Jitter is a time domain issue, so files won't have jitter, as you mentioned previously. Imagine if you were transcribing text from one document to another, it wouldn't matter if you sped up and slowed as you were doing it, the resultant document would still be the same.
You blame a lot of things on jitter, but it's probably something else. CDs have pretty amazing error recovery, but any unrecoverable or unmaskable errors may be audible.
The data (music, probably) you end up playing on your system could have been stored and transmitted and retransmitted across all kinds of media and links before it got to you. These would have had various error detection and recovery / retransmission mechanisms to ensure the reliability of the data.
The difference once you send it to your dac is that although there are error detection mechanisms, there isn't any retransmission or error recovery (either on USB or spdif).
However, this shouldn't be an issue with USB. I used to be product manager for x-ray imaging systems. The smaller systems utilised USB (the larger ones used lvds or GigE - search for CT XXL if you want to see an example). These devices sent real-time data (just like audio is), and should have used Isochronous transfer mode which does not support retransmission in case of errors since there is no time for this with real-time data. Some were incorrectly implemented to use bulk transfer mode. What this meant was that, when there was any error in the USB transmission the receiver would request a retransmit, the sender wouldn't be able to do this, and the system would freeze. In some systems this would never happen. In others it would happen every few days or weeks at the most, and with long cable runs in often harsh environments.
It shows that USB is pretty robust. Furthermore, modern USB DACs should be asynchronous, and have their own clocks and don't rely on timing from the USB interface, making them totally immune to any timing variation in the source. The prococol even includes flow control, so buffer under/overruns aren't an issue.
In isochronous mode, bandwidth is also reserved in the USB interface, so that isn't an issue, although some hardware implementations may have issues with internal bus bandwidth when there multiple high bandwidth devices.
TLDR: jitter probably isn't the problem.
I do wonder about the synergy between your HDVD800 and headphones, as doesn't that amp have an output impedance of 43ohms, and isn't that higher than the headphones' impedance?
I see your *points; I will believe that they are to further a discussion, and I see them mostly as to reassure me, or convince others that ‘jitter’ as I described doesn’t do the things I say it does.
Not getting into the semantics of whether equipment that isn’t made to be portable (and run off small batteries), and sell in a supermarket for $29 will have the same sorts of correction processes or ‘transmit quality’ being ‘bargain basement parts‘ built for a mass consumer world, as high end medical systems.
My takeaway from that discussion re: medical kit- is that USB ‘can’ be a solid performing method for data.
I have no awareness about bandwidth used (might be 10bit colour for all I know), but I wouldn’t use an understanding and experience with one system or product to attempt to ‘know about all others’. (things can be witnessed, yes… but I would never place any individual experience over the combined learnings (from multiple examples) of another system.. hence why I believe this MAY be a discussion).
Jitter as I have experienced it, for decades, regarding audio, has a range of effects, most of what I described, in terms of audio ‘change’, is due to the ‘goes hand in hand’ nature of WHEN DIGITAL ISN’T READ/TRANSMITTED
WELL.
We all assume it is. (digital is perfect on/off signal, hence perfect ‘zeros and ones’).
Sadly in the world beyond engineering documents, digital has ‘shades of grey’ they have to be resolved as black or white… and this is where error happens.
Actual errors.. that can affect the leading edge location (yes, in the ‘time domain’) of a note, through to the quality of bass notes…
The reasons, whilst not belonging to this thread, as they are a discussion that has been argued for aeons, mostly with regards to ‘digital cables’ passing an analogue wavelength that can lead to ‘misreads’; is beyond a ‘short post’, and isn’t something I feel is right to discuss here.
I could go to length to list my history with audio equipment and setups (spanning back to the eighties), or my expertise regarding many things digital (since the nineties), but feel that my ego doesn’t change the real world properties of which we speak.
Much of the basis isn’t ‘theory’ and is well understood by many, but the nature of observation bias, and humans ‘finding evidence to support any belief’ (it IS a large world we exist on!), is such that we have the potential for billions of different opinions.
Regarding ‘digital‘ audio, jitter exists. It exists for a range of reasons and is contributed to by a range of aspects of a digital playback chain.
And it is NOT DESIRABLE.
Maybe there is merit to most transports quoting their ‘jitter rate’ (usually as a sales point to convince pundits to invest in them, sure.)
For any who have learned to ‘see’ jitter and the undesirable effects that a high jitter rate playback can lead to, this is something useful to talk about.
For most people it is of ‘no concern’. (No reason to buy expensive kit that works hard to feed accurate digital, save the money… and, yes there is a reason flagship surround amps advertise ‘low jitter’ HDMI connections etc)
@SRKSRM thankyou for taking time to research the audio chain I was using yesterday.
I was using it to ‘learn’ about the output impedance and the effect it might have on those headphones. It required recabling my headphones back from 4.4mm ‘balanced’ etc.
I literally set the rig up to ‘check‘ whether the typical ‘bass bloat’ that comes by not following the 1/8th rule (for output impedance/headphone impedance), and, just like how the ‘real world’ does not always follow ‘engineering logic’, was actually a ‘great pairing’.
it is something I have learned to ‘stick to’ based on some Bowers and Wilkins P5 and P7 headphones that took years to ‘control’ (right).
The usual ‘good amp’, a Burson Conductor V2+ is on loan, so toying with the HDVD800 and Edition 5 combo was something to try..
Those headphones took nearly 1500 hours to truly settle (finish burn-in); never owned a set that required burn in so much.. (From hours 300-800 they were *unlistenable* due to high frequency screech), but they DO pair magically on the HDVD800.
That being said, a bass boost doesn’t hurt them, and the amp is deathly quiet, the resultant sound makes the Sony PHA3/MDRZ7 combo seem like unlistenable junk (never thought I’d say that).
I stand by my comment of the resultant sound from that chain being smooth (maybe the proof of the pudding IS in the eating!).
Given I could have simply plugged the headphones into the Diablo (it is a ‘nice’ amp, as well as being an excellent DAC), or even straight into the M11+ (those THX amp modules many people think are the ‘best there is’ (engineering and ‘white sheet docs’ no doubt!)..
I learned from Hans Van Beekhuyzen to ‘never judge a setup I have not listened too’. (for legitimate reasons).
Clearly your opinion on USB as a transport method is giving it much more credit than I do, hence why I do my absolute best to avoid it, as consistently COAX at 1/8th the price point betters the sound.. (USB ‘for audio’ has proven to require big investment to equal budget cables using ‘antiquated’ systems). Of course it depends on whether the parts are built to favour one input/over another. Most budget DACs serve the USB input as the highest performing part. So I can see how it would be easy to ‘test and see’ that USB is the ‘better method’, many pundits have no doubt experienced this.
The Diablo fed via COAX, means I have to give up on the ‘GTO filter’ (a reason to own an iFi DAC for those listening to redbook audio (16bit 44khz)), and I think DSD may not work either…
The organic sound that the temporary rig I setup yesterday offered was ’truly exceptional sound quality’. No matter what the math and rulebook says.
tl:dr jitter is the bane of good digital transmission. less jitter can make a big difference to some DAC/clock chip designs. YMMV (your mileage may vary)
I hope my observations have use for ^some^ head-fiers.