Does the quality of a digital signal matter?

p.s. I had another thought : )

It seems to me that, when it comes to audio, we clearly don't know everything yet !

It seems to me that the bit stream speed is independent of the bit content. If this is correct than should not the jitter be either constant of possible a function of the disc itself (like radial position or burn/pressing quality)?

That was assumed when CD players were first invented. However, many things can affect the accuracy of the clock signal in the DAC. And even the bitsream is variable - error bursts and misreads may be cyclical and perhaps only the digital "preamble" is fairly consistent - so the data may vary in a certain repeating patterns.

Provided jitter is random, it is in general a negligible problem. However when patterns - such as power supply oscillations due to cyclical laser servo movements to track the pits on the rotating disc occur - then we can get non-random jitter. Another major cause of non-random jitter may be the Phase Locked Loop between teh master and slave clock - in this case, the very act of trying to keep the slave clock in time with the master cause oscillatory patterns as the slave hunts back and forth trying to keep in time. These repetitive patterns in clock timing erros cause new oscillatory audio signals to appear in the analog music coming out of the DAC - sometimes called sidebands - non-harmonically related signals. It is these very small (-40 db) but 'correlated' sounds that become audible - usually as hash or lack of clarity in upper midrange and HF (although this may significantly affect the perceptive sound of percussive instruments with low frequencies - like piano or drums - due to the way we "hear")

Anyway - jitter is an analog problem - it only appears upon conversion to analog or, up front, when converting analog to digital.

If you have a perfect clock then you will not have jitter.
DAC's have evolved to have better clocks. Early designs like Meitner used patterns in the digital data called "preamble" to try and achieve a more accurate clock. Others like Lavry used algorithms to maintain a very slow correction pattern on the slave clock that could be filtered out. Since about 2002 the problem has been substantially addressed by "asynchronous DACs" - basically these type DACs ignore the master clock altogther - and in these designs the jitter is totally determined by the clock quality in the DAC along and nothing upstream of the DAC.

On thing to note is that the information in one of the long and detailed articles linked in this discussion is 17 years old. My EE friend pointed out that a 2x CD player was a big deal 17 years ago. I would hope that many of the problems described have been reduced or solved by now. When it comes to electronics, 17 years is a very long time for technology to develop.

I picked up an SACD player from a friend today to borrow for a few days. We'll see how much difference there is once I get a copies of a single album in both formats. It'll probably be Nora Jones since I already have the CD and know it's a quality recording.

I picked up an SACD player from a friend today to borrow for a few days. We'll see how much difference there is once I get a copies of a single album in both formats. It'll probably be Nora Jones since I already have the CD and know it's a quality recording.

Apples and oranges......

I suggest you compare the two players just using the CDs you have now. You should be able to hear a difference between the two players.

Post back your findings.
.

let me ask a related but slightly different question. I read often of a player being colored. THere is a thread right now about a music hall player being slightly "forward" and "bright". All things equal and in a world where jitter is the dominate error (or am I misunderstanding the tone of this thread?) how does a "color" creep in? Jitter seems like a strictly time domain issue and the coloration of music would seem like a level vs frequency issue? If so that may imply that the DAC or perhaps some outside electronics issue outside of bit reading is at play. Is that a reasonable assessment?

Sorry about the dumb questions. I'm your boy for mechanical engineering / physics / material science but I'm just about the "bang the rocks together" stage with digital electronics!