$695 USB cable wins TAS's editors' choice award

Re:$695 USB cable wins TAS's editors' choice award by rayiner on Wednesday January 15, 2014 @02:46PM (#7065661)

I'm not really talking about the cable, but the idea that digital interfaces like SPDIF/USB Audio/HDMI are "bit perfect." The digital phenomenon of bit perfection, as programmers see it, is a fiction created by buffering, retry, and error-correction. SPDIF doesn't do any of these, and USB Audio doesn't do all of these in all modes. SPDIF receivers (and depending on the mode, USB Audio receivers) will recover the clock signal from the input signal, and use that clock signal to drive a DAC. So it's not just the sequence of ones and zeros that is relevant (a digital phenomenon), but the timing of their appearance on the cable (an analog phenomenon). Variation from the ideal timing will cause a distorted clock to be recovered by the receiver, which will cause distortion in the analog signal produced by the DAC. Whether or not these distortions are audible, they are certainly measurable.

[ Parent | on HN ]

Re:$695 USB cable wins TAS's editors' choice award by rdale on Wednesday January 15, 2014 @04:20PM (#7066243)

In over 100 audiophile bashing posts, at last someone who actually understands the issues. USB cables have an 'eye pattern' which affects the timing of the signal presented to the DAC.

If USB cables make a difference, then there is probably an argument for doing something else (not ignoring the problem because USB cables must be 'perfect'). Like putting the DAC and computer driving the DAC in the same box and connected by I2S with the DAC as the master.

There are some good articles on the Audiostream site about why 'bit perfect' doesn't cut it in the context of audio:

http://www.audiostream.com/category/industry-voice

[ Parent | on HN ]

Re:$695 USB cable wins TAS's editors' choice award by Dylan16807 on Wednesday January 15, 2014 @05:41PM (#7066744)

What kind of insane hardware depends on CPU timings and doesn't have a hardware buffer that always plays at the same speed? Jitter in the intermediate steps doesn't matter when you're moving digital data from one place to another, as long as the buffers are larger than the amount of jitter.

DAC is hard. Moving a couple megabits of data five feet over a cable with multi-millisecond buffers is not hard.

I don't think it's even possible to use USB for 100%-throughput unbuffered data. I call bullshit on the eye pattern affecting playback.

Just because a difference technically exists in some circumstances with an oscilloscope does not mean it makes a difference in normal use cases.

Also wow that site called wireless a 'potential long-term health risk'.

[ Parent | on HN ]

Re:$695 USB cable wins TAS's editors' choice award by rayiner on Wednesday January 15, 2014 @09:24PM (#7067653)

What you're describing isn't as easy as it sounds. In order to have a "hardware buffer that always plays at the same speed" you need a high-quality locally generated clock, plus a decent amount of local buffering. But in a USB audio device or SPDIF device, you can't run a totally independent clock, because it will get out of sync with the clock on the host device sending the data. There are various tricks to get around this (e.g. having a feedback loop using a sideband), but it's actually a non-trivial problem. See: http://www.cypress.com/?docID=45044 for one approach. Most USB audio devices don't even attempt to do this. They try and recover the host machine's playback clock using the timing of USB SOF packets (which are sent every millisecond-ish) plus the timing of the USB audio data packets it receives.

[ Parent | on HN ]

Re:$695 USB cable wins TAS's editors' choice award by Dylan16807 on Wednesday January 15, 2014 @11:44PM (#7068119)

Yes, sure, those things are hard, and they have nothing whatsoever to do with the quality of passive components like cables, as long as they are not in the failure range. Or anything to do with working-transceiver quality except when it comes to the clocks that the transceivers are using.

When you talk about jitter introducing significant distortion based on packet timing, that's something that can only happen in two ways: either the audio receiver is using an absolute garbage clock, or it's using an absolute garbage and shortsighted algorithm for adjusting its clock rate, and not trying for a stable sync. The quality of transceiver and cable components is not a factor.

[ Parent | on HN ]

Re:$695 USB cable wins TAS's editors' choice award by Dylan16807 on Wednesday January 15, 2014 @03:27PM (#7065903)

Unless your cable is broken, there will be no retrying or high level error correction going on. And buffering will be independent of the cable chosen.

A passive cable can't change timing. The only question is whether the signal gets through in an intact form and can be locked onto. Any kind of difference you can have from working-cable construction is completely overshadowed by things like cable length.

[ Parent | on HN ]

Re:$695 USB cable wins TAS's editors' choice award by rayiner on Wednesday January 15, 2014 @04:14PM (#7066213)

I'm not saying that the cable can change timing. My post really wasn't about the cable. It was about the more general misconception that people have that digital audio/video interconnects (the whole system, cable + transceivers) are somehow "bit perfect."

[ Parent | on HN ]