USB audio cracked... finally!

[fas42]

Don't want to distract from the Lush story overly - but this piece, http://www.curiouscables.com/our-story.html, is another version of how people find out that "everything matters" ...

[pkane2001]

30 minutes ago, fas42 said:

Don't want to distract from the Lush story overly - but this piece, http://www.curiouscables.com/our-story.html, is another version of how people find out that "everything matters" ...

Quote

Best results are attained when the cable is used: 

• Between an external hard drive and computer/music server, …

Magic...it must work on magic. 

[jabbr]

2 hours ago, pkane2001 said:

Timing errors from more or less steep 0 to 1 transition are irrelevant

 

I take it that you've read the article that I referenced above in this thread, and disagree? Hmm... the article goes into substantial detail including models. What basis do you have to go against this? Is their math in error? See figure 20. Unreasonable to repeat this entire long thread.

[Speed Racer]

10 minutes ago, jabbr said:

 

I take it that you've read the article that I referenced above in this thread, and disagree? Hmm... the article goes into substantial detail including models. What basis do you have to go against this? Is their math in error? See figure 20. Unreasonable to repeat this entire long thread.

 

That guy is just trying to set cables. I wouldn't believe anything on that web page....

 

Oops...I though you mean't the article 3 posts above this one....

[pkane2001]

6 minutes ago, jabbr said:

 

I take it that you've read the article that I referenced above in this thread, and disagree? Hmm... the article goes into substantial detail including models. What basis do you have to go against this? Is their math in error? See figure 20. Unreasonable to repeat this entire long thread.

 

I believe I did, but this has been a long thread with a number of links posted. Can you please repost the link to the article you are referring to?

 

[rando]

14 minutes ago, pkane2001 said:

Magic...it must work on magic. 

 

The Dutch are not widely renowned for sorcery unless you are referring to stroopwafel vendors ability to slip kids cuttings under their parents nose.  The more openly this gets chewed on, and the less he feels appreciated, the better chances are PeterSt will quit handing out goodies to digest on the sly.  

[jabbr]

 

 

Well accepted in high speed digital design is the idea to match the correct rise time to the receiving circuit to avoid ringing. The "Micron" article in particular (and see that this issue is most certainly not unique to audio!)

 

On 7/17/2017 at 11:09 AM, jabbr said:

Yes. as you say, detailed in Howard Johnson's "Black Magic" and "Advanced Black Magic" books on high speed digital design  There are two issues:

1) differential pair impedance

2) termination  impedance -- (can be to ground)

 

https:\\www.micron.com\~\media\documents\products\technical-note\dram\tn4606_point_to_point-termination.pdf

Note Figure 20, as the rise time of the transition slows the SI improves ... this is an example of removing "hard edges".

 

https://fenix.tecnico.ulisboa.pt/downloadFile/3779579943588/[Bus Terminations] Ethirajan and Nemec - Termination Techniques for High Speed Buses.pdf -- see Figure 4 for use of capacitance along with resistance.

 

http://www.ti.com/lit/an/snla043/snla043.pdf gives an extensive discussion of the effects of and use of termination capacitance.

...

[Doak]

Suggestion: 

If you are actually interested in the item/topic under discussion here, please set about acquiring a LUSH USB cable, insert it into your music playback system for a while and then decide what you think about it.  

In the end, for YOU and your system, there is only one opinion that matters: YOURS.

 

Corollary: In MY system, MINE is the only one that matters. 

 

[pkane2001]

1 minute ago, jabbr said:

 

Ah, yes, I did read it, thank you for the reminder. 

 

The point is not that jitter cannot occur in USB transmission. The point is that jitter on USB side has zero effect on jitter on the I2S side using isochronous protocol. How ever the bits arrive on the receiver side, they are ultimately stored in a memory buffer and doled out from that buffer using an onboard clock that is not subject to jitter errors on the receiver side of the buffer. As long as timing errors are not so large as to cause an entire microframe to be dropped (resulting in a huge number of bit errors), jitter on the receiver side does not translate into jitter on the I2S side. 

[PeterSt]

2 hours ago, rickca said:

Peter, you said you can do the same thing as the Lush cable in XXHighEnd.  What software control produces the same effect (or does this give away too much)?

 

Hi Rick - I don't think I said that I can do the same thing by means of software; I said (or tried to say) that the influence the D/A process undergoes can also be done by software. Remember, it can only be the D/A process which is influenced one way or the other, unless we assume flipping bits which we thus NOT assume.

Doing it by software is a very very indirect way with a lot of guessing which causes what; noise as such seems the logical subject to think about, but it is always (spiky) current draw which causes noise (this was dealt with more in the beginning of the thread but now for the receiver's end).

Working on the interface itself, which is part of the transfer of the "noise" (but we really should not think about noise as such) would be the best, but useless because it is not what is common to each system for physical aspects. The cable we use is though ...

And to keep in mind (also said in this thread somewhere) : the ideas for this sprung from me working on the finalization of the outboard Phisolator in "re-gen" fahsion, which Uptone already came up with first, as a sheer alternative (I announced this literally on the Phasure forum). And a month or so later it worked.

 

We can say that the little current spikes from a CPU always performing different tasks won't be able to influence much, but then I'd have to remind people about the ~120W a PC comsumes in normal circumstances vs the 30W the very same (20 core hyperthreaded) PC consumes with the for audio "minimized" Operating System. 90 Watts ??? (of difference).

It must be one big mess of bouncing ground planes in there. Call it noise, but ... we were talking about poor little tiny transmitter chips with fragile wave form.

 

Not sure yet whether I am going to sell software or Phasure Stealth PCs today. But probably nothing because I am busy. 

[jabbr]

2 minutes ago, PeterSt said:

We can say that the little current spikes from a CPU always performing different tasks won't be able to influence much, but then I'd have to remind people about the ~120W a PC comsumes in normal circumstances vs the 30W the very same (20 core hyperthreaded) PC consumes with the for audio "minimized" Operating System. 90 Watts ??? (of difference).

 

XXHE works quite well on my 10W J1900 board thank you. Should I underclock this? -- its got many cycles to spare

 

[jabbr]

41 minutes ago, pkane2001 said:

The point is that jitter on USB side has zero effect on jitter on the I2S side using isochronous protocol. How ever the bits arrive on the receiver side, they are ultimately stored in a memory buffer and doled out from that buffer using an onboard clock that is not subject to jitter errors on the receiver side of the buffer. As long as timing errors are not so large as to cause an entire microframe to be dropped (resulting in a huge number of bit errors), jitter on the receiver side does not translate into jitter on the I2S side. 

 

This statement is is an assumption, aspirational and I agree that things should ideally work this way yet how do you know that the assumptions are correct? Do you have detailed knowledge of the state of the ground and power planes? Circuit? Do the onboard clocks share common power/ground? Does this have an effect? Are you sure? Have you made a detailed model of the entire circuit? Is the model accurate? I wouldn't be so certain...

[Speed Racer]

9 minutes ago, jabbr said:

 

This statement is is an assumption, aspirational and I agree that things should ideally work this way yet how do you know that the assumptions are correct? Do you have detailed knowledge of the state of the ground and power planes? Circuit? Do the onboard clocks share common power/ground? Does this have an effect? Are you sure? Have you made a detailed model of the entire circuit? Is the model accurate? I wouldn't be so certain...

 

All of that is why I said that you would need to "tune" the "Lush" cable for just about every DAC.

[pkane2001]

27 minutes ago, jabbr said:

 

This statement is is an assumption, aspirational and I agree that things should ideally work this way yet how do you know that the assumptions are correct? Do you have detailed knowledge of the state of the ground and power planes? Circuit? Do the onboard clocks share common power/ground? Does this have an effect? Are you sure? Have you made a detailed model of the entire circuit? Is the model accurate? I wouldn't be so certain...

 

Are you sure that tiny variations in timing will result in a detectable difference in analog domain? Remember, all digital circuits produce noise. All shift registers, all memory load/refresh/read operations, all FPGA logic, etc., all generate a constant level of noise at the ground plane. The sequence of these noise generating events is the same (or very, very nearly the same) with or without jitter in the same circuit. The only thing that changes is their timing if jitter is present.

 

Are you really ready to say that time differences at the level of a few hundred picoseconds in processing incoming bits will alter all that computation noise sufficiently to suddenly make an audible difference? I'd love to see a model that shows that. You'll also need to show this for more than one implementation, as @Speed Racer is suggesting, since different designs will generate wildly varying computation noise and have very different reaction to incoming jitter.

[manisandher]

OT

 

7 hours ago, fas42 said:

I did my own take on using LM3875 amps years ago, and it was highly successful. No excuses needed to be made for the sound - the only issues were needing more gain for some recordings, and that the chip would go into self-protection behaviour on sustained, big single note phrases.

 

I'm very far from being an expert... on anything really... but especially on this stuff. My understanding is that the LM3875 is very unhappy with speaker impedances below 8 Ohms. Perhaps this could have been your issue? Apparently, running four LM3875s per channel in a bridged parallel config overcomes this. (These chips are cheap, but getting hold of 16 BG N caps (two per chip) won't be easy or cheap now. I'm sure there are loads of alternatives though.)

 

In any event, I've tried both my 'cheap' stereo and 'expensive' mono gainclones on quite a few different speakers now with no issues. They are without doubt the 'cleanest'-sounding amps I've ever heard/owned.

 

Mani.

[fas42]

45 minutes ago, manisandher said:

I'm very far from being an expert... on anything really... but especially on this stuff. My understanding is that the LM3875 is very unhappy with speaker impedances below 8 Ohms. Perhaps this could have been your issue? Apparently, running four LM3875s per channel in a bridged parallel config overcomes this. (These chips are cheap, but getting hold of 16 BG N caps (two per chip) won't be easy or cheap now. I'm sure there are loads of alternatives though.)

 

In any event, I've tried both my 'cheap' stereo and 'expensive' mono gainclones on quite a few different speakers now with no issues. They are without doubt the 'cleanest'-sounding amps I've ever heard/owned.

 

Mani.

 

Errr ... because I ran them very loud !! They were using quite substantial heatsinks, but there's a point where there is insufficient contact area between the chip and the heat dissipation metalwork - and the over-temperature circuitry cuts in. Yes, a bridged-parallel arrangement would have given me far more headroom, but all the auxiliary bits would have blown the cost way higher - another time perhaps ...

[marce]

8 hours ago, fas42 said:

 

How do you know this?

What!

[marce]

8 hours ago, fas42 said:

 

That makes absolutely no sense at all, as a question.

Actually it does.

[marce]

6 hours ago, fas42 said:

If one really wanted to understand the mechanisms at play, then the 'correct' approach would be to 'artificially' generate the USB line signals, with complete control over every aspect of how the waveforms behave, in electrical terms - and see what happens at the analogue out. The best DAC and analogue reproduction chain is that which does not vary in behaviour one iota, while the USB lines are manipulated in such a fashion.

 

Alternatively, try combinations of cables and DACs, etc, and pick the the set which deliver the best sound ...

This is what one does signal integrity verify software, do you think digital data transmission is not UNDERSTOOD... 

[marce]

4 hours ago, Doak said:

Suggestion: 

If you are actually interested in the item/topic under discussion here, please set about acquiring a LUSH USB cable, insert it into your music playback system for a while and then decide what you think about it.  

In the end, for YOU and your system, there is only one opinion that matters: YOURS.

 

Corollary: In MY system, MINE is the only one that matters. 

 

We are discussing the concepts in finer detail than just listening... And whilst our own opinions matter for our systems the wider implications of what is happening, like all scientific and engineering endever cannot be left to personal preference alone, discussion and data (empirical) is what is needed. 

[semente]

5 hours ago, Doak said:

Suggestion: 

If you are actually interested in the item/topic under discussion here, please set about acquiring a LUSH USB cable, insert it into your music playback system for a while and then decide what you think about it.  

In the end, for YOU and your system, there is only one opinion that matters: YOURS.

 

Corollary: In MY system, MINE is the only one that matters. 

 

 

I don't use a USB connection but am still interested in the topic.

What should I do?

[PeterSt]

Just now, semente said:

I don't use a USB connection but am still interested in the topic.

What should I do

 

Get yourself the cheapest DAC with USB interface available and proceed from there ?

 

(not sure how serious you are with your question )

[semente]

2 minutes ago, PeterSt said:

 

Get yourself the cheapest DAC with USB interface available and proceed from there ?

 

(not sure how serious you are with your question )

 

 

I have been waiting for my bespoke PCM NOS DAC for ages but it seems that the builder has other things in hand...

[PeterSt]

Just now, semente said:

 

 

I have been waiting for my bespoke PCM NOS DAC for ages but it seems that the builder has other things in hand...

 

Ehm ... Which one would that be ? (somehow I am getting anxious)

[Doak]

56 minutes ago, semente said:

 

I don't use a USB connection but am still interested in the topic.

What should I do?

Purely theoretical discussion on an applied subject is such a drag.. .