Dither is a prime suspect

[StanD]

DACs, dithering and Amps, none of which change soundstage. Soundstage is composed of recorded spacial cues, e.g., phase, delay which can be perceptually affected by FR of which is flat in DACs as well as Amps so don't bark up those trees.

 

[FFBookman]

  DACs, dithering and Amps, none of which change soundstage. Soundstage is composed of recorded spacial cues, e.g., phase, delay which can be perceptually affected by FR of which is flat in DACs as well as Amps so don't bark up those trees.

DAC's render the data into electrical signal and amps (ahem) amplify so how could they not affect soundstage presentation?  Or are you talking not of build issues, but theoretical perfect operation?  I don't understand how they couldn't affect soundstage.  If 1 DAC is smearing or anti-aliasing something more than the other, there could be differences in the soundstage.
 
If the pre-delay of a certain sound is at a certain frequency it could be masked.   Amps introduce distortions that could be out of sync across the stereo channels. The final decay of a cymbal might end sooner and bring the overall soundstage size smaller.
 
I don't understand where you say "e.g., phase, delay which can be perceptually affected by FR of which is flat in DACs as well as Amps."
 
What is flat in DAC's and amps?

 

[StanD]

 
DAC's render the data into electrical signal and amps (ahem) amplify so how could they not affect soundstage presentation?  Or are you talking not of build issues, but theoretical perfect operation?  I don't understand how they couldn't affect soundstage.  If 1 DAC is smearing or anti-aliasing something more than the other, there could be differences in the soundstage.
 
If the pre-delay of a certain sound is at a certain frequency it could be masked.   Amps introduce distortions that could be out of sync across the stereo channels. The final decay of a cymbal might end sooner and bring the overall soundstage size smaller.
 
I don't understand where you say "e.g., phase, delay which can be perceptually affected by FR of which is flat in DACs as well as Amps."
 
What is flat in DAC's and amps?

The Frequency Response of DACs and Amps are flat within the human passband, Decay, phase and delay are not affected by DACs and Amps to be humanly perceptible. Phase and/or delay can be affected by Headphones (maybe not delay), Speakers and Acoustics (room or head). Any decent Amp has distortion well below human perception, read up on JDD. The rest of your statements are just part of a fishing expedition. 

 

 

[arnyk]

 
DAC's render the data into electrical signal and amps (ahem) amplify so how could they not affect soundstage presentation?Amps."
 
What is flat in DAC's and amps?

 
Any device whose alterations of frequency response, nonlinear distortion, noise and phase response are within high but readily achievable limits makes no audible alternations to the signals it processes. That includes timbre, pace, rhythm, timing and soundstage.
 
For example, consider 4 inches of 12 gauge bare (air insulated - all glass and porcelain knob and tube wiring) copper wire. Do you believe that its inclusion in an audio system would necessarily be audible?  I think that most persons would dismiss that idea because 4 inches of 12 gauge copper wire are very impressive robust, both mechanically and electrically. How much do we have to extend that piece of wire before the electrical alterations it makes becomes audible?  Now, we are having a reasonable discussion.
 
Your post seems to have at least two serious errors:
 
(1) It seems to be based on the idea that everything audibly changes every signal that it handles.
 
(2) It seems to be based on the idea that human perception is either impossible to measure or estimate, or such estimations have set the technical bar so high that no real world audio component can step up to the challenge.
 
At some level of detail nothing is flat, and its variations from perfection are readily measured and quantified. However, the ear and brain are relatively accepting of minor discrepancies, and simply don't notice them no matter how hard or long or what you listen to.

 

[FFBookman]

  The Frequency Response of DACs and Amps are flat within the human passband, Decay, phase and delay are not affected by DACs and Amps to be humanly perceptible. Phase and/or delay can be affected by Headphones (maybe not delay), Speakers and Acoustics (room or head). Any decent Amp has distortion well below human perception, read up on JDD. The rest of your statements are just part of a fishing expedition. 

 

I'm not going to out-labrat you guys, or out geek you. I am fishing, the thread is a fishing expedition accusing dither of affecting my mixes!  Perhaps because I have mixed music for awhile (and listen to older stuff) I hear things you don't?  The spacing, the air, the snare, the cymbal decay, the attack, the panning, the timing, the overall timbre/color of the music, the amount of voices it can present - these are things that I hear in my own and other people's mixes, and these are the things that are reduced as resolution is reduced.
 
I think many people see audio as 2D on a screen - 2 waveforms rolling by.  Audio is never on a screen and it is never 2D. It lives in real space, in real air, and in my opinion music can only really be studied fully in open air.  It is determined by the size of the room it was recorded in, the mic type, the amount of post-processing done, and yes the final resolution. 
 
Perhaps there's a dither out there that will successfully mask the reduction of a grid of 16,000,000 points down to 64,000, but I haven't heard it yet. That is the resolution reduction  when going from 24/192 to 16/44. 

 

[StanD]

 
I'm not going to out-labrat you guys, or out geek you. I am fishing, the thread is a fishing expedition accusing dither of affecting my mixes!  Perhaps because I have mixed music for awhile (and listen to older stuff) I hear things you don't?  The spacing, the air, the snare, the cymbal decay, the attack, the panning, the timing, the overall timbre/color of the music, the amount of voices it can present - these are things that I hear in my own and other people's mixes, and these are the things that are reduced as resolution is reduced.
 
I think many people see audio as 2D on a screen - 2 waveforms rolling by.  Audio is never on a screen and it is never 2D. It lives in real space, in real air, and in my opinion music can only really be studied fully in open air.  It is determined by the size of the room it was recorded in, the mic type, the amount of post-processing done, and yes the final resolution. 
 
Perhaps there's a dither out there that will successfully mask the reduction of a grid of 16,000,000 points down to 64,000, but I haven't heard it yet. That is the resolution reduction  when going from 24/192 to 16/44. 

Sorry buddy but I've been in professional audio environments / Recording Studios as far back as the 1970's so don't wave the I have and you haven't flag at me. I'm also an EE. Figure it out, dither is less than 1 bit, not enough to make a big deal of. It's the LSB.

 

[FFBookman]

   
Any device whose alterations of frequency response, nonlinear distortion, noise and phase response are within high but readily achievable limits makes no audible alternations to the signals it processes. That includes timbre, pace, rhythm, timing and soundstage.
 
For example, consider 4 inches of 12 gauge bare (air insulated - all glass and porcelain knob and tube wiring) copper wire. Do you believe that its inclusion in an audio system would necessarily be audible?  I think that most persons would dismiss that idea because 4 inches of 12 gauge copper wire are very impressive robust, both mechanically and electrically. How much do we have to extend that piece of wire before the electrical alterations it makes becomes audible?  Now, we are having a reasonable discussion.
 
Your post seems to have at least two serious errors:
 
(1) It seems to be based on the idea that everything audibly changes every signal that it handles.
 
(2) It seems to be based on the idea that human perception is either impossible to measure or estimate, or such estimations have set the technical bar so high that no real world audio component can step up to the challenge.
 
At some level of detail nothing is flat, and its variations from perfection are readily measured and quantified. However, the ear and brain are relatively accepting of minor discrepancies, and simply don't notice them no matter how hard or long or what you listen to.

re: what you see as 2 errors:   I do believe that everything audibly changes every signal that it handles, if the signal is analog. If it is digital data we are capable of making a perfect copy, but if it has to be re-rendered and apply any dithering or filtering then it is indeed altering the signal.
 
Yes your copper wire example is "audible" and could be damaged, or upgraded/downgraded based on cost and manufacturing decisions. To what extent  I care about this stuff inside of amps is very little.This debate for me is about format of distribution, not the quality of wire in your amplifier.
 
Human perception is impossible to measure accurately because it's so deeply tied to emotion. How much do you hate me? Can you assign a number? Not a real one. No scale for you to compare to.  What if you secretly love me? What if you have a nap and change your mind? No way to keep that measurement accurate.
 
That's why your side turns to ABX tests, to test the human perception abilities. But ABX tests only show that no one can pick anything correctly because they are trying to pick memory against real signal, and no one actually listens to music in that way.

 

[FFBookman]

  Sorry buddy but I've been in professional audio environments / Recording Studios as far back as the 1970's so don't wave the I have and you haven't flag at me. I'm also an EE. Figure it out, dither is less than 1 bit, not enough to make a big deal of. It's the LSB.

You walk around in professional audio environments believing nothing exists beyond 16/44?  Your head must explode hourly. Every single piece of gear in a studio goes beyond 16/44 these days.

 

[RRod]

 
You walk around in professional audio environments believing nothing exists beyond 16/44?  Your head must explode hourly. Every single piece of gear in a studio goes beyond 16/44 these days.

 
What's the frequency response of the speakers you use for mixing/mastering? Much flat past 20kHz?

 

[StanD]

 
You walk around in professional audio environments believing nothing exists beyond 16/44?  Your head must explode hourly. Every single piece of gear in a studio goes beyond 16/44 these days.

Higher resolution is used in the digital recording/production process to limit errors due to numeric processing. You should know this. The final target for human delivery is 16/44 which is after all that proceesing has been done and is well within out human limits. You have any more myths to explore or make up?

 

[Speedskater]

  I do believe that everything audibly changes every signal that it handles,

Do you have a special meaning for 'audibly changes' ?
 
I thought audible meant:
"Capable of being heard"

 

[dprimary]

 
I don't downsample and dither before sending to mastering engineer, not sure where you got that from.
 
I will occasionally downsample and dither w/o the mastering engineer to hear different dithering and downsampling on my own rig. But it's not high end converters like the mastering engineer has.
 
So I deliver in the highest native resolution I have and he downsamples and dithers. Occasionally he returns A,B,C versions of the file was created using different dithers, and occasionally he returns MP3 and AAC in that same 'almost blind' naming scheme.  I appreciate this stuff because I am a producer and it helps my ears.  I close my eyes and play them on the same rig that mixed them, and occasionally check them on other systems, then make my pick. Each of them sounds different. Each dither applies it's own character to the sound.
 
What part of that is so confusing?  I deliver 24/88 or 24/96, he returns 16/44 WAV, MP3, and AAC.

 
It is the third sentence in your first post
 
"Then I down-sample and apply dither. I have probably 4 different algorithms and each of them change the sound, especially the "size" of the sound -- the depth, width, clarity, and resolution of the various voices and instruments."
 
If you and the mastering engineer have gone though all this trouble, then what possible reason do you have to not compare the versions in ABX software? You are already preparing the recordings, you are comparing the results. It will take 30 more seconds to open ABX software and load the recordings.
 
I prefer to make many audio decisions blind I do not want price, numbers, or new and shiny to influence which mic's to use, which preamps, or anything else in the chain. 

 

[dprimary]

   
What's the frequency response of the speakers you use for mixing/mastering? Much flat past 20kHz?

Even many older studio monitors go out to 30k. Now flat might be a problem. Finding a way to measure acoustically past 20k without a DOD budget is not a small task either. I have not found any "off the shelf" under $10K way to do it

 

[dprimary]

 
Looking for soundstage with a scope --- no, I don't think I've ever tried such a thing. 
 
I don't currently own any scopes myself, just have used a few over the years in other people's rigs. I used to work at a facility with 3 TV and 4 radio stations and hung around the engineers. I was in programming and tech staff but not a radio/TV engineer myself. 
 
I wouldn't even know where to start to measure soundstage, measure delay and decays in recorded music, and to measure accuracy of signal of recorded music.  
 

 
Every recording engineer should have a scope, at the very least to check phase. Get all the mic's close in phase, delay compensated and everything gets louder and dynamic, with none of that horrible distortion called "compression" required. You gave me an idea I'm going to put a scope in every studio I design from now on they cost only a few hundred dollars.

 

[arnyk]

  I do believe that everything audibly changes every signal that it handles, if the signal is analog.

 
That is very easy to prove to be wrong in the real world. Just do a good listening test - one that avoids the usual 5 audiophile big mistakes that I have listed here many times.

 

Originally Posted by FFBookman /img/forum/go_quote.gif
 
Yes your copper wire example is "audible" and could be damaged, or upgraded/downgraded based on cost and manufacturing decisions.

 
That is also very easy to prove to be wrong in the real world. Just do a good listening test - one that avoids the usual 5 audiophile big mistakes that I have listed here many times.

 

Originally Posted by FFBookman /img/forum/go_quote.gif
 
Human perception is impossible to measure accurately because it's so deeply tied to emotion.

 
The above statement actually means nothing due to the use of the hedge word "accurately'. I can show a person who says things like that any kind of wonderful and precise study of human perception and they can say that it is not accurate enough.  It appears to be an example of solipisism, a very self-centered way of putting ones self into a logic-proof box.

 

Originally Posted by FFBookman /img/forum/go_quote.gif
 
That's why your side turns to ABX tests, to test the human perception abilities. But ABX tests only show that no one can pick anything correctly because they are trying to pick memory against real signal, and no one actually listens to music in that way.

 
The error here is the false idea that one can listen to music without relying on memory.  If you don't use your memory when you listen to music how do you remember that you every heard the music before?  How do you remember the circumstances in which you listened to this music before?
 
BTW the problem is not with ABX tests, but with any listening test that fails to make the following 5 fatal mistakes:
 
(1) Audiophile Sighted Casual Evaluations are not reliable evidence because they are not tests. That is, they do not involve comparison to a fixed, reliable standard.
 
(2) Audiophile Sighted Casual Evaluations are not  reliable evidence admissible because they involve excessively long switchover times, which makes them highly susceptible to false negatives because they desensitize the listeners.
 
(3) Audiophile Sighted Casual Evaluations are not reliable evidence because the do not involve proper level matching, which makes them highly susceptible to false positives because people report the level mismatches as sonic differences.
 
(4) Audiophile Sighted Casual Evaluations are  reliable evidence because they do not involve listening to the identical same piece of music or drama within a few milliseconds, creating false positives because people report the mismatched music as sonic differences in the equipment.
 
(5) Audiophile Sighted Casual Evaluations are not  reliable evidence because they constantly reveal the true identity of the UUTs to the listener, creating false positives because people report their prejudices and preconceived notions as sonic properties of the equipment