Asit, Thad, HiFiVision and others - the fundamental question here is different. The question is whether digital data moves from Point A to Point B without any loss (of any kind - drop, collisions, etc).
I checked with a few people in my office and they have not even heard of the concept of data loss. There is, of course, a huge age difference between them and me. I wonder if colleges have stopped talking about data losses in digital transmission systems completely as a waste of precious semester time.
Maybe --- because it probably
is a waste of time. And maybe because they didn't study computer science as a primary subject. Old technology made a certain familiarity with baud rates, parity, stop bits, CRC (It stands for Cyclic Redundancy Check, but hey, I never had any clue how it worked!) and so on, necessary: mostly, these days, it is not. Nor do you need (and you never did) to know about collisions on [unswitched] ethernet networks, because the system looks after them (and you ignored my reply: these collisions
do not cause data loss, enlarged upon by another poster who pointed out that
congestion screws up LANs, not collisions.)
It is a question of risk assessment. It is sensible to take care on the roads because a tiny proportion of people crossing roads get run down and killed. It is not so sensible to obsess about USB (and many other kinds of) cables because an occasional bit gets lost or mis-timed
unless the results are obvious.
it is an accepted fact that HDMI cables beyond 30 feet are problematic.
Then don't use them over 30 ft. There is a maximum defined length for most kinds of cable: for some it may be measured in thousands of km, or even tens of thousand, for others it may be in feet. Most of the cables that we use for PCs were probably designed to connect to stuff on the same desk, even cat5/6 ethernet has a specified maximum run of 100m. USB has a maximum of around 5m, it seems (
down-to-earth USB information). A quick look at the Wikkipedia page for hdmi leaves me with the impression, "depends on the cable". Attenuation happens. Your maximum-possible bandwidth from your ISP is determined by your distance from the exchange, because the signal looses strength. There is a mathematical relationship between length of copper cable and attenuation but it is made worse by the number of cable joins.
Work within the limits, and forget about data loss! Be conservative:
halve the published limits, and work within that!
Then we can return to what
should be the baseline for digital cables:
if a cable meets the specs, and has no faults, it will work. If it doesn't, then it
doesn't meet the specs, has intrinsic faults, or has been damaged in use. Throw it away, buy another one, but do not think that you need to spend ten or a hundred as much (unless you bought really cheap junk in the first place) as a fix. All you need is a cable that works.
(I had another bad cat-5 lead only yesterday. chucked it. didn't order a new one from Monster, just picked another one out of the bag that
did work)
I acknowledge that you are on the quest for perfection, and that is fair enough: it is a great quest ... but the stuff happening at either end of a digital link, given the above-mentioned criteria are met, has got to be infinitely more important than the link itself. As you know, your Windows operating system does not even deliver the same bytes it reads from a CD to your sound card, hence you load drivers that avoid that interference.
the medium and method of transmission will thus make a difference, right?
Yep. Did you use the internet in pre-www, pre-adsl days, over dial-up connection? When an interesting Usenet post disintegrated suddenly into a stream of gibberish, sometimes followed by that famous +++NO CARRIER+++ message? Now, as someone mentioned previously, we can stream not only text, but pictures, audio, video, across continents and oceans.
But yes, I agree that it is still possible for a poor-quality toslink cable to introduce substantial jitter. Imagine the differences between cloudy plastic, clear plastic, and the sort of materials (glass?) that carry digital signals across oceans. Glass is going to cost more than plastic; the protection needed for glass is going to be greater than the protection needed for plastic --- but all should be strictly within the
computer market pricing, and there is no need for the fear-mongering-among-audiophiles scum to get involved. Really, I regard them as on a par with e-mail scammers!.
Goble said:
...The total jitter in a system which will be transport jitter + transmission jitter + intrinsic jitter will add up ...
You'd have to include that the data on your CD might already be jittered by the ADC used to prepare the master digital source.
Except that apparently, jitter is not additive, and does not make in difference in the digital domain. It only matters at the DAC. Anyway, I've quoted this article before...
A lot of fuss is still made about jitter, but while it is potentially a serious issue it's rarely a practical problem these days simply because equipment designers and chip manufacturers have found very effective ways of both preventing it and dealing with it.
...
Another source of jitter (the strongest source these days) is cable-induced....
...
Fortunately, most modern D-As incorporate sophisticated jitter-removal systems to provide isolation between the inherently jittery incoming clocks embedded in the digital signal, and the converter stage's own reconstruction clock.
Sound on Sound
These days and
modern was back in 2008. Things got worse since then? Or the manufacturers forgot their effective ways of dealing with? I don't think so: I think that certain unscrupulous companies sold us paranoia and mis-information.
So, despite my aesthetic liking for smoky-clear-over-silver USB cables, I continue to assert that that cheap grey thing (as long as it
really is a USB2.0 cable) can be expected to work pretty well with your DAC or external interface. I accept that it is
just my view of reasonable.
.
hyeah: Wonder if anyone noticed though ... 
