USB cable and Sound Quality - Page 33

Quote:

Originally Posted by joeyjojo 

Interesting questions bigshot. As it might be semi-related could I point you clever lot at this thread http://www.head-fi.org/t/620094/dropouts-in-foobar-over-spdif-with-boinc-and-eist-solved

 

In short, I used to run intensive computing progs in the background while listening to music through an Asus DX -> CA Dacmagic. Occasionally you'd hear very tiny clicks, but they were so quiet and brief as to be ignorable and I didn't realise it was a problem. Then I swapped the DX out for an M-Audio 2496 and it was terrible! Clicks more frequent and more disruptive to the music. However with BOINC switched off and speedstep disabled it was fine. Recently I've been using an AMB gamma2 (gone back to the DX) and it must be far more robust (it has a clever asynchronous something-or-other) as it seems to be immune to any of these problems regardless of what else is running.

 

So what were these noises I was hearing? Perhaps they were this mysterious "jitter" caught in the act? Or buffer under runs? You can try the same thing yourself using a cpu stress test - something like prime95, Intel Burn Test, or LinX, and listen out.

My guess would be program thread priority, your CPU intensive programs where taking up CPU time such that the audio decoder got pushed down in priority which resulted in bad and/or interrupted decoding (it's pretty common back in earlier days when CPU weren't as powerful and we didn't have multi-core CPUs, like say 486 days, I can still remember interruption of music playback and introduction of distortion back then when I tried to fire up the browser or something intensive).  Also the way your CPU intensive programs taking up CPU would be relatively random depending on many random factors so you didn't get repeatable noise results.

There are so many more plausible explanations to this than "USB cable induced jitter" that I would investigate in first.  

My point was that it's a good way of hearing for yourself the kind of artifacts that appear when a digital system isn't working optimally, i.e. obvious clicks and gaps, not subtle things like "wow" factor.

Quote:

Originally Posted by nanaholic 

 

My guess would be program thread priority, your CPU intensive programs where taking up CPU time such that the audio decoder got pushed down in priority which resulted in bad and/or interrupted decoding (it's pretty common back in earlier days when CPU weren't as powerful and we didn't have multi-core CPUs, like say 486 days, I can still remember interruption of music playback and introduction of distortion back then when I tried to fire up the browser or something intensive).  Also the way your CPU intensive programs taking up CPU would be relatively random depending on many random factors so you didn't get repeatable noise results.

 

There are so many more plausible explanations to this than "USB cable induced jitter" that I would investigate in first.  

How about occasional DPC's taking 50% CPU-time like certain EMU-devices ? (0202/USB)

Quote:

Originally Posted by bigshot 

Audible jitter is extremely rare. It really isn't an issue.

Well, if the interfaces are within spec and reasonably well designed, there are issues other than jitter, that may have a stronger impact on performance... USB cables that meet the spec are not necessarily expensive... Monoprice sells them for about a $1 (may need to add tax and shipping though):

http://www.monoprice.com/products/product.asp?c_id=103&cp_id=10303&cs_id=1030301&p_id=8615&seq=1&format=2

http://www.monoprice.com/products/product.asp?c_id=103&cp_id=10303&cs_id=1030301&p_id=5437&seq=1&format=2

Most headphone's FR and distortion, along with poorly recorded music, are usually more problematic IMO... Well, it was for me anyway.

Quote:

Originally Posted by joeyjojo 

My point was that it's a good way of hearing for yourself the kind of artifacts that appear when a digital system isn't working optimally, i.e. obvious clicks and gaps, not subtle things like "wow" factor.

Yes. This. It works or it doesn't... it doesn't affect "soundstage" or the bass, or clarity - it drops out, clicks and gaps... that's all a USB cable or digital transfer protocol can be expected to work with.

Quote:

Originally Posted by liamstrain 

 

Yes. This. It works or it doesn't... it doesn't affect "soundstage" or the bass, or clarity - it drops out, clicks and gaps... that's all a USB cable or digital transfer protocol can be expected to work with.

Agree. Even if you take it to the extreme, packet drops won't get audible until there's some real slowdown, I mean, if you're hearing clicks and gaps, your system is really screwed up. On the other hand, a couple of milliseconds won't be noticeable at all.

Again, its still not related to USB cables.

Have you guys heard different USB transports - say at one end a cheapo older adaptive transfer one with single frequency oscillator compared to a newer USB transport with asynchronous transfer, dual high precision oscillators that measures well etc.  It's easier still to just compare the inbuilt USB on a DAC to one of the higher quality USB transports on the market.  If you are using a Benchmark DAC-1 that's cheating because of the asynchronous SPDIF inputs.  You can even just buy an old HIFace and compare the stock HiFace driver to the Young DAC driver, or compare with or without a USB cable extension.  I don't have proof that these sound different (in my experience I think they do), but I am trying to understand where you are coming from regarding 'dropouts or perfect audio' thing.

Quote:

Originally Posted by dleblanc343 


I can report the same clicking noises. It is frequent with a generic usb cable, maybe once or twice per three minutes of music playback. The audioquest carbon also does the same but myabe a clip every three minutes only. The tellurium q usb cable so far has been almost flawless; in maybe 30 hours of playback in the past 3 weeks and i can only recall noticing one clippage in sound. All cables are 1.5 meters

Ineteresting, can't say I have noticed anything similar on my system, but my setup is different as I am using USB transport with galvanic isolation connected to grounded computer and DAC on the same circuit.  So this is using USB from a laptop into a Bifrost DAC?  Without knowing what is inside the Tellurium Q cable it is hard to figure out what is going on there  If we could debug the USB receiver then maybe we could say whether or not the cable is affecting the buffer management causing buffer underrun or whether the dropout is from some sort of electrical interference.

The Bifrost reportedly clicks when the sampling rate changes.

As above if you're maxing out your cpu it's nothing to do with the cables, it's probably a buffer underrun. I was just pointing out that (working hypothesis) USB either works well, and no cable is going to change the sound; or it doesn't and you get big obvious noises like clicks and pops.

One exception may be if the device is also USB powered, but then you should look at getting a better power supply instead of a premium cable.

Quote:

Originally Posted by joeyjojo 

As above if you're maxing out your cpu it's nothing to do with the cables, it's probably a buffer underrun. I was just pointing out that (working hypothesis) USB either works well, and no cable is going to change the sound; or it doesn't and you get big obvious noises like clicks and pops.

 

One exception may be if the device is also USB powered, but then you should look at getting a better power supply instead of a premium cable.

The theory among people developing audiophile computer programs etc seems to be that you get best performance by ensuring a smooth flow of data rather than one with unreliable latency caused by dropped packets, hardware interrupts etc.  One might say that the solution is to just use a larger buffer, but computer audiophiles tend to prefer to use as small a buffer as possible for whatever reasons that are beyond my understanding of digital sound processing (other than input delay).  My favorite music software can playback with "direct link" buffer setting and still maintain skip free, even in windows with a browser window running.  With other music software like Foobar I have to use above 30 ms otherwise there are skips, then again the software I use pretty much decodes and buffers the whole song into ram ahead of time, and even then has different playback engines which have subtle differences in sound and yet are bit perfect...  Probably getting pretty far out of sound science territory here but got a bit carried away talking about buffer management...

Quote:

Originally Posted by drez 

 

The theory among people developing audiophile computer programs etc seems to be that you get best performance by ensuring a smooth flow of data rather than one with unreliable latency caused by dropped packets, hardware interrupts etc.  One might say that the solution is to just use a larger buffer, but computer audiophiles tend to prefer to use as small a buffer as possible for whatever reasons that are beyond my understanding of digital sound processing (other than input delay).  My favorite music software can playback with "direct link" buffer setting and still maintain skip free, even in windows with a browser window running.  With other music software like Foobar I have to use above 30 ms otherwise there are skips, then again the software I use pretty much decodes and buffers the whole song into ram ahead of time, and even then has different playback engines which have subtle differences in sound and yet are bit perfect...  Probably getting pretty far out of sound science territory here but got a bit carried away talking about buffer management...

You all need to realize that playback is not the same as recording ..

Obviously, when recording, any 'latency' is unacceptable -

YES, the guitarist,bassist, drummer whatever - will notice a 30ms delay from doing something till hearing it ..

It's enough to make the drummer off-beat !

When playing back a digital file from a computer - There is no such thing as 'latency' !

You are NOT playing a LP-record, 'decoded' in 'real-time' .

Ultimately, you will want the entire track read from disk to RAM and played from RAM without further disk-access .

CD's are almost read like LP's - That's why Sony made a load of money on that 'anti-skip' thing on the Discmans .

They did what foobar and all other computer-playback software does : Load it to RAM and play from there .