advertisement


USB audio coming of age?

Status
Not open for further replies.
OK then if we can only deal in pure facts (that is rather boring I have to say, must be fun in the pub...), how many DACs do we know of that use a separate board for the USB? It should be quite a number. I'm not saying this method is purposely used for SQ reasons, mostly it's probably because DACs often use an M2Tech, Adnaco etc board. Can we correlate reportedly good sounding DACs with USB daughter boards? I would post a link to the original posting but as it's on another forum its probably not correct to do so.

Here are some I know....there must be loads more
Metrum Octave and Hex
Caiman II
Ciunas

This discussion might seem pointless for some but it's got to be more useful than much of what's been said today.
Some facts would help.

You can buy small boards from China with a neat USB implementation to put into your DAC. You would do this because it solves a design problem conveniently rather than for performance per se.

So we need to know if what Beresford has done is simply hacksaw around the USB implementation on his board and use links rather than traces, or whether he has introduced a completely new and possibly rather better USB implementation. Or you could just speculate based on nothing.

Paul
 
Daniel,

Thanks for the honest response. It is great to have DAC designers such as yourself contributing to PFM.

Your DACS have a strong reputation for being immune to downstream computers and cables, yet user experience reported with other asynchronous USB DACS sometimes differs. Theres some examples in this thread and I have experienced first hand with the Audio research DAC8 - the sound improved if I used an external USB-SPDIF converter, which should not happen if the Asynch USB was working as it should. Another disappointment that contradicts the notion that Asynch DAC works perfectly was the Naim DAC when loading a file directly from a USB stick - it sounded crap!

A key question is whether you think the effectiveness of Asynch USB is variable across different manufacturers DAC's based on design decisions, or do you think any form of Asynch USB design is sufficient to make the DAC immune from upstream jitter?

I found the following quote from Mark Jenkins of Antipodes thought provoking. I'd be interested to hear your commentary on this:

(snip)

quoted from http://www.audiophilia.com/wp/?p=12895

A locally generated clock (like the one in a DAC with the input buffer scheme as described) of course can be contaminated with (asynchronous) signals. Such a sampling clock has to be considered to be an analog signal and thus is prone to all kinds of "contamination mechanisms". But it is simpler to keep a locally generated clock clean than to regenerate a low jitter clock from a jittery input clock.

Same with DACs with an async USB interface. The async scheme is a good idea, but it has to be properly implemented.

Daniel
 
A locally generated clock (like the one in a DAC with the input buffer scheme as described) of course can be contaminated with (asynchronous) signals. Such a sampling clock has to be considered to be an analog signal and thus is prone to all kinds of "contamination mechanisms". But it is simpler to keep a locally generated clock clean than to regenerate a low jitter clock from a jittery input clock.

Same with DACs with an async USB interface. The async scheme is a good idea, but it has to be properly implemented.

Daniel

Thanks, that makes sense.
 
Some facts would help.

You can buy small boards from China with a neat USB implementation to put into your DAC. You would do this because it solves a design problem conveniently rather than for performance per se.

So we need to know if what Beresford has done is simply hacksaw around the USB implementation on his board and use links rather than traces, or whether he has introduced a completely new and possibly rather better USB implementation. Or you could just speculate based on nothing.

Paul
Here's the one board prototype:
t1.jpg


Here's the daughterboard version:
t2.jpg
 
I'd like to see some measurements, it's intriguing why what appear to be the analogue outputs run so close to the USB interface. I'd also like to know what other board changes there are between the two.

Anyway this is engineering and completely out of place on this thread. Inquisitiveness is to be ridiculed.

Paul
 
So we need to know if what Beresford has done is simply hacksaw around the USB implementation on his board and use links rather than traces, or whether he has introduced a completely new and possibly rather better USB implementation. Or you could just speculate based on nothing.
A couple of years ago Beresford was spouting that async usb was a waste of time and just a marketing gimmick and that was the reason why he did not include the option on his dacs.:rolleyes:
 
A couple of years ago Beresford was spouting that async usb was a waste of time and just a marketing gimmick and that was the reason why he did not include the option on his dacs.:rolleyes:

I have no skin in the USB for audio game. My kit uses ethernet & TCP/IP.

But it strikes me that Beresford would have included async USB in his kit because the market demaded it. If it's a case of no async USB = no sales, you include async USB.

Chris
 
I have no skin in the USB for audio game. My kit uses ethernet & TCP/IP.

But it strikes me that Beresford would have included async USB in his kit because the market demaded it. If it's a case of no async USB = no sales, you include async USB.

Chris

Will be the same soon with DSD if rival products have it you have to have it as well if you want to compete in a crowded market.
 
I have no skin in the USB for audio game. My kit uses ethernet & TCP/IP.

But it strikes me that Beresford would have included async USB in his kit because the market demanded it. If it's a case of no async USB = no sales, you include async USB.

Chris

Agree about async USB being a buzzword that nobody understands.

You stream from a NAS to Linn thingy, don't you ?

If that's it, it's the files that are coming over the network, doesn't strike me as the same as sending a, for want of a better description, a real time digitised audio stream to a dac.

Or am I missing something ?
 
Agree about async USB being a buzzword that nobody understands.

You stream from a NAS to Linn thingy, don't you ?

If that's it, it's the files that are coming over the network, doesn't strike me as the same as sending a, for want of a better description, a real time digitised audio stream to a dac.

Or am I missing something ?

It's data, either way. There really is no such thing as a digitised audio stream. It's not audio until processed by the DAC.

Chris
 
I have no skin in the USB for audio game. My kit uses ethernet & TCP/IP.

But it strikes me that Beresford would have included async USB in his kit because the market demaded it. If it's a case of no async USB = no sales, you include async USB.

Chris
The Beresford DAC in question isn't equipped with async USB. I found the AoS thread about it. I don't find him an authority that inspires confidence. Others miles may vary.

Paul
 
I guess we'll have to differ in our view of things. For me, if you are sending files over your network to a your Linn streamer, it's not all the same as sending a data stream via usb in real time (or almost) to a dac.
 
Agree about async USB being a buzzword that nobody understands.

You stream from a NAS to Linn thingy, don't you ?

If that's it, it's the files that are coming over the network, doesn't strike me as the same as sending a, for want of a better description, a real time digitised audio stream to a dac.

Or am I missing something ?
You're right- async usb is different from TC/IP because it (this sort of usb) isn't really asynrchonous as the data has to flow roughly at the rate its played. AFAIK all you get with asynch usb is the receiving device saying "get a move on". That should be enough though to allow the dac to get on with converting the data according to its own clock.

They do have one thing in common though. Having basically solved the problem which was supposed to be such a big deal in audio (interface jitter) they still apparently leave the dac prey to interference down the wire or possibly by the wire.
 
I guess we'll have to differ in our view of things. For me, if you are sending files over your network to a your Linn streamer, it's not all the same as sending a data stream via usb in real time (or almost) to a dac.

Why not? you are sending digital information in both cases. In both cases, the analog domain is not involved. In both cases, the data is bufferred, re-clocked & fed to the DAC.

Chris
 
Even if there were/are good (Audible) reason to have an asynch DAC, some folk will still claim (Ala Item) that there are other problems which need solving :rolleyes:

Personally I think digital audio was perfected some time ago and only functionality, looks, price etc should be of concern.

Read too much foo propaganda though, and.....
 
Even if there were/are good (Audible) reason to have an asynch DAC, some folk will still claim (Ala Item) that there are other problems which need solving :rolleyes:

Personally I think digital audio was perfected some time ago and only functionality, looks, price etc should be of concern.

Read too much foo propaganda though, and.....

And he can sell you just the thing to solve them:)

Chris
 
You're right- async usb is different from TC/IP because it (this sort of usb) isn't really asynrchonous as the data has to flow roughly at the rate its played. AFAIK all you get with asynch usb is the receiving device saying "get a move on". That should be enough though to allow the dac to get on with converting the data according to its own clock.

They do have one thing in common though. Having basically solved the problem which was supposed to be such a big deal in audio (interface jitter) they still apparently leave the dac prey to interference down the wire or possibly by the wire.

This is a genuine question. What exactly is sent along the USB link?

My understanding was that it was encoded data, usually .wav.

Am I missing something?

Chris
 
This is a genuine question. What exactly is sent along the USB link?

My understanding was that it was encoded data, usually .wav.

Server to streamer/computer is reliable TCP/IP FTP protocol

Streamer/Computer then transcodes to WAV (if necessary)

Then data into DAC via half duplex USB 'pipe' via PCM...preferably asynchronously into a buffer
 
Status
Not open for further replies.


advertisement


Back
Top