TAS 194: The State of USB Audio
September 6th, 2009 -- by default
Source: The Absolute Sound
Source: The Absolute Sound
Q: Is USB the best-sounding PC output available?
A: No. In my tests, FireWire proved to be a far better audio interface than USB. As already noted, FireWire PC ports, though not universal, are commonplace. However, there is as yet no such thing as a FireWire DAC. So in order to assess FireWire, I turned to the pro recording market, where FireWire is the PC audio interface of choice. There, a large array of FireWire gear is available. Because many of these units incorporate S/PDIF outputs, they require nothing special on the DAC end. Many support resolution of 96/24 or higher.
I tested one example, the Focusrite Saffire ($499), and was knocked out not only by its lovely sound, but by the complete absence of the characteristics that spoil USB. (My review of the Focusrite Saffire, referenced above, contains a detailed description of the Saffire’s and FireWire’s sound.). While the Saffire alone cannot speak to FireWire’s ultimate sonic potential, it does establish a floor that is already higher than USB’s apparent ceiling. This interface holds great promise for audiophiles.
Conclusions
USB heralds a new era in the way we buy, store, and play music, but for now it is sonically a step behind. Coax S/PDIF is unquestionably a superior digital interface, capable of a natural presentation, rhythmic engagement, and tonal richness that elude USB. Given that S/PDIF is by far the more entrenched source-component interface—a state of affairs that hopefully will not drift wayward—most audiophiles need not worry about USB’s current limitations.
Yet the idea of a laptop acting as a music server has real merit, and here coax S/PDIF is rarely an option. Instead, the two most common interfaces are USB and FireWire. Here again, though, the existence of good-sounding USB-to-S/PDIF converters reduces the need for a USB DAC. Meanwhile, it has long been accepted dogma in the pro recording industry that FireWire sounds better than USB. My own tests, in a completely different context, confirm that consensus. Given that, high-end designers would do their customers a great service by producing a FireWire DAC. No doubt audiophiles would also welcome a FireWire-to-S/PDIF converter built to audiophile standards. The sonic possibilities of such products are truly exhilarating.
Despite the presence of alternatives and workarounds, the USB revolution will undoubtedly continue to build momentum, fueled by the interface’s ubiquity and simplicity. Considering the currently deficient state of USB audio, the high end must not merely embrace this revolution, but direct it toward the sonically meaningful standards our industry is built on.
All content, design, and layout are Copyright © 1999 - 2011 NextScreen. All Rights Reserved.
Reproduction in whole or part in any form or medium without specific written permission is prohibited.
Reproduction in whole or part in any form or medium without specific written permission is prohibited.
Comments
"Q: How does USB compare to S/PDIF?
A: There is no comparison. USB sounds much worse. All four of the DACs I tested support both USB and S/PDIF, making differentiating the interfaces a cinch. For each DAC, I compared the S/PDIF output of a good CD transport—as well as that of a budget CD player—to the USB output of a laptop, with both sources sending the same bits at the same rate (44.1/16) to the same DAC."
To be fair, this question should have read: "How does USB compare to S/PDIF" using this particular set of devices at this point in time"
IMO, jumping to generalized conclusions based on a small sample of products is like listening to 2 CD randomly selected players and declaring that all CD players sound harsh. There are other USB solutions out there and have been for many years that were not included in this shootout.
Alan! Full Steam Ahead!
How can you be staying the course with a reprint of this article when you have run aground with audio experts on so many fundamental points?
Is bad publicity better than nothing?
Alan,
The problem here is that you did not really prepare for the article. In the first place you are using totally outdated computers. From a background such as yours this just seems silly. Using a sub standard laptop, a totally outdated G series MAC... but here is the kicker... A Win2000 computer. Maybe you didn't read the fine print in the 2000 release notes but Win2000 is not even USB Class 1 compliant. Meaning it does not even support all the functionality required to reproduce audio.
I have had many reviewers use the same laptop Alan is using and said they could not even get 24/96 to operate. I even had one of these sent here and the USB port had error rates so high the output did not classify as MP3 quality.
Next Media Monkey... Alan come on it is a well known fact that Media Monkey is not even bit true.
Gang this is why I withdrew my Cosecant from this review. At the time Alan had never even used a computer for high end audio.
It just seems ashame from a magazine that totes Absolute Sound that they can't even get the basic needs of the computer right for this. Especially in an age were power cords are going for +$800 and he can't even spend that much for a decent computer to do the review with.
It would have been very easy and very appropriate to buy a maxed out Mini for about the same cost as the power cords he is using and merely dual boot the unit into Vista or OSX and have a stable platform and consistent for each operatiing system.
To the editors who I sent many emails before this train wreck occured and numerious afterwards.... I just throw up my hands.... what were you thinking????
Thanks
Gordon Rankin
Not here to plug my own products, just here because I believe in the technology and feel Absolute Sound did a total injustice to it.
Furthermore Taffel and perhaps even TAS have a beef to pick. There is no possibility IMHO, that Alan Taffel could write such an unscientific, biased article unless he was trying to boost sales for the "formerly mainstream" audio hardware manufacturers who support his rag. Have you seen a Wavelength or Ayre ad in TAS recently? No, TAS has refused to have an educated reviewer listen to and comment on asychronous USB Dacs. Notice how the Benchmark products which don't use Gordon Rankin's USB implementation, but are well reviewed nontheless, were panned by Taffel? He couldn't say they sounded good, it would blow the whole article to bits. (As if it could stand on its claims, anyhow). The fact is the newer, USB only DACs sound great. Many DACs that Taffel listened to, with SPDIF and USB inputs, sound poor through the USB inputs because they were placed as an afterthought. The ultimate responsibility here lies with Mr. Harley. This junk article should never have been printed. And to have Steven Stone print an article drooling on about the Music Streamer on the next page is more pathetic. Fact is, get your opinions at Computeraudiophile.com, Stereophile, your local dealer, audiosylum- anywhere but TAS.
Furthermore Taffel and perhaps even TAS have a beef to pick. There is no possibility IMHO, that Alan Taffel could write such an unscientific, biased article unless he was trying to boost sales for the "formerly mainstream" audio hardware manufacturers who support his rag. Have you seen a Wavelength or Ayre ad in TAS recently? No, TAS has refused to have an educated reviewer listen to and comment on asychronous USB Dacs. Notice how the Benchmark products which don't use Gordon Rankin's USB implementation, but are well reviewed nontheless, were panned by Taffel? He couldn't say they sounded good, it would blow the whole article to bits. (As if it could stand on its claims, anyhow). The fact is the newer, USB only DACs sound great. Many DACs that Taffel listened to, with SPDIF and USB inputs, sound poor through the USB inputs because they were placed as an afterthought. The ultimate responsibility here lies with Mr. Harley. This junk article should never have been printed. And to have Steven Stone print an article drooling on about the Music Streamer on the next page is more pathetic. Fact is, get your opinions at Computeraudiophile.com, Stereophile, your local dealer, audiosylum- anywhere but TAS.
I've spoken with several engineers who thoroughly understand USB and there should be nothing in USB, as a physical transport layer, that should affect audio quality. However, because I read this artical and tried a different setup with my system, I now know that for whatever reason, USB does affect quality. I have a PC running XP (in the basement with cable running through the floor so I don't hear the PC's fan and disc seeks) and iTunes. All of my CD are ripped to Apple lossless and I had been playing them via USB to a PS Audio DAC for the past two years. Before I purchased the PS Audio DAC I was using a CardDeluxe sound card. The PS Audio DAC via USB sounded so much better than the analog out of the CardDeluxe I kinda left it at that. I did do an experiment of plugging in a low end CD player (my good CD player was busted) to the DAC via coax, but the PC to USB to DAC sounded much, much better so I left it at that for the past two years. After reading this article I decide to try connecting the PS Audio DAC using a high end coax to the digital out of the CardDeluxe. OMG!! Night and Day. Immense improvement of clarity, soundstage, transients, frequency response...you name it...it all sounds better. Vocals, percussion, strings, everything. Its like I've never heard my CDs before.
I still don't understand why - USB physical layer? Window's USB audio-class driver? or what? But I sure do like it!!!
THANKS FOR WRITING THIS ARTICLE!
I've spoken with several engineers who thoroughly understand USB and there should be nothing in USB, as a physical transport layer, that should affect audio quality. However, because I read this artical and tried a different setup with my system, I now know that for whatever reason, USB does affect quality. I have a PC running XP (in the basement with cable running through the floor so I don't hear the PC's fan and disc seeks) and iTunes. All of my CD are ripped to Apple lossless and I had been playing them via USB to a PS Audio DAC for the past two years. Before I purchased the PS Audio DAC I was using a CardDeluxe sound card. The PS Audio DAC via USB sounded so much better than the analog out of the CardDeluxe I kinda left it at that. I did do an experiment of plugging in a low end CD player (my good CD player was busted) to the DAC via coax, but the PC to USB to DAC sounded much, much better so I left it at that for the past two years. After reading this article I decide to try connecting the PS Audio DAC using a high end coax to the digital out of the CardDeluxe. OMG!! Night and Day. Immense improvement of clarity, soundstage, transients, frequency response...you name it...it all sounds better. Vocals, percussion, strings, everything. Its like I've never heard my CDs before.
I still don't understand why - USB physical layer? Window's USB audio-class driver? or what? But I sure do like it!!!
THANKS FOR WRITING THIS ARTICLE!
Again, in computer to USB audio, quality depends on lots of factors in setup. If you got such a difference with the coax out of your sound card, then you probably needed some software or hardware tweaks to get comparable or better sound from USB. USB done right can sound as good or better than just about any soundcard output (even from COAX). But if you're happy with what you're doing now, that's all that matters for you.
I don't know your "engineer" acquaintances, but I'm willing to bet they have a computer and not a sound background. Computer people will tell you that there is nothing in USB that should affect sound. Audio people with USB experience will tell you that USB can sound great if you know how to use it and set it up for audio.
The problem with the above article is that the reviewer didn't know how to properly use USB for audio, and went on to conclude that "USB can't" give high level results.
I'm an oudiophile that is approaching the world of music on FLAC on hardrives and I'm reading as much as I can to get more info.
Well from all what I read in the previous comments we can assume that if on one hand the quality trhu SPDIF is a no brainer, on the other thru USB, unless the user has a solid foundation in informatic to implement all what it takes to have the USB the perform at it's best, the results are likely to be way below it's true potential and this is a serious issue that makes this technology not very user friendly so far. Is this correct?
It's not that simple. USB takes a little effort to learn the setup, but it's not rocket science.
See http://www.benchmarkmedia.com/wiki/index.php/Computer_Audio_Playback_-_S...
for the basics.
SPDIF is simple, but in many cases unless you have a pro level sound card like a Lynx, won't give sound at the level that properly setup USB will.
USB has some convenience factors in its favor also, once it is setup.
It's not that simple a dichotomy. USB takes a little effort to learn the setup, but it's not rocket science.
See http://www.benchmarkmedia.com/wiki/index.php/Computer_Audio_Playback_-_S...
for the basics - not hard to apply to any server.
SPDIF is simple, but in many cases (unless you have a pro level sound card like a Lynx), won't give sound at the level that properly setup USB will.
USB has some convenience factors in its favor also, once it is setup.