castleofargh
Sound Science Forum Moderator
- Joined
- Jul 2, 2011
- Posts
- 10,468
- Likes
- 6,105
That would only apply if listening to audio were an objective experience. Isn't listening to audio a subjective experience?
We all have one thing in common though: we all want the best listening experience possible.
You would think that a USB interconnect between your DAC and computer only carries bits having zero effect on sound quality, but that is not always the case. I think we all agree that not all DACS or USB busses or computers or music severs are created equal. Not to mention everyone having totally different listening environments with varying RF/EMI levels, or lack thereof.
I recall reading a white paper by Apogee (they make Pro studio recording digital equipment) a few years back explaining that for USB audio “speed” is crucial as time relationships within a digital stream are absolutely critical to the reconstruction of the analogue wave form that brings the sound experience into the analogue world. When playing audio, you’re depending on the DAC and the computer’s clock to avoid jitter. But because your DAC is connected to your computer via a cable, the cable itself can have an effect as well.
Time-based damage, jitter, negatively affects the experience of sound by making it sound flat, small, harsh and “foggy”. Yes your DAC, even an inexpensive one, has a USB chip on board that will correct for data loss, but it’s the computer’s and DAC’s internal word clock that must prevent jitter from corrupting the timing. That’s easier said than done and Apogee has even developed an expensive rack-sized device, Big Ben ( http://www.apogeedigital.com/products/big-ben ), a master word clock specifically for the purpose. Big Ben is a word clock that takes care of the jitter when recording audio in recording studios and guess what? It uses a special cable to ensure no jitter is introduced along the way.
subjectivity isn't an excuse. if we test sound, we should indeed test only sound, and use a listening test which removes, look, price, preconceptions... now if your idea is a better experience no matter the cause, including non audio ones, then go for it and enjoy your special cable(I'm not being sarcastic, even just a good looking thing can brighten the day). but that point of view takes away your right to make claims about the sound. which is what we're discussing in this topic.
if we're discussing sound, let's say the idea is "measurably better than average cables can improve the sound audibly". are you ok with that definition? it is clearly different from "all USB cables don't sound the same", which is more of a strawman as it includes all defective cables making the sound worst, and all the tests with a defective cable to begin with.
so if I want to test my idea, I need:
- a basic USB cable within USB specs(needs to be confirmed with measurements), and at least another cable that will make the output signal objectively "better than average cables"(also needs to be measured and confirmed).
- and of course I want to test audibility once those improvements are confirmed. to me that means blind test. anything else is testing way more than just sound so perceived differences cannot without any doubt be attributed to sound alone.
instead, you're already picking possible causes for the sound differences you think you heard, that you think were an improvement, on the cable you think is superior. but you didn't care to actually try and confirm any of it. that's putting the cart before the horse. (or for @GChief, putting the cat before the horse, while eating your desert ^_^).
even if you want to take demonstrability by the wrong end of logic, and discuss your hypotheses for objective causes leading to signal change, you will still have to come back to a proper testing to demonstrate audibility at a given magnitude and then have to measure your own gears to check if you're reaching audible magnitudes. so in the end you still need measurements and a proper listening test. else it's all conjecture.
RF noises aren't new, the guys creating the USB standards have to account for typical levels, and when they picked the specs for USB cables, they accounted that the attenuation would be sufficient. if they were wrong, it would suggest that USB is a poor standard. which begs questions far beyond changing the USB cable.
there is a paper on jitter looking to estimate the magnitudes we can hear, and the conclusion of that paper was that even at the time, consumer gears were expected to have jitter magnitudes below hearing threshold. if they're right and you get sound differences anyway, that puts you under conditions outside of the typical consumer gear in typical environment. so you have a very severe issue with RF noise or EMI or whatever, or we're back to Gregorio's idea that the DAC might not even reach average consumer quality when it comes to jitter. the solution would be to change the DAC. and if they're wrong, we'd like to see evidence of it. which yet again brings us back to proper listening tests.