"military grade" board. Likewise, a real mil. board (which will be $$$$)
will use ruggedised, higher spec parts. And, really, unless you are
going to be using the machine in the back of a truck while tearing
across an arabic sandpit, I think it would be right to question why you
think you need a military grade machine!

>> A legit copy of Windows 7 pro will hit you for something in the order of
>> £156 (Amazon.co.uk price),
> OEM version much cheaper.

You aren't supposed to use an OEM build on your *own* computer. You do
know what OEM stands for, right?

However, read this:

http://www.zdnet.com/blog/bott/is-it-ok-to-use-oem-windows-on-your-own-pc-dont-ask-microsoft/1561

>> and see if anything interesting turns up. Mostly it seems to be

>> overclockers that screw up, [...]

> Useful info. But I'm not into overclocking.

Nor am I, nor was I suggesting it. I was simply pointing out that if a
lot of results turn up for "xxx problems", you'll probably find a fair
few are overclockers that pushed the hardware too far, set fire to their
computer, and now want to think of it as a motherboard fault instead of
writing "i am a dick" one thousand times like they should...

>> Further questions are perhaps best directed to a more appropriate
>> newsgroup.
> Agreed. Starting here seemed to be the right place because of emphasis on
> RISC OS - VRPC.

But... you're asking about PC hardware! ;-)

>> which ones are 'best', and what socket types they go into. Indeed, for
>> me, my main criteria would be "can it play HD H.264?".
> Games?? Don't recognise that. Not into games.

HD H.264 is a video codec. Modern YouTube, Bluray, most animé at 720p or
the odd 1080p release... It leads to smaller filesizes and (objectively)
better quality, but it takes a *lot* of crunching to make it work.
My little eeePC can cope with regular HD video just fine in XviD format
(an older codec, more or less compatible with DivX), but often stalls,
stutters, and misses bits with H.264.

I'm not into games. The few I have run on an old PlayStation2 I got for
€20 last summer. On the PC? Duke Nukem (ancient, not the new one) and
Chuckie Egg for Windows. That's it, asides from the usual bundled stuff.

> What ever turns you on, Rick. 8-)

http://r-style.main.jp/aai/limages/laai10.jpg :-)

> My mobile was given to me and is only used for emergencies. First used in
> mid-Wales - nothing happened - no signal.

Err... That's Britain for you. I could argue that this place is more
remote than most of Wales, yet I can usually get EDGE in the house, and
if I step outside my phone switches to 'H' which runs around 2.5mbit
down and 1.8mbit up. It's rather ironic, given the wired internet only
manages 2mbit down and 256k up.

At any rate, the point wasn't the usefulness (or not) of having a mobile
phone, but rather - what it is I actually *need* vs what I can afford.
Sure, we'd all like the greatest sexiest bit of hardware around; however
with a good dose of circumspection, many people would realise that
"lesser" objects do all of the things they really need.

Indeed, as with the smaller eeePC, these lesser objects might in fact be
a better choice in the long run. Back to phones - what sort of impact
would you imagine a dual core processor would have on battery life? A
single core can do the same stuff, only a little slower, and if it gets
a few extra hours from its battery... well, the coolest hardware on the
planet is pretty damn useless when the juice runs out!

There's always more to consider than just "is this processor faster than
that?". As you have noticed with the 2Gb limitation.


Best wishes,

Rick.

[Brian Bailey]

Hi Rick

> > After the third ASUS board in the same machine I've had enough. It's
> > very picky over RAM and won't accept more than 2G - heavens knows why.

> Some memory controllers simply *cannot* address more. I think my eeePC
> may be likewise limited (but I only have 1Gb fitted so it's moot...).

Phrases like 'merchantable quality' come to mind.

> See http://ark.intel.com/products/chipsets/26558 (etc) and note the "Max
> Memory" entry.

> >> Mil grade? You mean military grade?!?
> > Yes, Military grade. My fingers got tired after Mil ....

> If the price compares with a standard board, it is probably *not* a true
> "military grade" board. Likewise, a real mil. board (which will be
> $$$$) will use ruggedised, higher spec parts.

Certainly, I do realise that, but something that has a modicum of
robustness wouldn't come amiss. Something that stands a chance of lasting
at least until I'm shoved in my box. 8-)

> And, really, unless you are going to be using the machine in the back of
> a truck while tearing across an arabic sandpit, I think it would be
> right to question why you think you need a military grade machine!

My brother recommended such an approach. Oh, and didn't I say that is a
customized machine, finish sprayed khaki?

> >> A legit copy of Windows 7 pro will hit you for something in the order
> >> of £156 (Amazon.co.uk price),
> > OEM version much cheaper.

> You aren't supposed to use an OEM build on your *own* computer. You do
> know what OEM stands for, right?

> However, read this:
>
> http://www.zdnet.com/blog/bott/is-it-ok-to-use-oem-windows-on-your-own-pc-dont-ask-microsoft/1561

[snip]

> >> which ones are 'best', and what socket types they go into. Indeed,
> >> for me, my main criteria would be "can it play HD H.264?".
> > Games?? Don't recognise that. Not into games.

> HD H.264 is a video codec. Modern YouTube, Bluray, most animé at 720p or
> the odd 1080p release... It leads to smaller filesizes and
> (objectively) better quality, but it takes a *lot* of crunching to make
> it work. My little eeePC can cope with regular HD video just fine in
> XviD format (an older codec, more or less compatible with DivX), but
> often stalls, stutters, and misses bits with H.264.

OK, with you now, didn't recognise it for what it is. Likewise, I wish to
be able use VLC, which I note, uses that codec.

[snip]

> At any rate, the point wasn't the usefulness (or not) of having a mobile
> phone, but rather - what it is I actually *need* vs what I can afford.
> Sure, we'd all like the greatest sexiest bit of hardware around; however
> with a good dose of circumspection, many people would realise that
> "lesser" objects do all of the things they really need.

Uh, uh, no thanks. I drive a modified VW and it's the best vehicle I've
ever had. Something more 'sexy' would be just a flaming expensive
liability.

> Indeed, as with the smaller eeePC, these lesser objects might in fact be
> a better choice in the long run. Back to phones - what sort of impact
> would you imagine a dual core processor would have on battery life? A
> single core can do the same stuff, only a little slower, and if it gets
> a few extra hours from its battery... well, the coolest hardware on the
> planet is pretty damn useless when the juice runs out!

Weeell, I can always go back to pencil and paper, a slide rule and seven
figure log tables if necessary.

> There's always more to consider than just "is this processor faster than
> that?". As you have noticed with the 2Gb limitation.

Agreed!

Regards

Brian

[Brian Bailey]

Hi All

[snip]

> > Be careful though. W7 32-bit version seems to be rare nowadays and is
> > limited to just over 3 GB of memory (which is probably enough anyway).
> > But the now-normal W7 64-bit version of Home Premium, although it can
> > use more than 3GB of RAM, has problems about Compatability Mode to
> > emulate XP.


> That's a good point. There are issues too with printer drivers for
> W7-65 - HP for example point you to the "Universal driver" for most of
> their older printers, and it can be a case of "one size fits none".

> Generally though I'd recommend looking at, for example, Misco.co.uk,
> and make sure you select W7 Pro and pick a good deal. You will get a
> reasonable machine including W7 pro that way for about the same money
> as a mainboard and windows license. You stll have to buy a processor,
> PSU, maybe a case etc... You will get a year's warranty too.

Many thanks to all who replied.

After receiving your kind advice I visited a generous friend who let me
poke around on his very fast Windows 7 Pro 64 bit machine for a while. I'm
fairly sure that my mind is now made up. I can see definite advantages
with that OS. If for no other reason than the enhanced colour management
features for photograph image manipulation, that in itself is a
significant gain.

Regards

Brian

[Rick Murray]

On 15/01/2012 18:46, Brian Bailey wrote:

>> Some memory controllers simply *cannot* address more. I think my eeePC
>> may be likewise limited (but I only have 1Gb fitted so it's moot...).
> Phrases like 'merchantable quality' come to mind.

It is merchantable, nobody is hiding the memory controller's
limitations. Should we get rid of the 6502 as it is only limited to 64K?
Or RISC OS 4/Select/Adjust/6 because it runs obsolete ARMs in a 26 bit
shared PC+PSR mode (thus meaning the CPU can only address 64Mb)?

If you want a board to address 16Gb of memory, look up the specs and
make sure it is capable.
Then make sure your OS is capable of using this memory.
Read: http://en.wikipedia.org/wiki/3_GB_barrier

> Something that stands a chance of lasting at least until I'm shoved
> in my box. 8-)

Unless you're Cheyne-Stokes-ing right now, I'd say your kit will
probably be extremely obsolete long before you kick the bucket.

Think about it, the Beeb era wasn't that long ago. I grew up with the
machines, and I am very familiar with the look and feel of the ceramic
16K EPROM. Now if I were to lay microSD cards side by side and stacked
to take the same sort of dimensions as an EPROM's body, do you realise
we'd be looking at something in the order of 192Gb of read/write storage
capacity (using 16Gb parts, double that if 32Gb).

Back when I was a teenager, ultra-low-res LCD televisions, monochrome,
would set you back a lot of dosh and they'd run for a couple of hours on
a set of AAs. My little Sony Watchman used a real CRT and it was awesome
enough to allow me to watch teletext on a 1" screen.

How? How about HD video on demand plucked from the air while standing in
a field in a rural area, in a gadget smaller than a CD case, resolution
better than most '90s display units, runs for hours on its own battery,
and can do a load of other stuff too. Apparently this object lets me
speak to people as well! ;-)

Once upon a time I jacked my modem into the world, called a BBS, and
watched the minutes tick by on a long download, and later I would pay
the price.
Now the box in the living room is always on, I can hook into The Wired
any time I want, I don't break into a cold sweat at an 800Mb download
for I know "CARRIER LOST" at 98% is no big deal, as it is unlikely to
happen and if it did, well downloads are frequently able to be resumed.
And I can use the facilities a little, or play 192kbit streaming radio
into my ears for hours and hours. Costs me the same either way.

See what I'm getting to? While so much of the march of technology seems
to be a fairly obvious progression (<cough> rounded corners </cough>),
the things we take for granted today were a dream, an impossibility,
just two decades ago.

In a way part of the charm of the pre-mobile world is lost. After all,
there have been many adaptions of Enid Blyton's books, but would it be
the same if Philip, Jack, Dinah, and Lucy-Ann went off in separate
directions and BBMed each other all the way? Why try to outsmart the bad
guys when you could video it and dump it on YouTube? <sigh>

> My brother recommended such an approach. Oh, and didn't I say that is a
> customized machine, finish sprayed khaki?

<facepalm>

> Likewise, I wish to be able use VLC, which I note, uses that codec.

VLC is useful to have around as it will often play stuff nothing else
can, but you will probably find SMPlayer to be a lot friendlier to use.
http://smplayer.sourceforge.org/ [I think!]

> Weeell, I can always go back to pencil and paper, a slide rule and seven
> figure log tables if necessary.

Look on the bright side, those who know how to do maths will be able to
restart the economy after the bomb drops. Those who can't do it without
a calculator in hand will just be everybody else's butt monkey.

After all, a bunch of bored scientists went and moved the minute hand on
the mid[night]mare clock one minute closer to our imminent destruction,
so we have to now be prepared for such eventualities (even though we
survived both Bush eras without anything getting nuked, go figure!).
[
http://www.thebulletin.org/content/media-center/announcements/2012/01/10/doomsday-clock-moves-1-minute-closer-to-midnight
]


Best wishes,

Rick.

[John Sandford]

Brian Bailey <bba...@argonet.co.uk> wrote:

> > A legit copy of Windows 7 pro will hit you for something in the order of
> > £156 (Amazon.co.uk price),
>
> OEM version much cheaper.

Not when you want to use it on a different computer and microsoft wont let
you as its an "oem version tied to the first machine" not an upgrade or
full.
Although I've successfuly convinced microsoft to issue the reset code on one
occasion, probaly safer to go for an upgrade version at least.

John

--
John Sandford
home

[John Sandford]

M Harding <ris...@mdharding.org.uk> wrote:

>
> I still haven't (as yet) managed to get it to talk to my Epson RX620
> printer's scanner. Nor does (?did) Nero work, though I've found a clone
> that does. It's the W7 Pro version for which you get all the pluses - but
> at great cost. But except for the Compatability Mode issue I'm as happy
> with W7 as with XP/3.
>
> Michael Harding Rev. Preb. M.D. Harding ris...@mdharding.org.uk

You need nero 8 minimum on windows 7, as for the scanner get Vuescan (
hamrick.com )about 24gbp, I had the same problem with my R300 no drivers.

[M Harding]

In article <mpro.lxvzx5000...@thesandfords.me.uk>,

John Sandford <li...@thesandfords.me.uk> wrote:
> M Harding <ris...@mdharding.org.uk> wrote:

> >
> > I still haven't (as yet) managed to get it to talk to my Epson
> > RX620 printer's scanner. Nor does (?did) Nero work, though I've
> > found a clone that does. It's the W7 Pro version for which you
> > get all the pluses - but at great cost. But except for the
> > Compatability Mode issue I'm as happy with W7 as with XP/3.

> You need nero 8 minimum on windows 7,

Mine is v.7 but I didn't upgrade it because (a) it works on the XP
desktop computer and (b) I read a review claiming that it was not yet
compatible with W7 64-bit - whereas the cheap Ashampoo version I
bought is.

> as for the scanner get Vuescan ( hamrick.com )about 24gbp, I had
> the same problem with my R300 no drivers.

I'll explore the Vuescan - though the RX620 also incorporates a slide
scanner and some nice software (which doesn't work in W7).

Thanks for the tips, John.

[Brian Bailey]

Hi Rick

> >> Some memory controllers simply *cannot* address more. I think my
> >> eeePC may be likewise limited (but I only have 1Gb fitted so it's
> >> moot...).
> > Phrases like 'merchantable quality' come to mind.

> It is merchantable, nobody is hiding the memory controller's
> limitations.

No, no, Rick. It's got four RAM slots, which I paid for, and it will only
accept two RAM cards, memory total 2G. It doesn't do what I paid for.
However, I win, I get my revenge, it's going for scrap. 8-)

[snip]

> > Something that stands a chance of lasting at least until I'm shoved in
> > my box. 8-)

> Unless you're Cheyne-Stokes-henge

Who he?? Not with you there Rick. What's that all about. Did a Wiki search
and got the abnormal breathing bit. But 'henge', you completely lost me.

> right now, I'd say your kit will probably be extremely obsolete long
> before you kick the bucket.

[snip]

> See what I'm getting to? While so much of the march of technology seems
> to be a fairly obvious progression (<cough> rounded corners </cough>),
> the things we take for granted today were a dream, an impossibility,
> just two decades ago.

Not quite such an obvious progression. I never imagined that British heavy
engineering could/would be so comprehensively scrapped.

Anyway, thanks for the intriguing and amusing diversion. When I had
completed my engineering training (as a mechanical engineer. The academic
side of the training course had a heavy steam engineering bias - still
relevant today) the transistor was still, I guess, a decade away from
commercial usage.

> In a way part of the charm of the pre-mobile world is lost. After all,
> there have been many adaptions of Enid Blyton's books, but would it be
> the same if Philip, Jack, Dinah, and Lucy-Ann went off in separate
> directions and BBMed each other all the way? Why try to outsmart the bad
> guys when you could video it and dump it on YouTube? <sigh>

Don't entirely agree, but I have attempted the latter.

> > Likewise, I wish to be able use VLC, which I note, uses that codec.

> VLC is useful to have around as it will often play stuff nothing else
> can, but you will probably find SMPlayer to be a lot friendlier to use.
> http://smplayer.sourceforge.org/ [I think!]

Thanks for that. I can see why you like it.

> > Weeell, I can always go back to pencil and paper, a slide rule and
> > seven figure log tables if necessary.

> Look on the bright side, those who know how to do maths will be able to
> restart the economy after the bomb drops. Those who can't do it without
> a calculator in hand will just be everybody else's butt monkey.

Ah, well, it's not just maths. We have lost so many skills. I used to be
able to file to 2 thous and turn to 0.25 thous and I was nowt special.
But, no more.

The old steam engineers amazed me. I was told, many years ago, about a
reciprocating steam engine delivered to a mill which when it came to be
erected was found to have a conical bore to the cylinder. No 'merchantable
quality' bit and return under warrenty. The mill craftsmen started out
with hammers and chisels and opened up the bore. They then went on to file
the bore truly parallel and circular.

> After all, a bunch of bored scientists went and moved the minute hand on
> the mid[night]mare clock one minute closer to our imminent destruction,
> so we have to now be prepared for such eventualities (even though we
> survived both Bush eras without anything getting nuked, go figure!). [
> http://www.thebulletin.org/content/media-center/announcements/2012/01/10/doomsday-clock-moves-1-minute-closer-to-midnight
> ]

Oh, dear. I was in the nuclear industry. 8-)

Regards

Brian

[John Sandford]

the more expensive vuescan pro does slide scanning.

[M Harding]

In article <mpro.lxw88w000...@thesandfords.me.uk>,

John Sandford <li...@thesandfords.me.uk> wrote:
> M Harding <ris...@mdharding.org.uk> wrote:


> >
> > I'll explore the Vuescan - though the RX620 also incorporates a
> > slide scanner and some nice software (which doesn't work in W7).

> the more expensive vuescan pro does slide scanning.

I've given up on its use as a printer (I use a mono laser normally
anyway) but had thought to replace it eventually with a scanner - but
the Vuescan Pro might be a lot cheaper! Thanks.

[Alan Adams]

In message <52527668...@argonet.co.uk>

Brian Bailey <bba...@argonet.co.uk> wrote:

> Hi Rick



>>>> Some memory controllers simply *cannot* address more. I think my
>>>> eeePC may be likewise limited (but I only have 1Gb fitted so it's
>>>> moot...).
>>> Phrases like 'merchantable quality' come to mind.

>> It is merchantable, nobody is hiding the memory controller's
>> limitations.

> No, no, Rick. It's got four RAM slots, which I paid for, and it will only
> accept two RAM cards, memory total 2G. It doesn't do what I paid for.
> However, I win, I get my revenge, it's going for scrap. 8-)

Would this be two pairs of slots, in two different colours?

If so, it's a very common configuration that lets you use either of
two different types of memory - but not both at onve. So you can have
one or two simms in the yellow slots, or one or two in the black
slots.

The memory controller limits how big those simms can be - in many
cases you can put 2 x 2GB in, but some only allow 2 x 1gb, or 1 x 2gb.

[snip]

[Rick Murray]

On 16/01/2012 14:08, Brian Bailey wrote:

> No, no, Rick. It's got four RAM slots, which I paid for, and it will only
> accept two RAM cards, memory total 2G.

Are they IDENTICAL memory slots? I saw a board recently they could take
"this type of memory" or "that type of memory" (but not both at the same
time).

> It doesn't do what I paid for.

What's the board make and model?

> However, I win, I get my revenge, it's going for scrap. 8-)

You lose. You paid for it. Now you're scrapping it. Double lose. :-)

>> Unless you're Cheyne-Stokes-henge
> Who he?? Not with you there Rick.

? I did not write "henge". Got my sent message right here, it says "-ing".

> Did a Wiki search and got the abnormal breathing bit.

Yup, that very specific gaspy irregular breathing that the dying do.


Point being, unless you're on the verge of dying, you WILL outlive your
equipment. No question. Sure, it might be functional, like a BBC B is
functional, but if you're looking for modern goodies, you'll be buying
another box soon. And then another. And then...

> I never imagined that British heavy engineering could/would
> be so comprehensively scrapped.

Years of mimanagement all around. And yet there's no viable alternative
for people to vote for? If anything, the Reds, Blues, and Yellows have
merged into a sort of murky purple colour.

As for America - a population of 312 million, and that bunch of d*ckwads
is the best the country has to offer for leadership? It's no wonder
western civilisation is flushing itself down the bog!

> completed my engineering training (as a mechanical engineer. The academic
> side of the training course had a heavy steam engineering bias - still
> relevant today)

It amuses me to think that the biggest, bestest, most ultra-modernest
nuclear (sorry, new-kew-ler) power station is... basically a giant water
heater to create steam to spin a turbine!

Steam rocks. I'd like a little Mammod engine, but it's a little out of
my price range. I'll have to make do with watching "steampunk" movies -
and of this I'd recommend "Sucker Punch". Multi-layered plot, cute girl,
mechanical steam nazis, loads of in-jokes, what's not to like?


> But, no more.

Ditto woodworking. How many people are actually much good at it these
days? For God's sake, I had to tell our builder how to mitre a joint
(and I worked that one out in my head on the fly!).

> The old steam engineers amazed me.

I guess there's a benefit in having something tangible and real to
understand. Not like this new-fangled "electric" stuff. ;-)


Best wishes,

Rick.

[Brian Bailey]

Hi Rick

> > No, no, Rick. It's got four RAM slots, which I paid for, and it will
> > only accept two RAM cards, memory total 2G.

> Are they IDENTICAL memory slots? I saw a board recently they could take
> "this type of memory" or "that type of memory" (but not both at the same
> time).

Point taken.

Anyway, I am assured by the supplier, it's broken.

> > It doesn't do what I paid for.

> What's the board make and model?

For the record, it's an ASUS M2N-E SLI.

> > However, I win, I get my revenge, it's going for scrap. 8-)

> You lose. You paid for it. Now you're scrapping it. Double lose. :-)

Probably, even certainly, but I have had some mileage out of it. 8-)

> >> Unless you're Cheyne-Stokes-henge
> > Who he?? Not with you there Rick.

> ? I did not write "henge". Got my sent message right here, it says
> "-ing".

So you did. Sorry about that. In relying I occasionally press a return in
the wrong place and Pluto immediately does a spell check and changes the
word, which I then don't notice. Damn!

> > Did a Wiki search and got the abnormal breathing bit.

> Yup, that very specific gaspy irregular breathing that the dying do.

Umm, I'd rather we didn't go down that route, please.

> Point being, unless you're on the verge of dying, you WILL outlive your
> equipment. No question.

Hey, that's great. Can I have that in (real) writing please.

Jokingly, I say that I have three objectives in life,
1 Live to the age of 92
2 Be a bloody old nuisance
3 Go into my box totally broke on that day

> Sure, it might be functional, like a BBC B is
> functional, but if you're looking for modern goodies, you'll be buying
> another box soon. And then another. And then...

> > I never imagined that British heavy engineering could/would
> > be so comprehensively scrapped.

> Years of mismanagement all around.

How I totally *totally* agree with you. I saw so much of it. I worked for
English Electric at one time and the writing was writ large on the wall.

[snip]

> It amuses me to think that the biggest, bestest, most ultra-modernest
> nuclear (sorry, new-kew-ler) power station is... basically a giant water
> heater to create steam to spin a turbine!

They are highly developed bits of kit.

> Steam rocks. I'd like a little Mammod engine, but it's a little out of
> my price range. I'll have to make do with watching "steampunk" movies -
> and of this I'd recommend "Sucker Punch". Multi-layered plot, cute girl,
> mechanical steam nazis, loads of in-jokes, what's not to like?

My one ambition, which I failed to fulfil, was to work on a North Eastern
Marine re-heat triple expansion marine engine. They were really highly
developed engines.

> > But, no more.

> Ditto woodworking. How many people are actually much good at it these
> days?

There are people, believe me. I once knew a really skilled chippy who
repaired Mosquito's during the war. Amazing stuff, even now.

> For God's sake, I had to tell our builder how to mitre a joint
> (and I worked that one out in my head on the fly!).

Gee whizz!

> > The old steam engineers amazed me.

> I guess there's a benefit in having something tangible and real to
> understand. Not like this new-fangled "electric" stuff. ;-)

Too right. 8-)

Regards

Brian

[Dave Symes]

In article <4f147528$0$2514$ba4a...@reader.news.orange.fr>,
Rick Murray <heyrickma...@yahoo.co.uk> wrote:
[Snippy]

> Ditto woodworking. How many people are actually much good at it these
> days? For God's sake, I had to tell our builder how to mitre a joint
> (and I worked that one out in my head on the fly!).

Ricky, Ricky... A builder builds things in brick, block, concrete etc,
they build to the nearest bend in the old boxwood folding ruler.

A Carpenter/Joiner is required for some precision work, particularly
things like mitres.

Mmmm! I wouldn't have thought a fly's head was big enough to do working
outs on. ;-) We used to use the insides of old fag packets.

Dave

--

Dave Triffid

[M Harding]

In article <mpro.lxvzx5000...@thesandfords.me.uk>,

John Sandford <li...@thesandfords.me.uk> wrote:
> M Harding <ris...@mdharding.org.uk> wrote:

> >
> > I still haven't (as yet) managed to get it to talk to my Epson
> > RX620 printer's scanner. Nor does (?did) Nero work, though I've
> > found a clone that does. It's the W7 Pro version for which you
> > get all the pluses - but at great cost. But except for the
> > Compatability Mode issue I'm as happy with W7 as with XP/3.

> You need nero 8 minimum on windows 7, as for the scanner get
> Vuescan ( hamrick.com )about 24gbp, I had the same problem with my
> R300 no drivers.

John, you've just saved me well over a hundred quid in buying a new
scanner!

Having examined that Vuescan URL it led me to the discovery that
though the W7 64-bit version doesn't support the RX620, it is in fact
possible to use the Vista version for it. So now I happily have the
laptop talking to the scanner/printer - and even to its slide scanner.

Very many thanks indeed for that lead, John. And I didn't even need to
buy Vuescan!

[John Sandford]

M Harding <ris...@mdharding.org.uk> wrote:


> John, you've just saved me well over a hundred quid in buying a new
> scanner!
>
> Having examined that Vuescan URL it led me to the discovery that though
> the W7 64-bit version doesn't support the RX620, it is in fact possible to
> use the Vista version for it. So now I happily have the laptop talking to
> the scanner/printer - and even to its slide scanner.
>
> Very many thanks indeed for that lead, John. And I didn't even need to buy
> Vuescan!
>
> Michael Harding Rev. Preb. M.D. Harding ris...@mdharding.org.uk

I'm glad to hear that, I did not look your printer up, so was not aware of
the vista software, For future reference anything that has a signed vista
driver seems in most cases to work with Win 7 64 bit.

I have an Epson perfection scanner which only has XP software, which would
not work even using compatibility mode, Vuescan fixed that problem here,this
was recommended by Epson themselves.
This is a bit off topic so perhaps we had better desist. :-)

[Brian Bailey]

Hi Rick

> > > Likewise, I wish to be able use VLC, which I note, uses that codec.

> > VLC is useful to have around as it will often play stuff nothing else
> > can, but you will probably find SMPlayer to be a lot friendlier to use.
> > http://smplayer.sourceforge.org/ [I think!]

> Thanks for that. I can see why you like it.

Except, that,

I'm curious about the relative acoustic performance of VLC, SMPlayer and
WMP, their codecs and the way they work.

I assume, that each player refers to a library of codecs, which, I also
assume, decompress the various file formats that are thrown at them.

You referred to a specific codec and that prompted me to see what the
library of each player might contain. Stay with me, I am trying to gain
some understanding of the way things work.

I've used WMP and have no difficulty in rejecting it.

I had the privilege, very recently, of attending a live performance of
Schubert's string quintet. And, I thought it would be interesting to
compare VLC and SMPlayer.

I was surprised to find that, in this instance, that there are a fair
number of options for decoding mp3 files in each player. Each selected a
different codec for the file being played. I assume that the skill in
designing a player is in enabling the selection of the /best/ codec for
the file in question.

Anyway, I had downloaded an mp3 file of the Schubert quintet. The
play-back through SMPlayer was barely acceptable whereas the play-back
through VLC was entirely credible and alive.

What is being looked for in the file format to enable the /correct/
selection of a codec?

Could you give some clue as to what is going on here, please.

Regards

Brian

[Jim Lesurf]

In article <52558b38...@argonet.co.uk>, Brian Bailey

<bba...@argonet.co.uk> wrote:
> Hi Rick

> > > > Likewise, I wish to be able use VLC, which I note, uses that codec.

> > > VLC is useful to have around as it will often play stuff nothing
> > > else can, but you will probably find SMPlayer to be a lot friendlier
> > > to use. http://smplayer.sourceforge.org/ [I think!]

> > Thanks for that. I can see why you like it.

> Except, that,

> I'm curious about the relative acoustic performance of VLC, SMPlayer and
> WMP, their codecs and the way they work.

> I assume, that each player refers to a library of codecs, which, I also
> assume, decompress the various file formats that are thrown at them.

Answering your questions is complicated by the lack of clarity in the ways
people often use terms like 'codec'. cf below. Alas, audio pros also often
muddle this by using the same word for distinct things.

At root each 'codec' is defined in terms of something like a file or stream
'format'. By which I mean a definition of the general process used for
encoding, and the way the bits in the resulting file or stream are to be
regarded by the decoder. However it may be left to each implimentor to
decide how to generate the output in the form that a decoder can understand
and decode. So you can expect that not all will be created equally well,
even when into the same 'codec' (format) and the same data bitrate.

So it is quite possible to have various different programs encode the same
LPCM source as a different set of say, 'mp3', data. And for different
decoders to generate different LPCM output when they are all given the same
'mp3' file or stream. Similar comments apply to many other coding systems.

Similarly, I've measured differences between 'dedicated hardware' that
'decodes' or 'plays' things like 'mp3'. Although in simple theory the 'mp3'
data uniquely defines what LPCM should be produced. The theory tends to
assume all 'compliant' decoders will give the same results fed a file in
the format they expect. No so, alas. Reality doesn't always know what
theory has told it to behave like. :-)

You can replace 'mp3' above with various other formats and the comments
will continue to be correct in many cases.

> You referred to a specific codec and that prompted me to see what the
> library of each player might contain. Stay with me, I am trying to gain
> some understanding of the way things work.

> I've used WMP and have no difficulty in rejecting it.

> I had the privilege, very recently, of attending a live performance of
> Schubert's string quintet. And, I thought it would be interesting to
> compare VLC and SMPlayer.

> I was surprised to find that, in this instance, that there are a fair
> number of options for decoding mp3 files in each player. Each selected a
> different codec for the file being played.

The above seems to confusing 'codec' (specification for the file or stream
format/container) with the choice of 'decoder' or 'decoder settings'. This
would not matter if all decoders for a given coding format worked in the
same way. But as you have found...

> Could you give some clue as to what is going on here, please.

The basic thing to bear in mind is that 'lossy codecs' all discard details
according to some specified set of 'judgement rules'. If you want to
minimise this, get out of the game. Use either LPCM or a system like FLAC
where the output can be a bit-perfect reconstruction of the original LPCM
that was 'encoded'.

Alas, there is of course more to it than that. Having recovered LPCM with
no loss or change due to the 'codec' it still isn't always certain that
your computer (or other hardware) will actually play this correctly. Ditto
for the playing application chosen. e.g. I've measured a version of
Audacious, (on Ubuntu) playing 24 bit LPCM Wave files with the least
significant bits set to zero. i.e. as 16 bit. Yet the same player and
computer plays 24 bit FLAC as 24 bit.

So play a 24bit Wave file and get *16bit* output sent to the DAC.
(Actually 24 bit values arrive, but with the least significant 8 bits
of every sample set to zero.) But play a flac of the same Wave data and it
comes to the DAC as 24bit without the problem. Weird!

If you just check if the DAC says "24 bit arriving" you don't notice the
problem as the it is in the 24bit having details lost. So any bells
or whistles visible to the user tell you all is well... when it isn't.
You have to actually capture the values reaching the DAC and check them
to find the problem.

Looks like whoever produced that version of the player simply worried about
flac. Perhaps they never tested Wave. Perhaps they had no way to even
measure the output data stream sent to a DAC. Alas my experience in
computer audio is that depressingly often "I can hear something" seems to
have been taken as an identity with "Sound output works perfectly". When it
was far from the case.

This comes down the the people who wrote the player and ensuring your
system is set up correctly.... and your choice of DAC or 'soundcard' or
whatever...

Short version: If worried, use LPCM or FLAC, but don't even then assume it
must be perfect.

One good thing, though. You've reminded me that I really should check the
sound from RPCEmu on my (Xubuntu) laptop. As yet I've not tried it.

Slainte,

Jim

--
Please use the address on the audiomisc page if you wish to email me.
Electronics http://www.st-and.ac.uk/~www_pa/Scots_Guide/intro/electron.htm
Armstrong Audio http://www.audiomisc.co.uk/Armstrong/armstrong.html
Audio Misc http://www.audiomisc.co.uk/index.html

[Brian Bailey]

Hi Jim

[snip]

Very probably!

> This would not matter if all decoders for a given coding
> format worked in the same way. But as you have found...

> > Could you give some clue as to what is going on here, please.

[snip]

> This comes down the the people who wrote the player and ensuring your
> system is set up correctly.... and your choice of DAC or 'soundcard' or
> whatever...

Thank you for your lucid explanation, Jim.

> Short version: If worried, use LPCM or FLAC, but don't even then assume
> it must be perfect.

Oh, I wasn't worried, I was just trying to reconcile the clearly audible
differences. If I had been worried there is nothing I could do have done
about it, anyway. 8-)

Oddly enough, in poking about in general, I found that both VLC and
SMPlayer use a generic audio output driver, which I hadn't been aware of.
I use an M-Audio card and reconfiguring each player to use the M-Audio
driver led to a quite startling transformation in audio quality, though
even then VLC sounded much superior.

> One good thing, though. You've reminded me that I really should check
> the sound from RPCEmu on my (Xubuntu) laptop. As yet I've not tried it.

Interesting stuff. Thanks, once again

Regards

Brian

[Rick Murray]

On Sun, 22 Jan 2012 12:44:14 +0000 (GMT), Brian Bailey
<bba...@argonet.co.uk> wrote:

> I'm curious about the relative acoustic performance of VLC,
SMPlayer and
> WMP, their codecs and the way they work.

Your best bet here would be to find a forum relevant to technical
discussion of the internal workings of codecs.

I 99% of cases, SMPlayer "just works". I did have a multilingual
EN/JP MKV that played silent, but picking a different MKV splitter
worked.
This weeks episode of "Moretsu Pirates" failed on SMPlayer, but
played on VLC. There are many variations of what might seem to be the
best thing.

> I assume, that each player refers to a library of codecs, which, I
also
> assume, decompress the various file formats that are thrown at them.

In general, yes. But note that "codec" is a very misunderstood word.
Meaning "COder/DECoder", is it correct to call a playback-only piece
of software a codec?
Waters further muddied by Philips who refer to their AIC23 chip as a
codec. It takes analogue audio and converts it to a McBSP data
stream, and vice versa, so technically their use of the term is
correct, though most people would think of a lump of software as a
codec...

> You referred to a specific codec

I did? I thought I referred to a movie player for a specific file
format.

If you look in SMPlayer's menus, File information or such, there is a
way to pick the "codec" you like. You'll notice there are three or
four for H.264 video (much is from ffmpeg).

> Stay with me, I am trying to gain
> some understanding of the way things work.

At its most basic, a file is a chunked set of stream data - in other
words, you have a big stream of data that is split into bits so there
can be some relationship between which bit of audio goes which which
bit of video. This is not necessary, one stream could follow another,
and on an AVI sitting on a harddisc it'll work okay, but the same on
a CD-R would be a disaster of endless seeks.

Note, by "stream" I mean a large sequence of data that describes a
picture or a sound. This isn't the same as "streaming".
Note, also, we have not yet mentioned the encoding.

So, you have this arrangement that is video and audio like
AVAVAVAVAVAVA[etc].
The next step is a "splitter" (also known as a "demuxer"). This
understands the file, but not the encoding. It's job is to separate
the audio data and video data into unique blocks of data. Typically
the decoder for each will work as an independent thread, so the video
decoder gets the video data, and so on. While file formats are
broadly similar (3gp vs MKV vs AVI vs mp4 vs flv...), they are not
compatible *even* *if* the data within is the same.
The final step is the decoder, what most think of as the codec. It
takes H.263 (XviD etc) or whatever and pastes it to the screen, while
a different decoder takes the mp3 (etc) and dumps it on the sound
driver. The end result? A movie to watch.

> I've used WMP and have no difficulty in rejecting it.

Correction - you've used WMP and it had no difficulty in rejecting
your files if they were in any way "unusual".

> I was surprised to find that, in this instance, that there are a
fair
> number of options for decoding mp3 files in each player.

Not a surprise. With a hardware player you get what you are given.
But there is no such thing as a one-size-fits-all. On my phone, for
example, I like strongly emphasised bass to compensate for the use of
earphones.

> I assume that the skill in designing a player is in enabling the
selection of
> the /best/ codec for the file in question.

Difficult. mp3 is a lossy codec, and you will find that for identical
settings, different encoders produce different results. There is a
lot of technology hidden behind phrases like "psychoacoustic" - in
essense, the decision of what is okay to throw away to make something
that "sounds the same" (as the input).
There are also differences like true stereo vs joint stereo (mono
channel plus a list of differences used to reconstruct stereo).

> Anyway, I had downloaded an mp3 file of the Schubert quintet. The
> play-back through SMPlayer was barely acceptable whereas the
play-back
> through VLC was entirely credible and alive.

Where might I find a copy of this to try it on my setup?

Check also your player settings (outside of the decoder choice). Is
it using the correct driver, or the Windows WAV fall-back? My system
has eight choices, I currently use "dsound". I have my SMPlayer set
to equalise audio, downmixed to stereo for headphones. Windows itself
has been told to mix for headphones.

This is important as different people will perceive sounds in
different ways. I could always tell Eagle FM (Guildford) on the dial
as the sound was very warm, like a valve radio, although technically
they must have been messing with it. It sounded better that way, a
purist might disagree. Then there is a battle to get the best output
from the smallest number of bits, and reconstruct that to an
acceptable representation of the original. Then there are technical
issues - domestic kit hasn't yet entirely caught on to 24 bit
sampling, so the file might be 24 bit, the audio output hardware
might be 24 bit, but the driver in the middle might chuck away some
of that, being stuck in a 16 bit world. It sounds silly, but then so
does a computer that tries to play everything at 48k sampling rate...

To be honest, however, if you appreciate music, you might find a
lossless format is better than mp3. Failing that, AAC at a high
enough bitrate (288kbit), and as a *last* resort, mp3 at 320kbit.
That's why I'd never buy music from Deezer, they were offering mp3 at
a mere 128kbit - I don't even consider that adequate for stuff I
record off the telly!

> What is being looked for in the file format to enable the /correct/
> selection of a codec?

Probably just a lookup table linking data type to what works on most
platforms. Just looking, mp3 is MPEG layer 3, which means I can pick
ffmp3, mp3, mad, or mp3acm. There might be others. The default is
ffmp3.

> Could you give some clue as to what is going on here, please.

Tell me where I can get the file and I'll try it on SMPlayer, VLC,
and WinAMP (with headphones) on my machine. Note, incidentally, that
the first two are designed to play video - audio only playback might
be seen as an "also" feature (ie, where's the eq? the stereo balance,
etc?).


Best wishes,

Rick.

--
Best wishes,

Rick.

[Brian Bailey]

Hi Rick

> > I'm curious about the relative acoustic performance of VLC, SMPlayer
> > and WMP, their codecs and the way they work.

> Your best bet here would be to find a forum relevant to technical
> discussion of the internal workings of codecs.

I was hoping that you, yourself would reply, also, hopefully, Jim. You
have both kindly responded. For which, many thanks.

> In 99% of cases, SMPlayer "just works". I did have a multilingual EN/JP

> MKV that played silent, but picking a different MKV splitter worked.
> This weeks episode of "Moretsu Pirates"

"Mouretsu Pirates" perhaps?? "Simon's Cat" is more my scene. 8-)

> failed on SMPlayer, but played on VLC. There are many variations of what
> might seem to be the best thing.

> > I assume, that each player refers to a library of codecs, which, I
> > also assume, decompress the various file formats that are thrown at
> > them.

> In general, yes. But note that "codec" is a very misunderstood word.

Certainly by me ...

> Meaning "COder/DECoder", is it correct to call a playback-only piece of
> software a codec? Waters further muddied by Philips who refer to their
> AIC23 chip as a codec. It takes analogue audio and converts it to a
> McBSP data stream, and vice versa, so technically their use of the term
> is correct, though most people would think of a lump of software as a
> codec...

> > You referred to a specific codec

> I did? I thought I referred to a movie player for a specific file
> format.

OK. I /thought/ that you were referring to a codec but then I am just
showing my lack of knowledge.

Counter correction, if I may. It didn't reject anything. I just didn't
like the quality of sound output at all; I also don't like the interface.
To my ears, it /sounded/ dull and lifeless.

> > I was surprised to find that, in this instance, that there are a fair
> > number of options for decoding mp3 files in each player.

> Not a surprise.

Well, to me it is, especially so if the default isn't optimal.

> With a hardware player you get what you are given. But there is no such
> thing as a one-size-fits-all. On my phone, for example, I like strongly
> emphasised bass to compensate for the use of earphones.

Not my preference, but I do take your point.

> > I assume that the skill in designing a player is in enabling the
> > selection of the /best/ codec for the file in question.

> Difficult. mp3 is a lossy codec, and you will find that for identical
> settings, different encoders produce different results.

I assumed that to be the case.

> There is a lot of technology hidden behind phrases like "psychoacoustic"
> - in essense, the decision of what is okay to throw away to make
> something that "sounds the same" (as the input). There are also
> differences like true stereo vs joint stereo (mono channel plus a list
> of differences used to reconstruct stereo).

> > Anyway, I had downloaded an mp3 file of the Schubert quintet. The
> > play-back through SMPlayer was barely acceptable whereas the
> > play-back through VLC was entirely credible and alive.

> Where might I find a copy of this to try it on my setup?

Firstly, if you are interested in following this slightly tortuous route,
go to Wikipeadia - String Quintet (Schubert) - under - External links -
click on Recording of the Quintet. It is free to download.

> Check also your player settings (outside of the decoder choice). Is it
> using the correct driver, or the Windows WAV fall-back?

It is now, though it wasn't until yesterday. I imagine it was using
Windows WAV as you suggest.

> My system has eight choices, I currently use "dsound". I have my
> SMPlayer set to equalise audio, downmixed to stereo for headphones.
> Windows itself has been told to mix for headphones.

Ah, that's interesting and tricky too. I've used various mix settings for
headphones (Sennheiser HD 598) and most of them just don't sound /right/.
I have settled on decent headphones that sound acceptable out of the box.

> This is important as different people will perceive sounds in different
> ways.

They do, they do!

> I could always tell Eagle FM (Guildford) on the dial as the sound
> was very warm, like a valve radio, although technically they must have
> been messing with it. It sounded better that way, a purist might
> disagree. Then there is a battle to get the best output from the
> smallest number of bits, and reconstruct that to an acceptable
> representation of the original. Then there are technical issues -
> domestic kit hasn't yet entirely caught on to 24 bit sampling, so the
> file might be 24 bit, the audio output hardware might be 24 bit, but
> the driver in the middle might chuck away some of that, being stuck in
> a 16 bit world. It sounds silly, but then so does a computer that tries
> to play everything at 48k sampling rate...

> To be honest, however, if you appreciate music, you might find a
> lossless format is better than mp3. Failing that, AAC at a high enough
> bitrate (288kbit), and as a *last* resort, mp3 at 320kbit. That's why
> I'd never buy music from Deezer, they were offering mp3 at a mere
> 128kbit - I don't even consider that adequate for stuff I record off
> the telly!

These days I can't discriminate to that extent. I used to be able to hear
right up to 28KHz, but very much less now.

> > What is being looked for in the file format to enable the /correct/
> > selection of a codec?

> Probably just a lookup table linking data type to what works on most
> platforms. Just looking, mp3 is MPEG layer 3, which means I can pick
> ffmp3, mp3, mad, or mp3acm. There might be others. The default is ffmp3.

OK. I wondered if that might be the case.

> > Could you give some clue as to what is going on here, please.

> Tell me where I can get the file and I'll try it on SMPlayer, VLC, and
> WinAMP (with headphones) on my machine.

As above. Hope you find it to be enjoyable not just a technical exercise.
8-)

> Note, incidentally, that the first two are designed to play video -
> audio only playback might be seen as an "also" feature (ie, where's the
> eq? the stereo balance, etc?).

Well, there is equalisation in the player and stereo balance via the sound
card. But, these days I am very much a minimalist and do not wish to have
to tweak what isn't initially acceptable eg I try to do photography on the
basis of getting the image right within the camera first time with very
little manipulation afterwards.

Thanks for the very interesting exchange.

Regards

Brian

[Jim Lesurf]

In article <5255e7e0...@argonet.co.uk>, Brian Bailey
<bba...@argonet.co.uk> wrote:
> Hi Rick

> > > Anyway, I had downloaded an mp3 file of the Schubert quintet. The
> > > play-back through SMPlayer was barely acceptable whereas the
> > > play-back through VLC was entirely credible and alive.

> > Where might I find a copy of this to try it on my setup?

> Firstly, if you are interested in following this slightly tortuous
> route, go to Wikipeadia - String Quintet (Schubert) - under - External
> links - click on Recording of the Quintet. It is free to download.

Interesting.

[Snip]

> > To be honest, however, if you appreciate music, you might find a
> > lossless format is better than mp3. Failing that, AAC at a high enough
> > bitrate (288kbit), and as a *last* resort, mp3 at 320kbit. That's why
> > I'd never buy music from Deezer, they were offering mp3 at a mere
> > 128kbit - I don't even consider that adequate for stuff I record off
> > the telly!

> These days I can't discriminate to that extent. I used to be able to
> hear right up to 28KHz, but very much less now.

Erm... most of the audible impact of low bitrates or poor (use of) codecs
tends to be at much lower frequencies than you may think. What the systems
tend to rely upon is the listener not having a clear idea of that the
'original' sounds like, or it being 'processed' music. The devil is all in
the detail. The BBC 320k AAC stream for Radio 3 can sound excellent. But
I've heard 'mp3' net radio at 256k sound awful in ways that you could hear
even if you hearing cut off at 3kHz!

As you've twigged from your own experience. It isn't enough to have a 'good
codec' or a 'high bitrate'. You also need the system to be employed with
due care and understanding. The BBC iPlayer provides this. The engineers
involved have put a lot of care into tweaking the setup to optimise the
results. And since they provide the browser plugin they also supply/control
your 'decoder'. My experience is that they have consistently hacked away at
any problems I've detected, and keep trying to make further improvements.

In contrast, some websites that sell 'high resolution' flac/wave downloads
have actually sold crude upconversions of CD standard material. It says
'96k/24bit' on the tin, but what is inside is 44k/16bit with added
aliasing!

> > > What is being looked for in the file format to enable the /correct/
> > > selection of a codec?

> > Probably just a lookup table linking data type to what works on most
> > platforms. Just looking, mp3 is MPEG layer 3, which means I can pick
> > ffmp3, mp3, mad, or mp3acm. There might be others. The default is
> > ffmp3.

> OK. I wondered if that might be the case.

Most of the relevant file/stream formats have a specific 'header' for the
file, or at intervals along the stream. The replay software looks for this.
Or at least it should. Some software 'guesses' when given a stream, and
every now and then gets it wrong. And a stream may contain a section that
fools/confuses the player. On top of this a file may have a declared type,
etc. This may or may not be used as a guide.

[Brian Bailey]

Hi Jim

[snip]

> > > Where might I find a copy of this to try it on my setup?

> > Firstly, if you are interested in following this slightly tortuous
> > route, go to Wikipeadia - String Quintet (Schubert) - under - External
> > links - click on Recording of the Quintet. It is free to download.

> Interesting.

> [Snip]

> > > To be honest, however, if you appreciate music, you might find a
> > > lossless format is better than mp3. Failing that, AAC at a high
> > > enough bitrate (288kbit), and as a *last* resort, mp3 at 320kbit.
> > > That's why I'd never buy music from Deezer, they were offering mp3
> > > at a mere 128kbit - I don't even consider that adequate for stuff I
> > > record off the telly!

> > These days I can't discriminate to that extent. I used to be able to
> > hear right up to 28KHz, but very much less now.

> Erm... most of the audible impact of low bitrates or poor (use of)
> codecs tends to be at much lower frequencies than you may think. What
> the systems tend to rely upon is the listener not having a clear idea of
> that the 'original' sounds like, or it being 'processed' music. The
> devil is all in the detail. The BBC 320k AAC stream for Radio 3 can
> sound excellent. But I've heard 'mp3' net radio at 256k sound awful in
> ways that you could hear even if you hearing cut off at 3kHz!

> As you've twigged from your own experience. It isn't enough to have a
> 'good codec' or a 'high bitrate'. You also need the system to be
> employed with due care and understanding.

Yes, absolutely. I mentioned my experience re the live Schubert Quintet
because it is a work I know reasonably well, the performers were all
professionals and I was about 20 feet away from them. It was a fine
performance and I wondered just how well it might compare with a digitised
rendition, or vice versa, with readily available software as the
performance was still fresh in my mind.

I must say that the digitised version came out of it really quite well.

> The BBC iPlayer provides this.
> The engineers involved have put a lot of care into tweaking the setup to
> optimise the results. And since they provide the browser plugin they
> also supply/control your 'decoder'. My experience is that they have
> consistently hacked away at any problems I've detected, and keep trying
> to make further improvements.

I'll look into that further. Thanks.

> In contrast, some websites that sell 'high resolution' flac/wave
> downloads have actually sold crude upconversions of CD standard
> material. It says '96k/24bit' on the tin, but what is inside is
> 44k/16bit with added aliasing!

> > > > What is being looked for in the file format to enable the
> > > > /correct/ selection of a codec?

> > > Probably just a lookup table linking data type to what works on most
> > > platforms. Just looking, mp3 is MPEG layer 3, which means I can pick
> > > ffmp3, mp3, mad, or mp3acm. There might be others. The default is
> > > ffmp3.

> > OK. I wondered if that might be the case.

> Most of the relevant file/stream formats have a specific 'header' for
> the file, or at intervals along the stream. The replay software looks
> for this. Or at least it should. Some software 'guesses' when given a
> stream, and every now and then gets it wrong. And a stream may contain a
> section that fools/confuses the player. On top of this a file may have a
> declared type, etc. This may or may not be used as a guide.

Noted.

Regards

Brian

[Tony Moore]

On 23 Jan 2012, Brian Bailey <bba...@argonet.co.uk> wrote:

[snip]

> go to Wikipeadia - String Quintet (Schubert) - under - External links -
> click on Recording of the Quintet. It is free to download.

Many thanks for the alert. It turns out that the file is part of a large
collection of chamber music - all free to download - hosted by the
Isabella Stewart Gardner Museum
http://www.gardnermuseum.org/music/listen/music_library?filter=composer

Tony

[Brian Bailey]

In that case, try,

soniventorum.com/soniventorum_archives.html

Super stuff.

Brian

[Tony Moore]

On 24 Jan 2012, Brian Bailey <bba...@argonet.co.uk> wrote:

[snip]

> In that case, try,
>
> soniventorum.com/soniventorum_archives.html
>
> Super stuff.

Indeed, another interesting archive. Many thanks again.

Tony

[Brian Bailey]

> [snip]

> > In that case, try,
> >
> > soniventorum.com/soniventorum_archives.html
> >
> > Super stuff.

> Indeed, another interesting archive. Many thanks again.

Can't resist adding,

www.classicalarchives.com

A fantastic search/research/listening/download archive.

Unfortunately, subscription for full service, but very useful,
nonetheless.

Brian

[Tony Moore]

On 29 Jan 2012, Brian Bailey <bba...@argonet.co.uk> wrote:

[snip]

> www.classicalarchives.com
>
> A fantastic search/research/listening/download archive.
>
> Unfortunately, subscription for full service, but very useful,
> nonetheless.

The website looks to be interesting but, since it makes extensive use of
Javascript, NetSurf won't play.

Tony

[Brian Bailey]

In article <5252437a...@argonet.co.uk>, Brian Bailey
<bba...@argonet.co.uk> wrote:
> Hi All

> [snip]

> > > Be careful though. W7 32-bit version seems to be rare nowadays and
> > > is limited to just over 3 GB of memory (which is probably enough
> > > anyway). But the now-normal W7 64-bit version of Home Premium,
> > > although it can use more than 3GB of RAM, has problems about
> > > Compatability Mode to emulate XP.


> > That's a good point. There are issues too with printer drivers for
> > W7-65 - HP for example point you to the "Universal driver" for most of
> > their older printers, and it can be a case of "one size fits none".

> > Generally though I'd recommend looking at, for example, Misco.co.uk,
> > and make sure you select W7 Pro and pick a good deal. You will get a
> > reasonable machine including W7 pro that way for about the same money
> > as a mainboard and windows license. You stll have to buy a processor,
> > PSU, maybe a case etc... You will get a year's warranty too.

> Many thanks to all who replied.

> After receiving your kind advice I visited a generous friend who let me
> poke around on his very fast Windows 7 Pro 64 bit machine for a while.
> I'm fairly sure that my mind is now made up. I can see definite
> advantages with that OS. If for no other reason than the enhanced colour
> management features for photograph image manipulation, that in itself is
> a significant gain.

Just a brief update, if I may.

I didn't go down the complete route that I had intended to follow. My home
central heating system packed up completely on 3rd Jan (at the same time
that the computer failed) and as it too needed a substantial re-build
during very cold weather there was a certain restriction on the amount of
folding stuff available for the computer. 8-((

Anyway, my machine has been completely re-built and customized with AMD
stuff and Win 7 Pro. At the same time my network was up-graded, including
an NAS drive. It is therefore timely for me to publicly offer my grateful
thanks to several people for equipment supplied and detailed advice given,
namely,

R-Comp, CJEMicros, GeneSys, Virtual Acorn,

and your good selves for the advice given, too.

If I might mention a small company from the 'dark side', who know nothing
of RISC OS, but, in my experience, actually listen and do their best to
offer good service,

Jamie at 'Delta Nine' www,deltanine.co.uk

Up to now, as far as I can tell, VRPC is working OK and I've now got so
many backups that I'm not sure where they all are. 8-)

Oh, before I forget, Win 7 colour management is much more complex than I
had at first thought and a chat with Permajet, followed by a custom
profile for my Canon iP6700D printer, worked wonders. In any event it
bypasses Win 7 colour management.



Regards

Brian

[Brian Bailey]

Hi All

[snip]

> Anyway, my machine has been completely re-built and customized with AMD
> stuff and Win 7 Pro. At the same time my network was up-graded,
> including an NAS drive. It is therefore timely for me to publicly offer
> my grateful thanks to several people for equipment supplied and detailed
> advice given, namely,

A brief addendum, if I may. Certain activities were still overly slow,
like boot time and loading large image files, so I have added a Crucial
Adrenaline SSD Cache - not too expensive.

[snip]

> Up to now, as far as I can tell, VRPC is working OK and I've now got so
> many backups that I'm not sure where they all are. 8-)

VRPC is now *very* responsive. Which was the point of the exercise.

[snip]

Regards

Brian