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 6  

post #76 of 256

You can, if you believe it hard enough wink.gif

 

I know that has nothing to do with science but it's a common thing around the rest of Head-fi.


Edited by RazorJack - 1/10/12 at 9:59am
post #77 of 256

Digital audio playback and digital file transfer is not the same thing.  There is no checksum involved in streaming data to the D2A - what's going to do the checksum, the D2A? When?  It never has the whole file, let alone the HW/SW needed to generate and compare the checksum.  Audio playback over USB has nothing in common with file transfer via TCP/IP, so forget all about that sort of error correction and retransmission request- it's irrelevant in this context.  USB audio has little in common with USB file transfer - in that case there is error checking and correction.

 

(Most) D2As don't store files in memory, nor does re-clocking work like many seem to think.  The hardware necessary simply isn't there.  A D2A asks for data and receives it, either on the PC's schedule or its own.  It can't collect it all, look at some magic parameter in the data (it's not there!) and reassemble the file in the proper sequence and time.  As long as the data is not too out of step, the D2A can make adjustments but if it's too late, nothing the D2A can do will make it early or even on time.

 

As far as cables go, don't make the mistake of thinking your experience is the sum total of all knowledge on the (any) subject.  Every day, mankind learns new explanations for phenomena it either thought it understood or never understood.  The fact that you (and I) don't know how to measure or explain a given phenomenon neither proves nor disproves that phenomenon's existence. I remember when no one in the audiophile world had heard of jitter and had to be convinced that 1) it existed and, 2) was meaningful.  Jitter can be introduced many places in the signal chain.  Poorly constructed cables are a common source.  Materials, construction techniques, build quality - all affect the audio data stream. 

 

And yes, it is a stream, not a file transfer.  The PC delivers the data to be converted by the D2A via USB.  Making the data stream subject to the same problems a s/pdif streamed or a BNC streamed signal are vulnerable to.  There is no decoding taking place in the D2A - that all happens in the PC in the player software.  So yes, subtle changes are exactly what you get between various USB cables used for audio; just like you'll observe when comparing any nominally similar audio cables that possess the base physical properties needed to complete the electrical circuit.

post #78 of 256
Quote:
Originally Posted by SeeHear View Post

Digital audio playback and digital file transfer is not the same thing.  There is no checksum involved in streaming data to the D2A - what's going to do the checksum, the D2A? When?  It never has the whole file, let alone the HW/SW needed to generate and compare the checksum.  Audio playback over USB has nothing in common with file transfer via TCP/IP, so forget all about that sort of error correction and retransmission request- it's irrelevant in this context.  USB audio has little in common with USB file transfer - in that case there is error checking and correction.

 

(Most) D2As don't store files in memory, nor does re-clocking work like many seem to think.  The hardware necessary simply isn't there.  A D2A asks for data and receives it, either on the PC's schedule or its own.  It can't collect it all, look at some magic parameter in the data (it's not there!) and reassemble the file in the proper sequence and time.  As long as the data is not too out of step, the D2A can make adjustments but if it's too late, nothing the D2A can do will make it early or even on time.

 

As far as cables go, don't make the mistake of thinking your experience is the sum total of all knowledge on the (any) subject.  Every day, mankind learns new explanations for phenomena it either thought it understood or never understood.  The fact that you (and I) don't know how to measure or explain a given phenomenon neither proves nor disproves that phenomenon's existence. I remember when no one in the audiophile world had heard of jitter and had to be convinced that 1) it existed and, 2) was meaningful.  Jitter can be introduced many places in the signal chain.  Poorly constructed cables are a common source.  Materials, construction techniques, build quality - all affect the audio data stream. 

 

And yes, it is a stream, not a file transfer.  The PC delivers the data to be converted by the D2A via USB.  Making the data stream subject to the same problems a s/pdif streamed or a BNC streamed signal are vulnerable to.  There is no decoding taking place in the D2A - that all happens in the PC in the player software.  So yes, subtle changes are exactly what you get between various USB cables used for audio; just like you'll observe when comparing any nominally similar audio cables that possess the base physical properties needed to complete the electrical circuit.

 

Great! Now you just have to prove a few things:

  • that reduced jitter does everything everyone claims USB cables do, like improves sound stage and bass extension
  • that high end cables have enough jitter reduction to be audible
  • what materials and construction techniques matter

 

Then your statements will cease to be assumptions and assertions, and become facts.

post #79 of 256
Quote:
Originally Posted by SeeHear View Post

I remember when no one in the audiophile world had heard of jitter and had to be convinced that 1) it existed and, 2) was meaningful. 

 

Sigh, here we go again. Nobody here doubts the existence of jitter. The issue is at what level does it become audible. To date the evidence for audibility of sub-ns jitter in controlled listening tests is a nice round zero. There are few published papers (Benjamin and Gannon, 1998; Ashihara et al, 2005) empirically examining the connection between the magnitude of jitter and its verifiable audibility but all that do exist point to levels of jitter well above normally competent digital kit.

 

 

 

Jitter can be introduced many places in the signal chain.  Poorly constructed cables are a common source (citation ?).  Materials, construction techniques, build quality - all affect the audio data stream. 

 

And yes, it is a stream, not a file transfer.  The PC delivers the data to be converted by the D2A via USB.  Making the data stream subject to the same problems a s/pdif streamed or a BNC streamed signal are vulnerable to.  There is no decoding taking place in the D2A - that all happens in the PC in the player software.  So yes, subtle changes are exactly what you get between various USB cables used for audio (citation ?) ; just like you'll observe when comparing any nominally similar audio cables that possess the base physical properties needed to complete the electrical circuit.

 

Several folks here (including PIO2001) have measured analog audio cables,  the differences in the user-end parameters such as FR, Noise and distortion have to date been found to be extremely small and (so far) generally below the threshold of audibility.



 

 

post #80 of 256


 

Quote:
Originally Posted by SeeHear View Post

Digital audio playback and digital file transfer is not the same thing.  There is no checksum involved in streaming data to the D2A - what's going to do the checksum, the D2A? When?  It never has the whole file, let alone the HW/SW needed to generate and compare the checksum.  Audio playback over USB has nothing in common with file transfer via TCP/IP, so forget all about that sort of error correction and retransmission request- it's irrelevant in this context.  USB audio has little in common with USB file transfer - in that case there is error checking and correction.

 

(Most) D2As don't store files in memory, nor does re-clocking work like many seem to think.  The hardware necessary simply isn't there.  A D2A asks for data and receives it, either on the PC's schedule or its own.  It can't collect it all, look at some magic parameter in the data (it's not there!) and reassemble the file in the proper sequence and time.  As long as the data is not too out of step, the D2A can make adjustments but if it's too late, nothing the D2A can do will make it early or even on time.

 

As far as cables go, don't make the mistake of thinking your experience is the sum total of all knowledge on the (any) subject.  Every day, mankind learns new explanations for phenomena it either thought it understood or never understood.  The fact that you (and I) don't know how to measure or explain a given phenomenon neither proves nor disproves that phenomenon's existence. I remember when no one in the audiophile world had heard of jitter and had to be convinced that 1) it existed and, 2) was meaningful.  Jitter can be introduced many places in the signal chain.  Poorly constructed cables are a common source.  Materials, construction techniques, build quality - all affect the audio data stream. 

 

And yes, it is a stream, not a file transfer.  The PC delivers the data to be converted by the D2A via USB.  Making the data stream subject to the same problems a s/pdif streamed or a BNC streamed signal are vulnerable to.  There is no decoding taking place in the D2A - that all happens in the PC in the player software.  So yes, subtle changes are exactly what you get between various USB cables used for audio; just like you'll observe when comparing any nominally similar audio cables that possess the base physical properties needed to complete the electrical circuit.



I'll agree that the real-time transmission protocol does not ensure absolute data integrity in the same way that bulk data transfer does, but I'll disagree with you on almost every other point. Drawing attention to the limitations of science is all very well, but just because science is limited doesn't mean that ignoring it is fully justifiable. As for jitter, it was known about long before the audiophiles got overexcited about it. It isn't very difficult to deal with in a DAC to the point where suggesting it would be audible is farcical (if commonly done), whilst the effect cables have upon it is utterly insignificant.

Jitter is an issue, but its importance has been grossly overstated. People now look at quite stupidly small amounts of jitter and declare it audible, despite the complete lack of any evidence to support their assertions. This is nothing new: before jitter intermodular distortion was the favourite audiophile thing to harp on about.

 

EDIT: Beaten to it.

post #81 of 256

Sorry - my intent is not to support or deny jitter's effect or any aspect of how it affects (or doesn't) one's perception of a music playback system.  I used it only as an (obviously poorly chosen) example of how knowledge evolves.  Nor am I saying that I know how/what to measure to predict the change in performance of a playback system imparted by changing one or several parts (cables, in this case) in that system.  In other words: Dude, I have no idea what jitter sounds like, only that it is measurable!

 

However, I do assert that when I hear differences in my system when I make changes, I will attribute said differences to the cable if that is what was changed.  I have heard differences in digital cables - I don't know what physical property of the cable is responsible, but I know which cable I preferred at that time with that set of components.  If you must know, I think that any well built digital cable sounds pretty much the same as any other.  Some, however, really screw up the sound.  I don't keep those.

 

Incidentally, I don't use USB from my PC; I like s/pdif because, with the PC and components I have at my disposal, it gives better results.

 

Having been raised by scientists, I reject assertions based on "I can't measure it, so it can't be."  All science starts with observation, as technology allows, we use it to observe.  Go back far enough in time and you will find proof that many things we "know" today are wrong, based on then-available measuring tools and techniques.  My point being we can hardly assume that we are Blessed With All The Knowledge There Is To Know About Everything just because now we are alive to know it.

 

And, Nick_Charles, at the risk of inflaming you again, who says that "FR, Noise and distortion" are the relevant characteristics affecting what the listener heard?  One could argue that your experiment's measurements have merely proven they aren't. 

 

And, for the record, I didnt say anyone here doubts that jitter exists - I said I remember when it was an unknown phenomenon.  I believe that you think it is inaudible, and I hope you believe me when I say I have no idea what its affect sounds like or even if it's audible at all.  I do know that in digital products I like, claimed jitter tends to be lower - when I've bothered to check the specs.  Though, I must confess, when it comes to audio equipment, I pay little attention to specs unless I'm trying to explain somethng I didnt expect or hadn't encountered before.

 

I have come to believe that a lot of the dissatisfaction I experienced early on with PC based audio stemmed from sample rate conversion and bit depth mismatches imparted by the crappy Creative software and sound cards I used.  The cables had very little, comparatively speaking, to do with the sound.  I know that, regardless of which USB cables I tried, the USB D2As I tried all sound unacceptable in my system.  I also know that people whose opinions I respect claim to get great results with USB audio and hear relevant differences in USB cables - usually they like the less expensive ones. Go figure. Great for them, I have no interest in reconciling my experience with theirs, and less interest in proving them wrong.

post #82 of 256

I'm probably gonna be beheaded for commenting on this but I did notice a slight difference when going between the stock cable included with my Fiio E7 DAC and when I used the one provided with my DACport LX. It was definitely more clear using the cable from the DACport. I'm not sure why this is but I could clearly hear it. I didn't attempt to fool myself as other may say or imagine it (for that matter I didn't even attempt to test this cable out to see if it was clearer). It was just a clear difference. I don't give a rats-@ss about the negative comments nor am I trying to troll. Just stating my opinion on what I heard.

post #83 of 256
Quote:
Originally Posted by lee730 View Post

I'm probably gonna be beheaded for commenting on this but I did notice a slight difference when going between the stock cable included with my Fiio E7 DAC and when I used the one provided with my DACport LX. It was definitely more clear using the cable from the DACport. I'm not sure why this is but I could clearly hear it. I didn't attempt to fool myself as other may say or imagine it (for that matter I didn't even attempt to test this cable out to see if it was clearer). It was just a clear difference. I don't give a rats-@ss about the negative comments nor am I trying to troll. Just stating my opinion on what I heard.


You don't choose to succumb to placebo.

post #84 of 256
Quote:
Originally Posted by Head Injury View Post


You don't choose to succumb to placebo.



I beg to differ. You're much more likely to succumb to placebo when you are expecting a result; in which I was not. I just grabbed a random cable to hook up my device. Later when I went back and hooked up the stock cable I noticed the difference.


Edited by lee730 - 1/10/12 at 3:11pm
post #85 of 256
Quote:
Originally Posted by lee730 View Post

I beg to differ. You're much more likely to succumb to placebo when you are expecting a result. In which I was not. I just grabbed a random cable to hook up my device. Later when I went back and hooked up the stock cable I noticed the difference.


You should take an Implicit Association Test. It's a pretty interesting concept.

 

If you're interested in learning about all the ways our subconscious screws with us, you could take whole psychology courses. I took a social psychology course and a lot of it could be applied to audio. I imagine a cognitive psychology course would work equally well.


Edited by Head Injury - 1/10/12 at 3:15pm
post #86 of 256
Quote:
Originally Posted by Head Injury View Post


You should take an Implicit Association Test. It's a pretty interesting concept.

 

If you're interested in learning about all the ways our subconscious screws with us, you could take whole psychology courses. I took a social psychology course and a lot of it could be applied to audio. I imagine a cognitive psychology course would work equally well.



I took 3 of those tests, the age, race and weight one. Scored the same on all of them. That I have no automatic preference to any of them. 

post #87 of 256
Quote:
Originally Posted by SeeHear View Post

However, I do assert that when I hear differences in my system when I make changes, I will attribute said differences to the cable if that is what was changed.  I have heard differences in digital cables -

 

Have you tried measuring the differences between the cables. To date nobody has been able to pass a DBT with digital cables, a world renowned stats expert (Wavoman) here set up some trials between bog-standard and high quality digital cables and none of his subjects were able to tell them apart (blind) nor to the best of my knowledge has anybody else. Hearing a difference in a casual listening test is very different from hearing a difference when unaware of the identity of the component under test. Back in the 80s Masters and Clark ran DBTs between amplifiers (with stable loads and running below clipping) in the sighted portion several listeners heard clear and definable differences between the amps. When the test became blind nobody managed a statistically significant level of discrimination between several very different amps. I am not being rude when I say that imagination is a powerful deceiver.

 

And, Nick_Charles, at the risk of inflaming you again, who says that "FR, Noise and distortion" are the relevant characteristics affecting what the listener heard?  One could argue that your experiment's measurements have merely proven they aren't. 

 

You mistake a northern world-weariness for incandescence. See me when I am dealing with cheating undergraduates. I used these as examples. instead of RLC,  of things we might hear, we cannot hear inductance, capacitance or resistance, but we can add crosstalk, group delay and phase distortion if you wish, I am unaware of any tests of "conventional" cables where significant (and potentially audible) differences in these parameters have been found. But this is all meaningless until someone actually passes a DBT between two nornal cables.

post #88 of 256

 

Quote:
Originally Posted by RazorJack View Post

You can, if you believe it hard enough wink.gif

 

I know that has nothing to do with science but it's a common thing around the rest of Head-fi.

 

C'mmon now, don't be unkind.  Some folks just don't know any better.  They may have fallen for hyped ad copy, and don't realize that it's truly impossible for any "subtle" sonic, or for that matter ANY "sonic" alterations to take place.  Bit errors might be interpreted differently by different codecs but the sonic difference would be anything but subtle.

post #89 of 256

If it were analog I'd think about how the change would occur but in digital, its a straight NO.

 

A digital 1 is anything above 3V. So if the cable gives a 3.001 or 3.003 its still a 1.

Most of the times the standard followed is 3.3V to ensure a clear differentiation between high and low.

 

If the cable were to change a 1 into 0 (make it below 3V) , I doubt you'd even recover the signal.

 

 

post #90 of 256

Don't forget, the 0's and 1's are anything but static and a good USB 2.0 cable must be able to pass 240MHz, to faithfully deliver the data.

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?