USB Clock Upgrade My Experiences (so far)

[xp9433]

148 to Michael Clark! Come on Australia. Are you watching the NZ  v West Indies cricket match at the same time - like me?
Frank

[acg]

No Frank but those Kiwis are certainly showing the Windies how it is done.

Nice morning for Aussie cricket though...

[PeterSt]

As I said or implied or whatever, the two clocks running on an ALWAYS different speed is the importance here. Jitter gives the noise coming from that another pattern, or IOW oscillation (resonance) varies (at high speed).

When using one oscillator this just can not happen. And the thing can be as jittery as it likes. Well, this is what I think so far; can be proven to be diferent later.

Peter

[PeterSt]

Nick,

Yea, nice tuning facility you made yourself there. Of course I did not know this (trust me) but as often things seem to come together by again very different routes; you just prooved in advance that this is going to work (one oscillator). Still Coen has the credits for my part.
Well, assumed that is able to give the best sound which (of course and sadly) depends on the "pattern" created by "nothing" (no oscillation).

My part of the game here is the means HOW to realize this one oscillator. THAT will be the new stuff ...
So Anthony, no sarcasm at all here.

Regards,
Peter


PS: I think I am going to move this topic to the NOS1 board. That should make it serious, right ?

[acg]

So Anthony, no sarcasm at all here.

Regards,
Peter


PS: I think I am going to move this topic to the NOS1 board. That should make it serious, right ?

That does sound serious Peter.

[Scroobius]

Ha ha isn't that typical of my luck (blown Dexa power supply) with only one clock working I cannot even play with my "heavenly trimmer". I wonder what my wife will make of that.

This is a really interesting stuff - we could soon know whether noise or data transfer errors are the culprit. We all know what Peter believes and my initial thoughts were that it could not possibly be data transfer errors could it?

I just cannot get used to this step backward in sound quality - I only heard two Dexa's for a few days - where is the bass, where is the crystal easy clarity?

Can't wait to get that Dexa PS due Monday ggrrhhh!!!

Paul

[Nick]

Ha ha isn't that typical of my luck (blown Dexa power supply) with only one clock working I cannot even play with my "heavenly trimmer". I wonder what my wife will make of that.

This is a really interesting stuff - we could soon know whether noise or data transfer errors are the culprit. We all know what Peter believes and my initial thoughts were that it could not possibly be data transfer errors could it?

I just cannot get used to this step backward in sound quality - I only heard two Dexa's for a few days - where is the bass, where is the crystal easy clarity?

Can't wait to get that Dexa PS due Monday ggrrhhh!!!

Paul

Its a cruel drug not to have  once tried

The heavenly trimmer  could still work with the paul pang card. Just remember where it was set in case it needs to go back. It might be an interesting result.

Best Nick

[Scroobius]

So Peter - Just a thought  -  with a Dexa at the NOS1 and a standard pcie USB3 at the other end - if I tweak the Dexa "heavenly trimmer" and I am lucky enough to be able to match the Dexa clock frequency to the pcie clock frequency then in theory it should sound as good as two Dexa clocks if the noise theory is correct?

Paul

[CoenP]

So Peter - Just a thought  -  with a Dexa at the NOS1 and a standard pcie USB3 at the other end - if I tweak the Dexa "heavenly trimmer" and I am lucky enough to be able to match the Dexa clock frequency to the pcie clock frequency then in theory it should sound as good as two Dexa clocks if the noise theory is correct?

Paul

Paul,

The jitter won't be the same and they have both different tempraturesensitivity (and different temperatures during the year).

But it is closer to ideal than having an unoptimised clock sync.

Regards, Coen

[Nick]

So Peter - Just a thought  -  with a Dexa at the NOS1 and a standard pcie USB3 at the other end - if I tweak the Dexa "heavenly trimmer" and I am lucky enough to be able to match the Dexa clock frequency to the pcie clock frequency then in theory it should sound as good as two Dexa clocks if the noise theory is correct?

Paul

Paul hi,

Just to clarify, that is what I'm suggesting above, use the dexa trim pot. This is the set up I had years ago with the m-audio card and audiocom clock. You will just be setting the best sounding frequency on the dexa relative to the Paul pang card.  I could make a big difference. Listen for the sound stage expanding whilst keeping good dynamics and no sharpness.

There are a number of ways of thinking about the effect of relative clock speed. I won't go into the details but equal speed clocks (or slaved) may not be the best setting. There could very well be an optimum difference in speed that suits the data transmision rate and the buffer size at the reciving end.

I think its imporant to understand the relationship so i have equipment on order to test explicitly these theories. Will post back with results, then we will know of the trouble of transmitting a clock 1.5m whilst maintaining clock performance is worth the effort.

Cheers,

Nick.

[Scroobius]

Hi Coen,

Appreciate about the jitter not sure about temperature as that is a longer term drift effect so for a quick check should not be a problem. Anyway I shall have a play later on.

Nick apologies missed your post - look forward to seeing the results of your tests.

P

[Nick]

Paul hi,

Do give it a go, it's the trim pot next to the xtal can on the dexa. It's a variable cap with one full turn of adjustment. If the speeds are out it could REALLY make you smile (the 5.1 type sound stage is back again here  )and it takes moments to adjust.

Cheers,

Nick.

[AlainGr]

Hi Nick,

Trying to read about the "master-slave" relation between two USB ports... Is it something that is established normally between the sender and the receiver in the PC USB port and the NOS1 USB port or is it something that has to be "programmed" first ? If so, I would have thought that the sender would be the master ? By that I mean the "A" type USB as the master and the "B" type USB as the receiver ?

And would this achieve what you, Paul, Cohen and Peter are talking about ?

Regards,

Alain

[Scroobius]

Hi Alain - NOS1 is "asynchronous" but the way the data is transferred is maybe not strictly that. I think (and I am sure Peter will correct me if I am wrong) that when NOS1 is ready (i.e. its receiving buffer is running low) it requests a block of data to fill it ("bulk data transfer"). So Nos1 is the Master and the PC is the slave. That is why on the face of it, it is hard to understand how jitter in the USB link can be a problem because (I think) the data is transferred with timing information and so should be immune to the effects of jitter. I am sure that is why Peter thinks this is all about noise.

Another possible way of transmitting over USB is Isochronous which means that the data is transmitted equally spaced - but I am pretty sure that is not the way NOS1 works and I am also sure Peter will very soon correct me if I am wrong.

Cheers Paul

[PeterSt]

Maybe a few scattered responses, with for Nick the message that I did not read your lengthy email yet. But it makes me more objective.

So Peter - Just a thought  -  with a Dexa at the NOS1 and a standard pcie USB3 at the other end - if I tweak the Dexa "heavenly trimmer" and I am lucky enough to be able to match the Dexa clock frequency to the pcie clock frequency then in theory it should sound as good as two Dexa clocks if the noise theory is correct?

Paul

Paul,

The jitter won't be the same and they have both different tempraturesensitivity (and different temperatures during the year).

But it is closer to ideal than having an unoptimised clock sync.

Regards, Coen

Maybe not all the way through on the subject, but the Dexa is a TCXO or IOW temperature controlled. Of course the PCIe card is not, so what you say is correct in the end. With a Dexa at each end this is different though;

I think Nick already said it, but we won't be able to exactly "match" anyway. With real slaving (VCXO - Voltage Controlled) this is different, but still at a low frequency (like a few times per second) the oscillator at the other end will be corrected (adjusted in speed to match the master) and for NOISE your miles will vary as much - but probably things will even get worse. This is about the low oscillation (low change of frequency) which now also will be completely repeatebale (each second the same pattern repeats). Thus, in my view haveing them both much off related to eachother would be the better thing.

Of course this is all theories and how the real work out will be is to be seen (heard).

This is how I think that NO oscillation will at least imply a steady situation which we can depend on, and then still it is to be seen whether it is for the better. This too might not be for the better at all, because now we don't have nothing to trim ...
Also (and once again) the noise patteren there will conists of less freqencies (nothing runs into eachother) so it could be worse to begin with.

But I am going to try and I know it can be setup.

Must go now, but will be back.
Peter