USB-S/PDIF Converter or USB DAC?


I've decided it's time to use a music server with my main system. I currently use one (iBook) with my headphone system with the Grace m902 headphone amp with built-in USB DAC. I intend to get a new Intel iBook when they are released (March/April).

I currently have an unused Wavelength Audio S/PDIF DAC that I could use with a USB-S/PDIF converter (such as Empirical or Red Wine Audio) or get a USB DAC (WA Brick, Stello, ???). Is a dedicated USB DAC preferred over the hack USB-S/PDIF converter route?
budrew
Yes, the lower jitter equates to better sound.

Not sure about the SB jitter measurements. Most stock units have relatively high jitter IME.

Steve N.
Thank you very much for your answers Steve and Hagtech!

Does the lower jitter correlate to better sound...I am dubious there...

How does the Squeezebox works? Is it bidirectonal? I know it has about 5 minutes of buffer, does that help with jitter? What are the jitter figures of the Squeezebox2?

I have compared the Squeezebox2 to a transport and it was not even close, yet the jitter is lower in the SB2... Again: Does lower jitter means better sound?

All the Best
"-Isoncronous (isosyncronus?) data transfer.."

Isochronous means that the bandwidth and access is guaranteed. This helps eliminate underruns. Most USB controllers use this mode.

"-We also have the new DVD players with buffer systems to do the error correction,"

Errors are not the issue, jitter is.

"-then we get the external clocks like the BigBen or the DCS verona that enslave transport and DAC to the same clock, so no error from 2 clocks there, supposed to be great also!
Everyone claims to be the best..."

The Big Ben does a good job of asynchrnously reclocking the S/PDIF data stream and does reduce jitter. Only as good as the clock in the Big Ben though.

"What about a huge overbuilt transport like VRDS, CEC or Forsell..."

I've modded many of these to improve jitter, including CEC. Still much more jitter than a well-implemented USB or Wi-Fi interface, even with mods.

"I have yet to try the external clock and a USB DAC but I am getting less and less motivated by this findings..."

Again, it's in the implementation, the design. Computer audio has the potential to beat all transport-based systems.

Steve N.
Empirical Audio
Manufacturer/modder
The isochronous data transfer is basically a one-way packet sent from PC to DAC via USB. It is sent 1000 times per second, and does not have error correction. This is the basic USB audio that most of us use. If a packet has an error or gets dropped, then there is a glitch or pause in the output. No buffering of data. Think of it as sort of a pulsed S/PDIF stream. Any glitch is immediately heard in the output, no error corrections.

The output word clock from most of the USB chips on the market is derived from this 1000Hz packet rate. Some xtal in the PC sets this timing reference (can anyone say jitter?). The USB receiver then uses an asynchronous clock to catch the data. It is then re-clocked (from packets to continuous I2S or S/PDIF) using a new clock derived from a PLL based synthesizer that uses the 1ms packet rate as a reference. Both the USB clock and 1ms clocks impact the jitter performance. The onboard PLL does help to remove some of the reference jitter. Results can actually be pretty good.

The other USB mode commonly in use by pro-audio companies (Edirol, M-audio, etc.) is to create their own custom USB interface. This require special software drivers but allows them to run other data rates (24/96), multiple channels (more than stereo), error correction, or even clocking schemes where the receiver is master. It allows for a lot of potential performance.

Most of us use the PCM270x or PCM290x series of chips from TI. They are easy to use, and relatively plug-and-play with both XP and OS-X. Our products are limited to 16 bit 44.1k/48k resolution. At the moment. Someday I expect TI and the operating system folks to take this to the next level (I could be wrong).

That said, in my opinion, the isochronous USB audio isn't really inherently superior to S/PDIF. They both have issues. The both can have jitter. To me, neither is vastly superior over the other. In fact, they both can be implemented quite well. My solution to both is to perform the reclocking afterwards. That is, right at the converter output. This essentially compensates for most of the signal deterioration incurred in the data transmission between boxes.

To me, I find such jitter-reducing reclocking to make a big difference in the sonics. Everything becomes smoother and more relaxed. In a way, the music almost slows down. Any harshness or edge or agressiveness disappears. I guess the most noticable attribute is that long-term listening fatigue is eliminated. For me anyway, that's the way it is.

jh
i share most of jsadurni's confusion and would benefit from some good explanations. it would seem to justify a new thread.
I am intrigued by USB as soooo many posts claim its superiority, but can somebody please explain this:
-Isoncronous (isosyncronus?) data transfer..
-and bidirectional with error corrrection.
-We also have the new DVD players with buffer systems to do the error correction,
-then we get the external clocks like the BigBen or the DCS verona that enslave transport and DAC to the same clock, so no error from 2 clocks there, supposed to be great also!
Everyone claims to be the best...
What about a huge overbuilt transport like VRDS, CEC or Forsell...

I would only trust my ears because there are a lot of technical explanations that sound soooo logical but when the time comes to hook the thing up to your system they just dont sound that good...
I had in my system Squeezeboxes modded and not which have Ethernet connection, I dont think you can go more bidirectional than that! Unless they are not using the function (isoncronous maybe?) It does sound nice and very comfortable BUT not audiophile quality really (compared to say a Forsell transport)
I had a intensly modded DVD player with the buffer thingy and extra clock etc, it just dont cut it (compared to say a CEC transport)
I have yet to try the external clock and a USB DAC but I am getting less and less motivated by this findings...
I am really intrigued by the Wavelenght DACs mostly. I hear a lot of good things about them!
Anyone compared some of these in their own system?
Thanks
Hmm... I'm intrigued by the wireless USB hub. I don't think I'd worry about UWB as a RF technology--its a very, very low level signal; we are talking about impulses here. Anyway, I have a hard time believing your are going to get interference problems with your hardware.

B'sides, the upside is huge--if the Belkin lit is to be believed, its freeking fast. USB 2.0 is 3mbps, and they say its 100x faster. Plenty of bandwidth to do very interesting things, including 24/96.

With reference to some other prior comments, and I may be wrong here, I don't think pops and clicks are a USB issues. When my DAC loses sync, it takes at least a second to re-sync. Anything resolving as a pop or click strikes me as being a source problem, power problem or the like, not a USB falling off the BER cliff problem.

I've used USB repeater cables and that sounded one helluva lot better to my ear than going toslink for long distances... Granted, it was a crappy toslink cable, but the difference in audiophile speak was "not subtle."
I did A/B comparisons for a long time with my WA Brick and Gordon is right. Going longer than one 5 m USB cable degrades the sound. I could hear a clear difference when trying three and even two USB cables.
Belkin stated that the release of the wireless USB hub will be in Sept. The chip maker backed out of the deal.
Gordon Rankin at Wavelength claims that his DACs are not isochronous, he says they are bidirectional with error corrrection.

Even those that are, if they properly re-clock the data, should be about as jitter free as you can get.

Steve, heard anymore about the Belkin? Seems to have dropped off the radar.
So guys, what about the fact that all known USB DACs use isoncronous data transfer, which makes the major advantages of generic USB: it's bidiretional and the signal is error corrected at both input and output, to lose its value? It works pretty much simlar to SPDIF, and prone to even higher jitter... Why would anyone want it?
Not sure about the spurious emissions. It is evidently the new ultrawideband technology in the Belkin device, which is low-power but wide frequency emissions. If there is an RF technology that might cause problems, this is it.

If you want something safer, then go with the AirPort Express. This is 802.11G. I have a version of this, Off-Ramp WIFI with S/PDIF or AES output, and soon hopefully I2S output.

Steve N.
Empirical Audio
Manufacturer
Thanks. I think I get it. This could solve a major problem for me, which is how to locate a laptop at my listening position and feed a DAC across the room without running a physical cable. I'd stick one of these hubs near the audio rack. I wonder if the frequency this thing operates at will pose any problems to an audio system.
See this link, not 802.11x:
http://www.belkin.com/pressroom/releases/uploads/01_03_06CableFreeUSB.html
>wirless USB hub

What does that mean, exactly? Is it a separate USB network (with the hub as access point and wireless-enabled USB receivers or devices communicating with the hub) or part of an 802.11x network?
Belkin will also be offering a wireless USB Hub in July 06. This will have the bandwidth for 24/96.
For your USB extension, check out this optical cable:

http://www.usbstuff.com/opticis.html

I use a 65 foot optical usb extensnion with no loss or degradation of the signal.

If you are using a PC, the following ASIO driver is superior to the ASIO4ALL

http://www.asio2ks.de/index.html

Using this driver, I have no pops or ticks with the Cosecant. The settings on J. River and Foobar are minimum buffering.

I hope this helps you,

Steve
I had the same problem with the WA Brick and Gordon gave me the advice of using long ICs. I had bought two 5m USB extenders and it´s what I am using, but unfortunately it crackles every now and then. The IC route is way too expensive and I guess over time I will put the computer next to my system.
For more than 5m, use a USB-SPDIF converter first, then you can send a long way. Across the house. To the neighbors, even.

jh
If any buffering involved, you may have delays resulting in pops and clicks...

So what do you guys do if the distance is more then 5m?
5M is the maximum length.

There's a great web site called "Google" where you can find out things like this or, it seems, just about anything.

Not sure what you mean about timing.
What is the maximal reasonable length for USB connection? AFAIK after some length you have to put USB repeaters. It won't distort the data, but it may mess up timing, isn't it?
Does anyone have any experience or opinions about Toslink to coax converters (maybe it's an adapter and not a converter)?
Budrew -- fyi, since you sold your Cosecant, he's come out with a new version -- same guts, but he tweaked the cosmetics -- check his web site. -- David
Thanks, I'm going the USB DAC route. I actually had an early Cosecant, but did not use it much at the time so I sold it. But now computer audio is really useful for me and I like the character of Gordon's DACs. I'll probably spring for the Cosecant.
Budrew -- I have the Wavelength Cosecant (after trading up from the Brick), and have to second Terrat's observations. Both are the best digital I've ever heard, and while the Cosecant is significantly better than the Brick, I'd have to say the latter is more of a bargain. (Both are basically the same USB DAC design, but the Cosecant has an output transformer and a better power supply).

I have no experience with other USB DAC's. I'm fairly sure that a lot depends on how well the USB controller is implemented and how good the actual DAC part is. For example, Gordon Rankin uses no up or oversampling. I don't say this because it is necessarily better or worse, just to indicate that different USB DAC's are likely to sound different, like everywhere else in audio. Make sure you buy something that sounds good to you in reality, not just in concept. End of lecture ;<) Hope that helps. -- David
Budrew, you are heading in the right direction. YES! USB has many advantages over SPDIF. One big advantage is that USB is bidiretional in it's signal transfer. The signal is error corrected at both input and output, resulting in a bit for bit transfer of data. Not to mention the added benefit of having an exact bit for bit copy of your all your CDs at your fingertips. Just don't forget to back-up your data.