The State of USB Audio by Alan Taffel

[Audio_ELF]

I guess my point isn't the exact cost ... it's that many people on forums such as this, forget how many other people can feel about computers. They may want to use it as a audio source, but are frightened of "what's in the box". To you or I and probably most people who visit the forum, "just open the box and stick 2x 2GB RAM sticks in" is an easy task, done in half an hour or so. To many audiophiles though, the prospect of opening their MacMini is scary, and (while admittedly I've not read the article but have read TAS before) these are the people that were being addressed in the article. Maybe that makes the omission of testing async USB and no mention of Weiss Minerva even worse, but so far these are minority products (with all due respect to Gordon and Daniel) and not from the companies that Audiophiles will generally turn to. As I think I commented before, they'll go off, buy their Bryston DAC (for example) and plug it in via USB, maybe finding it lacking, plug a CD transport into SPDIF and find that better, and so give up on Computer Audiophile, blaming the "computer" aspect before they even properly begin. At least (what I've heard) the article will maybe expand the readers mind beyond USB as the only way to connect their PC to their DAC - though as has been said more information can be gathers at Computer Audiophile forums in a short time - maybe enough to confuse the hell out of them. :-)

 

Eloise

 

[CG]

Lars,

 

Well put.

 

I'd add that at least around where I live, not exactly out in the wilderness, there are many, many more reputable shops that will upgrade any Mac than will set up a turntable.

 

Not only that, at least it's possible to do the Mac update through the mail if you choose. That's kind of hard with any decent turntable installation.

 

[icebreaker]

I do not think that Mr. Rankin should be required to subject his State of the Art asynchronous USB DAC products to a superficial survey that takes a layman's approach to the field of computer audio. How and whether it is possible for a highly educated person in the field of computer and info technology to suspend his knowledge and do so is another question. Yet that was the approach taken by Mr. Taffel... a survey of the common person's computer and audio equipment. Yet the title and one would assume objective of the article was to discover the lofty "State of USB Audio". Mr. Taffel says it is bleak. Mr. Rankin's products demonstrate otherwise. That has been the buzz in audiophile circles for some time. It is up to Mr. Taffel to provide the right showcase for Mr. Rankin's products. Perhaps then he can pronounce (from a layman's perspective) on the "State of USB Audio".

 

[The Computer Audiophile]

Alan Said, "I consciously approached the topic of USB audio from the perspective of a typical high end audiophile (that is, a typical TAS reader), who is not necessarily a computer whiz. My aim was to determine, from this perspective, the results that could be achieved from USB both on an absolute and relative basis."

 

 

Let's not forget the role of a good audio dealer. There should be no more mystery involved when purchasing a music server and USB DAC than there is purchasing a CD player. The same dealer that helps setup a pair of speakers should help setup and obtain the best sound out of any USB DAC. This is also the case for SSDs and any music server tweaking that should take place. There are dealers around that make music servers fairly easy for customers and the sound achieved is very good.

 

[cfmsp]

 

"that wasn't the starting point for this article."

 

according to who? There was no mention of a 'basic' or 'beginners' view as being the starting point of this article. The title of the article was "The State of USB Audio". The author made sweeping statements about the use of the USB AS A protocol.

 

"Q: How does USB sound?

A: I regret having to report that, at this stage in its development, USB does not rise to the level of an audiophile-quality interface"

 

Note: this last bit (USB does not rise to the level of an audiophile-quality interface) was the penultimate statement of the article, being shown in bold red letters.

 

"Q: How does USB compare to S/PDIF?"

A: There IS no comparison. USB sounds much worse"

 

note: the emphasis on 'IS' was the authors

 

Given TAS' predilections for absurdly expensive products and the title, it seems unlikely that a typical TAS subscriber would believe this article was intended as a basic primer, for beginner's only. If you're gonna run with such an attention grabbing title, and yet not expect people to believe this is an all out assault, you've gotta set the context much better than was done.

 

Indeed, after reading the article, it's hard (for me) to believe much (and certainly not all) of the 'revisionist' historical context that Alan provided in defense of his words.

 

 

 

clay

 

 

[rlodad]

Well ladies and gents, now TAS is blaming the backlash against the article on the industry. No wonder Ayre and wavelength don't want anything to do with the idiots. Here's the note i just got from the editor:

 

Dear Dr. ****:

 

Thank you for writing. It appears that there's an orchestrated effort on the part of one or more manufacturers of USB DACs to attempt to discredit Alan Taffel's article. Below is Alan Taffel's reply to such criticisms.

 

Best regards,

 

Robert Harley

 

Apparently, Mr. Harley doesn't realize it was Mr. Taffel who pissed some of us off, and instead chose to blame it on "the big bad manufacturers." Grow up! You're a businessperson, and we are your clients, you fool!

 

 

[Gag Halfrunt]

...that's the problem it faces.

 

We aren't TAS's readership. As Alan Taffel said earlier, TAS's readership considers USB the stuff of the 'computer whiz', while everyone else considers it USB. OK, it's not quite as simple as that, but given there's very little support for TAS on the boards, even on AVGuide's board, the editors might be right.

 

I suspect the typical TAS reader really does consider anything that happened after CD to be 'computer whiz' material. As such, we aren't TAS's target audience anymore. We're already lost to them.

 

What an horrific state to be in, an audience so old, it doesn't even 'do' the internet!

 

[Lars]

I have been a subscriber to the TAS from day 1: long before Robert Harley was on the scene. I am surprised that he is trying to shift "fault" to manufacturers. This is the height of arrogance. I will not renew my subscription. Given the size of the magazine today, I don't think they are long for this world.

 

[CG]

Nah. It's all more sinister than that.

 

The owner of TAS and HiFi-Plus used to be VP of marketing for Dell. They make PCs. Michael Dell has had plenty of thoughtful comments about Apple and The Mac over the years.

 

Get it???

 

Who says that computer audio can't have it's own conspiracy theory?

 

[satkinsn]

Lars wrote:

 

"I purchase The Absolute Sound to be informed on the status of high end audio. If I want the common man's approach, I'll purchase Consumer Reports. Would you play the Wilson Maxx3 speakers with a $200 amp from WalMart? That's the equivalent of what Alan did in his report."

 

I'm different. Even though I bought a subscription as part of learning audio, I'm very far on the other side - I'm looking for the best sound as cheaply as possible. That's why I find this dust-up so interesting - I'll be very disappointed if using USB requires the same deep pockets regular old audio equipment does.

 

I don't have Alan's piece handy, and I don't remember how much detail he gave about the computer system he was using. Was it enough to conclude the system really was not up to stuff, or was it simply less than Gordon's recommended Mac?

 

Scott A.

 

[DavidL]

I first joined this forum when I was looking for a Firewire audio interface, primarily to do AtoD conversion. Using the info I got from this forum and Audio Asylum I bought an RME Fireface 400 - there was a strongly supportive user group but no audiophile reviews so I went out on a limb and bought it blind.

It has been doing the job required very well (24bit / 192kHz LP archiving).

I have been using the DAC in the FF400 for monitoring my archived results and the output seems at least as good as that from a MF Trivista DAC and better than the USB DAC in my MF X-CANv8.

I was very pleased when the HiFi News review of the PS Audio PWT / PWD combination found that the output from the DAC in the RME FF400 matched or in some cases exceeded the sound quality from the PWT / PWD combination on hi-res music.

So my experience supports the notion that external firewire DACs can be very good; certainly better than the USB alternative I had available.

Can we have some more audiophile reviews of these firewire DACs please?

 

[Lars]

Scott,

 

The iMac was an old non-intel CPU unit. It will not run the new Snow Loepard Mac OSX operating system coming in Sept. I don't know the specs, but given the age of the unit, its not very impressive. Alan was also using a PC running Windows 2000. This operating system is very old. XP, Vista, and now Windows 7 is coming soon. He did have a an HP laptop runnign Vista. How could he compare OSX on an ancient computer with his laptop running Vista?

 

Computer DACS are just like other high end components. It all depends on how much you want to spend. You can get a good USB DAC from Wavelength for $900 (the Proton ) or a firewire DAC from Apogee.

 

There are other DACs that Chris has spoken about. This forum is a great place to learn about computer audio.

 

Thanks to Chris for making this place a reality.

 

[rlodad]

1. I am old (42)...

2. I still know quite a bit about Macs and PCs, though I Majored in History and have a doctorate in medicine.

3. I WAS a TAS subscriber.

Arrogance is too nice a word. It implies an IQ compatible with logic.

 

[rlodad]

David,

FWIW, I own both of the "DACs" you mentioned. The trivista is a butiful sounding piece of equipment. The USB interface in the X-can V8 must have been an afterthought (part of the problem in the TAS article). It is a great headphone amp, but the usb DAC stinks, IMHO.

 

Brad

 

[rlodad]

David,

FWIW, I own both of the "DACs" you mentioned. The trivista is a beautiful sounding piece of equipment. The USB interface in the X-can V8 must have been an afterthought (part of the problem in the TAS article). It is a great headphone amp, but the usb DAC stinks, IMHO.

 

Brad

 

[cfmsp]

 

"Was it enough to conclude the system really was not up to stuff, or was it simply less than Gordon's recommended Mac?"

 

Scott,

 

I don't recall the Windows systems, but the Mac Alan used is an iMac G4. There were manufactured from 2002 thru Aug 2004. So the computer is at least 5 years old, perhaps as much as 7. Remember when CPUs were single core and measured in Mhz? remember when memory was measured in Mb?

 

The processor speed on the single core PowerPC processor could be 700Mhz up to 1.25 Gb. If the memory has been maxed out, it could be 1Gb, if not, something more like 256Mb was standard I think.

 

Gordon had a point.

 

clay

 

 

 

 

[satkinsn]

...that there's a meaningful difference between PCs and Macs, nor between the last several iterations of the operating systems.

 

Obviously, with XP (and 2000, if memory serves) you have to deal with the kmixer issue. But once you get past that, I don't know of any particular superiority in Vista or the forthcoming Windows 7. In fact, XP and 2000, in particular, have smaller memory footprints.

 

(Aside - I think 7 is shaping up as the 2000 of this generation. 2000 was, for Windows, a fast, light OS. 7 has that same feel to it.)

 

Any Mac OS since core audio was implemented should yield the same results.

 

If there's an argument to be made here, it might be in the area of memory, as Gordon mentions. But without some study, I'm hard pressed to believe any mainstream PC post, say, 2002 would be an impediment.

 

s.

 

[CG]

An iMac G4 or a Mini G4 will work, but probably not nearly as well as a more modern system using an Intel processor. The main concern to me would be the operating system. There's been at least two major revisions of the software since that vintage was released, with significant changes in the audio system.

 

I don't think it's unreasonable to think that a "high end audiophile" would apply the same level of system optimization for the computer than s/he would for speaker placement, phono cartridge VTA, and 278 other factors. But, I are but an amateur.

 

[arin]

I was so happy to read this forum topic!

 

When I read the Taffel article I thought, here we are again, always the same old story: computer audio will never be so hi-end as the "real" hi-end!

But TAS is not alone in that respect: every hi-fi magazine (American, English, French or Italian) I read in the last 2 or 3 years told the same big lie about computer audio.

No matter what computer, what operating system and what program you use, according to them they will never ever sound as good as a hi-end CD player.

 

A pity because it was the same TAS which more or less a year ago published an little and brave article about the superior playback of audio from a hard disk!

And in that very article they started the "jitter war"...

And then back to the usual articles&reviews!

 

Ciao!

 

Arin

 

[bottlerocket]

- The DACs TAS reviewed sound better with SPDIF than USB.

 

- Gordon removed his DAC from the survey.

 

- TAS completely missed the outcome of a lot of discussion on this site - Good DACs are either firewire, asyncronous USB or SPDIF. They apparently reviewed syncronous USB and compared that to SPDIF and came to a conclusion showing that they have a firm grasp of the obvious.

 

Is that so bad though? The next logical step is to highlight the alternatives and that will be a good thing.

 

I do think the average user wants USB. Its ubiquitous. Its simple. Most computer users think that is where you plug in the accessories so they'd understand the fact that all they need to do is plug in a DAC. That being said, if MACs sound better than PCs and MACs come with firewire, why aren't we all just using firewire DACs up to 24/192?

 

 

 

[Gag Halfrunt]

The new editor at Hi-Fi+ in the UK seems to be more interested in computer audio than most. I spoke to him at the Bristol Show and he said that he wants to do more computer audio products, but he'd already received a barrage of negative comments at the suggestion from the regular readers at the show.

 

His first issue was full of products like the Benchmark and the HRT Streamer, but not so much in the next.

 

Perhaps his readers got to him...

 

[satkinsn]

Do we know - either through studies or through math - exactly how taxing lossless file playback is?

 

We know, for example, that a limited, old computer can play back a cd or an mp3 file.

 

The question becomes, (I think), is a computer that can play back a given file all you need? In other words, if a given processor and memory combination can do it at all, is it doing it perfectly?

 

Does it make any difference whether you're dealing with a new quad core, a two year old dual core or a circa 2000 Power PC G4?

 

Obviously, the quality of the DAC, the interconnects (if you believe in such things), the software, and the speed of the hard drive could all affect what happens. But computers have been very powerful for a while now; I can send an email from from a circa 2003 Dell in my office as easily as I can from the late 2007 Mac Mini I'm writing this on. The web loads just as quickly.

 

Is music playback so taxing, does it stress a computer's resources so much that we are only now (in, say, the dual core era) getting enough power to do the job?

 

Is anybody measuring computers for their adaquacy?

 

s.

 

 

 

 

[idiot_savant]

Hello all,

 

my take on the whole PC/Mac being FAST enough is pretty simple - if the CPU can't process enough audio, it doesn't degrade in a subtle way. Packets ( which are probably of the order of 100ms or so ) will not reach the output in time - this won't cause "a collapse of soundstage" or "an increase in sharpness" - it will cause a big pop, bang or crackle - the data is there (good), or it's gone missing or is late ( bad )- imagine stitching together bits of tape 1/10 of a second long with variable gaps in it. Smaller packets ( which require MORE processing time ) will case smaller pops, but it will be obvious.

As to whether or not the CPU running at more utilization affects the audio, I just don't have the answer for that. I know what I think ( i.e. if the interface is done properly, and is external to the PC, it's hard to see how it makes a difference ), but there seems to be lots of subjective evidence to the contrary, but also very little objective evidence in affirmation.

 

PCs/Macs deal with BUFFERS of audio data in real time, not single samples... Having said that, adaptive USB is probably the most sensitive audio interface you can imagine... ( apart from I2S over a couple of feet )

 

My issues ( here we go ) with the article ( which I read today ) are as follows:

 

Most audiophiles, upon receipt of a new piece of hardware spend some time and effort investigating the best way to set it up, be it by ear, or reading articles, and will tend to play about with positioning, the various settings on the new piece, cabling, etc. etc. The whole "just plug in a USB DAC and everything will be perfect" slant he puts on seems an antithesis to serious audiophiles. The job of a reviewer in this case should be to help the user differentiate the signal from the noise - adding 16Gb of RAM is likely to be less useful than checking that the sample rate is correct on Vista using iTunes, for instance, and a unit using a PCM270x will likely be not very good ( which you can tell simply by the sample rates/bit depths it supports ).

 

The second issue I have is that the article is very broad in it's sweep of dismissing USB "in it's current state", yet the author who claims he is aware of the differences between USB modes fails to mention anything to do with the implementation between his sources, OR that there is a better implementation available, and has been for some time ( Wavelength well over a year, dCS well over 6 months, Ayre must have been available at review time ). Indeed, if he IS aware of the different implementations ( which, unlike a lot of claims in audio CAN be measured ), why does he not mention them at all?

 

Thirdly, a swipe which is often taken at USB, and furthered by this article is "it's designed to connect printers and stuff like that" - originally it was, but the Isochronous mode is actually extremely similar to the firewire implementation ( don't believe the rubbish you read about firewire "streaming" and USB "packets" - firewire and USB both "stream" "packets" in iso mode in EXTREMELY similar ways ).

 

your friendly (annoyed) neighbourhood idiot

 

 

[BobH]

I think I tend to agree with Mr. Taffel. Not his methodology, or the broad sweeping statements, but with the general notion that usb, as an audio interface, is rubbish. Granted, it can be made to be very good indeed - or so the reviews of well implemented usb dac's would indicate. My own experience, however, is that, whilst it can be good, it is nowhere near as good as coax and not even in the same territory as networked interfaces.

 

The three most respected UK audiophile kit manufacturers, namely Linn, Naim and Meridian, all use 'networking' to deliver the data. Where usb ports are provided they are purely for external data storage. These manufacturers make expensive bits of kit, with a great deal of R&D development time put into them. The fact that they all chose to produce networking streamers, and none of them produce usb dac's, cannot be a coincidence.

 

My humble set-up, of Netbook/Squeezebox/Promitheus NOS Dac, allows me to directly compare computer-usb-->dac, computer-co-ax spdif -->dac and network-coax spdif -->dac. Sparing the hyperbole, the latter's superiority is clearly audible to me. When I have tested and scored the various delivery methods they have always come out, in order of preference, Network/Coax Spdif/USB.

 

USB is an ubiquitous interface and, as such, is an attractive option to those seeking a 'computer audio' market. But usb was designed for the simple attachment of hardware devices. The networking ports were designed for the streaming of data. I think it shows.

 

IMHO the computer is ideal for control/delivery. It is nowhere near being the best audio interface, not even close. In that respect I think the TAS article came to the correct conclusion - it just didn't do it very well!

 

[idiot_savant]

Hi,

 

I'm sorry but I'm going to have to take issue with this...

 

Networking ( Ethernet, anyway ) WAS NOT DESIGNED to stream data - do you think any interface that was designed to stream realtime data would not have better QoS schemes in place? Try playing a few videos over a network, and see what happens - do you get a "sorry, I can't get enough bandwidth to do this properly", or stuttering video?

 

I think that everyone knows (now) that there are a handful of USB implementations done properly - does your extensive testing involve any of these? USB can ( like firewire, or ethernet ) provide measureably better jitter performance than SPDIF ( on it's own ) - IF done properly. Of course, if you're using a USB->SPDIF, any advantages USB may have in jitter performance will depend on the SPDIF side of things, but we'll let that lie, eh?

 

My issue isn't about USB, or firewire, or ethernet, or computer audio - it's that the facts are not represented fairly, and one interface is done down due to some poor implementations, whereas the good ones are completely ignored.

USB, firewire and ethernet all have the capability of providing excellent data with excellent timing ( they obviously don't actually do this for every implementation ). They all have pros and cons, so to rule one of them out for no good reason than ( apparently ) the magazine having the grumps with a couple of manufacturers ( Wavelength & Ayre we know of ) is wrong.

 

Oh, and what networking does Meridian do ( or do you mean Sooloos )? And can you replace your Linn DS with an older NaimNet DAC?

 

your friendly neighbourhood idiot

 

I'm sorry for the rather confrontational aspects of this post, but I'm a rather annoyed idiot...

and what was ethernet designed for, again?