R/E/P Community

Please login or register.

Login with username, password and session length
Advanced search  

Pages: [1] 2  All   Go Down

Author Topic: Volume Attenuation in Digital vs. Analog Domains  (Read 13009 times)

Rob Darling

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 294
Volume Attenuation in Digital vs. Analog Domains
« on: December 18, 2004, 04:40:09 PM »

I have a client moving toward being 100% digital in their studio, so the console, which has been their jukebox and volume control for some time, will be eliminated.  There will be multiple inputs to their DAW, and two outs to their monitors.  2 track will be iTunes.  Not an unusual setup at all.

Now here is the question.  At what point will using digital attenuation inside the mixer of their soundcard (in this case a Lynx L22 feeding an external convertor tbd) to control volume, or possibly some kind of plug-in that does metering, etc. inside of AudioHijack Pro, not live up to what analog attenuation will do?  And that's theoretical.  In practice, just how good does the analog attenuation have to be to warrant not using digital?  What are the issues in the tradeoffs between the two?

Logged
____________________
rob darling
rob@robdarling.net
www.robdarling.net

Albert

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 56
Re: Volume Attenuation in Digital vs. Analog Domains
« Reply #1 on: December 19, 2004, 12:25:24 PM »

I've found it more practical to have a volume control situated post DAW outputs, between the DAW and the monitors.

A speaker switcher box with volume control is ideal for this, and there are a number on the market. I use a Presonus Central Station, but there are higher end units from companies like Grace Designs, Crane Song, Coleman Audio and many others that will do a great job as well.

There is a market for these kinds of units since so many people have done exactly what your client is doing, ditch their analog board and mix in the box. So these outboard monitor sections are pretty popular right now.

I went 100% digital a while back and found it a nightmare to try to control the master volume from software. I really didn't like it. If the Lynx box has a volume pot on the front of it, make sure the box is within reach, because that in all likelyhood is what will end up getting used to control master volume. Especially if your client is used to grabbing knobs.

Incidentally, I'm in the process of switching back to analog mixing. I found my productivity declined considerably when I was all digital. I was never happy with the sound either, particularily when it came to mixing. However, I had amazing routing, and could do anything I wanted in that regard. It was just such a time-suck to use the system, always troubleshooting, always some stupid something or other needed attention. I came to the realization that I don't need every option in the world, all I need is what I need. So I've identified that. The setup I'm in the process of changing over to is analog, and designed to be flexible but also relatively simple and fast to work on.

Your client might be different, but I was not happy with an all digital rig. I know that's not what you were asking, but I thought I'd throw that in there. Best of luck to you and your client.
Logged

danlavry

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 997
Re: Volume Attenuation in Digital vs. Analog Domains
« Reply #2 on: December 19, 2004, 04:08:49 PM »

[email

robdarling@mail.com[/email] wrote on Sat, 18 December 2004 21:40]I have a client moving toward being 100% digital in their studio, so the console, which has been their jukebox and volume control for some time, will be eliminated.  There will be multiple inputs to their DAW, and two outs to their monitors.  2 track will be iTunes.  Not an unusual setup at all.

Now here is the question.  At what point will using digital attenuation inside the mixer of their soundcard (in this case a Lynx L22 feeding an external convertor tbd) to control volume, or possibly some kind of plug-in that does metering, etc. inside of AudioHijack Pro, not live up to what analog attenuation will do?  And that's theoretical.  In practice, just how good does the analog attenuation have to be to warrant not using digital?  What are the issues in the tradeoffs between the two?




A “purely” digital attenuation (via computation, not via digitally controlled hardware)? One should be very careful with it. Say for example I have a CD (16 bits) and I wanted to lower the volume by 6dB. The digital way of doing it is to multiply the digital signal be 1/2  and that will amount to shifting all the bits by one so that the most significant bit (MSB) data is now the MSB-1 location and so on… You are now playing the same music with a 15 bit digital audio… You see, the “new” MSB is fixed at zero, and the old LSB (least significant bit) has no where to shift to (there is no 17 bits, only 16).

The same concept is true for say -10dB, which is not a multiple of  power of 2. It shifts one bit (as above) and than you lose about 4 more db on the LSB side… If you want to attenuate by 50dB, you have less than 6 bit quality…

And of course the other issue is dither. When you do such digital attenuation, you are in fact reducing word length so dither is needed. In fact the further you go the more it is needed. That will of course cost you in the dynamic range department. If you do not dither, you will end up with noise modulation and distortions.

Simply put, with a rather crude example, if you have 24 bits digital material, and the noise it say -110dB (about 18 bits), but the digital bus is 100 bits wide, multiplying and/or shifting by say 10 bits is great, but the noise “moved from bit 18 to bit 28” (This is very crudely stated but stated to illustrate the point). At the end you have to remove the lower 76 bits, and end up with 24 bits. The release format does not allow you to have 28 bits to express the same dynamic range, so you lost 4 bits. The top 10 bits are now set to 0, and you 24 bits is gone, It is now at a 14 bits performance level… That is in theory. What happens if your DA can not track the lower bits bits of that 24 word? Say it is a 18 bit DA, than you lose 6 more bits and you are now at 8 bit digital audio. With dither it is even less. Of course that 10 bits example is 60dB attenuation, pretty extreme, but it does illustrate the point...

In my view, analog attenuation is not problem free either, nor is it a walk in the park, but it beats the pure digital attenuation “blind folded with hands tied behind the back”. One (that is not a golden ear) may be able to “get by” with a small range of digital attenuation - each 6dB attenuation cost you 1 bit in theory, and that is a lot of damage.

BR
Dan Lavry
Logged

Rob Darling

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 294
Re: Volume Attenuation in Digital vs. Analog Domains
« Reply #3 on: December 19, 2004, 09:38:29 PM »

Hmm.  Ok, this is where I get confused.  So if I have a 16 bit cd playing back and the convertor is 24 bits, why don't I have 8 bits to play with (48 db) before I get down to where I'm no longer using 16 bits to represent the audio?   (btw, the mixer app is 48bit and has selectable dithers if this info matters.)
Logged
____________________
rob darling
rob@robdarling.net
www.robdarling.net

bobkatz

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 2926
Re: Volume Attenuation in Digital vs. Analog Domains
« Reply #4 on: December 19, 2004, 09:54:35 PM »

There are some excellent digital "volume controls", but

a) these should be integrated with the finest, lowest noise "24-bit" DAC available

and

b) you have to carefully balance the gain structure of the system so that the dither noise which is required in the digital volume control is at an acoustic level significantly below threshold of audibility.

The DCS Elgar DAC is extremely transparent and incorporates a (64 bit float I believe) digital volume control. As long as you apply not too much analog gain post the DAC then the volume control can be run at a fairly high range so that the distortion that the calculation creates is well below the dither noise floor.

TC Electronic's system 6000 contains a monitor matrix with a 48 bit volume control which is then dithered to 24 bits on its output and directly feeds its DAC. You then adjust the analog output level of the DAC for optimum SPL at the loudspeakers. I cannot see any quantization distortion on the FFT of the TC until the attenuator approaches -20 dB so there is a lot of leeway, and you try to run it so the attenuator sits between about -10 and 0 dB.

I was very happy using the TC as my primary volume control for a long time. I did a month's worth of A/B comparisons between that and a stepped active analog attenuator at matched levels and finally came to the conclusion that the TC is audibly transparent when as I said the gain structure is so optimized.

The main reason I wanted the TC is for a remote-controlled volume control, because I could and did easily build an analog stepped control that was not remote controlled.

There is no such thing as a free lunch, and I eventually "graduated" to the Cranesong Avocet, which is a "purist" analog remote controlled volume control. The topology is minimalist using discrete opamps, and gold-sealed relays switching resistive attenuators in front of the opamps. There are no VCAs or MDACs in the Avocet signal path. When you get something that transparent, it can sound as good or better than the digital approach and you instantly get several advantages:

1) ergonomics. Instant comparison of up to 6 sources and ability to match levels between them. It has a built in DAC and a router that switches 3 digital PCM sources and also 3 analog sources with a direct analog path.

2) compare any source, analog or digital, at any sample rate, without breaking down the sample rate of your system. With the TC, since it is locked to the master sample rate of your system, you are forced to be at the sample rate of your system, and if someone comes in with a CD and you're at 96 kHz, you're up the creek. Of course you could dedicate the TC to be slaved to whatever source you wish, but then you lose the ability to use the other engines as processors. And what if you want to evaluate a "pure" analog source such as a phonograph record or analog tape or cassette? Should you be penalized and be forced to do an A/D conversion to use your digital volume control?

For these and many other reasons, I'm sticking with my "well-made" analog volume control.

Hope this overview helps!


BK
Logged
There are two kinds of fools,
One says-this is old and therefore good.
The other says-this is new and therefore better."

No trees were killed in the sending of this message. However a large number of
electrons were terribly inconvenienced.

Timeline

  • Full Member
  • ***
  • Offline Offline
  • Posts: 215
Re: Volume Attenuation in Digital vs. Analog Domains
« Reply #5 on: December 20, 2004, 11:41:44 AM »


I have found that when I keep the levels near peak on my 2 MOTU HD192's the sound is brighter and punchier on the analog OPs.  

I mix in my DAW and combine tracks analog outside the box. I was told long ago that if I was to output all tracks individually at near the top of the digital range out the IO to analog, sound and clarity would improve.  My ending track counts are usually about 48 audio and some virtual midi's. So far to do that I would need two more 12 channel IO' so I mix in comped pairs through a pair of 12 channel IOs. Close but mixing within a DAW changes digital levels pre the DA converter. I think this is how most AD/DA's work, is that right?

Why don't manufacturers provide digitally controlled DAC level mixing controls within their IOs. Would this improve the sound?

It seems this part of the digital landscape is far from optimized.

Thanks
Gary Brandt
Producer Engineer.
TL
Logged
Gary Brandt
Timeline

danlavry

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 997
Re: Volume Attenuation in Digital vs. Analog Domains
« Reply #6 on: December 20, 2004, 12:27:36 PM »

[email

robdarling@mail.com[/email] wrote on Mon, 20 December 2004 02:38]Hmm.  Ok, this is where I get confused.  So if I have a 16 bit cd playing back and the convertor is 24 bits, why don't I have 8 bits to play with (48 db) before I get down to where I'm no longer using 16 bits to represent the audio?   (btw, the mixer app is 48bit and has selectable dithers if this info matters.)


There are at least 2 points that enter the equation:
1. How many bits in the final format?
2. How many "real bits" does you DA provide?

When you attenuate, you need to make sure that the data content in the lower bits can:
1. be there in the final format - have enough bits...
2. be above the noise floor (of the DA or anything else in the chain)

I guess my first post did not state it clearly enough.

Yes, IN THEORY you could take say a 16 bit material and shift it by 8 bits and play it back through your 24 bit DA (which is 144dB dynamic range) into the amp and speakers (that are 144dB dynamic range)...

The ear is more forgiving at very low levels (where you do not hear as well). If you attenuate a significant amount digitally, than amplify back in analog, you will hear the impact of reduced bits much more clearly. If you do not amplify with analog, the effect is more subtle, but it is certainly there, and one does not have to be a golden ear to make that observation.

Regards
Dan Lavry

Logged

Rob Darling

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 294
Re: Volume Attenuation in Digital vs. Analog Domains
« Reply #7 on: December 20, 2004, 01:11:01 PM »

Thanks Dan.  I just wanted to make sure I was clear.  My understanding is as follows:  Assuming 16 bits as a final destination to be monitoring for, technically, if I put 16 bits out through a 24 bit conversion, I have 8 bits, 48 db, before I get down to losing real information digitally.  Where I run into a problem is that the analog noise floor of my convertor, if it is very good, will probably be somewhere around -118, so I really only have 4 bits, 32db, before I get into my 16 bits of info.   Is this correct?  

My one additional question- do we continue to hear low-level info behind the noise of the convertor, or is it completely lost?  I've assumed this is analog noise, but is there a digital component to the noise that makes it an actual barrier instead of a mask?

Where this came up was at the point that Bob was discussing.  My ears were saying something bad was happening before I got to my theoretical breakpoint, but I didn't have any way of really testing it to get any empirical data.  In my practical tests of this mode of operation at my own place, I felt like I was starting to feel like I was missing something when attenuation reached the mid 20's, using the Mytek 896.  But with the amps I'm using and gain structure of my system, this happens at a point where the volume sensitivity of my ears is crossing over.  I was wondering what of this was my ears and what was real degradation of the sound.  The idea of quantization distortion and when it would set in isn't really something I understand and I was wondering at what point it would become an issue.

I will try lowering my reference level so that the max output level is more closely matched to the maximum value of my amps to see what kind of results this gives me and see if I can reduce the degree of attenuation the system will ultimately need.

Best,

Thanks again for all the great stuff in these forums,

rob.
Logged
____________________
rob darling
rob@robdarling.net
www.robdarling.net

danlavry

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 997
Re: Volume Attenuation in Digital vs. Analog Domains
« Reply #8 on: December 20, 2004, 02:26:53 PM »

"Thanks Dan. I just wanted to make sure I was clear. My understanding is as follows: Assuming 16 bits as a final destination to be monitoring for, technically, if I put 16 bits out through a 24 bit conversion, I have 8 bits, 48 db, before I get down to losing real information digitally….”

If you put 16 bits out through 24 bit conversion? I assume you mean if you start with 24 bits AD. Well, you seem to insist on using theoretical gear. There is no such thing as a real 24 bit AD. The last few bits are noise, and they do not count. In fact, very often a real mic- mic pre – AD does not even yield 16 bits. That is what matters! There is NO WAY you have 48 dB extra!!! In most cases you have 0dB to spare!!! Many if not most CD’s out there do not meet the 96dB noise floor. And than comes the DA, and the dither…

“Where I run into a problem is that the analog noise floor of my convertor, if it is very good, will probably be somewhere around -118, so I really only have 4 bits, 32db, before I get into my 16 bits of info. Is this correct?”

If your final format destination is 16 bits, a DA with 17-18 bits performance will not be the limiting factor. The 16 bit format will be the limiting factor.  

”My one additional question do we continue to hear low-level info behind the noise of the convertor, or is it completely lost? I've assumed this is analog noise, but is there a digital component to the noise that makes it an actual barrier instead of a mask?”

No you do not! I saw a lot of such claims about hearing below the noise floor, and that is marketing BS. You do not hear under the noise floor unless you are willing to play marketing tricks with definitions. When we talk about 6dB/bit, or 96dB dynamic range for a 16 bit machine, we are using a definition that “lumps” everything into one number.
That 96dB number is the COMBINED NOISE ENERGY AT ALL THE AUDIO FREQUENCIES. If you look at an FFT plot, the noise level at each individual frequency is at about -136dB. So when you put say a -100dB 1KHz tone there, it is much higher than -136dB and you can see it on the FFT. You can also hear it if the volume is turned up. But it does not mean you can hear under the noise.

Think of a forest with 100 young trees, each one is 1 foot tall. The combined height is 100 feet, and it may serves to describe the general amount of wood. In fact, you can have 50 trees at 1/2  foot and 50 at 1.5 foot and have the same general number 100 feet combined.  
Do you want to claim you are taller than a forest with 100 feet height?

So assuming we are talking about n bits, and also assuming the FFT noise floor is flat (equal length trees),  the FFT noise will be about 43dB lower (for 22KHz audio bandwidth  20*log(sqrt(22050))=43dB) than what we normally refer to as noise (the number combining the noise at all the frequencies).    

“The idea of quantization distortion and when it would set in isn't really something I understand and I was wondering at what point it would become an issue.”

Absolutely, quantization distortion will become an issue and that is why I said you will need to have dither added. If you want to understand it better, there are some papers on my web www.lavryengineering.com. Look at “Do you need 20 bits” (under the support section).

Regards
Dan Lavry
Logged

Albert

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 56
Re: Volume Attenuation in Digital vs. Analog Domains
« Reply #9 on: December 21, 2004, 08:52:23 PM »

Getting back to the original question, is there any possible advantage to digital master volume control? Why go through all these technical hoops and risk possible signal degredation with digital volume control when doing it in analog is much simpler and avoids the pitfalls?

Whether it's analog or digital, the user still has to input the information to the system regarding what volume level is required. There is no advantage like time savings, as the act of changing the volume will take the same amount of effort whether digital or analog. So I don't see the advantage of going through all that math to get to the same or worse audio as an analog volume pot. This seems like a "why bother" type scenario. Maybe I'm missing something?

As I've mentioned in another thread, I'm a composer not an engineer, and view studio issues from a practical but probably simplistic point of view.
Logged

danlavry

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 997
Re: Volume Attenuation in Digital vs. Analog Domains
« Reply #10 on: December 22, 2004, 01:26:40 PM »

Albert wrote on Wed, 22 December 2004 01:52

Getting back to the original question, is there any possible advantage to digital master volume control? Why go through all these technical hoops and risk possible signal degredation with digital volume control when doing it in analog is much simpler and avoids the pitfalls?

Whether it's analog or digital, the user still has to input the information to the system regarding what volume level is required. There is no advantage like time savings, as the act of changing the volume will take the same amount of effort whether digital or analog. So I don't see the advantage of going through all that math to get to the same or worse audio as an analog volume pot. This seems like a "why bother" type scenario. Maybe I'm missing something?

As I've mentioned in another thread, I'm a composer not an engineer, and view studio issues from a practical but probably simplistic point of view.


As I've mentioned in another thread, I'm a composer not an engineer, and view studio issues from a practical but probably simplistic point of view.”

One should make a distinction between 2 different types of volume adjustments:

The first type is the for adjusting the volume level for listening, such as on your stereo, TV and much more … and for audio pro’s in the monitoring of audio production. That type of volume control is best achieved (from sound quality standpoint) by analog circuitry.

The second type of volume level adjustment is restricted to the audio production environment. It is about placing the audio data properly within the bounds of a format. Too much level will cause clipping of the signal. Too low a level gets the signal too close to the noise floor (reduces the dynamic range).

People that do recording need to set the level so the peaks (loudness wise) of the music is reasonably high, close to the maximum but not clipping. We can talk about that later. If someone “falls short”, the way to correct for it is to multiply the digital values. Also, adding digital tracks, mixing tracks and other processes may end up with “too much” or “too little” in terms of digital level. thus requiring digital gain or attenuation.

Analog does not do any good when the process must be digital. Digital may emulate analog gain or attenuation but will cause the problems we mentioned. In my view, analog attenuation and gain is better whenever it can be done. Digital attenuation may be a necessary evil of the production process, but digital attenuation lowers the dynamic range, and digital gain amplifies not just the signal but also the noise floor… You do what you need to do, but only when you have to…

Regards
Dan Lavry    



Logged

sdevino

  • Full Member
  • ***
  • Offline Offline
  • Posts: 153
Re: Volume Attenuation in Digital vs. Analog Domains
« Reply #11 on: December 26, 2004, 12:00:11 AM »

This thread feels like a trip back in time. I cannot believe anyone is still advocating tracking as close to full scale as possible. This makes no sense. Am I misunderstanding your position Dan?

The example of reducing a 16 bit signal by 6 dB resulting in a 15 bit signal is not a real world situation and is misleading.

First , in a well designed contemporary digital mixer there are 48 fixed bit values.
The 16 bit word at full scale is using a group of bits somewhere in the middle of that range. There are 8 bits ( about 48 dB) of headroom for gain before clipping and 8 bits (48 dB) of room above the noise floor. So you would have to reduce the signal by 48 dB or more before you start loosing bits.  If you have to do either of these things you should be fired for gain structure miss management. Of course there is no additional error detectable unless you re-amplify.

Second, at the point of conversion (ADC). The resolution of the ADC is not effected by the input signal level.  A 24 bit converter is always capable of capturing 256 times the detail of a 16 bit converter whether the input signal is at full scale or -60DBFS. This is defined by the smallest discernible voltage the converters can detect. A 24 bit converter will detect voltage changes 144 dB below full scale or  (in reality) somewhere a few dB below the analog noise floor (at about -90-100dBFS). This is as good as any analog system can possibly do.

Third: Analog attenuation while easier to implement and make convenient will always be much nosier from a purist standpoint than digital. Digital scaling in the DAW can be managed with errors as small as -288 dB. But analog attenuation results in the generation of noise caused by power dissipation within the attenuator. It also typically results in RFI and suffers the inaccuracies of the very expensive but not very accurate precision resistors used to create a stepped attenuator. The best resistor in the world is probably more than 10,000x less accurate than a cheap 16 bit converter.

Am I just reading your comments out of context Dan?
Logged
Steve Devino

Granite Rocks Recording Studios
Studio gear design and setup

Jason Phair

  • Full Member
  • ***
  • Offline Offline
  • Posts: 140
Re: Volume Attenuation in Digital vs. Analog Domains
« Reply #12 on: December 26, 2004, 12:42:31 PM »

I mix in Samplitude 6, all ITB.  When summing all of my tracks together, the two-buss ALWAYS goes over, and I generally have to attenuate it by anywhere from 6-10dB.  I've been doing this at the master buss instead of each individual tracks.  Is this then, a BAD thing?  Is attenuation more "destructive" at the 2-buss or an individual track?  I suppose the same could be asked on a smaller scale of subgrouping, because I do a lot of that ITB as well.  Thank you.  

PS I learn a hell of a lot from this particular forum, but damn it makes me feel stupid sometimes.
Logged
Jason Phair
Advanced Production Group
Dunkirk, NY
http://www.advancedproductiongroup.com

Formerly:
Sound Services
SUNY Fredonia


Get that fucking thing off my vocal will ya?

Thanks.

sdevino

  • Full Member
  • ***
  • Offline Offline
  • Posts: 153
Re: Volume Attenuation in Digital vs. Analog Domains
« Reply #13 on: December 26, 2004, 02:01:32 PM »

The master fader works by changing a multiplication factor that is applied to all the channels feeding it in the mixer.

So changing the master fader is essentially the same as turning down all the individual channel levels at once.

Master faders implemented in this fashion do not consume any additional CPU or TDM power which is pretty cool.

Steve
Logged
Steve Devino

Granite Rocks Recording Studios
Studio gear design and setup

bobkatz

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 2926
Re: Volume Attenuation in Digital vs. Analog Domains
« Reply #14 on: December 26, 2004, 05:00:48 PM »

phairphunk wrote on Sun, 26 December 2004 12:42

I mix in Samplitude 6, all ITB.  When summing all of my tracks together, the two-buss ALWAYS goes over, and I generally have to attenuate it by anywhere from 6-10dB.  I've been doing this at the master buss instead of each individual tracks.  Is this then, a BAD thing?  Is attenuation more "destructive" at the 2-buss or an individual track?  I suppose the same could be asked on a smaller scale of subgrouping, because I do a lot of that ITB as well.  Thank you.  



Samplitude, like most native programs, computes in floating point. Therefore it doesn't matter if you attenuate in the master or in the inputs. It doesn't matter where you attenuate, anywhere else in the program EXCEPT where "the rubber meets the road", where you interface with the outside world. That is, if you are feeding an external converter or a digital processor via AES/EBU or SPDIF, then you should at least look at the levels at that point in the circuit and see that they are peaking up to, let's say, -10 dBFS, and no harm as far as I can see, up to -3 dBFS.

Hope this helps,

BK
Logged
There are two kinds of fools,
One says-this is old and therefore good.
The other says-this is new and therefore better."

No trees were killed in the sending of this message. However a large number of
electrons were terribly inconvenienced.
Pages: [1] 2  All   Go Up
 

Site Hosted By Ashdown Technologies, Inc.

Page created in 0.05 seconds with 19 queries.