greenelec
greenelec's picture
Offline
Last seen: 9 years 11 months ago
Joined: Feb 10 2006 - 12:37am
USB vs. Firewire
CharlyD
CharlyD's picture
Offline
Last seen: 12 years 8 months ago
Joined: Jul 20 2006 - 4:01pm

It's all in the implementation. There are some very good sounding USB DACs as well as several I'd rather not live with. The best sounding USB DACs compete very well against the best sounding Firewire models.

struts
struts's picture
Offline
Last seen: 2 years 10 months ago
Joined: Feb 1 2007 - 12:02pm


Quote:
I have seen comments on the web, by respected mastering engineers, that state that firewire is audibly better than USB for an audio interface.


I suspect that these are cases of 'proof by example' ("this Firewire product sounded better than that USB product ergo Firewire must be superior to USB"). That such generalizations are incanted by 'respected mastering engineers' compounds one fallacious argument with another, argumentum ad verecundiam.

The USB is a general-purpose interface (the clue is in the name: Universal Serial Bus) designed to enable the connection of a wide range of computer peripherals. Pretty much every printer, mouse, webcam, etc. has a USB interface whereas Firewire is generally confined to high-capacity disk drives, and audio/video equipment. Firewire was designed for high-speed comms and bulk data transfer so one could argue that theoretically it is a more appropriate interface for high resolution audio applications. Remember, some folks claim to be able to 'hear' theoretical superiority, while others link cause and effect a bit too selectively.

That said, the USB does provide support for several Audio Device Classes and the needs of audio applications were anticipated in its design. Several synchronization schemes are supported for isochronous data transfer, including asynchronous (to the USB clock that is) which can be used to implement a scheme whereby the receiver (DAC) can control the rate at which the sender (PC) sends the data.

Asynchronous mode has only recently been exploited in high-end audio applications (most notably by Gordon Rankin of Wavelength Audio) and has been shown to enable a significant reduction in jitter - see Ayre's White Paper on Rankin's 'Streamlength' technology used in their QB-9 USB DAC and JA's review of same in 'phile for more detail. However as CharlyD says, in practice the interface used for data transfer is just one of the factors that will influence the sound quality, a myriad of other design and implementation decisions will be of equal, if not greater, significance.

Perhaps in part because of its flexibility USB has become pretty much ubiquitous: you will find USB-ports on pretty much all laptops, netbooks and most desktop PCs - including Macs, whereas Firewire ports are provided as standard on Macs (it was after all Apple that invented the standard) and a small minority of PCs. So the addressable market for a USB DAC will be considerably greater than that for a Firewire DAC and this undoubtedly influences product planners, even at the high end.

linden518
linden518's picture
Offline
Last seen: Never ago
Joined: Dec 12 2007 - 5:34am

It's also worth a mention that the USB DACs will support up to 24/69 whereas the Firewire can do 24/192, etc.; who knows, that may have some kind of a ramification down the line.

struts
struts's picture
Offline
Last seen: 2 years 10 months ago
Joined: Feb 1 2007 - 12:02pm


Quote:
It's also worth a mention that the USB DACs will support up to 24/69 whereas the Firewire can do 24/192, etc.

Oops, looks like you missed Friday's newsflash, SD. See Charlie Hansen's comment to this post.

linden518
linden518's picture
Offline
Last seen: Never ago
Joined: Dec 12 2007 - 5:34am

Wow, I stand corrected, strutsie! What a brave new world we live in.

greenelec
greenelec's picture
Offline
Last seen: 9 years 11 months ago
Joined: Feb 10 2006 - 12:37am


Quote:

I suspect that these are cases of 'proof by example' ("this Firewire product sounded better than that USB product ergo Firewire must be superior to USB"). That such generalizations are incanted by 'respected mastering engineers' compounds one fallacious argument with another, argumentum ad verecundiam.

Well, you're wrong. And your insulting. Barry Diament(an actual mastering engineer)has stated that he has tried all available DACs USB, Firewire, optical. And that he finds no USB that sounds as good as a quality firewire DAC.

struts
struts's picture
Offline
Last seen: 2 years 10 months ago
Joined: Feb 1 2007 - 12:02pm


Quote:
Well, you're wrong. And your insulting. Barry Diament(an actual mastering engineer)has stated that he has tried all available DACs USB, Firewire, optical. And that he finds no USB that sounds as good as a quality firewire DAC.

Well I've been wrong before so it will no doubt happen again! I'm sorry that you find my post insulting but I can assure you it wasn't intended to be.

Your original question was essentially whether the industry was using an inferior technology to pander to the masses instead of pursuing ultimate sound quality. In response (and I apologize for the repetition) I was simply trying to point out:

  • In practice products based on USB are capable of excellent sound quality and measured performance (along with some data to back that assertion)
  • Given the above it makes perfect business sense to provide a USB interface in a high-end product since it increases the addressable market for that product considerably.

There are lots of design choices and implementation decisions in a DAC and they pretty much all influence the resulting sound. For instance something as seemingly insignificant as routing a signal track near to a transformer, power rail, or even IC can all induce jitter in the signal, quite regardless of the interface used.

Moreover seizing on one design choice to explain perceived sound quality differences is a very hazardous business. Product designers do this by reading spec sheets and then exhaustively testing the alternatives (both subjectively and with measurements) under controlled conditions, in other words changing one thing at a time and auditioning/measuring the effect. When one is changing a thousand things at once, which is effectively the case when comparing different products, it is simply impossible to ascribe any perceived differences to one particular parameter, even if one has seemingly compelling circumstantial evidence to point to. If you find this notion insulting then I recommend you ignore any future posts by me.

If I am wrong then I am afraid you haven't presented any argument that convinces me of the fact. Even if Mr Diament has auditioned "all available DACs USB, Firewire, optical" (wow, that's a whole heap of DACs!) and "he finds no USB that sounds as good as a quality firewire DAC" that still doesn't prove that the signal interface is the critical determinant of sound quality, let alone that Firewire is superior to USB. He may have a good ear, even a 'golden ear', but unless he has considerable theoretical and practical expertise with these circuits (and I mean pulling them apart and playing with them, not just using them) he is still only really able to observe, not link cause and effect. If he has some compelling rationale that explains this assertion I will be genuinely fascinated to hear it.

This discussion reminds me of similar recent shenanigans such as those concerning upsampling and the AK 4396 "miracle DAC", all equally spurious in my view. Presenting one particular design choice or component as "the answer" is the preoccupation of marketeers not engineers - not even 'actual mastering' ones.

ncdrawl
ncdrawl's picture
Offline
Last seen: 12 years 11 months ago
Joined: Oct 18 2008 - 9:18am

I agree, Struts..one persons observations mean nothing, no matter how "golden" the ears are.

Id be willing to bet any amount of money that Diament has NOT tried all available USB/Firewire DACs

Log in or register to post comments
-->
  • X