mike_mccue

I thought that data packet transmissions systems such as USB corrects it's errors by requesting a replacement packet. Yes? No?

http://www.beyondlogic.org/usbnutshell/usb4.shtml

I think the answer is Yes and that there is no way for a "audiophile" USB cable to improve some portion of the audio stream or spectrum. When you read the review on the guys web page that speaks of enhanced clarity in the bass and dynamics... that's the "marketing speak" that makes people look like fools.

If there is some uncorrectable error caused by a USB cable the buffer runs dry and you get nothing.

It's go-no go... if you have a problem you are going to hear a stutter or a splat... or silence.




Schemes like the Reed Solomon system we use with CD's actually "fixe" errors with statistically likely corrections... which seems to result in nearly perfect results as well.

At least the guys trying to minimize errors on CDs have a legitimate concern in that the data is being slightly altered to the extent that the Reed Solomon scheme may not guess if it's a 0 or a 1 on rare occasions.



BTW FAX is considered secure because the manufacturers have made it is easy to use an encryption appliance. Encryption is easy in email too... but there are so many options that the FAX appliances seems easier by comparison. Also email copies linger in places where people that have the time to break encryption can easily find the email and work at leisure. None of that has much to do with error correction.


best regards,
mike

+1
 
this is correct.
 
while proper shielding and good solid connections and quality made wire are important, the built cable has nothing to do with error correction or handshaking.

You guys are missing it.

Audio quality is not what drives audiophools. They are driven primarily by pretension. If something is ridiculously impractical, difficult, esoteric and (especially) super expensive, it's better.

Not because it provides better audio quality, but because they had to jump through more hoops (financial or otherwise) to get there. That's what drives them - spare no expense, bear any burden, chase down esoteric solutions that "science doesn't understand", do anything and everything possible, no matter how small or questionable the benefit.

A cheap cable that sounds identical will never be as good to them. And not because it's cheap, but because it's easy, common and generic. It's not the kill, it's the thrill of the chase.

I want 2 of these chords like Jonbouy's.

Just THINK of all the 1 and 0 envy your friends will have once you get one of these cables...

I often feel I'm being held back by ethics. Why do we cringe at the idea of stealing from gullible people? Is it not just natural selection in action?

Can the leek really do that?

You all are thinking about this wrong.
Sure, $3500 is a lot for a USB cable, but if you send just 1GB worth of data over that cable then it's only 3.5 ten thousand's of a penny per bit.  That's just $0.0000035.  

Isn't that mere pittance worth paying to ensure that every single one of your precious bits arrives in perfect and pristine condition?!!!!


(Hmmmm, maybe I should've pursued a career in marketing.... at least marketing of overpriced goods.  ;-)   Honestly, this seems like an excellent example of going waaaay past the point of diminishing returns. )

I don't how it works but with that cable, I'm sure it's working.

mike_mccue

I thought that data packet transmissions systems such as USB corrects it's errors by requesting a replacement packet. Yes? No?

During the initial device detection stage yes, according to the website link you posted, but I'm not seeing how USB handles streaming data such as audio. That's the key difference. It's different than transmitting a file of known size such as a .DOC, .PDF, .JPG ... etc. If you have an inferior cable and there is excessive packet loss because of the cable, I could see how a better quality cable could help, in the same way a shielded cable is recommend for USB 2.0. I didn't see any details on this, but I'm assuming audio is transferred via USB very similar to how picture data is on a fax because everything else seems to be very similar according to that web site. And I know during the picture transfer stage of a fax, if you get packet loss, db loss, or noise, it effects the quality of the image when it's printed on the receiving end.

http://www.beyondlogic.org/usbnutshell/usb4.shtml

That's a great web site. What it explains is very similar to how fax machines work.

I think the answer is Yes and that there is no way for a "audiophile" USB cable to improve some portion of the audio stream or spectrum. When you read the review on the guys web page that speaks of enhanced clarity in the bass and dynamics... that's the "marketing speak" that makes people look like fools.

It can't improve the audio stream or spectrum, but it can improve the integrity of that data and how accurately it is sent to your audio device.

If there is some uncorrectable error caused by a USB cable the buffer runs dry and you get nothing.

I would think it would depend on which stage of the audio data transfer you're at. Again, I'm assuming that since the other similarities to facsimile data transmission are almost dead on, that audio data is transferred very similarly also, and on a fax, that stage can definitely be altered without the transmission being truncated.

BTW FAX is considered secure because the manufacturers have made it is easy to use an encryption appliance. Encryption is easy in email too... but there are so many options that the FAX appliances seems easier by comparison.

I wouldn't say it's a matter of being easy, but rather a matter of a PDF not being intercepted while in transmission. At least, that's what the IT Directors would tell me at the banks I used to service. But like you said, this has nothing to do with EC. I got off on a tangent there. :)

I think we all agree that $3500 for a USB cable is absolutely ridiculous. :) Do I think a better made cable could make an improvement, yes, but not to the extent that web site and the reviewer is claiming.

Streaming USB data is still buffered and packetized, and still contains CRC redundancy error checking.

The difference is, when streaming, missing or problematic packets are not re-transfered.

When doing packetized digital transfers, it is impossible to get "more accurate" transfers, unless you are comparing to a bad transfer with dropped packets or uncorrected errors. And the result of dropped packets or uncorrected errors would not likely be subtle in the real world.

bitflipper


I'll only buy these cables if they're endorsed by John Vestman.

You know when I read that article I found it to be full of errors.

I suspect it must be because he used cheap cables to connect his keyboard and mouse to his PC  when he wrote it.

I know squat about USB, but sin't there a checksum at the end of each digital packet, or whatever? And, since RME sent me a very nice-looking cable with my $2100 UFX, I think they'd put a decent enough one in the box.. I mean, the thing gets rave revues without any mention of "nice unit, shame about the cable".. :O)

bitflipper



And don't think you're too smart to be fooled by expectation bias. None of us are.

I expected you'd say that. 

Jonbouy

Beagle

bitflipper



And don't think you're too smart to be fooled by expectation bias. None of us are.

I expected you'd say that. 

I did too but I thought it was going to sound better than it did, must be the cables across the Atlantic reducing the quality...

It did sound better, you just didn't bias it right.

I always like to say, "There's a word for people who think they're too smart to be fooled by expectation bias - dumb".

I'm dumb enough not to be fooled.