The Great USB cable debate poll

I assure you .. if someone else did the cable switching, you would not hear a difference.

 

I'll acknowledge this isn't totally true. It's definitely mostly true though. I gave an anecdote in the other running USB cable thread about using the cheap cable that came with my Schiit Fulla and how when I'd have my mobile phone right next to it it would introduce noise into the signal. Of course, the grain of salt to take that with is that I didn't check to see if it was the cable or the device itself, which obviously is too small and cheap to be shielded.

Another example I made earlier is Cat5 cable, which uses a twisted pair topology for noise rejection, and comes in a shielded version for EFI/RFI rejection. But Cat5 is designed for much, much longer runs then USB, and given that these days it's generally used alongside TCP/IP I'm not sure how much it matters these days.

 

I assure you you are wrong. The difference was huge.

 

If you happen to have a 100,000 watt transmitter in your back yard, then you might be getting interference issues. If so, you would not hear softer highs, or a veil over the sound, or a smaller sound stage. You would hear ugly, nasty, digital spitting. If you have ever seen DirecTv or Dish network in a rainstorm ... its the audio equivalent.

 

Well, when you go down this line of thinking you first have to get a sense for how many errors there actually are. I can't find a link to an authoritative source for this, but I've seen it said that in 24/7 testing of isochronus USB transfers the error rate averaged 1 a month.

But the main thing you have to make sure you don't do is apply analog rationales to digital. That's the biggest problem in this space. People are saying things like the highs were higher, the lows were lower, better pace and flow, etc. This isn't how "fewer errors/guesses between 1 and 0" would manifest. It would be more like "the jarring skips in the music replaced by harsh obvious noise went away".

 

Huge. Well .. that's a lot.

 

My primitive understanding of this is the digital filter fills in the blanks so to speak. The algorithms have evolved to do it so well most people would probably not notice. If they can do it with complex genomes they can certainly do it with music quite easily I would think.

 

They must have bumped a knob, or changed another setting. Sorry, it just doesn't work that way.

 

With respect, I think you're missing the point.

 

And this is where things get weird.

70% of us say NO DIFFERENCE.

You're not willing to say the difference was subtle. Your word is "huge". HUGE differences with a digital cable that the rest of us don't hear AT ALL.

What the hell do you suppose is going on here? 30% of us are mutants with super hearing? 70% of us have hearing loss?

 

Sure. I do wonder though if there is a small crack in the window of possibility. There seem to be jitter circuits etc. (and some say these don’t matter) but the more these are working and taxing the power supply, well, it seems that there is enough variation in the receiving devices and the transfer environment that blanket statement are not as easy as someone with a strong technical understanding of the transfer protocol might imagine.

One a month is negligible, but in a strong RFI situation, could that be more? Could some people have really bad situations and have better sound because the issues they have are addressed in an over designed cable?

I suppose the 1s and zeros are voltages, and the DAC has to manage/correct the data stream in real time using hardware that draws on the power supply. I really don't know enough about the various implementations and various real-world environments to say a cable couldn't make a difference.

 

The green USB cables sound the best.

 

I'd hate to imagine the environment someone could be living in where they're surrounded by enough RFI that the signal once outside of their computer needs shielding. At that point I think I'd worry about cancer and move.

Think of the digital signal like a square wave being generated by to different voltages. One representing the peak and one the valley. The peak is a 0 and the valley is 1 and the longer the specific voltage maintained the more 1s or 0s generated. One of the beauties of a digital signal is its inherent ability to transmit its information despite interference. In other words to hit a ,1 say the .5mv that are needed, can be dirty and scrambled but as long as the receiving device notes the potential the 1 is received. We are not talking about the delicate nuances of an analog signal. Even though in all honesty outside of a phono cable the signals in your home stereo are really not that delicate

 

I would say the noise was due to interference in the analog signal (audio) side of the device, not the corruption of the digital signal nor its transfer between devices.

 

It's absolutely possible, reference my anecdote about having my cellphone actually touching the cable/DAC, but the aspect of critical importance is how the errors manifest - it's in no way subtle. It's not your music sounding bland or flat or something, it's jarring noise.

 

I suspect you're right, at any rate I did absolutely nothing to eliminate variables.

 

The effect a USB cable may have is dependent on the equipment it's being used with. USB cables may make a larger difference if the DAC's input is not asynchronous, or if the server's 5V source is very noisy.

 

You did absolutely nothing to address the content of my post. Literally nothing at all.

Unless of course you're saying that fancy USB cables only matter if the DAC isn't asynchronous, and naturally you're not.

But hey, what let that stop you? You're deep in the game, after all.

 

Is the "improvement" people claim to hear just limited to audio, or do the pricier USB cables also improve non-audible data transmission in some way?

Hopefully some of the folks here who have don A/B tests on different USB cables sound characteristics have also taken the time to test digital data transmission accuracy outside of music.

Seems strange that what has been described as a "huge" difference by some would be limited to just the audible realm.

 

Could You possibly post something that will show what will happen, and also some samples of how the difference will in practice manifest itself in different sound?

 

Something that has to be kept in mind is that objective testing in this space is somehow considered a bad idea. I've pointed out how trivial it would be to measure USB performance.

On one hand, we have some talking about the absurd "objectivists" and their silly little charts and graphs.

On the other hand, we have some claiming that anybody who doesn't hear tangible improvements in expensive USB cables simply don't understand engineering. Try to wrap your head around that one. I haven't had much luck personally.

 

Can anyone record and post samples of x vs y USB cable where they find a difference between them? - same as a needledrop? Surely that's a straightforward thing to do if you know how, and any differences should be audible...

 

You're going to get an answer heavy on theory and absent practical application.

 

It has been done, the defect cable sounded awful. Can´t remember at the moment by whom.