Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

DIY Preamp Kits

224 views
Skip to first unread message

Paul

unread,
May 6, 2013, 11:00:02 AM5/6/13
to
Quite a few to choose from:

http://www.diyrecordingequipment.com/preamp/

I'm pretty sure the "clean wire with gain" is the
route for me, and have whittled it down to these
two:

http://fivefishstudios.com/diy-kits

http://www.seventhcircleaudio.com/C84/c84_about.htm

Some people claim the C84 is one of the cleanest.

eth...@ethanwiner.com

unread,
May 6, 2013, 11:15:53 AM5/6/13
to
On Monday, May 6, 2013 11:00:02 AM UTC-4, Paul wrote:
> I'm pretty sure the "clean wire with gain" is the
> route for me

I agree with that approach, and the key to knowing how good these preamps are is by their specs. But Five Fish offers no specs at all, and the specs for Seventh Circle are incorrectly listed. For example, the distortion can't be less than 0 percent. Most of the other specs are not useful either. This might be a great preamp! But you can't tell from those numbers. If you contact those companies and get valid specs, please post them here.

--Ethan

Paul

unread,
May 6, 2013, 11:27:07 AM5/6/13
to
Didn't notice that. Yeah, less than 0% THD....how the hell did
they measure that?

Imagine a negative THD! It would take a distorted signal, and
clean it up! haha!

The C84 looks great, and will fit 8 pres in a 2 space rack (the
Fivefish is only 4 units in a 2 space). However, the C84 doesn't
have the LED bargraph, which is a useful feature of course.....



PStamler

unread,
May 6, 2013, 4:36:50 PM5/6/13
to
On Monday, May 6, 2013 10:15:53 AM UTC-5, eth...@ethanwiner.com wrote:
> On Monday, May 6, 2013 11:00:02 AM UTC-4, Paul wrote:
>
> > I'm pretty sure the "clean wire with gain" is the

> > route for me
>
>
>
> I agree with that approach, and the key to knowing how good these preamps are is by their specs.

Not necessarily. Most specs quoted by manufacturers are horsefeathers, and won't tell you which one sounds closest to a straight wire with gain. Noise figures, if honestly measured and not fudged, can be useful, and high-frequency IM specs, again if all the conditions are properly specified, can tell you things. But99% of the specs quoted in spec sheets are, I repeat, horsefeathers. The ones that are meaningful are almost never given.

Peace,
Paul

Paul

unread,
May 6, 2013, 4:44:39 PM5/6/13
to
Perhaps that's why Fivefish doesn't have them.

Many people think specs are completely useless when it comes to
evaluating audio gear, and that only your own ears will
let you know if you like it or not. I would tend to agree....





Scott Dorsey

unread,
May 6, 2013, 9:39:54 PM5/6/13
to
In article <km94ha$8ki$1...@dont-email.me>, Paul <Quill...@gmail.com> wrote:
> Many people think specs are completely useless when it comes to
>evaluating audio gear, and that only your own ears will
>let you know if you like it or not. I would tend to agree....

That's because they don't know how to make the right measurements.

But, it's true that the manufacturer won't give you much in the way of
useful measurements, so you have to make your own. But looking at
small signal distortion spectrum is going to tell you a lot about how
a preamp sounds. But it won't tell you anything at all about how a
microphone sounds into that preamp's load, and sometimes that is the
most important part.

My suggestion to you is to get some THAT 1512, put them in a box with
careful filtering and decoupling and good layout, and see how you like
it. It's not the greatest thing in the world but it's inexpensive and
inoffensive and most of the hard work of designing the front end has been
done for you. Don't expect it to be a Millennia, but it's probably an
improvement over what you're using now.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

Trevor

unread,
May 6, 2013, 11:59:11 PM5/6/13
to

"Paul" <Quill...@gmail.com> wrote in message
news:km94ha$8ki$1...@dont-email.me...
> Many people think specs are completely useless when it comes to
> evaluating audio gear, and that only your own ears will
> let you know if you like it or not. I would tend to agree....

Gee I wish my ears were better than my measurement equipment! I only listen
to a short list once I've compared specs, then measure for myself.
Speakers and microphones are a problem of course, but not microphone pre's
IMO.

Trevor.


Paul

unread,
May 7, 2013, 2:30:35 AM5/7/13
to
The C84 is supposedly something of a Millenia clone:

http://www.groupdiy.com/the-lab/most-millenia-like-d-i-y-preamp-circuit/

"The Millennia is an optimised version of the double balanced microphone
amplifier design presnted by Graeme John Cohen in AES preprint 2106
(available on the AES website for 20 bucks). Lots of pres are based on
a similar topology the green, some Ameks, the Millennia, Seventh Circle
C84 etc. Fred Forssell has a schemo on his site for a version, though
he noted in one of the boards that the design didn't meet his sonic
standards."

And here:

http://www.diyrecordingequipment.com/topic/how-big-of-a-difference-is-there-between-these-pre-amps-a-fivefish-sc-1-mk-3-or-a-seventh-circle-c84/

"you can compare the C84 to the Millennia HV3. If clean is what you�re
after, the C84 will not disappoint."


Looks like the C84 wins.


PStamler

unread,
May 7, 2013, 3:30:36 AM5/7/13
to
I'm not sure I'd go that far, but the specs *that manufacturers publish* are mostly useless. Check out th opamp tsts that Samuel Groner has done for examples of measurements done right.

I'd like to see Sam do some tests to see how high-frequency IM tests correlate with his various measurements, and how output stage biasing affects the measurements. A couple of very crude tests I've done say that output stage biasing affects the results in ways that may correlate with better sound.

Peace,
Paul

Scott Dorsey

unread,
May 7, 2013, 9:30:32 AM5/7/13
to
In article <kma6rv$uun$1...@dont-email.me>, Paul <Quill...@gmail.com> wrote:
>
> The C84 is supposedly something of a Millenia clone:

No, it's not.

>http://www.groupdiy.com/the-lab/most-millenia-like-d-i-y-preamp-circuit/
>
>"The Millennia is an optimised version of the double balanced microphone
>amplifier design presnted by Graeme John Cohen in AES preprint 2106
>(available on the AES website for 20 bucks). Lots of pres are based on
>a similar topology the green, some Ameks, the Millennia, Seventh Circle
>C84 etc. Fred Forssell has a schemo on his site for a version, though
>he noted in one of the boards that the design didn't meet his sonic
>standards."

This is pretty much the standard transformerless mike input stage that
everybody uses. Including Mackie and Behringer, not just Amek and
Millennia.

It's a good topology. If it weren't good, it wouldn't have become almost
universal. But topology is not everything.

>http://www.diyrecordingequipment.com/topic/how-big-of-a-difference-is-there-between-these-pre-amps-a-fivefish-sc-1-mk-3-or-a-seventh-circle-c84/
>
>"you can compare the C84 to the Millennia HV3. If clean is what you�re
>after, the C84 will not disappoint."

You spend too much time listening to people on the internet who have
probably never even used the stuff they are making comparisons to. Go
try some equipment and see for yourself.

Also be aware that sometimes the most transparent device is not necessarily
the best, for example when it's picking up taxicab radios.

Paul

unread,
May 7, 2013, 9:44:21 AM5/7/13
to
On 5/7/2013 6:30 AM, Scott Dorsey wrote:
> In article <kma6rv$uun$1...@dont-email.me>, Paul <Quill...@gmail.com> wrote:
>>
>> The C84 is supposedly something of a Millenia clone:
>
> No, it's not.
>
>> http://www.groupdiy.com/the-lab/most-millenia-like-d-i-y-preamp-circuit/
>>
>> "The Millennia is an optimised version of the double balanced microphone
>> amplifier design presnted by Graeme John Cohen in AES preprint 2106
>> (available on the AES website for 20 bucks). Lots of pres are based on
>> a similar topology the green, some Ameks, the Millennia, Seventh Circle
>> C84 etc. Fred Forssell has a schemo on his site for a version, though
>> he noted in one of the boards that the design didn't meet his sonic
>> standards."
>
> This is pretty much the standard transformerless mike input stage that
> everybody uses. Including Mackie and Behringer, not just Amek and
> Millennia.
>
> It's a good topology. If it weren't good, it wouldn't have become almost
> universal. But topology is not everything.
>
>> http://www.diyrecordingequipment.com/topic/how-big-of-a-difference-is-there-between-these-pre-amps-a-fivefish-sc-1-mk-3-or-a-seventh-circle-c84/
>>
>> "you can compare the C84 to the Millennia HV3. If clean is what you�re
>> after, the C84 will not disappoint."
>
> You spend too much time listening to people on the internet who have
> probably never even used the stuff they are making comparisons to. Go
> try some equipment and see for yourself.
>
> Also be aware that sometimes the most transparent device is not necessarily
> the best, for example when it's picking up taxicab radios.
> --scott
>

But, but....someone wrote it on the internet, so it's gotta be true!

:P

I might order a couple of C84s, and see what I think.....


Paul

unread,
May 7, 2013, 1:20:24 PM5/7/13
to
On 5/7/2013 6:30 AM, Scott Dorsey wrote:
> In article <kma6rv$uun$1...@dont-email.me>, Paul <Quill...@gmail.com> wrote:
>>
>> The C84 is supposedly something of a Millenia clone:
>
> No, it's not.
>
>> http://www.groupdiy.com/the-lab/most-millenia-like-d-i-y-preamp-circuit/
>>
>> "The Millennia is an optimised version of the double balanced microphone
>> amplifier design presnted by Graeme John Cohen in AES preprint 2106
>> (available on the AES website for 20 bucks). Lots of pres are based on
>> a similar topology the green, some Ameks, the Millennia, Seventh Circle
>> C84 etc. Fred Forssell has a schemo on his site for a version, though
>> he noted in one of the boards that the design didn't meet his sonic
>> standards."
>
> This is pretty much the standard transformerless mike input stage that
> everybody uses. Including Mackie and Behringer, not just Amek and
> Millennia.
>
> It's a good topology. If it weren't good, it wouldn't have become almost
> universal. But topology is not everything.
>
>> http://www.diyrecordingequipment.com/topic/how-big-of-a-difference-is-there-between-these-pre-amps-a-fivefish-sc-1-mk-3-or-a-seventh-circle-c84/
>>
>> "you can compare the C84 to the Millennia HV3. If clean is what you�re
>> after, the C84 will not disappoint."
>
> You spend too much time listening to people on the internet who have
> probably never even used the stuff they are making comparisons to. Go
> try some equipment and see for yourself.
>
> Also be aware that sometimes the most transparent device is not necessarily
> the best, for example when it's picking up taxicab radios.
> --scott
>


Here's what SCA told me:

"The C84 does employ a variation on the Cohen topology, but it is not
intended to be a copy or clone of the Millennia HV3 circuit. Based on
measured performance alone, the C84 meets or beats the HV3 in every way.
We believe it's actually superior in one important way: greatly
increased output current capability.

The best way to evaluate the C84 is in your own room on your own sources.
Try one out for two weeks and return it if not satisfied."


OUR product is always better than THEIRS, right? Haha!




eth...@ethanwiner.com

unread,
May 7, 2013, 1:29:25 PM5/7/13
to
On Monday, May 6, 2013 9:39:54 PM UTC-4, Scott Dorsey wrote:
> That's because they don't know how to make the right measurements.

Clearly that's the case here with Seventh Circle. :->)

> looking at small signal distortion spectrum is going to tell
> you a lot about how a preamp sounds.

Yes, the notion that specs are not as useful as listening is for the most part a flawed view. Specs can tell us everything needed, as long as all the specs are given. And not just with small signals, but at higher levels too if the device has tubes and / or transformers.

If two devices have an audibly high level of distortion, then specs won't tell you which distorted sound will be more pleasing to you. But I'd never buy a preamp expecting it to serve as a fuzz tone. I prefer my audio gear to be perfectly clean, and I'll add color separately when mixing when I can hear everything in context. So in that sense specs can tell you everything: If the frequency response is flat over the audible range, and the distortion and noise are too soft to hear, then the device is audibly transparent. And one transparent device sounds just like every other transparent device. So then you choose based on price and features.

--Ethan

Scott Dorsey

unread,
May 7, 2013, 1:54:11 PM5/7/13
to
In article <kmbcuc$c9v$1...@dont-email.me>, Paul <Quill...@gmail.com> wrote:
>
>"The C84 does employ a variation on the Cohen topology, but it is not
>intended to be a copy or clone of the Millennia HV3 circuit. Based on
>measured performance alone, the C84 meets or beats the HV3 in every way.
> We believe it's actually superior in one important way: greatly
>increased output current capability.

Could be.

The Great River MP-2 has an output impedance of something like nine ohms... it
will drive a small monitor speaker directly. Very cool, and often handy for
driving insanely long lines.

>The best way to evaluate the C84 is in your own room on your own sources.
>Try one out for two weeks and return it if not satisfied."

This is most definitely true.

hank alrich

unread,
May 7, 2013, 2:29:00 PM5/7/13
to
<eth...@ethanwiner.com> wrote:

> Specs can tell us everything needed, as long as all the specs are given.

Everything needed by anyone who doesn't care what it sounds like,
because specs will not yet tell us that.

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

Don Pearce

unread,
May 7, 2013, 3:06:14 PM5/7/13
to
On Tue, 7 May 2013 11:29:00 -0700, walk...@nv.net (hank alrich)
wrote:

><eth...@ethanwiner.com> wrote:
>
>> Specs can tell us everything needed, as long as all the specs are given.
>
>Everything needed by anyone who doesn't care what it sounds like,
>because specs will not yet tell us that.

A proper set of specs will certainly tell you if it doesn't sound like
anything - which for me is the important thing.

d

Trevor

unread,
May 7, 2013, 10:18:43 PM5/7/13
to

"hank alrich" <walk...@nv.net> wrote in message
news:1l2hzu4.14a0nju1aoxkioN%walk...@nv.net...
> <eth...@ethanwiner.com> wrote:
>
>> Specs can tell us everything needed, as long as all the specs are given.
>
> Everything needed by anyone who doesn't care what it sounds like,
> because specs will not yet tell us that.

But you had the snip the part where Ethan says he doesn't want it to sound
like anything! (other than what goes in).

Here it is again :
----------------
>>So in that sense specs can tell you everything: If the frequency response
>>is flat over the audible range, and the distortion and noise are too soft
>>to hear, then the device is audibly transparent. And one transparent
>>device sounds just like every other transparent device.
-----------------

Trevor.


Trevor

unread,
May 7, 2013, 11:11:56 PM5/7/13
to

"Scott Dorsey" <klu...@panix.com> wrote in message
news:kmavlo$sin$1...@panix2.panix.com...
> Also be aware that sometimes the most transparent device is not
> necessarily
> the best, for example when it's picking up taxicab radios.

That's a fair comment, all the measurements and pre purchase listening you
do might not tell you everything that you only find out later. That's where
others experience found on the internet can sometimes be very useful though.

Trevor.


Paul

unread,
May 8, 2013, 12:02:48 AM5/8/13
to
Yes, it may be easier for known professionals to "try before you
buy", but for the rest of us, it's a good idea to ask around about
what has worked for other people.

Especially in my case: Seven Circle accepts returns on
PRE-ASSEMBLED amps, but if it's a kit, and you touch it with a
soldering iron, it's yours to keep and you cannot return it.



Mike Rivers

unread,
May 8, 2013, 7:42:20 AM5/8/13
to
On 5/7/2013 3:06 PM, Don Pearce wrote:

> A proper set of specs will certainly tell you if it doesn't sound like
> anything - which for me is the important thing.

But it seems that, at least with audio equipment, the only way to get a
proper set of specs is to do your own testing. If you're going to all
that trouble, you might as well just listen to it and decide whether you
like it or not.


--
For a good time, call http://mikeriversaudio.wordpress.com

Don Pearce

unread,
May 8, 2013, 12:26:57 PM5/8/13
to
On Wed, 08 May 2013 07:42:20 -0400, Mike Rivers <mm1...@yahoo.com>
wrote:

>On 5/7/2013 3:06 PM, Don Pearce wrote:
>
>> A proper set of specs will certainly tell you if it doesn't sound like
>> anything - which for me is the important thing.
>
>But it seems that, at least with audio equipment, the only way to get a
>proper set of specs is to do your own testing. If you're going to all
>that trouble, you might as well just listen to it and decide whether you
>like it or not.

Maybe true, but that is maybe a reason to put pressure on the
manufacturers to publish useful specs, and not marketing hype

I know this is a lost cause in the consumer market, but it should not
be so in the professional world.

d

Scott Dorsey

unread,
May 8, 2013, 12:38:44 PM5/8/13
to
Don Pearce <sp...@spam.com> wrote:
>On Wed, 08 May 2013 07:42:20 -0400, Mike Rivers <mm1...@yahoo.com>
>wrote:
>
>>On 5/7/2013 3:06 PM, Don Pearce wrote:
>>
>>> A proper set of specs will certainly tell you if it doesn't sound like
>>> anything - which for me is the important thing.
>>
>>But it seems that, at least with audio equipment, the only way to get a
>>proper set of specs is to do your own testing. If you're going to all
>>that trouble, you might as well just listen to it and decide whether you
>>like it or not.
>
>Maybe true, but that is maybe a reason to put pressure on the
>manufacturers to publish useful specs, and not marketing hype

I would tend to agree, but then I go to AES standards committees where
vendors basically come in an explain that we have to make the standard
THIS measurement made THIS way (because that makes their product look
better than the competition) and they can't possibly put THIS OTHER
measurement on the data sheet.

>I know this is a lost cause in the consumer market, but it should not
>be so in the professional world.

The problem is that there is now an enormous overlap between the two,
with a huge number of self-styled "professionals" who have no understanding
of audio theory, measurements, or why the measurements would be made the way
they are.

Mike Rivers

unread,
May 8, 2013, 1:08:32 PM5/8/13
to
On 5/8/2013 12:26 PM, Don Pearce wrote:

>> But it seems that, at least with audio equipment, the only way to get a
>> proper set of specs is to do your own testing. If you're going to all
>> that trouble, you might as well just listen to it and decide whether you
>> like it or not.
>
> Maybe true, but that is maybe a reason to put pressure on the
> manufacturers to publish useful specs, and not marketing hype

It doesn't work. The high end manufacturers generally make equipment
that doesn't make you wonder what the specs are, and they're the ones
most likely to publish usable, if not totally complete, measured
performance data.

The low end manufacturers don't publish detailed performance data for
their products because they don't want their customers to be confused or
bewildered. It's the truth. I've worked with some like that. Then
there's the other reason - they don't know. I've worked with some of
those, too.

And by the way, a specification is what you lay down before you design
and build the product. After you've built it, you test to verify that it
meets the specification. Specifications aren't test or performance data.
Functional specifications usually originate from the marketing
department, but sometimes things, often very useful things like maximum
output level or maximum input sensitivity are simply not specified, but
are stated generically as "+4 dB".

Don Pearce

unread,
May 8, 2013, 1:51:28 PM5/8/13
to
On Wed, 08 May 2013 13:08:32 -0400, Mike Rivers <mm1...@yahoo.com>
wrote:
Well you start with the spec - the design target, if you like. Then
when you are done you generate the data sheet. If you've done your job
well, this can claim performance well ahead of the spec for some
parameters. But you do need to beware of the sheets that use words
like "typical" instead of "maximum".

I know that when I am designing, my eye is always primarily on
achieving the best possible performance regardless of the spec, then
balancing that against the real world necessity of works cost price.

d

Don Pearce

unread,
May 8, 2013, 1:54:41 PM5/8/13
to
All too true, unfortunately. In the past I have come up with a set of
spec parameters together with recommendations for how to measure them.
This has worked well enough in my own radio industry where the specs
carry the authority of ETSI and ITU-R which makes them enforceable.
But of course in the audio industry none of this applies and everybody
does what makes him look good.

d

Scott Dorsey

unread,
May 8, 2013, 2:38:36 PM5/8/13
to
Don Pearce <sp...@spam.com> wrote:
>
>Well you start with the spec - the design target, if you like. Then
>when you are done you generate the data sheet. If you've done your job
>well, this can claim performance well ahead of the spec for some
>parameters. But you do need to beware of the sheets that use words
>like "typical" instead of "maximum".

The design target is usually something like "make a microphone
with TWO tubes in it instead of just one, for under $500 retail"
unfortunately.

Don Pearce

unread,
May 8, 2013, 2:46:08 PM5/8/13
to
On 8 May 2013 14:38:36 -0400, klu...@panix.com (Scott Dorsey) wrote:

>Don Pearce <sp...@spam.com> wrote:
>>
>>Well you start with the spec - the design target, if you like. Then
>>when you are done you generate the data sheet. If you've done your job
>>well, this can claim performance well ahead of the spec for some
>>parameters. But you do need to beware of the sheets that use words
>>like "typical" instead of "maximum".
>
>The design target is usually something like "make a microphone
>with TWO tubes in it instead of just one, for under $500 retail"
>unfortunately.
>--scott

Not if I was writing the target it wouldn't. I would automatically put
in voltage and current requirements, noise figure, gain range,
distortion at several levels, headroom, frequency flatness etc. Those
and a bunch of others would cover performance. Then would come the
human interface requirements and connectivity. An hour or two of that
sort of study and you understand what it is you are going to design.

then you start throwing some candidate architectures at it, and
usually one will stand out above the others. That is the one you
pursue.

If you have done the first part of the job sensibly, all of the spec
points are doable - the question then becomes how much better than
spec you are.

d

Scott Dorsey

unread,
May 8, 2013, 4:04:13 PM5/8/13
to
Don Pearce <sp...@spam.com> wrote:
>On 8 May 2013 14:38:36 -0400, klu...@panix.com (Scott Dorsey) wrote:
>>Don Pearce <sp...@spam.com> wrote:
>>>
>>>Well you start with the spec - the design target, if you like. Then
>>>when you are done you generate the data sheet. If you've done your job
>>>well, this can claim performance well ahead of the spec for some
>>>parameters. But you do need to beware of the sheets that use words
>>>like "typical" instead of "maximum".
>>
>>The design target is usually something like "make a microphone
>>with TWO tubes in it instead of just one, for under $500 retail"
>>unfortunately.
>
>Not if I was writing the target it wouldn't. I would automatically put
>in voltage and current requirements, noise figure, gain range,
>distortion at several levels, headroom, frequency flatness etc. Those
>and a bunch of others would cover performance. Then would come the
>human interface requirements and connectivity. An hour or two of that
>sort of study and you understand what it is you are going to design.

And that, in short, is why engineers don't write design targets anymore,
except at a VERY few engineering-driven companies.

It's changing too at many of the old-line engineering-driven companies too,
as the people who originally set the companies up and drove their culture
are dying off and being replaced by people who really couldn't care less
about the product itself.

Don Pearce

unread,
May 8, 2013, 4:17:28 PM5/8/13
to
Fortunately there are still some of us who do it that way. I design
seriously cheap consumer satellite equipment - both receive and
transmit. Cheap means high volume, and high volume means the highest
standards of engineering - vital because a production yield less than
about 99.5% is unsupportable.

The start of every project is the DRS - design requirements
specification. This goes into great detail about every aspect of the
product, from electrical performance to UV stability of paint. It
becomes the design bible, and usually undergoes a couple of minor
revisions as the design progresses.

A design cycle takes five months, from concept to the production line
in Shanghai producing volume - maybe 100,000 a month. You can't afford
mistakes with that kind of production commitment.

My company is led and run by engineering.

d

geoff

unread,
May 8, 2013, 4:48:03 PM5/8/13
to

"Don Pearce" <sp...@spam.com> wrote in message
news:518a7c8...@news.eternal-september.org...
A published 'low' spec could certainly be believed and may indeed give an
indication of sound (distortion, noise, limited freq respoanse, etc). But I
guess these are rare....

geoff


eth...@ethanwiner.com

unread,
May 8, 2013, 4:57:46 PM5/8/13
to
On Tuesday, May 7, 2013 10:18:43 PM UTC-4, Trevor wrote:
> But you had the snip the part where Ethan says he doesn't want it to sound
> like anything! (other than what goes in).

Exactly. Further, I qualified that specs won't tell if you'll like the sound when a device is not transparent, and that was snipped too. Here's what I said:

"If two devices have an audibly high level of distortion, then specs won't tell you which distorted sound will be more pleasing to you."

It seems some people just see red when their beliefs are challenged, then ignore what was actually stated!

--Ethan

Mike Rivers

unread,
May 8, 2013, 9:17:27 PM5/8/13
to
On 5/8/2013 1:51 PM, Don Pearce wrote:

> Well you start with the spec - the design target, if you like. Then
> when you are done you generate the data sheet. If you've done your job
> well, this can claim performance well ahead of the spec for some
> parameters.

If you have a realistic specification, you'll know the deisgn is
finished when it meets that specification. There should be no need to
exceed it. Now if exceeding the spec comes essentially free, then you
can advertise it of course. But it's incorrect to make something,
measure it, and then publish the results of your measurements and call
it a specification. It's performance measurement data.

But then, what do the customers know? The marketing department needs to
use the terminology (and advertise the numbers) that matches up with
that of the competition.

Sean Conolly

unread,
May 8, 2013, 11:08:41 PM5/8/13
to
"Mike Rivers" <mm1...@yahoo.com> wrote in message
news:kmet8f$e3i$1...@dont-email.me...
From a marketing & sales viewpoint, the litmus test is always "how many more
will we sell if we did that?" . If I publish number that are great, whether
real or imagined, will it actually result in more sales? Or opportunities
for competitors to refute your claims?

In fact, if your market can't tell the difference why bother trying to make
the product better at all? Just make it good enough and lie as far as the
law will allow. That's why they don't use specs and just say things like
"everyone agrees it's the best sounding poo ever!" It can't be disputed
because it's entirely subjective.

Sorry - I'm feeling rather cynical today. I should get some Beats (R)
headphones, that will cheer me up.

Sean


Don Pearce

unread,
May 9, 2013, 12:36:54 AM5/9/13
to
Exceeding the spec normally doesn't involve extra work, it just comes
as a bonus of good engineering. It is never a question of stopping
when you have designed sufficiently.

d

Trevor

unread,
May 11, 2013, 3:53:25 AM5/11/13
to

"Scott Dorsey" <klu...@panix.com> wrote in message
news:kmdv2k$aie$1...@panix2.panix.com...
>>I know this is a lost cause in the consumer market, but it should not
>>be so in the professional world.
>
> The problem is that there is now an enormous overlap between the two,
> with a huge number of self-styled "professionals" who have no
> understanding
> of audio theory, measurements, or why the measurements would be made the
> way
> they are.

Wel that disqualifies them from being considered professional, regardless of
wether that's the market they are targeting or not.

Trevor.



Scott Dorsey

unread,
May 11, 2013, 7:25:48 AM5/11/13
to
By Olympics standards, they are very professional.

Mike Rivers

unread,
May 11, 2013, 9:25:45 AM5/11/13
to
On 5/11/2013 3:53 AM, Trevor wrote:

> Wel that disqualifies them from being considered professional, regardless of
> wether that's the market they are targeting or not.

How important is "professional" to you? Do you ignore anything that you
don't consider professional? Are you willing to use something that does
the job you need, as good as you need it done, for a price that you can
afford? Or would you just prefer to do without it?

I think it's silly to characterize equipment as "professional." It's the
users who are professional or not, and they'll choose the equipment that
works for them. The more "professional" the user is, the more likely
he'll make the wisest choice, _all_ things considered.

Trevor

unread,
May 12, 2013, 12:45:50 AM5/12/13
to

"Mike Rivers" <mm1...@yahoo.com> wrote in message
news:kmlglu$dfn$1...@dont-email.me...
> On 5/11/2013 3:53 AM, Trevor wrote:
>> Wel that disqualifies them from being considered professional, regardless
>> of
>> wether that's the market they are targeting or not.
>
> How important is "professional" to you? Do you ignore anything that you
> don't consider professional? Are you willing to use something that does
> the job you need, as good as you need it done, for a price that you can
> afford? Or would you just prefer to do without it?

Not sure what any of that has to do with a company being professional. I
often buy from companies that are not, and I realise they are *not*.


> I think it's silly to characterize equipment as "professional." It's the
> users who are professional or not, and they'll choose the equipment that
> works for them. The more "professional" the user is, the more likely he'll
> make the wisest choice, _all_ things considered.

Again you missed the point, it's the company that I said wasn't
professional. Their equipment and their users may well be. But if they have
no idea of proper measurements, then it's probably more good luck than good
management.

Trevor.


Mike Rivers

unread,
May 12, 2013, 9:30:58 AM5/12/13
to
On 5/12/2013 12:45 AM, Trevor wrote:

> Again you missed the point, it's the company that I said wasn't
> professional. Their equipment and their users may well be. But if they have
> no idea of proper measurements, then it's probably more good luck than good
> management.

I didn't miss the point, because you didn't make the point until now by
qualifying a non-professional company as one who has no idea of proper
measurements. I don't think these companies that we consider "non-pro"
don't make any measurements or make them incorrectly. It's about the
marketing. They're targeting their products at less sophisticated users
who would be confused by detailed measurements, particularly
measurements that looked worse on paper than what they think is
equivalent measurements by the competition.

I suppose it's fair to call a company "professional" if they publicize
their equipment's performance characteristics in a way that someone who
understands what that parameter means can interpret how good the product
performs. A good example, since we're talking about mic preamps here, is
EIN (equivalent input noise). Marketing departments like to put this in
their product spec sheets because it's a really low number (usually
around -127 dBu) that has "noise" in its name. Some just say it as "-127
dB" which is just plain incorrect. But then, from the Mackie 1604
VLZ-Pro manual:

Equivalent input noise (EIN):
Mic in to Insert Send Out, maximum gain
150 ohm termination
-129.5 dBm unweighted

It's unusual, though correct in this case to use dBm rather than dBu.
But if you calculate voltage from that power into 150 ohms and express
the voltage as dBu, unless I slipped a bead on my abacus, it comes out
to be -126 dBu which is pretty close to my own measurement on that
mixer. But, you know, most people who would buy a Mackie mixer want to
compare one number, not read that the manufacturer actually

They specified where it was measured, under what conditions, and came
out with an accurate number. The Marketing part, however, is still there
in spades:

The number expressed in dBm is lower than when it's expressed in dBu, so
that makes them look better than the folks who specify "-127" Further,
EIN is a useful measurement to designers, but isn't isn't all that
practical for comparison except at maximum gain, and different devices
of the same category have different maximum gains.

So does specifying EIN correctly make Mackie a "professional" company?

hank alrich

unread,
May 12, 2013, 11:09:03 AM5/12/13
to
Plenty of people making a living or a portion thereof using Mackie kit.
My 1202 won't die. It tried, then I cleaned and lubed the pots and
swtiches, thinking maybe I'd get another year out of it. That was
several years ago.

Professional means one thing in marketing, and another thing at the
bank.

eth...@ethanwiner.com

unread,
May 12, 2013, 1:13:18 PM5/12/13
to
On Sunday, May 12, 2013 9:30:58 AM UTC-4, Mike Rivers wrote:
> They're targeting their products at less sophisticated users
> who would be confused by detailed measurements, particularly
> measurements that looked worse on paper than what they think is
> equivalent measurements by the competition.

If only everyone used the same specs, there would be less confusion. I proposed a workshop to address exactly this for the AES show in NYC this coming October. Mike, if I get the gig, would you like to be on the panel?

> So does specifying EIN correctly make Mackie a "professional" company?

Yes! :->)

--Ethan

John Hardy

unread,
May 12, 2013, 1:43:38 PM5/12/13
to
I always thought Mackie was a "professional" company because their XDR
preamps have "high frequency resolution that defines cymbals, triangles
and bells down to the molecular level"

I mean, down to the MOLECULAR LEVEL. Wow!

John Hardy

Mike Rivers

unread,
May 12, 2013, 2:38:15 PM5/12/13
to
On 5/12/2013 1:13 PM, eth...@ethanwiner.com wrote:

> If only everyone used the same specs, there would be less confusion.

True, but equally important, they should measure performance in the same
way. There are a few things that are fairly easy to compare like mic
preamp gain, and it's pretty well understood that amplifier power is
measured into a resistive load, but when you see a spec sheet for a
powered speaker that lists its power at 1,000 watts and yet you can plug
two of them, and the rest of the PA system, into a 15 amp circuit, it
makes you wonder how long they measured that 1,000 watts for.

> I proposed a workshop to address exactly this for the AES show in NYC
> this coming October. Mike, if I get the gig, would you like to be on
> the panel?

Sure, that would be fun. Keep me posted.

Mike Rivers

unread,
May 12, 2013, 2:40:11 PM5/12/13
to
On 5/12/2013 1:43 PM, John Hardy wrote:

> I always thought Mackie was a "professional" company because their XDR
> preamps have "high frequency resolution that defines cymbals, triangles
> and bells down to the molecular level"

They certainly have professional writers for their marketing material.
Always have.

Here In Oregon

unread,
May 12, 2013, 3:30:30 PM5/12/13
to
"Mike Rivers" wrote in message news:kmonfd$4u4$2...@dont-email.me...

On 5/12/2013 1:43 PM, John Hardy wrote:

> I always thought Mackie was a "professional" company because their XDR
> preamps have "high frequency resolution that defines cymbals, triangles
> and bells down to the molecular level"


>>>>They certainly have professional writers for their marketing material.
>>>>Always have.

Including you if I am not mistaken.

Scratch that, I think you wrote their mixer manual or sumthin.





hank alrich

unread,
May 12, 2013, 4:05:46 PM5/12/13
to
Dustbin blather today, John. Now all the molecules are digital.

Ladies and Gentlemen, John Hardy. He'll be here as long as he can. Try
his preamp and see what a professional-grade device delivers.

Thank you. Not kidding.

hank alrich

unread,
May 12, 2013, 4:05:48 PM5/12/13
to
Mike Rivers <mm1...@yahoo.com> wrote:

> On 5/12/2013 1:13 PM, eth...@ethanwiner.com wrote:
>
> > If only everyone used the same specs, there would be less confusion.
>
> True, but equally important, they should measure performance in the same
> way. There are a few things that are fairly easy to compare like mic
> preamp gain, and it's pretty well understood that amplifier power is
> measured into a resistive load, but when you see a spec sheet for a
> powered speaker that lists its power at 1,000 watts and yet you can plug
> two of them, and the rest of the PA system, into a 15 amp circuit, it
> makes you wonder how long they measured that 1,000 watts for.

Until the hamsters keeled over.

> > I proposed a workshop to address exactly this for the AES show in NYC
> > this coming October. Mike, if I get the gig, would you like to be on
> > the panel?
>
> Sure, that would be fun. Keep me posted.

Yea, Team!

John Hardy

unread,
May 12, 2013, 5:18:10 PM5/12/13
to
I just happened to be looking at a pdf from Mackie entitled
"VLZ3_vs_VLZpro.pdf. I just realized that the XDR2 mic preamps are
"better" than the XDR ("molecular level") preamps. Perhaps Mackie has
gone BEYOND THE MOLECULAR LEVEL. (Is that the Monolith that I see...)

John Hardy

William Sommerwerck

unread,
May 12, 2013, 8:53:01 PM5/12/13
to
"Here In Oregon" wrote in message news:kmoqdp$ncb$1...@dont-email.me...
I thought the mixer manual was well-written, but badly organized. I asked Bob
Mackie if I could rewrite it, but he turned me down. They were apparently very
fond of their writer, and unwilling to replace him.





Mark

unread,
May 12, 2013, 9:07:01 PM5/12/13
to

>
> I just happened to be looking at a pdf from Mackie entitled
> "VLZ3_vs_VLZpro.pdf. I just realized that the XDR2 mic preamps are
> "better" than the XDR ("molecular level") preamps. Perhaps Mackie has
> gone BEYOND THE MOLECULAR LEVEL. (Is that the Monolith that I see...)
>
> John Hardy

if the quantum theories turn out to be correct, then "the world"
really is digital and not analog after all.


Mark

hank alrich

unread,
May 12, 2013, 11:04:52 PM5/12/13
to
Whole new twist on "affairs of state".

Trevor

unread,
May 13, 2013, 12:40:15 AM5/13/13
to

"Mike Rivers" <mm1...@yahoo.com> wrote in message
news:kmo5bm$qsf$1...@dont-email.me...
> On 5/12/2013 12:45 AM, Trevor wrote:
>> Again you missed the point, it's the company that I said wasn't
>> professional. Their equipment and their users may well be. But if they
>> have
>> no idea of proper measurements, then it's probably more good luck than
>> good
>> management.
>
> I didn't miss the point, because you didn't make the point until now by
> qualifying a non-professional company as one who has no idea of proper
> measurements.

Yes, it was in reply to the original post which said exactly that. You'd
have to re-read the post before you snipped it of course.
Not when that's all they get right :-)

Trevor.


Trevor

unread,
May 13, 2013, 1:03:34 AM5/13/13
to

"Mike Rivers" <mm1...@yahoo.com> wrote in message
news:kmonfd$4u4$2...@dont-email.me...
> On 5/12/2013 1:43 PM, John Hardy wrote:
>> I always thought Mackie was a "professional" company because their XDR
>> preamps have "high frequency resolution that defines cymbals, triangles
>> and bells down to the molecular level"
>
> They certainly have professional writers for their marketing material.
> Always have.

Yea professional writers, not necessarily professional technical writers
though :-(

Trevor.


Trevor

unread,
May 13, 2013, 1:08:41 AM5/13/13
to

"Mark" <mako...@yahoo.com> wrote in message
news:aca49970-3aab-478e...@v9g2000yqo.googlegroups.com...
> if the quantum theories turn out to be correct,

Correct enough already to make most of our modern technolgy work. Of course
always plenty more to discover.

>then "the world" really is digital and not analog after all.

Like everything in the universe, it's all in how you define those terms.

Trevor.




Message has been deleted

Mike Rivers

unread,
May 13, 2013, 7:52:18 AM5/13/13
to
On 5/12/2013 3:30 PM, Here In Oregon wrote:

> Scratch that, I think you wrote their [Mackie] mixer manual or sumthin.

I wrote manuals and Technical Reference for the HDR24/96 and The Mackie
Compact Mixer Reference Guide (everyone should read it)

Mike Rivers

unread,
May 13, 2013, 7:53:46 AM5/13/13
to
On 5/12/2013 8:53 PM, William Sommerwerck wrote:

> I thought the mixer manual was well-written, but badly organized. I
> asked Bob Mackie if I could rewrite it, but he turned me down.

No wonder. Bob Mackie designed Cher's dresses. Greg Mackie used to run
Mackie the audio company.

William Sommerwerck

unread,
May 13, 2013, 10:23:27 AM5/13/13
to
"Mike Rivers" wrote in message news:kmqk1a$tqu$2...@dont-email.me...
On 5/12/2013 8:53 PM, William Sommerwerck wrote:

>> I thought the mixer manual was well-written, but badly organized.
>> I asked Bob Mackie if I could rewrite it, but he turned me down.

> No wonder. Bob Mackie designed Cher's dresses. Greg Mackie
> used to run Mackie the audio company.

Well, costume design requires a good sense of organization, doesn't it?

I knew there was something wrong when I wrote that, but couldn't figure out
what it was. Both Bob and Greg Mackie ran companies named "Mackie Design",
which caused all sorts of confusion.

Les Cargill

unread,
May 13, 2013, 1:22:47 PM5/13/13
to
I don't think any of Greg's designs eve4r involved a curtain rod.

--
Les Cargill

William Sommerwerck

unread,
May 13, 2013, 1:29:31 PM5/13/13
to
"Les Cargill" wrote in message news:kmr6ui$hl7$1...@dont-email.me...
William Sommerwerck wrote:
> On 5/12/2013 8:53 PM, William Sommerwerck wrote:

>> I knew there was something wrong when I wrote that, but couldn't
>> figure out what it was. Both Bob and Greg Mackie ran companies
>> named "Mackie Design", which caused all sorts of confusion.

> I don't think any of Greg's designs ever involved a curtain rod.

Not even as a buss bar?

(One of the great moments in TV comedy.)
Message has been deleted

Here In Oregon

unread,
May 13, 2013, 3:23:52 PM5/13/13
to
"Mike Rivers" wrote in message

On 5/12/2013 3:30 PM, Here In Oregon wrote:

> Scratch that, I think you wrote their [Mackie] mixer manual or sumthin.

>>>I wrote manuals and Technical Reference for the HDR24/96 and The Mackie
Compact Mixer Reference Guide (everyone should read it)

>>>For a good time, call http://mikeriversaudio.wordpress.com



Wonderful body of work and years ago it gave me some really good ideas.

http://www.mackie.com/support/compactmixer/

I guess that was when you made your first million.


HIO
Here In Oregon
"For a really good time let's hook up"
Wordpress Copyright 2006
All rights reserved
Legal pending
Pro Se



Scott Dorsey

unread,
May 13, 2013, 3:42:19 PM5/13/13
to
In article <kmqk1a$tqu$2...@dont-email.me>, Mike Rivers <mm1...@yahoo.com> wrote:
>On 5/12/2013 8:53 PM, William Sommerwerck wrote:
>
>> I thought the mixer manual was well-written, but badly organized. I
>> asked Bob Mackie if I could rewrite it, but he turned me down.
>
>No wonder. Bob Mackie designed Cher's dresses. Greg Mackie used to run
>Mackie the audio company.

I hate to point this out, but Cher's dresses, unlike some others, never
came apart on-air. Unlike the SR24....
--scott


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

Frank Stearns

unread,
May 15, 2013, 1:25:29 PM5/15/13
to
John Hardy <john...@johnhardyco.com> writes:

snips

>I just happened to be looking at a pdf from Mackie entitled
>"VLZ3_vs_VLZpro.pdf. I just realized that the XDR2 mic preamps are
>"better" than the XDR ("molecular level") preamps. Perhaps Mackie has
>gone BEYOND THE MOLECULAR LEVEL. (Is that the Monolith that I see...)

Quark resolution, of course.

And after that, the next bottom-feeder manu will claim that you can hear the
"strings" from the current String Theory thinking in sub-atomic physics.

Sigh. Give me one of John's preamps any day.

Frank
Mobile Audio
--
.

Frank Stearns

unread,
May 15, 2013, 1:26:48 PM5/15/13
to
walk...@nv.net (hank alrich) writes:

>Mark <mako...@yahoo.com> wrote:

>> >
>> > I just happened to be looking at a pdf from Mackie entitled
>> > "VLZ3_vs_VLZpro.pdf. I just realized that the XDR2 mic preamps are
>> > "better" than the XDR ("molecular level") preamps. Perhaps Mackie has
>> > gone BEYOND THE MOLECULAR LEVEL. (Is that the Monolith that I see...)
>> >
>> > John Hardy
>>
>> if the quantum theories turn out to be correct, then "the world"
>> really is digital and not analog after all.
>>
>>
>> Mark

>Whole new twist on "affairs of state".

Just so long as they are "solidified." (ducking, running)

Frank Stearns

unread,
May 15, 2013, 1:29:35 PM5/15/13
to
Jeff Henig <yom...@yomama.com> writes:
>Does this dress make my diode look fat?

Depends on which direction you're looking.

And, btw, "fat" (silly me, I mean "phat") is always a good thing, right?
Message has been deleted

Scott Dorsey

unread,
May 18, 2013, 6:37:14 PM5/18/13
to
Don Pearce <sp...@spam.com> wrote:
>On Tue, 7 May 2013 11:29:00 -0700, walk...@nv.net (hank alrich)
>wrote:
>
>><eth...@ethanwiner.com> wrote:
>>
>>> Specs can tell us everything needed, as long as all the specs are given.
>>
>>Everything needed by anyone who doesn't care what it sounds like,
>>because specs will not yet tell us that.
>
>A proper set of specs will certainly tell you if it doesn't sound like
>anything - which for me is the important thing.

Unfortunately not. A proper set of specs will only tell you if it DOES
sound like something. It cannot completely rule out some other artifact you
might not be measuring for.

These days we can get pretty good, but listening tests are still needed to
make sure you're measuring the right things properly. Likewise, you need
measurements to know what you're listening to.

eth...@ethanwiner.com

unread,
May 19, 2013, 1:01:42 PM5/19/13
to
On Saturday, May 18, 2013 6:37:14 PM UTC-4, Scott Dorsey wrote:
> A proper set of specs will only tell you if it DOES
> sound like something. It cannot completely rule out some
> other artifact you might not be measuring for.

Maybe we're saying the same thing only differently? I want my gear to be audibly transparent. Specs can definitely tell us that. As long as distortion plus noise is 80 dB or more below the signal, and the response is flat within 1 dB or better over the entire range, that's transparent enough for me. In that case the gear has no sound, which is what I want. Now, if specs show the gear is *not* transparent, then you might need to listen to know if that coloration is pleasing to you.

--Ethan

Scott Dorsey

unread,
May 19, 2013, 2:00:36 PM5/19/13
to
<eth...@ethanwiner.com> wrote:
>On Saturday, May 18, 2013 6:37:14 PM UTC-4, Scott Dorsey wrote:
>> A proper set of specs will only tell you if it DOES=20
>> sound like something. It cannot completely rule out some
>> other artifact you might not be measuring for.
>
>Maybe we're saying the same thing only differently? I want my gear to be au=
>dibly transparent. Specs can definitely tell us that. As long as distortion=
> plus noise is 80 dB or more below the signal, and the response is flat wit=
>hin 1 dB or better over the entire range, that's transparent enough for me.=

Not at all! Let's say the thing makes an unbearable screeching noise every
ten minutes.

The THD+N number is great, the response is flat. But the device is unusable.

No matter how many things you can measure for, there's always one insidious
artifact that will pop up.

In the eighties, we had digital systems that measured great but sounded awful.
It took folks some time to figure out what was going on, and what to measure
that correlated with the system performance, but they did. Once folks learned
what to listen to, they could learn what to measure, then they learned how to
fix it and now most digital gear sounds remarkably good.

In the seventies, we had amplifiers that had an order of magnitude lower
distortion than previous generations of amplifiers, and an order of magnitude
more power, and yet they sounded awful. Folks learned to look at actual
distortion spectra and not just a scalar number, they found the problem,
they fixed it, now there are plenty of solid-state power amps that sound great.

As soon as we get a new technology, whether that be a new circuit topology
or a new device or a totally new concept like digital recording or FM
recording, we get new weird artifacts that we hadn't had to measure before
and didn't think about.

> In that case the gear has no sound, which is what I want. Now, if specs sh=
>ow the gear is *not* transparent, then you might need to listen to know if =
>that coloration is pleasing to you.

You can only show that the gear has none of the colorations that you know to
look for. You can't show that it is devoid of all possible colorations until
you know what all possible colorations are, and every time we get a big advance
in technology we get new and different kinds of distortion.

William Sommerwerck

unread,
May 19, 2013, 2:33:03 PM5/19/13
to
> I want my gear to be audibly transparent.
> Specs can definitely tell us that.

How?

> As long as distortion plus noise is 80 dB or more below
> the signal, and the response is flat within 1 dB or better
> over the entire range, that's transparent enough for me.

I can name at least one amp that has very low measured distortion, but sounds
horrible.

If amps sound different, then they /must/ measure different. But we don't
definitive data on this, because no one wants to do the hard work to establish
the needed correlations.

Scott Dorsey

unread,
May 19, 2013, 2:49:21 PM5/19/13
to
William Sommerwerck <grizzle...@comcast.net> wrote:
>> I want my gear to be audibly transparent.
>> Specs can definitely tell us that.
>
>How?
>
>> As long as distortion plus noise is 80 dB or more below
>> the signal, and the response is flat within 1 dB or better
>> over the entire range, that's transparent enough for me.
>
>I can name at least one amp that has very low measured distortion, but sounds
>horrible.

Very low measured _THD_. I bet a nickel there's a measurement
that will find your problem, and that measurement could be as simple
as 1 Khz square wave response. But then again, it might as as complex
as a high order TIMD measurement.

>If amps sound different, then they /must/ measure different. But we don't
>definitive data on this, because no one wants to do the hard work to establish
>the needed correlations.

But we DO have definitive data on it, going back to Paul Klipsch in the
1940s. The problem is that every time the measurement guys get a good
measurement to consistently detect a problem, the development guys get
a new technology with a new problem.

Rick Ruskin

unread,
May 19, 2013, 2:57:03 PM5/19/13
to
Maybe it's time to learn that if a thing works properly, we shouldn't
try to fix it.
Rick Ruskin
Lion Dog Music - Seattle WA
http://www.liondogmusic.com

Mike Rivers

unread,
May 20, 2013, 8:19:13 AM5/20/13
to
On 5/19/2013 1:01 PM, eth...@ethanwiner.com wrote:
>
> Maybe we're saying the same thing only differently? I want my gear to
> be audibly transparent. Specs can definitely tell us that. As long as
> distortion plus noise is 80 dB or more below the signal, and the
> response is flat within 1 dB or better over the entire range, that's
> transparent enough for me. In that case the gear has no sound, which
> is what I want. Now, if specs show the gear is *not* transparent,
> then you might need to listen to know if that coloration is pleasing
> to you.

A device that measures to your requirements (your specs) for
"transparency" under lab conditions may not actually perform the same
way in real life. Or maybe it will. It depends on what's inside the
black boxes - the device under test, the source that's feeding it, and
the measurement device. An input transformer may only meet your spec for
flat frequency response when fed from one source and not when fed from
another source. In the field of specsmanship, you find the ways to make
measurements that make the product look best to the customer.

Since a preamp is of no practical use without a source and a
destination, you may not, in practice, get the performance that someone
measured in the lab.


--
"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

Drop by http://mikeriversaudio.wordpress.com now and then

eth...@ethanwiner.com

unread,
May 20, 2013, 2:27:25 PM5/20/13
to
On Sunday, May 19, 2013 2:00:36 PM UTC-4, Scott Dorsey wrote:
> Let's say the thing makes an unbearable screeching noise every
> ten minutes.

Well sure, but in that case it's broken, not poor specs. This is not unlike Richard Heyser's special box that spec'd perfectly but sounded awful:

http://www.theabsolutesound.com/articles/insider-with-robert-harley-what-a-test-sounds-like-1/

But it's a parlor trick, not an example of not knowing what or how to measure.

> No matter how many things you can measure for, there's always one insidious
> artifact that will pop up.

It seems to me that most modern gear is clean enough that this is not an issue, but I'm glad to consider a (realistic) example.

> In the eighties, we had digital systems that measured great but sounded awful.

Many great sounding recordings were made with early digital:

http://www.ethanwiner.com/early_digital.htm

> You can only show that the gear has none of the colorations that you know to
> look for.

A null test reveals all added artifacts, including those you might not know to look for. Nulling has been around since the 1940s! See here:

http://www.radiomuseum.org/forum/hewlett_pa_330b_distortion_analyzer.html#1

So if there really was some type of artifact audio engineers didn't know about, it would have been revealed years ago in a null test.

--Ethan

eth...@ethanwiner.com

unread,
May 20, 2013, 2:33:23 PM5/20/13
to
On Sunday, May 19, 2013 2:33:03 PM UTC-4, William Sommerwerck wrote:
> I can name at least one amp that has very low measured distortion,
> but sounds horrible.

Then the distortion was not measured properly. Obviously measuring distortion of a power amp at 1 KHz and 1 milliwatt is not a complete enough test. But measuring can indeed tell us whether a device is audibly transparent if the right things are measured. There's also the null test I've mentioned many times.

> If amps sound different, then they /must/ measure different. But we
> don't definitive data on this, because no one wants to do the hard
> work to establish the needed correlations.

I've done many such tests, which is how I determined that gear is transparent if the distortion plus noise is at least 70 dB below the music, and the response is flat within (let's say) half a dB over the audible range. If you have a concrete example showing otherwise, please share. If I'm wrong, I'd like to know!

--Ethan

eth...@ethanwiner.com

unread,
May 20, 2013, 2:35:52 PM5/20/13
to
On Monday, May 20, 2013 8:19:13 AM UTC-4, Mike Rivers wrote:
> An input transformer may only meet your spec for flat frequency
> response when fed from one source and not when fed from another
> source. In the field of specsmanship, you find the ways to make
> measurements that make the product look best to the customer.

I agree with this. Transformers especially are prone to vary with different input and output impedances. Of course, most gear these days doesn't use transformers.

--Ethan

William Sommerwerck

unread,
May 20, 2013, 3:11:14 PM5/20/13
to
>> If amps sound different, then they /must/ measure different. But we
>> don't definitive data on this, because no one wants to do the hard
>> work to establish the needed correlations.

> I've done many such tests, which is how I determined that gear is
> transparent if the distortion plus noise is at least 70 dB below the
> music, and the response is flat within (let's say) half a dB over the
> audible range. If you have a concrete example showing otherwise,
> please share. If I'm wrong, I'd like to know!

Please tell us the details of these tests.

eth...@ethanwiner.com

unread,
May 21, 2013, 12:36:39 PM5/21/13
to
On Monday, May 20, 2013 3:11:14 PM UTC-4, William Sommerwerck wrote:
> Please tell us the details of these tests.

My AES Audio Myths Workshop video explains a lot about this stuff, and the part that assesses audibility starts around 24:00 into the video:

http://www.youtube.com/watch?v=BYTlN6wjcvQ

All of the original Wave files from that video are available for download on my web site:

http://www.ethanwiner.com/aes/

These articles address related topics:

http://www.ethanwiner.com/loop-back.htm
http://www.ethanwiner.com/audibility.html
http://www.ethanwiner.com/perception.htm
http://www.ethanwiner.com/levels.htm
http://www.ethanwiner.com/dither.html
http://www.ethanwiner.com/converters.html
http://www.ethanwiner.com/believe.html
http://www.ethanwiner.com/audiophoolery.html

--Ethan

Scott Dorsey

unread,
May 21, 2013, 12:54:12 PM5/21/13
to
<eth...@ethanwiner.com> wrote:
>On Sunday, May 19, 2013 2:00:36 PM UTC-4, Scott Dorsey wrote:
>> Let's say the thing makes an unbearable screeching noise every
>> ten minutes.
>
>Well sure, but in that case it's broken, not poor specs. This is not unlike Richard Heyser's special box that spec'd perfectly but sounded awful:
>
>http://www.theabsolutesound.com/articles/insider-with-robert-harley-what-a-test-sounds-like-1/
>
>But it's a parlor trick, not an example of not knowing what or how to measure.

It _is_ a parlor trick, but we still encounter similar problems in the real
world.

Take truncation effects in early digital gear. The full power distortion
measurements were great. But, the systems didn't sound right, people
complained of a "lack of spaciousness" which was really just the reverb
tails being affected by low-level nonlinearity.

So... in the mid-eighties people started taking low level distortion
measurements seriously... and they were coming in pretty awful! The
THD on the PCM1610 was great with a full scale tone, but at -60 dBFS it
was nearly 10%.

So, some folks came up with dither, and by the early 1990s everybody was
using it and the problem was solved.

>Many great sounding recordings were made with early digital:
>
>http://www.ethanwiner.com/early_digital.htm

Yes, and I was making recordings with that stuff at that time, and I assure
you it was not easy. What came out of the PCM-F1 wasn't the same as what
went in, even when it wasn't losing lock...

eth...@ethanwiner.com

unread,
May 22, 2013, 1:28:49 PM5/22/13
to
On Tuesday, May 21, 2013 12:54:12 PM UTC-4, Scott Dorsey wrote:
> Take truncation effects in early digital gear. The full power distortion
> measurements were great. But, the systems didn't sound right, people
> complained of a "lack of spaciousness" which was really just the reverb
> tails being affected by low-level nonlinearity.

People complain about all sorts of stuff that mysteriously disappears when compared in a proper blind test. But obviously I agree that dither reduces distortion. Then again, 10 percent distortion at -60 dB is probably not audible unless you raise the volume to an unnaturally high level. If you set full scale to play back at 100 dB SPL, that puts 10% distortion below -60 at 20 dB SPL. This is much softer than the background noise in most rooms!

--Ethan

Paul

unread,
May 22, 2013, 3:34:50 PM5/22/13
to
People also sometimes mysteriously favor cheap microphones over
expensive ones when they aren't told which wav file is which!

:)





david gourley

unread,
May 22, 2013, 3:38:44 PM5/22/13
to
Paul <Quill...@gmail.com> said...news:knj6eb$shl$1...@dont-email.me:
If it wasn't mentioned already, you may be interested with Lynn Fuston's
Mic / Pre CD. http://www.3daudioinc.com/mics.html


I just read a rave review by him on Ethan Winer's latest book, kudos all.

david

S. King

unread,
May 22, 2013, 3:47:10 PM5/22/13
to
I read Lynn's review as well. Kudos to Ethan. I was surprised that Lynn
ended his review by saying it was a great book for those new to the
business or owrds to that effect. I imagine that I have forgot enough that
I once knew to qualify as a member of that target audience;-)

Steve King

Don Pearce

unread,
May 22, 2013, 4:03:50 PM5/22/13
to
Talking of microphones, for days I have been trying to get decent
audio from Skype using a Sennheiser omni on a desk stand. No luck, it
just booms wherever I put it. Long story short, I have found the place
there is no boom. I took it out of the stand and laid it on the desk.

I suppose now it is officially a PZM. They can really work where all
else fails.

d

Scott Dorsey

unread,
May 22, 2013, 5:45:23 PM5/22/13
to
<eth...@ethanwiner.com> wrote:
>On Tuesday, May 21, 2013 12:54:12 PM UTC-4, Scott Dorsey wrote:
>> Take truncation effects in early digital gear. The full power distortion
>> measurements were great. But, the systems didn't sound right, people
>> complained of a "lack of spaciousness" which was really just the reverb
>> tails being affected by low-level nonlinearity.
>
>People complain about all sorts of stuff that mysteriously disappears when =
>compared in a proper blind test. But obviously I agree that dither reduces =
>distortion. Then again, 10 percent distortion at -60 dB is probably not aud=
>ible unless you raise the volume to an unnaturally high level. If you set f=
>ull scale to play back at 100 dB SPL, that puts 10% distortion below -60 at=
> 20 dB SPL. This is much softer than the background noise in most rooms!

That was the point being made at the time, yes. But people were noticing
a "lack of space" that disappeared consistently when the dither was turned on.
I didn't believe it myself at the time until I heard it.

Scott Dorsey

unread,
May 22, 2013, 5:47:01 PM5/22/13
to
In article <knj6eb$shl$1...@dont-email.me>, Paul <Quill...@gmail.com> wrote:
> People also sometimes mysteriously favor cheap microphones over
>expensive ones when they aren't told which wav file is which!

Absolutely, but that's not always because the cheap one is better, but because
it's the sound they are used to. Give folks a snare drum recorded with a 57
and they'll tend to pick it over just about anything else, because they have
heard so many recordings made that way that they expect recordings to be made
that way. In fact, a spotted snare is inherently unnatural.

Les Cargill

unread,
May 22, 2013, 6:20:31 PM5/22/13
to
Distortion is differently audible than noise. I think it's more
audible than noise.

--
Les Cargill

Scott Dorsey

unread,
May 23, 2013, 9:34:35 AM5/23/13
to
Les Cargill <lcarg...@comcast.com> wrote:
>
>Distortion is differently audible than noise. I think it's more
>audible than noise.

Depends on what kind of distortion it is, too. You might not notice as much
as 2% third harmonic distortion, but sixth harmonic you'll notice nearly two
orders of magnitude lower. This is why a scalar THD measurement is worse than
useless unless you are comparing things with identical spectra.

eth...@ethanwiner.com

unread,
May 23, 2013, 1:16:13 PM5/23/13
to
On Wednesday, May 22, 2013 3:47:10 PM UTC-4, S. King wrote:
> I read Lynn's review as well. Kudos to Ethan. I was surprised that Lynn
> ended his review by saying it was a great book for those new to the
> business or words to that effect. I imagine that I have forgot enough that
> I once knew to qualify as a member of that target audience;-)

I was surprised by that too, because there's a lot of content even experienced professionals will benefit from. My book does start at the beginning including everything a newbie needs too, but that's just to be as complete as possible so anyone at any level can jump in.

--Ethan

Trevor

unread,
May 24, 2013, 1:55:33 AM5/24/13
to

<eth...@ethanwiner.com> wrote in message
news:660ec134-b25f-40f2...@googlegroups.com...
--------------------------

True, but if you had 10dB headroom when recording, and 110dB playback level
(not uncommon) then you are up to 40db SPL and definitely audible. Of course
since that has been more than fixed so long ago, the debate is all rather
pointless now.

Trevor.


0 new messages