24bits and computer power

[dan lavry]

When dealing with audio, Where does one need to put a mic pre, an AD,
DA and their stable LOW JITTER clock? Inside the computer? Outside? Of
course outside. The insides of a computer is a bad enviornment for
analog signals. Both from electromanetic interfearance and from
conducted (power supply and grounds) stand point. That is why external
devices for analog yield significantly beter performance than sound
card.

But computers don't supply high voltage. Desktop comuters tend to make
fan noise, which is even less desirable for recording. Laptops make
less noise, and also offer portability, which is an advantage to some.
But the amount of power allowed to be taken out of a computer port is
always limited, especialy when using laptop battery in the field.

So while dramaticaly better than a sound card, designers try and get
the most out of a 5V noisy supply with limited power.

I am surprised at the "hutspa" of folks that try and sell 24 bits
recirders and playback made out of 5V supply.

Lets examine pro gear with say +/-15V DC supply. Given that 24 bits
amounts to 144dB dynamic range, the lowest noise amplifier I can buy
at any cost, With BALANCED 12.28V peak to peak signal (Full scale
AES/EBU) yields a 34.72V peak to peak. That is a huge signal. If I
want the noise to be low enough to
have 144dB, it needs to be about 5 nano volts/square root Hz, for a
20KHz audio!

This is all a theoretical excersize, because no microphone is going to
yield me a 34.72V peak to peak signal! One needs to amplify the small
microphone signal. If one uses say only 24dB gain, than the signal to
noise ratio is down by 24dB which is 3 bits degredation. You are now
in 21 bit world
(126dB range)... That is extreamly good result for high quality pro
gear!

In other words, 24 bits is not just about the digits we use. Some of
those digits are "just there doing nothing". It is like counting real
money with 4
digits for cents. Instead of saying $1.27, we say $1.2700. The right 2
digits are always 00.

Now, lets go into the computers world, With 5V analog supply
enviornment, instaed of the +/-15V above (equivalent to 30VDC). We
must reduce the
signal range by about 6:1 and lost about 15.5dB in signal to noise...
Another 2.5 bits gone...

That too is a terribly optimistic picture. That 5V from the USB bus or
other port is so dirty, that by the time you clean it, you are
probably starting at
3.3VDC, not 5V... and you are entering the 16 bit performance world.
Do not feel bad. Most CD's are less than 16 bits in terms of signal to
noise, and some of them are great sounding!

But, don't spend your money on BS hype. 24bit USB device for $200? You
can not get real 24 bits for $200000. For reasonably low cost, you
will do
very well to really achive 16 bits.

I lot of folks are complaining that the music industry is not what it
used to be, that the customers are not into quality. Maybe we should
clean house first, before complaining about the BS elsewhere. We are
still talking the 24 bits talk and the 192KHz talk. Both are LIES, and
many keep following that that blind path.

Say I try to do the right thing, and tell you that my 24 bits AD is
"only" 21 real bits. Someone else tells you that their unit is a
"great 24 bit". What would you choose? Often it is that "great 24 bit"
that "wins". Even if it performs as 17 bits a 1Khz going to 15bits at
5KHz...

I am also rather disapointed with the low level of consumer
expectations. But I am on the consumers side on the issue od 192KHz
and 24 bits. I am glad to see they do not buy into that level of BS. I
am sad to see that so many of the so called pro's do buy into that
level of BS. It makes me wonder. It should be about good music, good
sound, and if one wishes to talk technical, about 21 real bit and no
more than 96KHz, both are an overkill, but can be be achived at some
effort and cost.

BR
Dan Lavry
www.lavryengineering.com

[Neil Henderson]

>I am also rather disapointed with the low level of consumer
>expectations. But I am on the consumers side on the issue od 192KHz
>and 24 bits. I am glad to see they do not buy into that level of BS. I
>am sad to see that so many of the so called pro's do buy into that
>level of BS. It makes me wonder. It should be about good music, good
>sound, and if one wishes to talk technical, about 21 real bit and no
>more than 96KHz, both are an overkill, but can be be achived at some
>effort and cost.

I use 24-bit @ 88.2k, and that's good enough for me.

NeilH

[Geoff Wood]

"dan lavry" > wrote in message
om...
> When dealing with audio, Where does one need to put a mic pre, an AD,
> DA and their stable LOW JITTER clock? Inside the computer? Outside? Of
> course outside. The insides of a computer is a bad enviornment for
> analog signals. Both from electromanetic interfearance and from
> conducted (power supply and grounds) stand point. That is why external


Why do you put so much effort in writing a huge treatise based on a
fundamentally flawed assumptions.

Many computer interfaces use external electronics boxes with their own
power supplies that could be 1000v , if desired.

However some of the very best interfaces are internal PCI models.

BTW there is +/- 12VDC available from the computer PSU, and there are such
devices as DC-DC converters that will give you pretty much whatever rails
you want, if 'necessary'.

I would have though better of you.


geoff

[Tommi]

"dan lavry" > wrote in message
om...

> In other words, 24 bits is not just about the digits we use. Some of
> those digits are "just there doing nothing". It is like counting real
> money with 4
> digits for cents. Instead of saying $1.27, we say $1.2700. The right 2
> digits are always 00.

I think most engineers have accepted the fact that even though their
converters say "24 bit", they know the dynamic range is always lower than
that 144dB.
I don't see many new converters labeled "16 bit", which is one obvious
reason
why no-one has any converters labeled 16 bit at their disposal. There was
discussion about the 24-bit word lenght some time ago here, and some people
pointed out that even if the last 3 or 4 bits would just be idling there, it
is allegedly easier to keep the word lenght at 24 since computers can handle
the three eight-bit bytes a bit(no pun)easier.

> But, don't spend your money on BS hype. 24bit USB device for $200? You
> can not get real 24 bits for $200000. For reasonably low cost, you
> will do
> very well to really achive 16 bits.

I think most guys know this.

> I lot of folks are complaining that the music industry is not what it
> used to be, that the customers are not into quality. Maybe we should
> clean house first, before complaining about the BS elsewhere. We are
> still talking the 24 bits talk and the 192KHz talk. Both are LIES, and
> many keep following that that blind path.

I don't see the 24-bit thing as big a lie as the 192kHz thing. No-one has
converters with 144dB dynamic range, most have converters with over 96dB
dynamics overall, so if its only 17bit really, it's still better than a 16
bit 96dB converter would be.

> Say I try to do the right thing, and tell you that my 24 bits AD is
> "only" 21 real bits. Someone else tells you that their unit is a
> "great 24 bit". What would you choose? Often it is that "great 24 bit"
> that "wins". Even if it performs as 17 bits a 1Khz going to 15bits at
> 5KHz...

About this I don't know anything. I am sorry to hear if your converters are
ditched just because you're trying to be honest and the other guy is lying.
It shouldn't be that way.

> I am also rather disapointed with the low level of consumer
> expectations. But I am on the consumers side on the issue od 192KHz
> and 24 bits. I am glad to see they do not buy into that level of BS. I
> am sad to see that so many of the so called pro's do buy into that
> level of BS. It makes me wonder. It should be about good music, good
> sound, and if one wishes to talk technical, about 21 real bit and no
> more than 96KHz, both are an overkill, but can be be achived at some
> effort and cost.

But we have to have some standards! If your converter is really 21-bit, and
the other guy's product is really 19.5 bits, what would it be good for to
decimate the final files into the computer as "19.5 bit(that's not even
possible)), 44.1 kHz" and "21 bit, 44.1 kHz"? Personally I think it's wise
to keep the file bit depth at 24-bits, no matter what. Even if your tracks
would have only 70dB's of dynamic range.

[Ethan Winer]

Dan,

Good post, thanks for taking the time to write it. I agree with Geoff that 3
VDC from a USB port is not an issue. Otherwise you hit all the right notes
perfectly in tune.

--Ethan

[Dan Lavy]

"Ethan Winer" <ethanw at ethanwiner dot com> wrote in message
...
> Dan,
>
> Good post, thanks for taking the time to write it. I agree with Geoff that
3
> VDC from a USB port is not an issue. Otherwise you hit all the right notes
> perfectly in tune.
>
> --Ethan
>

Geoff seems to miss my point, which was specificaly aimed at 5V. I obviously
know that one can make converters with AC line voltage. I have designed,
manufactured and sold thousands of channels to just about any high end
facility, many are used with computer worstations.
I wrote this one with 2 reasons in mind:

First, there are a number of devices that use 5V from USB ports, and I make
such a device. I am finding others to not only claim 24 bits, but also to go
through the unbelivable trouble of writing drivers for 24 bits, which in in
my opinion is just so that one can claim that 24 bits. Those drivers are
often pretty unstable, comparing to the generic drivers.

Second, I took the opertunity to elaborate some on that 24 bit issue, which
some understand and others do not. I did take the time to write a "long
post" for those that wish to follow some details.

I have been in this NG for a short time, and feel that some appreciate my
posts. I do not know Geoff and I have to assume he does not know who I am. I
find his post to be by far the most unappreciative and rather disrespecfull
in tone. He said:

"Why do you put so much effort in writing a huge treatise based on a
fundamentally flawed assumptions".

I started the thread and it was about a subject I chose: computer powered
devices using 5V and the 24 bits calim. I feel that Geoff did not respect my
effort. That is OK, I get plenty of respect from folks I respect. But short
of public clear apology from that person (without qualifications) I will
resrict my presence in rec.audio.pro to occasional short anouncments.

Best Regards

Dan Lavry

[Dan Lavy]

> Why do you put so much effort in writing a huge treatise based on a
> fundamentally flawed assumptions....
>
> I would have though better of you.
>
>
> geoff


Mr goeff,

I belive that you chose to go over my email too fast, thus not to understand
that I was addresing specific cases. That is fine. Folks can have
misunderstandings, and differences of opinions. But I find your last
remark - "I would have though better of you" a personal one and very
offensive one. You are way out of line to talk to a grown up person, to a
stranger, in such a way. It is one thing to deal with content. It is another
thing to put out statements about the person.

"I WOULD HAVR THOUGHT BETTER OF YOU" was out of place.

If you do not appologise, I will resrve the right to think of you as one of
those guys that tries to bring themselves up by talking down to someone
else. I am not really waiting for an apology. I think I have you figured
out. Am I wrong?

Dan Lavry

[Geoff Wood]

"Dan Lavy" > wrote in message news:XUoFb.436

> I started the thread and it was about a subject I chose: computer powered
> devices using 5V and the 24 bits calim. I feel that Geoff did not respect
my
> effort. That is OK, I get plenty of respect from folks I respect. But
short
> of public clear apology from that person (without qualifications) I will
> resrict my presence in rec.audio.pro to occasional short anouncments.

Jeepers - touchy,,,,

I unreservedly apologe for not reading sufficiently thoroughly to realise
you were talking specifically about USB-powered devices that do not use
DC-DC up-conversion to provide a sufficient power rail.

As you state, your arguments regarding the 'real bitdepth' applies equally
to all interfaces, irrespective of their powering arrangements, and would
mean that *no* converters could ever be described as "24 bit".

Do you advertise your products as "21 Bit" ?

geoff

[Geoff Wood]

"Dan Lavy" > wrote in message news:yysFb.825

> I belive that you chose to go over my email too fast, thus not to
understand
> that I was addresing specific cases. That is fine. Folks can have
> misunderstandings, and differences of opinions. But I find your last
> remark - "I would have though better of you" a personal one and very
> offensive one.

Clearly now, the point of you whole post was *not* what I had assumed: that
(pretty much all) converter companies, especially those with 24-bit
USB-powered ones, were deceitful in placing their products in the generic
"24 bit" category, and now the 'unstable drivers' extension.

"24 bits" is a valid description for all these devices. It is a converter
data path specification, and a valid broad categorisation of the type of
device. Most manufacturers include a s/n and dynamic range specification,
which are the relevant AUDIO specs. As the objective is audio these are
what should be considered most important. The bittage is a broad generic
slot in which to classify a device. I think most serious users understand
this.


geoff

[Dan Lavy]

"Geoff Wood" -nospam> wrote in message
...
> I unreservedly apologe for not reading sufficiently thoroughly to realise
> you were talking specifically about USB-powered devices that do not use
> DC-DC up-conversion to provide a sufficient power rail.

You missed the point again. The thing I did not like was your statment
"I WOULD HAVR THOUGHT BETTER OF YOU"
I wrote it in capital letters. This was a personal remark about a preson
(I). And it was out of place.
I do not care if you read my post or not. I do not expect everyone to agree
with my points.
But you are out of order to announce to the world that you do not like or
respect someone you do not know.

> Do you advertise your products as "21 Bit" ?
>
> geoff

I have been trying to educate the world about this for 10 years. I was the
FIRST ONE to talk about jitter in AD's for audio. I am making a big point
about the 192KHz issue. I talk about it "all the time".

When I broke the real 20 bit berrier (122dB range), I called it AD122. The
next year I was competing with every Tom Dick and Hary's 24 bits AD's, and
most of them were far from 18 bits! The sales guys wanted it to say 24 bit
just to be able to compete. Next, I was asked to do a 192K. I refused. I go
and post here that there is no 24 bits, and I am a converter maker.

I do not advertize my AD122 MKII. It is on the web. Those that want it know
where to find it, and there are a lot of them out there in good places. And
yes, I brag that it is a true 21 bits and I do explain that there is no 24
bit performance out there, and that 192KHz is a crock. I get a lot of
suport, and at times un called for agravation. There are a lot of fine folks
here on the NG and I do not think it is "fair" to leave all because of a
couple of unpleasent comments, but I am inclined to do so. I am too good an
engineer and too old a man to recive "grades" or unpleasent remarks about
being respected less or other non sense.

Did you get this time? you can argue with the content and technical merit
all day long. It was your personal tone I had problem with. I am NOT your
extanded family and it was your comment
"I WOULD HAVR THOUGHT BETTER OF YOU"
that was out of place.

To which you now added "TOUCHY"

Dan Lavry

[Geoff Wood]

"Dan Lavy" > wrote in message news:dqvFb.1001
>
> You missed the point again. The thing I did not like was your statment
> "I WOULD HAVR THOUGHT BETTER OF YOU"
> I wrote it in capital letters.

I didn't write the capital letters.

I had (maybe mistakenly) assumed that the real point of your otherwise
interesting, detailed, and concise post, was to brand yourself as the
epitome of straight-up honesty, and by implication pretty much all other
manufacturers as liars.

Through your discussion of internal computer noise and USB-power
limitations, I maybe mistakenly assumed you were trying to discredit not
only portable interfaces as inherently inferior, but also all PCI-card based
products. I accept that this was not your intention.

> This was a personal remark about a person
> (I). And it was out of place.
> I do not care if you read my post or not. I do not expect everyone to
agree
> with my points.
> But you are out of order to announce to the world that you do not like or
> respect someone you do not know.

You were the one referring to claims of 24 bit and/or192KHz specs as "lies"
and "BS", and presumably those quoting those attribute as LIARS (my word, my
capitalisation), instead of merely having a different interpretation of
terminology to you. .

24/96/192/whatever are real specs in the digital domain. The audio
equivalence does not approach it. Most here realise that, and accept the
broader generic spec in that context.

What does the "24" in your product description and model numbers actually
refer to ? I do give you credit for the focus of your specs being on the
actually audio performance, as in the AD122 - yes a real milestone..

> Did you get this time? you can argue with the content and technical merit
> all day long. It was your personal tone I had problem with. I am NOT your
> extanded family and it was your comment
> "I WOULD HAVR THOUGHT BETTER OF YOU"

No, sorry, I still don't get it.

> To which you now added "TOUCHY"

geoff

[Geoff Wood]

"Ben Bradley" > wrote in message
>
> This (acoustic noise generated by equipment) is a separate
> discussion. I think many or most recording machines make noise. I've


Dan was referring to electrical noise, which is a valid point, but can also
be all but eliminated by good design and manufacturing, as evidenced by Lynx
and others.

But Dan is talking about the very top end - the last half-bit and decibel -
of the very best equipoment in the world. Where I lose the plot is the
damning of everthing and everybody else for using the accepted generic terms
for general specs.

geoff

[Tommi]

"Geoff Wood" -nospam> wrote in message
...

> "24 bits" is a valid description for all these devices. It is a converter
> data path specification, and a valid broad categorisation of the type of
> device. Most manufacturers include a s/n and dynamic range specification,
> which are the relevant AUDIO specs. As the objective is audio these are
> what should be considered most important. The bittage is a broad generic
> slot in which to classify a device. I think most serious users understand
> this.


I agree in a sense that 24-bit data path is the standard, and also most guys
know that no-one has a 144dB converter yet.
However, it isn't enough that soundcard and converter makers just give a s/n
ratio+dynamic range, there's still dozens of ways to hide the truth about a
device's quality; they can release the specs A-weighted, give very vague THD
values, etc..
I'm sure Mr. Lavry knows most if not all of the different specifications
where the manufacturers of cheap soundcards generally try to hide their
system's inadequacies. It also gets very messy since most things are
frequency-dependant.

[Ethan Winer]

Dan,

> short of public clear apology ... I will restrict my presence <

I'll stay out of the personal feud - there's lots of great technical issues
to argue about! But as someone who has received plenty of abuse for my own
audio views, just make your points and let people say what they will. No big
deal. If you have something to say, go ahead and say it. I'll listen. :->)

--Ethan