digital faders distort sound

Status
Not open for further replies.
G

Guest

Guest
Archived from groups: rec.audio.pro (More info?)

I've noticed that the further down you go with a digital fader
(attenuation) the more lifeless and digitized the signal becomes.
when you have a full mix and obviously you need to keep the faders
down a bit this is a concern, I believe.

opinions?
 
G

Guest

Guest
Archived from groups: rec.audio.pro (More info?)

In article <25150933.0410190415.773e8144@posting.google.com>,
maxdm <maxdimario@aliceposta.it> wrote:
>I've noticed that the further down you go with a digital fader
>(attenuation) the more lifeless and digitized the signal becomes.
> when you have a full mix and obviously you need to keep the faders
>down a bit this is a concern, I believe.
>
>opinions?

This is a sign you have some sort of low-level nonlinearity issues going
on. This should not happen unless you have something nasty like truncation
going on, or if you have a buss that is way too low resolution. What sort
of console have you noticed this on?
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."
 
G

Guest

Guest
Archived from groups: rec.audio.pro (More info?)

All the bits are not being used and you get a degrigation in signal, are you
getting the signal from a tape machine or your computer? THat also makes a
difference. The tape machine should not have ny distortion, if it does your
mixer crossover is doing that. If it is a digital signal, try to add some
dither to the signal, maybe that could help.

"maxdm" <maxdimario@aliceposta.it> wrote in message
news:25150933.0410190415.773e8144@posting.google.com...
> I've noticed that the further down you go with a digital fader
> (attenuation) the more lifeless and digitized the signal becomes.
> when you have a full mix and obviously you need to keep the faders
> down a bit this is a concern, I believe.
>
> opinions?
 
G

Guest

Guest
Archived from groups: rec.audio.pro (More info?)

maxdm wrote:

> I've noticed that the further down you go with a digital fader
> (attenuation) the more lifeless and digitized the signal becomes.
> when you have a full mix and obviously you need to keep the faders
> down a bit this is a concern, I believe.
>
> opinions?

Define the terms Lifeless and Digitized as you used them here please.
Also on what makes/models of mixers do you notice this effect?
I have not noticed this effect, but I'm primarily a data and power
guy.

--Dale
 
G

Guest

Guest
Archived from groups: rec.audio.pro (More info?)

"maxdm" <maxdimario@aliceposta.it> wrote in message
news:25150933.0410190415.773e8144@posting.google.com

> I've noticed that the further down you go with a digital fader
> (attenuation) the more lifeless and digitized the signal becomes.

I've noticed the same thing with analog faders. Don't you think that might
be due to the Fletcher Munson effect?

> when you have a full mix and obviously you need to keep the faders down a
> bit this is a concern, I believe.

What might be an example of a "full mix"?

> opinions?

When it comes to human perception, psychoacoustics rules.
 
G

Guest

Guest
Archived from groups: rec.audio.pro (More info?)

> What might be an example of a "full mix"?

When you are using all the channels on your board you must turn the gain
down somewhere. Either on your channels or your master gain.
 
G

Guest

Guest
Archived from groups: rec.audio.pro (More info?)

"Chip Borton" <chip@cybermesa.com> wrote in message
news:xLadnbr_Dc6A7ujcRVn-oA@comcast.com
>> What might be an example of a "full mix"?
>
> When you are using all the channels on your board you must turn the
> gain down somewhere. Either on your channels or your master gain.

Thinking about modern digital consoles, I think about products like the
Yamaha 01V, 02R, DM1000, DM2000 etc. These products are all have
properly-dithered 24 bit busses and faders which have in excess of 140 dB
dynamic range.

AFAIK, its pretty difficult to achive more than 120-130 dB with the
corresponding analog circuitry.

In either the case of good analog or digital faders, the signal disappears
into the noise floor as you turn it down. The digital faders and busses have
the advantage of at least 10 dB more dynamic range, or as one might say, a
10 dB lower noise floor.

Please explain why a digital noise floor that would be at least 10 dB lower
than analog would some kind of sonic disadvantage.
 

Co

Distinguished
Oct 19, 2004
8
0
18,510
Archived from groups: rec.audio.pro (More info?)

This is true.......

Digital is an aproximation of the signal already, when attenuating it doesnt
get any better.
I have noticed on my behringer crossover DCX2496 that the main faders do
major damage but the crossover functions are "not" noticable. I geuss the
had to compromise because of proccecing power........

CO


"maxdm" <maxdimario@aliceposta.it> wrote in message
news:25150933.0410190415.773e8144@posting.google.com...
> I've noticed that the further down you go with a digital fader
> (attenuation) the more lifeless and digitized the signal becomes.
> when you have a full mix and obviously you need to keep the faders
> down a bit this is a concern, I believe.
>
> opinions?
 
G

Guest

Guest
Archived from groups: rec.audio.pro (More info?)

CO wrote:

> I have noticed on my behringer crossover DCX2496 that the main faders do
> major damage

Please expand on that and explain in more detail what you're hearing and
what settings deliver "major damage". I'm curious.

--
ha
 
G

Guest

Guest
Archived from groups: rec.audio.pro (More info?)

"Sylvester Malik" <sylvestermalik@sympatico.ca> wrote in message news:<yBadd.22979$J16.888601@news20.bellglobal.com>...
> All the bits are not being used and you get a degrigation in signal, are you
> getting the signal from a tape machine or your computer? THat also makes a
> difference. The tape machine should not have ny distortion, if it does your
> mixer crossover is doing that. If it is a digital signal, try to add some
> dither to the signal, maybe that could help.
>
Please, use your ears, and let me know.
I notice a big difference from -15 to -30 dB etc.

try lowering your master if you ain't got a full range track.
it sounds different to me and to another friend (who has sold 15 million records)
 
G

Guest

Guest
Archived from groups: rec.audio.pro (More info?)

CO wrote:
> Digital is an aproximation of the signal already,

So is analog...

- Logan
 
G

Guest

Guest
Archived from groups: rec.audio.pro (More info?)

In article <w9cdd.25055$rY1.7345@fe2.texas.rr.com>,
Logan Shaw <lshaw-usenet@austin.rr.com> wrote:
>CO wrote:
>> Digital is an aproximation of the signal already,
>
>So is analog...

It's right. Only live music is any good at all.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."
 
G

Guest

Guest
Archived from groups: rec.audio.pro (More info?)

Arny Krueger wrote:

> Don't you think that might be due to the Fletcher Munson effect?

There's a new plugin to compensate for that, the Meltcher Bunsen plug,
which warms up the signal considerably and puts lots more life into
playback.

--
ha
 
G

Guest

Guest
Archived from groups: rec.audio.pro (More info?)

"Arny Krueger" <arnyk@hotpop.com> wrote in message news:<bIydnRYjq5-h8ejcRVn-oA@comcast.com>...
> "maxdm" <maxdimario@aliceposta.it> wrote in message
> news:25150933.0410190415.773e8144@posting.google.com
>
> > I've noticed that the further down you go with a digital fader
> > (attenuation) the more lifeless and digitized the signal becomes.
>
> I've noticed the same thing with analog faders. Don't you think that might
> be due to the Fletcher Munson effect?

Not exactly.
Try this:

Place the cursor at - 20 db and RAISE THE CONTROL ROOM VOLUME 20 dB so
that the listening volume is the SAME.

You will notice loss of detail etc. and on some of the DAW you will
notice a different spectral balance (brighter darker etc.)
this won't happen with a proper analog 600/600 attenuator (fader) but
a similar effect *might* come through on an improperly biased VCA I
guess.
 
G

Guest

Guest
Archived from groups: rec.audio.pro (More info?)

Arny Krueger wrote:
> I've noticed the same thing with analog faders. Don't you think that might
> be due to the Fletcher Munson effect?
:
:
> When it comes to human perception, psychoacoustics rules.

Oh, speaking of that, I've been wondering about something for a while,
and I'm curious if maybe someone here has any insight.

Sometimes I find myself hearing music at a level that is quite low.
Not quite down to the level of barely perceptible, but more at the
kind of level where I can hear the music well enough to recognize
parts of it but not well enough to necessarily follow along. For
instance, this happens in my car, where the aftermarket CD unit I
installed has a habit of resetting its volume level to something
way too low whenever I turn off the ignition. It is low enough
that, because of ambient noise, I often don't even notice that it's
playing.

Anyway, if I'm listening at this low level, I get a certain idea of
the pitch, i.e. the key that the song is in. If I then turn the
volume up so that it's normal listening level (or even below normal
listening level but high enough that I can clearly make everything
out), then the pitch seems to increase by something on the order
of 1/2 of a semitone to a full semitone! If I then lower the volume
again, it doesn't seem to go back down in pitch and instead stays
where it was at the higher volume, as if my brain has somehow "locked
on" to the proper pitch once it got a taste of it.

Perhaps coincidentally or perhaps not, I am trying to develop perfect
pitch, and I have a habit of coming in about 1 semitone low or maybe
a little less than that. For instance, I'll hear a song in my head
and start singing it, and I'll be in C#, but then I'll put the CD in
and play it to compare, and the song is actually in D.

So am I a brain-damaged freak or is this a known phenomenon? :)

- Logan
 
G

Guest

Guest
Archived from groups: rec.audio.pro (More info?)

On Tue, 19 Oct 2004 08:15:56 -0400, maxdm wrote
(in article <25150933.0410190415.773e8144@posting.google.com>):

> I've noticed that the further down you go with a digital fader
> (attenuation) the more lifeless and digitized the signal becomes.
> when you have a full mix and obviously you need to keep the faders
> down a bit this is a concern, I believe.
>
> opinions?

If lower volume is achieved by fewer bits (and it doesn't HAVE to be that
way), you end up with fewer bits describing each sound. That makes for grainy
sound.

You may also be being tricked by the fletcher munsen curve.

Regards,

Ty Ford



-- Ty Ford's equipment reviews, audio samples, rates and other audiocentric
stuff are at www.tyford.com
 

Co

Distinguished
Oct 19, 2004
8
0
18,510
Archived from groups: rec.audio.pro (More info?)

> > I have noticed on my behringer crossover DCX2496 that the main faders do
> > major damage
>
> Please expand on that and explain in more detail what you're hearing and
> what settings deliver "major damage". I'm curious.



OK, major is a bit overdone.........

On both the in and outputs the software algorithm is of a much lesser
quality than the crossover attenuation possibilities (which can cover full
bandwidth, and also digital) on my modified (powersupply) behringer DCX2496
OS version 1.16 iirc. Thats why i think they had to put the DSP power where
they needed it most.

Even a 0,1db change (compared to off) gives a noticable loss of ambient
sound on a high resolution setup.
Indeed it becomes a little lifeless.

I cant say all digital attenuation is bad perse..

Resistive fading could do even more damage though depending on the
brand/materials/setup.
There are a couple of pots that standout though........

Using a transformer to do this retains te highest quality though.

Regards, Coolin
 
G

Guest

Guest
Archived from groups: rec.audio.pro (More info?)

"CO" <nomail@plea.se> wrote in message
news:10nc2jmho9etraa@corp.supernews.com
>>> I have noticed on my behringer crossover DCX2496 that the main
>>> faders do major damage
>>
>> Please expand on that and explain in more detail what you're hearing
>> and what settings deliver "major damage". I'm curious.

> OK, major is a bit overdone.........

> On both the in and outputs the software algorithm is of a much lesser
> quality than the crossover attenuation possibilities (which can cover
> full bandwidth, and also digital) on my modified (powersupply)
> behringer DCX2496 OS version 1.16 iirc. Thats why i think they had to
> put the DSP power where they needed it most.

> Even a 0,1db change (compared to off) gives a noticable loss of
> ambient sound on a high resolution setup.

In any resonable comparison, no way, Jose'

> Indeed it becomes a little lifeless.

Anybody who rants and raves about the audibility of 0.1 dB changes is either
dealing with grotesquely defective equipment or some kind of biased
listening test.


> I cant say all digital attenuation is bad perse..

> Resistive fading could do even more damage though depending on the
> brand/materials/setup.
> There are a couple of pots that standout though........
>
> Using a transformer to do this retains te highest quality though.

This very much depends on the quality of the transformer. Since your
baseline DCX2496 is budget-priced digital, it's logical to use a
budget-priced transformer for the comparison. BTW, the budget priced
transoformer will still cost far more. The sound quality losses in
budget-grade transformers can easily be so egregious that the claim that
"Using a transformer to do this retains te highest quality though." is
summarily demolished.
 
G

Guest

Guest
Archived from groups: rec.audio.pro (More info?)

On Wed, 20 Oct 2004 02:23:49 -0700, maxdm wrote:

> "Arny Krueger" <arnyk@hotpop.com> wrote in message news:<bIydnRYjq5-h8ejcRVn-oA@comcast.com>...
>> "maxdm" <maxdimario@aliceposta.it> wrote in message
>> news:25150933.0410190415.773e8144@posting.google.com
>>
>> > I've noticed that the further down you go with a digital fader
>> > (attenuation) the more lifeless and digitized the signal becomes.
>>
>> I've noticed the same thing with analog faders. Don't you think that might
>> be due to the Fletcher Munson effect?
>
> Not exactly.
> Try this:
>
> Place the cursor at - 20 db and RAISE THE CONTROL ROOM VOLUME 20 dB so
> that the listening volume is the SAME.
>
> You will notice loss of detail etc. and on some of the DAW you will
> notice a different spectral balance (brighter darker etc.)
> this won't happen with a proper analog 600/600 attenuator (fader) but
> a similar effect *might* come through on an improperly biased VCA I
> guess.

That effect could be your D/As not working so well with less bits?

Try setting up some routing in a DAW so the same signal is routed
both to the master outs and through as many channels and busses as
possible, all with their faders at different levels. Then, put a polarity
inversion plug on the last bus, and adjust the level so it nulls with the
original signal.

In Cubase SX the faders don't seem to lose information, so even with the
most tortuous routing, it will still null.
 
G

Guest

Guest
Archived from groups: rec.audio.pro (More info?)

In article <10nc2jmho9etraa@corp.supernews.com>, CO <nomail@plea.se> wrote:
>
>On both the in and outputs the software algorithm is of a much lesser
>quality than the crossover attenuation possibilities (which can cover full
>bandwidth, and also digital) on my modified (powersupply) behringer DCX2496
>OS version 1.16 iirc. Thats why i think they had to put the DSP power where
>they needed it most.
>
>Even a 0,1db change (compared to off) gives a noticable loss of ambient
>sound on a high resolution setup.
>Indeed it becomes a little lifeless.

This is still very bad, and I would not consider this acceptable. Have you
submitted a bug report to the Behringer guys?

>I cant say all digital attenuation is bad perse..

It shouldn't be, and there is really no reason for it to be bad at all because
CPU time is cheap. It takes so little to do it right that there is no reason
not to.

>Resistive fading could do even more damage though depending on the
>brand/materials/setup.
>There are a couple of pots that standout though........
>
>Using a transformer to do this retains te highest quality though.

I disagree. A good resistive attenuator should be a lot cleaner than an
autotransformer fader. If you don't like pots, stepped attenuators are just
fine.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."
 

Co

Distinguished
Oct 19, 2004
8
0
18,510
Archived from groups: rec.audio.pro (More info?)

> > Even a 0,1db change (compared to off) gives a noticable loss of
> > ambient sound on a high resolution setup.
>
> In any resonable comparison, no way, Jose'


Your missing my point. I'm not saying 0.1db is audible but the lesser
quality digital fader kickes in at -0.1 db as compared to 0 db (as in OFF,
as in not in the loop.)
So my point is if not well executed it can be harmfull. To a lesser degree
top quality digital attenuation might do harm as well.


> > Using a transformer to do this retains te highest quality though.
>
> This very much depends on the quality of the transformer. Since your
> baseline DCX2496 is budget-priced digital, it's logical to use a
> budget-priced transformer for the comparison. BTW, the budget priced
> transoformer will still cost far more. The sound quality losses in
> budget-grade transformers can easily be so egregious that the claim that
> "Using a transformer to do this retains te highest quality though." is
> summarily demolished.

Indeed, expensive transformers.......
 
G

Guest

Guest
Archived from groups: rec.audio.pro (More info?)

"CO" <nomail@plea.se> wrote in message
news:10ncpsgh49ikpbd@corp.supernews.com

>>> Even a 0,1db change (compared to off) gives a noticable loss of
>>> ambient sound on a high resolution setup.

>> In any resonable comparison, no way, Jose'

> Your missing my point.

Really?

> I'm not saying 0.1db is audible but the lesser
> quality digital fader kickes in at -0.1 db as compared to 0 db (as in
> OFF, as in not in the loop.)

Good quality digital faders can be shown to be completely undetectable when
set for up to a 1-2 0.1's of a dB because of the audibility of the change
for larger amounts of attenuation. Good digital faders are inaudible when
compared to a good quality analog fader set for the identically same amount
of attenuation. I've done this experiment many times, but blind testing can
be very helpful to avoid listener bias, particularly among listeners who are
analog bigots or simply poorly-informed.

Digital attenuators can be readily built with incredibly large amounts of
resolution for example 100's of dBs, while analog attenuators have
real-world resolution issues in the 120-130 dB that I described in the part
of my post that you decided to ignore. The digital attenuators in my
favorite mixer have dynamic range around 1,000 dB. Of course there is no
audible difference between dynamic range of 120 dB or 1000 dB.

> So my point is if not well executed it can be harmfull. To a lesser
> degree top quality digital attenuation might do harm as well.

The sky might fall. So much for global claims including the word "might".

>>> Using a transformer to do this retains te highest quality though.
>>
>> This very much depends on the quality of the transformer. Since your
>> baseline DCX2496 is budget-priced digital, it's logical to use a
>> budget-priced transformer for the comparison. BTW, the budget priced
>> transoformer will still cost far more. The sound quality losses in
>> budget-grade transformers can easily be so egregious that the claim
>> that "Using a transformer to do this retains te highest quality
>> though." is summarily demolished.

> Indeed, expensive transformers.......

....still corrupt the signal far more than well-executed but far less
expensive active and passive approaches, given that we're talking about
circuitry inside the console.

Transformers can be justified as interfacing devices to the outside world
when that world is atypically noisy, or as a means to obtain desired kinds
audible coloration.
 
G

Guest

Guest
Archived from groups: rec.audio.pro (More info?)

"Logan Shaw" <lshaw-usenet@austin.rr.com> wrote in message
news:0Indd.22261$Rf4.19316@fe2.texas.rr.com...
> Arny Krueger wrote:
>> I've noticed the same thing with analog faders. Don't you think that
>> might
>> be due to the Fletcher Munson effect?
> :
> :
>> When it comes to human perception, psychoacoustics rules.
>
> Oh, speaking of that, I've been wondering about something for a while,
> and I'm curious if maybe someone here has any insight.
>
> Sometimes I find myself hearing music at a level that is quite low.
> Not quite down to the level of barely perceptible, but more at the
> kind of level where I can hear the music well enough to recognize
> parts of it but not well enough to necessarily follow along. For
> instance, this happens in my car, where the aftermarket CD unit I
> installed has a habit of resetting its volume level to something
> way too low whenever I turn off the ignition. It is low enough
> that, because of ambient noise, I often don't even notice that it's
> playing.
>
> Anyway, if I'm listening at this low level, I get a certain idea of
> the pitch, i.e. the key that the song is in. If I then turn the
> volume up so that it's normal listening level (or even below normal
> listening level but high enough that I can clearly make everything
> out), then the pitch seems to increase by something on the order
> of 1/2 of a semitone to a full semitone! If I then lower the volume
> again, it doesn't seem to go back down in pitch and instead stays
> where it was at the higher volume, as if my brain has somehow "locked
> on" to the proper pitch once it got a taste of it.
>
> Perhaps coincidentally or perhaps not, I am trying to develop perfect
> pitch, and I have a habit of coming in about 1 semitone low or maybe
> a little less than that. For instance, I'll hear a song in my head
> and start singing it, and I'll be in C#, but then I'll put the CD in
> and play it to compare, and the song is actually in D.
>
> So am I a brain-damaged freak or is this a known phenomenon? :)

I don't know if you could say it's a known phenomenon exactly as you
describe it, but it's similar to an effect that happens to singers who are
trying to sing while monitoring with full cans (i.e. NOT leaving one ear
open/off), or even sometimes onstage/live... if the music's too loud -
relative to their own vocal volume in the mix - they tend to go sharp, if
the music's too soft - again, relative to their vocal volume - they tend to
go flat. This is certainly not universal, but it's common enough to where
I'd bet most everyone here has seen it happen.

Neil Henderson
 
G

Guest

Guest
Archived from groups: rec.audio.pro (More info?)

> Try setting up some routing in a DAW so the same signal is routed
> both to the master outs and through as many channels and busses as
> possible, all with their faders at different levels. Then, put a polarity
> inversion plug on the last bus, and adjust the level so it nulls with the
> original signal.
>
> In Cubase SX the faders don't seem to lose information, so even with the
> most tortuous routing, it will still null.

Digital distortion appears above 5KHz to my ears.
try doing the same thing above with a FULL RANGE WELL RECORDED SIGNAL
(a group playing or a solo performance that has some high end to it).

the fletcher munsen curve is volume--dependent, and the whole idea is
to listen to the sound always AT THE SAME LEVEL by using an external
volume control you trust to be reasonably distortion -free

AF sine waves are not music, they are absolutely the easyest waveform
to reproduce, and this is why manufacturers love sine wave tests so
much!
 

Co

Distinguished
Oct 19, 2004
8
0
18,510
Archived from groups: rec.audio.pro (More info?)

> I don't know if you could say it's a known phenomenon exactly as you
> describe it, but it's similar to an effect that happens to singers who are
> trying to sing while monitoring with full cans (i.e. NOT leaving one ear
> open/off), or even sometimes onstage/live... if the music's too loud -
> relative to their own vocal volume in the mix - they tend to go sharp, if
> the music's too soft - again, relative to their vocal volume - they tend
to
> go flat. This is certainly not universal, but it's common enough to where
> I'd bet most everyone here has seen it happen.


Ive read something about not being able to judge pitch precisely at higher
volumes.

I think this is the reason when you see live (loud) music on TV you think
boy that sounds false,
but if you were there at the same concert you wouldnt have noticed.

Coolin
 
Status
Not open for further replies.