Originally Posted by gregorio
....I always try to minimise my system jitter but it always plays on the back of my mind there's a possibility that less jitter does not necessarily mean better sound quality. Even though logically this doesn't make sense and there's no evidence I've seen which would justify it.
What is your view?
I do not know what the issue was with the 888, there are a lot of possibilities to have SASU (something all screwed up).
But I would suggest that jitter is never better, and less jitter is always better.
It is difficult to argue with subjective taste. One may like more bass, or the type of distortions of some tube. Those kind of alterations or distortions tend to be "fixed". They are there "all the time" when you use the particular gear.
But jitter is a different animal. The impact of jitter varies all the time, moment by moment, as the music changes. So how can one like such a thing?
In fact, "Fixed non linearity" (tubes, transformers and some electronic circuits) present a "stationary transfer characteristics", meaning the behavior does not change in time, and it "scales with the signal". Of course I am not talking about long term aging of components; the time frame is seconds not month and years. If you use a tube for a year it may sound different.... Say you feed a tube a 1KHz signal and it generates additional 2KHz and 3KHz. If you feed the tube a 2KHz, it will yield 4KHz and 6KHz. In both case it is responding to the input with adding 2nd and 3rd harmonics. If the tube amplitude response is flat, the harmonic amplitudes ratios will stay the same... In other words, there is some "predictability" here.
When you feed a steady tone into a fixed non linearity, the distortions are always at HIGHER frequencies, and the distortion energy falls on the harmonic frequencies of the tone. There are no sub harmonics. So while the tone is distorted (not the original sound) there in some sense some "musicality" to the outcome. The distortions are at harmonic locations of the original tone. Clearly, when you feed multiple tones into a non linearity, you end up with a mess (sums and differences of the various frequencies). That is why I do not advocate non linearity. But at the end of the day, one can argue that they like non linearity, because taste is subjective, and one is entitled to like any thing, even the sound of finger nails on a black board.
However, jitter is NOT a stationary behavior. If your jitter is a 120Hz tone, and you feed a signal of say 1KHz, you end up with 1.12KHz, as well as 880Hz. If you feed it 2KHz, the additional tones are at 2.12Khz and 1.88Khz. The distortions are non harmonic, and their relationship changes with the signal itself! It also changes with the jitter type itself. I used 120Hz (typical rectified AC line frequency in the US). The same gear in Europe will yield 1.1K and 900Hz instead of 1.12KH and 880Hz.... And all that that was for the simplest case - a pure fixed sine wave tone, and a steady fixes jitter pure tone. When you have complex music, all hell breaks loose. One can not like it, it is a constantly moving target.
And indeed jitter is a much more complex subject. There is random jitter (covering the whole spectrum), often due to electronic component noise, there is more specific jitter such as due to line frequency or some electromagnetic pickup, and there is also jitter that is due to coupling of digital audio signal into the analog path.
Then there are different circuits, and they respond to various jitter differently. A PCM converter is hardly effected by jitter when the signal is very low. A PCM converter is hardly effected when the signal is very low frequency. But a sigma delta is effected all the time, even the noise floor with no signal at all rises with increased jitter...
Jitter is a very complex subject. But in all cases, even the simplest steady pure sine wave, it makes for very unmusical distortions that change moment by moment all over the place. It is best to have lower jitter.
I am responding to your post, Gregorio, and this post is even more technical then the previous one. I still wonder if my posting here is too technical for most. I do not want to break the flow for the majority here.