Head-Fi.org › Forums › Equipment Forums › Sound Science › USB cable supposedly improving DAC sound quality? How can I take other posts seriously after that?
New Posts  All Forums:Forum Nav:

USB cable supposedly improving DAC sound quality? How can I take other posts seriously after that? - Page 2  

post #16 of 256
Quote:
Originally Posted by 00940 View Post

Oh please, not again. It's not just "ones and zeroes". That's a gross oversimplification of what a digital connection is.
I now see this kind of post as trolls...saying on an audiophile forum that all digital cables sound the same either proves that:
-no real world experiment has been conducted whatsoever, and somehow ppl want to convince themselves that everything sounds the same...you know, the same kind of thread that says that a $100 DAC and a $1K DAC are impossible to DBT on transparent headphones afrojojo.gif
-low end gear was used
-the testers either aren't trained for analytical listening or have hearing problems.

if all USB/coax/toslink cables sounded exactly the same, this would be such a wonderful world \o/

OTOH, the major issue is that some sellers obviously make 2/3 figures markups...they make you pay for their trial and error at finding the best sounding cables(to them, which might be a different story for you...YMMV as usual in the audio world).

IMHO and IME, the best cable is the shortest..less signal attenuation and less interferences/jitter. And my best sounding cables were cheap too. It's not because all cables sound different that a $100 USB cable will sound better than a $5. It's all about build quality and synergy.

The problem w/ snake oil cables is that those companies need to provide a lot of free review samples and bribe (british?) audiophool magazines to backup their bs arguments. You're paying to be conned basically.

Digital signals are very fast sinewaves, not quite 0's and 1's only. A million things can go wrong in a digital audio cable, believe it or not: http://en.wikipedia.org/wiki/Coaxial_cable#Issues

And unlike HDMI, USB audio and S/PDIF don't use ECC...what's lost *is* lost.
Edited by leeperry - 3/27/11 at 6:20am
post #17 of 256
Quote:
Originally Posted by leeperry View Post
Quote:
Originally Posted by 00940 View Post

Oh please, not again. It's not just "ones and zeroes". That's a gross oversimplification of what a digital connection is.
I now see this kind of post as trolls...saying on an audiophile forum that all digital cables sound the same either proves that:
-no real world experiment has been conducted whatsoever, and somehow ppl want to convince themselves that everything sounds the same...you know, the same kind of thread that says that a $100 DAC and a $1K DAC are impossible to DBT on transparent headphones afrojojo.gif
-low end gear was used
-the testers either aren't trained for analytical listening or have hearing problems.

if all USB/coax/toslink cables sounded exactly the same, this would be such a wonderful world \o/

OTOH, the major issue is that some sellers obviously make 2/3 figures markups...they make you pay for their trial and error at finding the best sounding cables(to them, which might be a different story for you...YMMV as usual in the audio world).

IMHO and IME, the best cable is the shortest..less signal attenuation and less interferences/jitter. And my best sounding cables were cheap too. It's not because all cables sound different that a $100 USB cable will sound better than a $5. It's all about build quality and synergy.

The problem w/ snake oil cables is that those companies need to provide a lot of free review samples and bribe (british?) audiophool magazines to backup their bs arguments. You're paying to be conned basically.

Digital signals are very fast sinewaves, not quite 0's and 1's only. A million things can go wrong in a digital audio cable, believe it or not: http://en.wikipedia.org/wiki/Coaxial_cable#Issues

And unlike HDMI, USB audio and S/PDIF don't use EEC...what's lost *is* lost.

Nice post, but I can't resist...  Digital signals are square waves...beerchug.gif  Which technically can be reconstructed from an infinite number of varying frequency sine waves...but I digress  _|¯¯|__|¯¯|__|¯¯|__|¯¯|__|¯¯|__|¯¯|__|¯¯|__|¯¯|__|¯¯|_
 

 

post #18 of 256

more or less?

 

QA_5.jpg

post #19 of 256
Quote:
Originally Posted by leeperry View Post

more or less?

 

QA_5.jpg


Yup, more or less.  The better the line driver, the better the cable, the more the waveform will resemble a square wave.  Anyway, FWIW, my best USB cables work better than my cheapie USB cables.  When one is transferring data from point A to point B, if the cable works at all, it is ok.  The transceivers will initiate a resend if data corruption happens along the way.  In digital audio signal transmission, missed data due to corruption is missed data.  That missing data is "guessed" at (interpolated) by one's system and clocked out.  If the guess that fills in the missing data is wrong, then you have a dropout or glitch and it is VERY audible.  My good cables seem to be glitch free and that's all I care about, because after all, it's digital.  wink_face.gif  With many of my cheapie cables, I get glitches in my audio.  Cables ARE important, just keep everything within reason.  If your chepie cables allow you to play your digital library error free (no audible glitches) then you're good to go.

 

post #20 of 256

Well, IME even glitch-free digital cables can sound very different...I've had USB cables that really sounded "colored", in a bad and boring way. My best sounding USB cable sounds much clearer and far less colored.

 

OTOH, all my USB coax transports work in adaptive mode. I realize that a properly async transport shouldn't make cables sound different, but they still seem to matter.

post #21 of 256
Quote:
Originally Posted by leeperry View Post

Well, IME even glitch-free digital cables can sound very different...I've had USB cables that really sounded "colored", in a bad and boring way. My best sounding USB cable sounds much clearer and far less colored.

 

OTOH, all my USB coax transports work in adaptive mode. I realize that a properly async transport shouldn't make cables sound different, but they still seem to matter.


Now I have to ask, does this same USB cable make your pictures clearer and more in focus? If it affects the sound going through it, I am going to assume it affects everything that passes through it. Try hooking it to your camera and see if it enhances your megapixels. rolleyes.gif


You know this is the sound science forum, and we want to see proof that it actually changes the sound. Your subjective opinion is not proof. And before you go there, yes, my gear is "low end" therefore I cannot "hear" the differences. I know how this goes. I've seen it a million times. I just know from all of the reading I have done, that the making no difference argument is MUCH stronger than the makes a difference one. A bunch of people looking at cables, listening to them, and then saying there is a difference, is not proof.
post #22 of 256

well, it's on a strictly IME/YMMV level...maybe tiny impedance mismatches/interferences make a hell of a difference on the cheapo Tenor chip internal DSP, few ppl have the right gear to measure jitter anyway.

 

last time I saw some jitter measurements for a coax cable, it was different depending on which direction it was used: http://www.stereophile.com/content/transport-delight-cd-transport-jitter-page-4

 

that was a mere coax cable, the more strands the more problems start arising: skin effect, intra-pair skew, inter-pair skew, Far End Crosstalk(FEXT), Near-End Crosstalk (NEXT), etc etc...and the worse it gets when you start embedding data and power cables together within the same sleeve.

 

tbh, I'm not up for 20 pages of sterile debating again.


Edited by leeperry - 3/27/11 at 10:17am
post #23 of 256
Quote:
Originally Posted by nick_charles View Post


But are any of these differences verifiably audible ?

 

To date the empirical evidence is no except in the case of actually faulty cables, to date published evidence of audibility of jitter places audibility at levels well above any sligthly iffy cable viz several nanoseconds, digital kit measured by stereophile at 14, 9, 3 and 4 ns of peak-to-peak jitter has been subjectively reviewed as fine, i.e these massive levels of jitter have not caused reviewers to flee clutching their ears. Published peer reviewed AES articles from researchers at Dolby Labs and NHK (Japan)  have tested listeners' ability to detect jitter, in both cases very high levels of jitter (up to 250ns random jitter, 10 - 20ns correlated jitter) were extremely hard to detect.

 

I'd simply not worry about jitter until I had better evidence that it is audible at levels found in competent digital audio kit.



you are the first person i have ever seen here who spoke about jitter in a way that didnt seem rediculous.  are there any threads you could point us to that contain more information of this kind.  most of what i have read here about jitter clearly came form ppl who had no idea about it, and made it out to be some boogie man of audio.  I have still yet to see a post where someone actually said they heard jitter before, or can describe what it sounds like.

 

 

 

post #24 of 256

 

Originally Posted by Br777 View Post

describe what it sounds like.


harmonic distortion in the time domain...there's a simple test to "hear" it: use a Firestone Bravo to reclock ±1000ps toslink to 50ps coax through WM8804: http://hifiduino.blogspot.com/2010/02/programming-wm8804.html

 

Done \o/


Edited by leeperry - 3/27/11 at 8:10am
post #25 of 256

@the OP, if you're still around read this: It's nonsense. Audiophiles are always looking for something to increase their sound and will justify and defend their expensive purchase to the death. Psychosomatics play a huge role as well. There's a perceived difference because they expect and believe there will be. Also anything from Stereophile I've ever read is null and void because they endorsed this product:

 

http://www.musicdirect.com/product/73520

post #26 of 256

I have no idea what that does, but it stands to reason that if you (re)move charges around a LP, you can make it attract less dust, and this would lead to a better sounding disc. This being said, though, the claims in that site seem ridiculous.

post #27 of 256
Quote:
Originally Posted by tmars78 View Post

Now I have to ask, does this same USB cable make your pictures clearer and more in focus? If it affects the sound going through it, I am going to assume it affects everything that passes through it. Try hooking it to your camera and see if it enhances your megapixels. rolleyes.gif

 

You're assuming things that a 5 min search through the USB specifications white paper (and the annex about USB audio) would prove wrong. USB protocols are not the same for USB audio and for a camera (isochronous adaptive vs bulk mode). Different protocols have different vulnerabilities. In the case of USB audio, the problem is not even a question of data but mostly of timing (the accuracy of the reconstructed analog sinewave being function of data and timing), a factor utterly irrelevant to your pictures example. This kind of posts with the "digital is perfect" motto and so on are so easily dismissed that they actually reinforce audiophiles in their beliefs.

 

There are good reasons why one could dismiss the use of "audiophile" USB cables. One could say that they're not the weakest link in the usb audio chain and that any degradation that they would introduce is below the native problems of the interface (pretty much my opinion for now). One could say that the performance of USB audio is usually good enough anyway (nick_charles's argument).

 

I could imagine though one reason why different USB cables could have an impact. If the USB source is rotten wrt noise (high level of HF noise in the 5V USB line, it's quite frequent), different cables constructions/geometry might either attenuate the crap getting to the DAC side or at the contrary couple some of that noise into the PS lines of the DAC (or maybe even into the data lines). Such a model has many variables: quality of the computer PS and USB interface, the cable, the internal construction of the dac (ICs used, decoupling, pcb layout, etc.) and its design (usb powered or not). Not much of digital question btw, but, well, an USB cable is a dual use one: data and power.

post #28 of 256
Quote:
Originally Posted by 00940 View Post

 

You're assuming things that a 5 min search through the USB specifications white paper (and the annex about USB audio) would prove wrong. USB protocols are not the same for USB audio and for a camera (isochronous adaptive vs bulk mode). Different protocols have different vulnerabilities. In the case of USB audio, the problem is not even a question of data but mostly of timing (the accuracy of the reconstructed analog sinewave being function of data and timing), a factor utterly irrelevant to your pictures example. This kind of posts with the "digital is perfect" motto and so on are so easily dismissed that they actually reinforce audiophiles in their beliefs.

 

There are good reasons why one could dismiss the use of "audiophile" USB cables. One could say that they're not the weakest link in the usb audio chain and that any degradation that they would introduce is below the native problems of the interface (pretty much my opinion for now). One could say that the performance of USB audio is usually good enough anyway (nick_charles's argument).

 

I could imagine though one reason why different USB cables could have an impact. If the USB source is rotten wrt noise (high level of HF noise in the 5V USB line, it's quite frequent), different cables constructions/geometry might either attenuate the crap getting to the DAC side or at the contrary couple some of that noise into the PS lines of the DAC (or maybe even into the data lines). Such a model has many variables: quality of the computer PS and USB interface, the cable, the internal construction of the dac (ICs used, decoupling, pcb layout, etc.) and its design (usb powered or not). Not much of digital question btw, but, well, an USB cable is a dual use one: data and power.


I said it in jest, which is why I had the eyeroll at the end of it. I wasn't at all being serious. Sorry if it came off that way. I have done lots of reading on these things, especially since I have been a member here. I have learned boatloads that I did not know before.
post #29 of 256



 

Quote:
Originally Posted by Br777 View Post





you are the first person i have ever seen here who spoke about jitter in a way that didnt seem rediculous.  are there any threads you could point us to that contain more information of this kind.  most of what i have read here about jitter clearly came form ppl who had no idea about it, and made it out to be some boogie man of audio.  I have still yet to see a post where someone actually said they heard jitter before, or can describe what it sounds like.

 

Nobody here is qualified to do this as nobody here has the proven capability to detect known levels of jitter in a blind test.

 

 

 


 

Hydrogen Audio has a few threads dedicated to jitter audibility including some technical info, and there are a few papers on the topic. For theory the works of Malcolm Hawksford and Julian Dunn are good but neither of these did controlled listening tests.

 

The best starting point for actual proper (non anecdotal) listening tests is:

 

Theoretical and Audible Effects of Jitter on Digital Audio Quality, Eric Benjamin and Benjamin Gannon, 105th AES Convention 1998

 

sadly this is only available via the AES at a cost of $5 for members or $20 for non members, if you are interested and are a student you can get AES membership cheaply.

 

On the web you can find

 

Detection threshold for distortions due to jitter on digital audio, Ashihara et al., Acoustic Science & Technology, 26,1 2005

 

- this papers takes a different tack evaluating the audibility of random jitter.

 

There is no credible empirical evidence whatsoever that jitter in the sub-nanosecond p-p range is audible.

 

The degradation due to jitter is easily measured and 10ps will degrade a 16 bit signal to being effectively a 15 bit signal at 20khz , but this is at a range where  human hearing is not sensitive enough for it to make a difference, at lower frequencies the level of jitter required to degrade a signal increases until it becomes several 100 ps, Julian Dunn did some excellent work on this but his estimates (never empirically tested) are based on people listening at 120db above background noise !

 

The shocking truth is that as detectors of audible defects we humans are really far less sensitive than a $29 ADC connected to a set of RCA analog outputs or a freeware audio program!

 

The other thing worth doing is to do look at Stereophile's jitter measurements, any digital component that is not utter pants shows distortion sidebands that are at least 90db down on signal, my ancient 1998 Entech does even better (-117db) and the mass-produced $350 Marantz CD5004 has jitter sidebands that are over 124db down on signal, how audible do we really think this is likely to be ?

 

For the poster child of really bad measured jitter see...

http://www.stereophile.com/content/mcintosh-ms750-music-server-measurements

 

yet even this was deemed okay until the reviewer was told just how bad it was wink.gif

 


Edited by nick_charles - 3/27/11 at 11:37am
post #30 of 256

 

Quote:
Originally Posted by tmars78 View Post

I said it in jest, which is why I had the eyeroll at the end of it. I wasn't at all being serious. Sorry if it came off that way. I have done lots of reading on these things, especially since I have been a member here. I have learned boatloads that I did not know before.



My apologies then. I thought your ironic eyeroll was directed against the other side.

New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Sound Science
This thread is locked  
Head-Fi.org › Forums › Equipment Forums › Sound Science › USB cable supposedly improving DAC sound quality? How can I take other posts seriously after that?