Pushing the Computer Audio Boundaries :)

[jkeny]

Here's another question, John: if second-order distortion sounds good, can you introduce some to your DAC? Just so we're clear, I'm not joking!

(If you could introduce it in a controllable way, the listening tests could be very interesting.)

I'm not so sure that 2nd harmonic distortion is "good" - I just think that it is more benign & less discordant than higher order distortion, particularly odd order harmonics. So, I wouldn't go along with deliberately adding it to a device unless it were beneficial in somehow masking another distortion!

[Diapason]

So anyway, this is the threat correlation function of the ear. It may also explain some issues that people have with some digital - that uncomfortable, uneasy feeling where long term listening to some digital is impossible. Digital audio, in my opinion, made a fundamental mistake in focusing, almost exclusively, on frequency accuracy & ignoring perhaps the more important timing element of signals. It's only now beginning to rectify this mistake by paying more attention to it.

That's a compelling argument. Can you explain more about digital ignoring the timing aspect, or point me in the direction of some discussion online? I might find myself out of my depth very quickly, but I'll take the risk!

[jkeny]

Right then, another piece of my audio "knowledge" has been seriously upended. It's pretty stark too:
"for me the issue is now settled. Unless and until somebody comes up with a "magic" pattern of nonlinearity that truly enhances sound quality, I will believe euphonic distortion to be a fantasy."

As Ciaran said, I don't know if there is a "beneficial" kind of distortion. Having said that, what we are doing when listening to audio is listening to an illusion of a sound stage with breadth & depth. Sound stage breadth is an interesting phenomena because certain types of distortion can give rise to an illusion of a wider sound stage. This is a good thing, right? Well, I would question that. if it isn't on the recording then I would consider it not a good thing, it's a distortion even if we might prefer it on some types of music, it will negatively effect other types of music.

Also, bear in mind that a number of people prefer the sound of distortion to none, in the same way as some people prefer the taste of food with MSG than food "au natural". The sound of jitter is a case in point. There is a tendency to think of digital as being very fast & sharp & revealing & OK, it can be over the top & strident but this is just the price of such realism. Wrong! This ultra-realism is in fact the sound of the distortion of certain types of jitter but people who have been immersed for so long in it, think very low jitter digital sounds too polite & mellow. I had one French (ex)customer return my DAC saying it sounded too British. I was so insulted on so many fronts that I didn't reply :)

[jkeny]

That's a compelling argument. Can you explain more about digital ignoring the timing aspect, or point me in the direction of some discussion online? I might find myself out of my depth very quickly, but I'll take the risk!

Well, the fundamental error is that the focus was on how many bits & what sample rate is required to accurately represent the analogue waveform that is being recorded. This is still an argument that rages today - is 32bit audibly better than 24bit better than 16bit. What is missing here is the other aspect of the waveform, timing. I can't remember what the original jitter specification was for SPDIF but it was considered to be accurate enough - I believe it was in the 100s of nanoseconds. There are still papers quoted today of experiments where 200nS is the threshold for perceptible jitter. This is rubbish & falls into the trap of treating jitter as one number (the same mistake used in THD measurements). Jitter has many forms, some more benign than others, and it also has a spectrum. Both of these are of huge importance. For instance jitter that is correlated to the signal (data correlated jitter) causes much more intrusive distortion. This sort of jitter is audible in the tens of picoseconds (a 100 to a 1,000 times more sensitive).

Edit: The above 200nS jitter is random jitter which generally produces the distortion called noise & we are fairly immune to noise as a distortion i.e look at vinyl :). Data correlated jitter (or more correctly, deterministic jitter) produces more objectionable & audibly noticeable distortion

Remember the wrong waveform delivered at the right time is distortion of the original signal. BUT, the correct waveform delivered at the wrong time is also distortion of the original signal.

[DaveF]

Dave, I'm not sure why you appropriate intractable complexity to the functioning of the ear?

well I wasnt really, was just asking the question. I really havent a clue how simple or otherwise the ear is. I guess where I'm bringing the complexity into the question is how the brain interprets what the ear is delivering. Is that well understood? I'm all ears. :-)

[jkeny]

Dave, I'm not sure why you appropriate intractable complexity to the functioning of the ear?

well I wasnt really, was just asking the question. I really havent a clue how simple or otherwise the ear is. I guess where I'm bringing the complexity into the question is how the brain interprets what the ear is delivering. Is that well understood? I'm all ears. :-)

OK, Dave, misinterpreted your posts. But leave the brain out of it for the moment - it just complicates matters - as you say be "all ears" :)

[Diapason]

Very interesting. I should read around more of this stuff. Last time I looked into the more technical stuff I was still in college.