To Reclock or Not to Reclock ...

[audioengr]

"To bad about the m2tech hi-face though...no linux drivers!"

 

Marco says he plans to work on it. Market is fairly small though.

 

All my USB interfaces are reclocked.

 

Steve N.

 

[Wavelength]

Jest,

 

Actually I have to agree with you up until here. I think much of the hoopla up to this point was because people did not understand what you were saying.

 

~~~~~

 

Gang... interesting point... I was asked by some of the companies I develop for why Asynchronous USB was better than Adapative (with maybe the use of an upsampler to cure the high jitter at the dac chip). Having a sample of a product on hand we tested the Adaptive jitter using the WaveCrest DTS system. The thing these companies that depend on me did not realize was that there was so much jitter on the I2S feed into the upsampler that THERE was the problem. Not the jitter to the dac, but actually jitter induced errors were forming before the upsampler.

 

This is what Jest is talking about. Jitter can happen anywhere on the I2S feed and this is the only jitter we should be talking about.

 

For example years ago I was working on a dual input I2S and picked up a standard logic gate for multiplexing 2 sets of 4 inputs (74xx157). This is a simple part used a ton in the old days for DRAM and other miscellaneous digital circuits. I noticed that this part actually contributed to the jitter getting to the dac chip.

 

Actually any IC in the chain can contribute to the jitter seen by the dac chip. This is why careful circuit design is really critical... also some way to test it.

 

~~~~~~

 

Jest,

 

In general this is the way we are working the Asynchronous USB electrically:

 

Two low jitter fixed oscillators are put next to the DAC chip. The selected output oscillator is then buffered and sent to the USB chip to create the I2S. The I2S feed is then again reclocked before it hits the DAC chip.

 

Software wise there are many differences in the way the buffering is handled. In Class 1/Class 2 USB Audio the provisions for flow control are pretty basic. If you adhere (I do not) to this it will work for you well, but not great.

 

In the use of bulk streaming (requires device driver to make it look like and audio device) the interface (DAC receiving unit) basically buffers coming in until the interface has room for the data and then accepts it. The computer will retransmit these buffers until the interface the buffer which means it is received.

 

In either one of these cases if the computer neglects to send data in time and the interface is starved of data (or is overrun in the Class 1/2) the dac will hiccup... but then again this has nothing to do with jitter and it should be understood that while jitter is one variable to look at, there are tons more than people should consider.

 

~~~~~~~~~

 

Steve,

 

Believe me... I was doing computer audio in 1994. I actually got asked then to design and make an Audio Computer. This was the 4th mother board I designed for companies wanting my design services. I have shipped 5.4M computers with my mother boards. Since then I have designed one more, but no more of that...way too much time for little return. I designed my first USB Audio controller 8 years ago. I think I have a little edge right now. I control all my source code for everything and have tons of test gear to verify my findings. Can you say the same?

 

Thanks

Gordon

 

 

[vortecjr]

thanks for talking to him, but IMO Marco is wrong on the market share aspect of it! He is not taking into account the 7500 registered users of vortexbox alone. It would be nice to have options, but I'm not worried as other devices based on Gordon's code and the Centrance code are supported without drivers.

 

Jesus R

www.sonore.us

 

[audioengr]

"I think I have a little edge right now. I control all my source code for everything and have tons of test gear to verify my findings. Can you say the same?"

 

I rely on third parties for this, as this is not my expertise. Likewise, I defer the clock and regulator designs to others whose expertise is in these areas. My expertise is in high-performance innovative digital. This is where I have the edge.

 

I have known about the contributors to jitter from the beginning. I wrote three white-papers for high-end audio that address the subject, one on jitter, one on S/PDIF interface and one on computer audio, all published by Positive-Feedback. I have been a proponent of using fast slew-rate gates/logic, proper terminations and proper power delivery from the beginning. I was told by other designers that slower logic families sounded better etc., which I knew was hogwash.

 

Steve N.

 

 

[SteveChicago]

Ask a seemingly simple question, and a hockey game breaks out!

 

Guess I have the knack for stirring up trouble without even trying.

 

Cheers,

 

Steve

 

[hankmallard]

As a Canadian, I just luv hockey fights, but this was more like a bench clearing brawl. Some people were already fined (posts deleted). Who do you think won?

 

I noticed one thing though as I read through this very educational posting,and that is that many of the combatants are listed/reviewed in the C.A.S.H list. Empirical is not.

 

I have read many favorable reviews on Empirical and am on this forum to decide what to buy for a computer based system (people tell me this is THE site for that).

 

Empirical and PSaudio are two I am looking at. Given they actually are noted for making good stuff, then how come none of it is listed in the C.A.S.H. List? Am I missing something here?

 

Hank

newbie

 

[The Computer Audiophile]

Hi Hank - The new PS Audio components are finally ready for review. PS and I agreed to wait until the Bridge was released as a production part of the components. A review will be forthcoming. I can't say whether PWD will make the C.A.S.H. list until I've had plenty of time reviewing it.

 

[hankmallard]

Great.

I listened to the PW/PWD combo last week up here in canada. To tell you the truth, the dealer did not really understand this system, but I had him hook the PW to the PWD with the $800 I2s PSAudio cable he had on the wall and loaded some of my 96/24 files (he had none) and listened for about 30 minutes. We had problems with the remote, and the system would sometimes "pop" when we changed tracks and there were two times we got harsh static noise switching tracks?????

 

Other than that, it sounded very good playing 96/24 downloaded files and also good with straight CD's.

 

What about Empirical, you did not say anything about them? Have you ever tried their stuff?

 

[The Computer Audiophile]

I've heard Empirical stuff at shows.

 

[barrows]

Hank, I think it is important to understand that the CASH list is a list of recommended components. Chris can only recommend components which he has had extensive experience with, and he cannot possibly be expected to have had this type of experience with all the products available. What you should take from this is that if a component is on the CASH list, it probably has something good on offer, but that exclusion from the CASH list means nothing, good or bad, regarding a given component.

Certainly, if I was in the market for a (computer capable) DAC, the PS Audio stuff would be on my list to try, as would the Ayre QB-9, the new Aesthetix Pandora, the Audio Research DAC 8, and at higher prices, the dCS Debussy (Chris are you reviewing this?) the Weiss DAC202, and the Resolution Audio Cantata. There are a lot of good products out there that deserve a listen these days! It is a good time to get into computer playback.

 

[The Computer Audiophile]

Hi Barrows - No plans to review the Debussy. Just adding it as one of my reference DACs in my system :~)

 

[hankmallard]

and.........

 

[ziggyzack]

'I have read many favorable reviews on Empirical and am on this forum to decide what to buy for a computer based system (people tell me this is THE site for that).

 

Empirical and PSaudio are two I am looking at. Given they actually are noted for making good stuff, then how come none of it is listed in the C.A.S.H. List? Am I missing something here?'

 

Yes. Under-represented on this site are the class of dacs which connect via an ethernet network. PS Audio's bridge does this, Linn and Naim both have ranges of players, and there are other makes too. There are some great advantages to this approach.

 

ZZ

 

[tgb]

up for this subject.

 

i read reviews about reclockers and large benefits on SQ.

It would be to add between my drive and the DAC.

A good one at competitive price is now the Mutec mc3+, <1kUSD.

Did you experience such products ? with good results ? do you know other products of that kind ?

thanks

rgds

[iago]

A good one at competitive price is now the Mutec mc3+, <1kUSD

 

There is the Pace Car from Empirical Audio, now replaced with Synchro-Mesh. The price is similiar to the Mutec's, but probably easier to source in the USA.

 

I believe that external upsamplers like dCS Purcell are also working as re-clockers.

 

The Mutec MC-3+ is discussed in this thread: Mutec MC-3+

 

I'm using one with external clock source.

[maelob]

Wyred4sound just released one for 399