[googleborg]

nice links, using an electron microscope to laser cut traces on the silicon...sounds fun ^_^
 
part 3 is still there btw, links at the bottom of part two
 
http://www.eetimes.com/design/audio-design/4009528/Meeting-the-Distortion-Goals-The-D-A-Diaries-Part-3-of-3

 

[DanielCox]

Quote:

nice links, using an electron microscope to laser cut traces on the silicon...sounds fun ^_^

Trust me, it isn't

 

[MrGreen]

Difference in USB cable threads should be banned because the medium actually checks itself for accuracy.
 
The only difference is an asynch usb driver which allows for a higher sampling resolution

 

[Prog Rock Man]

All cable threads have one side who find physical and other differences in cables, such as jitter and then say that such are the reason why they hear a difference in such cables. But to do that other reasons within themselves are ignored.
 
There is nothing in this thread that shows the 'differences' in USB cables are audible.
 

 

[ex0du5]

As a computer engineer who has worked with USB and even much lower levels of digital transfer, I cry folly as to the "differences" in USB cables.
 
Digital data sent over the USB cables is error corrected and buffered. As long as there is a moderate buffer, any errors that occur during the initial transmission will be corrected before being output from the buffer to the source.
 
I can guarantee 100% that there won't be a difference. I have yet to see any cable fail to deliver 100% accurate data (which, working with custom FPGA designs, would be absolutely critical to success).
 
The jitter in the signal will come from the source, not the cable. And even then, the jitter is only an issue if the receiving end is not buffered...which would be a crying shame in any digital system.
 
The only time a high quality USB cable would be needed would be for an extremely fast and unbuffered connection that must be transmitted quickly...but in that case you'd be more likely to use a serial connection, as USB just has too much overhead.

 

[googleborg]

so what if your a computer engineer with 'experience', have you actually bought a £1000 usb cable and listened to it, otherwise i don't see how you can possibly comment.  the difference is revelatory, i assure you.  inky blacks.
 
 
 
i jest

 

[googleborg]

Quote:

Quote:

nice links, using an electron microscope to laser cut traces on the silicon...sounds fun ^_^

Trust me, it isn't

 
but lasers are fun
 
 

 

[zenpunk]

Surely, nobody is that stupid to spend that amount of money on some  wires!...

 
Quote:

so what if your a computer engineer with 'experience', have you actually bought a £1000 usb cable and listened to it, otherwise i don't see how you can possibly comment.  the difference is revelatory, i assure you.  inky blacks.
 
 
 
i jest

 

[googleborg]

i think it might be a form of self-worth or something, supreme confidence in themselves and their abilities to hear things, and it's always men, oddly enough, so perhaps some kind of aspergers/OCD  obsessive thought processes, basically weaknesses in peoples/the male characters that force them to buy such things to ease a kind of existential pain?  /ramble
 

 

[kite7]

True story, happens all too often here. The "who cares but did you buy blah blah and try dah dah" attitude
 
Quote:

so what if your a computer engineer with 'experience', have you actually bought a £1000 usb cable and listened to it, otherwise i don't see how you can possibly comment.  the difference is revelatory, i assure you.  inky blacks.
 
 
 
i jest

 

[ex0du5]

Quote:

so what if your a computer engineer with 'experience', have you actually bought a £1000 usb cable and listened to it, otherwise i don't see how you can possibly comment.  the difference is revelatory, i assure you.  inky blacks.
 
 
 
i jest

 Well played, sir. Well played.

 

[00940]

Quote:

Originally Posted by ex0du5 /img/forum/go_quote.gif
 
Digital data sent over the USB cables is error corrected and buffered. As long as there is a moderate buffer, any errors that occur during the initial transmission will be corrected before being output from the buffer to the source.

Did you read the links I provided ? USB audio is working with extremely small buffers, too small for error correction. They're made just big enough to account for the variations in the clocking of the arriving packets.
 
 
edit: and I don't "believe" in USB cables making any differences. The interface is inherently poor, can't see a cable degrading the situation much.

 

[ex0du5]

Quote:

Quote:

Originally Posted by ex0du5 /img/forum/go_quote.gif
 
Digital data sent over the USB cables is error corrected and buffered. As long as there is a moderate buffer, any errors that occur during the initial transmission will be corrected before being output from the buffer to the source.

Did you read the links I provided ? USB audio is working with extremely small buffers, too small for error correction. They're made just big enough to account for the variations in the clocking of the arriving packets.
 
 
edit: and I don't "believe" in USB cables making any differences. The interface is inherently poor, can't see a cable degrading the situation much.

Why do they work with such small buffers? There's no reason, other than to save costs...heck, just stick some internal memory in the device and buffer it there before having it go through the DAC. Audio is not something that requires immediate response, so the buffer can be as big as need be. And a buffer to receive 100% correct information on a measily 1441kbps stream would not have to be very big at all...
 
My Maverick Audio D1 DAC is unusable on the USB output due to an apparent lack of buffering, as it will actually skip/crackle from time to time. That is what you might expect from a bad or unbuffered digital stream, gaps in the sound. You won't get slightly degraded sound quality. Much like a bad HDMI signal will result in blank screens or dropouts, not a slightly worse picture.
 
I don't understand why more DACs don't seem to implement proper care in handling the digital streams. Don't want to rely on the source clocks? Transfer it and store it internally before having it interact with the DAC component, keeping all your timings internal. 
 
edit: after reading through that article, here's what stood out to me:
 
 
Quote:

  There are four USB transmission modes (please see Table 1). The two of those that are used for sending large quantities of data are (1) Isochronous Mode - A fixed number of packets is guaranteed to be sent and received. This mode is used with multimedia data such as images and audio. (2) Bulk Mode - A fixed quantity of data is sent at one time. If for some reason some of the data, is lost it is resent.
 
For data storage or printer applications the bulk mode is best because speed is of utmost importance and, through retransmission, data errors will be eliminated. But for audio data, real-time transfer is even more important than occasional missing data. (Noise is more tolerable than interruptions in the data. Of course pops and clicks are intolerable, but even more unpleasant is an intermittence of the data.) In this case, the isochronous mode is used. In other words, a real-time transmission scheme, with no re-sending of packets, is used for audio data, which streams from the PC in an RS-232-C-like manner.

 
 
In a very cheap DAC, I might expect the design to use isochronous data transmission. My proposal: use bulk transmission, and store the 100% error free signal in internal memory. From there, you have a much quicker and controlled transmission environment.
 
Given that some DACs may use isochronous transmission, I'll admit that it's a possibility that USB cables 'could', theoretically, make a difference. However, it begs the question whether errors occuring during transmission are the cause of the transmission medium (the cable), or the transmission itself (outputting USB port).
 
I really don't think asynchronous transmission should be used for anything but the most time sensitive tasks. Audio is not one of these tasks.

 

[00940]

Quote:

Why do they work with such small buffers? There's no reason, other than to save costs..
 
My Maverick Audio D1 DAC is unusable on the USB output due to an apparent lack of buffering, as it will actually skip/crackle from time to time. That is what you might expect from a bad or unbuffered digital stream, gaps in the sound. You won't get slightly degraded sound quality. Much like a bad HDMI signal will result in blank screens or dropouts, not a slightly worse picture.
 

Die area for a FIFO is scarce when your usb receiver must fit into a small smd package. But yes, cost is a big deal.
 
The unbuffered stream has another consequence: in isochronous USB audio, the output clock is derivated from the frequency at which the packets arrive (by a PLL). Some usb receivers are absolutly awul at that : see for example: http://www.diyhifi.org/forums/viewtopic.php?f=2&t=1360
 
 
Edit: the PCM2707 are isochronous receivers. However their jitter is low  enough(measured at 2 or 3ns from memory) and TI spact system is good enough not to get interruptions. So I just used that followed by a SRC4192, which clears up things nicely (see : http://www.diyaudio.com/forums/digital-source/46413-any-feedback-new-cs8421-high-res-asrc.html#post528312  , according to the same Bruno Putzeys the SRC4192 will stay in "narrow mode" for such low amounts of jitter).

 

[JerryLove]

I've never seen signal interference high enough on USB cables under normal conditions for the cable to result in data-loss.
 
The discussions of USB controller chips and timing signals is relevant to good sound: but should be solved by any compitent designer and problems should be ID'd in QA.