A first test and some food for audio thought

[cfmsp]

 

Paul,

 

I have no experience with computer audio on Windows - nor do I plan to acquire any - and so my observations almost never include disclaimers/caveats about not being applicable to Windows-based computer audio.

 

It's my understanding that high quality audio is available on PCs, it's just that it's not as easy to sort out as on OS X. iTunes on Windows is apparently not as good sounding as on OS X, and of course there are other reasons that all things are not equal between the two OSes.

 

XXHE would be my first (and hopefully only) stop if I HAD to setup a PC-based music server.

 

cheers,

clay

 

[cfmsp]

 

 

"Firewire is a superior protocol for data transfer for many technical reasons, but USB is ubiquitous. BetaMax/VHS type of think, I guess."

 

Agreed with your analogy. I haven't found a single advantage USB has over Firewire for computer audio, at least for end users.

 

Even the claimed ubiquity is not there with respect to computer audio, as there are no more Async USB DACs than Firewire.

 

clay

 

[The Computer Audiophile]

"... I haven't found a single advantage USB has over Firewire for computer audio, at least for end users."

 

A major benefit of USB is it can use the operating system's built-in device drivers. This should not be underestimated. There are always issues with drivers and updating them on a regular basis when something changes. For example there is a current thread about how to uninstall the hiFace drivers. I know when a new OS version comes out TC Electronics has to fix its FireWire chipset drivers. This leads to impatient users and frustrated users who elect to go the beta route.

 

I do like both interfaces :~)

 

[cfmsp]

 

"A major benefit of USB is it can use the operating system's built-in device drivers."

 

Chris, I view this as being primarily a benefit for the manufacturers, but then again, I'm not fussed about having to install drivers once in a blue moon.

 

 

So, I'll politely disagree on considering it major. Even if I did consider it so, it pales in comparison to the advantages of Firewire, IMO.

 

Competent manufacturers provide updated drivers with the release of new OS X versions (at least the leading OS X-only Firewire/audio product vendors do).

 

I"ve not experienced any issues of this sort, and I routinely (some might say obsessively) install new releases immediately.

 

 

 

It's funny (to me) how often this (claimed benefit of native USB drivers) is referred to as an advantage, yet it's still an achilles heel from where I sit, e.g., where is the 192 kHz support for Async USB?!

 

 

"For example there is a current thread about how to uninstall the hiFace drivers"

 

HiFace is USB, not firewire, non?

 

I rest my case! ;0

 

respectfully,

clay

 

 

 

 

[Paul R]

To be fair Clay, under Windows, USB is a lot less trouble than Firewire, since Microsoft keeps the USB drivers up to date but ignores the FW drivers.

 

FireWire is still the superior protocol though.

 

If you ever want to have an amusing couple of hours, get around some people engineering a third party product for Windows and listen to them gripe about supporting Firewire, SCSI, Fibre protocols, and so forth. You would think to listen to them that USB is God's gift to the chosen ones.

 

When those guys have to do a Mac product, they usually get amazed at how much cleaner FireWire is to implement, and especially how things "just work!" under Mac OS. Most of the really good ones tend to go buy Macs after their first hardware project! (IMNSHO of course)

 

-Paul

 

 

[audiozorro]

it wasn't that long ago when many questioned whether the Firewire days were numbered. Apple had dropped the FW interface from their Macbook laptops and many loyal Mac customers complained loudly with what they perceived to possibly be a continued decline of firewire. Isn't it bad enough that Apple dropped FW from iPods, and didn't include it in the iPhones and iPads.

 

http://www.zdnet.com/blog/apple/apple-drops-firewire-from-video-ipod/13

 

 

With USB 3 coming to market who needs firewire and all the associate proprietary baggage associated with it. Yes I know that Firewire is the superior technology (BetaMax) to the competing USB (VHS) common standard. But technology marches on and we have graduated to DVD and will soon largely adopt Blu-ray. I don't mean to start a war and I do own Apple products including several iPods, an iPhone and MacBook Pro. But I think this supposed technological superiority is overblown and not completely accurate.

 

[Audio_ELF]

Audiozorro commented... "With USB 3 coming to market who needs firewire and all the associate proprietary baggage associated with it. Yes I know that Firewire is the superior technology (BetaMax) to the competing USB (VHS) common standard. But technology marches on and we have graduated to DVD and will soon largely adopt Blu-ray. I don't mean to start a war and I do own Apple products including several iPods, an iPhone and MacBook Pro. But I think this supposed technological superiority is overblown and not completely accurate."

 

But your analogy with video is missing a step here ... We didn't move from BetaMax and VHS to DVD and onto BluRay, but we moved from BetaMax to BetaCam (FireWire 400 to FireWire 800) and from VHX to S-VHS to Digital-VHS (USB to USB2 to USB3).

 

In the same way, what may finally bring about the change to DVD and BluRay, uniting the formats so to speak, is not USB3 ... but Intel LightPeak, the next generation interface which combines elements of many different interfaces not just USB and FireWire.

 

The actual superiority about FireWire is nothing to do with speed - but more about philosophy and the way FireWire is actually a peer to peer system rather than Host lead - which unfortunately makes FireWire devices more expensive though better. We are talking about audio devices and that there (may) be more ubiquity of USB on devices - but this is consumer devices - the vast market is for professional multi-channel devices and here FireWire does rule supreme.

 

Eloise

 

PS. USB3 doesn't offer anything new for Audio use, and there is also FireWire 3200 to consider.

 

[Paul R]

Firewire devices also allow for allocated bandwidth, meaning they can sustain data transfer despite CPU load, while USB transfers always vary in bandwidth from moment to moment, depending upon CPU load.

 

I wonder if indeed that is why a lot of the sonic differences seem to disappear when using Firewire vice USB.

 

-Paul

 

 

[cfmsp]

 

 

"I wonder if indeed that is why a lot of the sonic differences seem to disappear when using Firewire vice USB. I wonder if indeed that is why a lot of the sonic differences seem to disappear when using Firewire vice USB."

 

Paul, we're wondering the same thing.

 

 

clay

 

 

[Paul R]

In my opinion only, one need be very careful when reading information from anything ZDNET publishes - twice as careful if that information regards any non-Windows system, and thrice as careful when they mention Macs or Apple.

 

Those people seem to have a serious grudge against Apple, and express it by spreading disinformation.

 

Just my opinion, YMMV, not the views of the management, etc.

 

-Paul

 

 

[cfmsp]

Eloise,

 

Agreed.

 

USB3 offers nothing (that I can see) for the computer audio user, and will never be cause of the "death" of Firewire.

 

Indeed, usage of Firewire DACs for computer audiophiles has increased since the 'death of Firewire' comments peaked some time ago.

 

When I arrived here at CA, no one was talking about Firewire interfaces.

 

Lightpeak will be the eventual death of Firewire. It will also kill off USB, IMO.

 

As for Firewire 3200, it is backwards compatible with firewire 800 ports, so will work with existing machines without requiring that all devices be upgraded to F3200.

 

I'll repeat - IMHO, NOTHING about USB is superior over Firewire, either now OR in their planned future(s), except that there are more USB devices out there.

 

 

Indeed, IMO, it's quite possible that USB's reliance on the core processor capabilities of the computer (rather than Firewire's "self contained" processing) contributes to the differences in sound of bit-perfect players when transmitting via USB (even Async USB). ;0

 

[oohhh, this ought to get the USB advocates going!]

 

 

 

http://en.wikipedia.org/wiki/IEEE_1394_interface

 

"Although current high-speed USB 2.0 (introduced in 2001) is quoted as running at a higher signaling rate (480Mbps) than legacy FireWire 400 (400 Mbps, available since 1995), data transfers over S400 FireWire interfaces generally outperform similar transfers over USB 2.0 interfaces. Few if any USB 2.0 device implementations are able to saturate the entire 480 Mbps, but this can be achieved with multiple devices on the same bus. Typical USB PC hosts rarely can sustain transfers exceeding 280 Mbit/s, with 240 Mbit/s being more typical. This is likely due to USB's reliance on the host processor to manage low-level USB protocol, whereas FireWire delegates the same tasks to the interface hardware (requiring less or no CPU usage). For example, the FireWire host interface supports memory-mapped devices, allowing high-level protocols to run without loading the host CPU with interrupts and buffer-copy operations.[5]

 

Besides throughput, other differences are that FireWire uses simpler bus networking, provides more power over the chain and more reliable data transfer, and is less taxing of a CPU.[28]"

 

 

cheers,

clay

 

 

 

[cfmsp]

 

The ZDNet article referenced is from October 2005, only 5 months short of being 5 year old information.

 

so much for the death of Firewire.

 

Don't get me started.

 

oops, too late, someone already did.

 

clay

 

 

 

[Audio_ELF]

clay said... "I'll repeat - IMHO, NOTHING about USB is superior over Firewire"

 

It could be argued that the ability to produce an asynchronous transfer DAC (such as Wavelength's Streamlength Code) under the standard Class 1.0 drivers is a superiority over FireWire, but I can't think of anything else. And as we've seen there's lots of hassles with scaling this to 24/192 (bugs in Mac OS X drivers, no support in Windows, etc).

 

Eloise

 

(Just giving some fair, unbalanced reporting)

 

[cfmsp]

 

 

"It could be argued that the ability to produce an asynchronous transfer DAC (such as Wavelength's Streamlength Code) under the standard Class 1.0 drivers is a superiority over FireWire, but I can't think of anything else. And as we've seen there's lots of hassles with scaling this to 24/192 (bugs in Mac OS X drivers, no support in Windows, etc)."

 

Understood.

 

Chris and I debated this point upstream.

 

I'll not view this as a superiority until it's fruition is achieved, i.e. when the special code required doesn't also provide all the same (sort of) issues that Firewire drivers suffer, e.g. code updates, bugs, etc.

 

But I see your point.

 

clay

 

 

 

[cfmsp]

 

 

"(Just giving some fair, unbalanced reporting)"

 

Eloise,

 

I believe this may be copyrighted by Fox News here in the US.

 

 

clay

 

 

 

 

[Audio_ELF]

Fox News ... isn't that part (or fully) owned by News Corp. owned by the completely balanced and fair Rupert Murdock?? :-)

 

Hosted by Kent Brockman??

 

Eloise

 

[idiot_savant]

I've seen a lot of statements on here that seem to be commonly taken truths with respect to audio performance...

 

(a)Firewire400 > USB2.0 in terms of speed and/or latency

(b)USB is designed for printers, mice, etc whilst firewire is not

©Many professional devices use firewire, so it must be the best

(d)You need USB 3.0 to do 24/192

(e)the host->peripheral architecture is inferior to peer->peer

(f)the physical interface is better in firewire ( isolation, clock, power etc) than USB.

(g)Driver issues

(h)Availability of interface

 

Any comments on the above statements? I'm aiming for a bit of an overall view here, rather than Firewire>USB>Firewire>USB>.........

 

your friendly neighbourhood idiot

 

 

 

 

 

[Audio_ELF]

The truth about cats and dogs ... well FireWire and USB ... as I see and understand it.

 

(a)Firewire400 > USB2.0 in terms of speed and/or latency

Theoretically, USB2.0 is quicker than FireWire 400 (for data transfers, etc) however the processor overheads and inbuilt inefficiency of USB2.0 means that real world transfer rates are quicker on FireWire.

 

(b)USB is designed for printers, mice, etc whilst firewire is not

USB was designed as a replacement for Serial, Parallel and PS/2 ports for keyboards, mice, printers, etc. This is fact. It was later expanded (USB1.1) to allow for high speed devices such as disk drives, etc.

 

©Many professional devices use firewire, so it must be the best

Not necessarily true. professionals use devices because they are reliable a lot more than because they are best. This is true of SQ as much as any other - ask a sports photographer which is BEST out of a EOS1D Mk II or a Nikon D3 and the Canon had the potential for better quality photographs, but it was less reliable than it's Nikon counterpart and so lost sales.

 

(d)You need USB 3.0 to do 24/192

No ... as we've heard before M2Tech do 24/192 with the HiFace using proprietory drivers and Gordon is doing 24/192 over USB2.0 using Class 2.0 Audio device.

 

(e)the host->peripheral architecture is inferior to peer->peer

I'm not sure in Audio situations it matters.

 

(f)the physical interface is better in firewire ( isolation, clock, power etc) than USB.

Seams that given the right motivation both can be equal. Though there does seam to be something in USB which means isolating the 5v power from the computer is impossible.

 

(g)Driver issues

This swings both ways. The idea of a Universal driver is good, until it wont do what you need it to or on the OS you need it to work on.

 

(h)Availability of interface

The most common isn't necessarily the best - see Beta vs VHS above.

 

Not sure if this is 100% accurate or backable in truth ... but it's how I see it which (I think) was what I_S was asking.

 

Eloise

 

[idiot_savant]

thanks for your input - this is precisely the kind of debate I'm trying to encourage,

 

your friendly neighbourhood idiot

 

[cfmsp]

How are things, I_S, good to read your words here, as always!

 

 

(a)Firewire400 > USB2.0 in terms of speed and/or latency

 

As Eloise says - and Wikipedia confirms - USB can rarely deliver on it's spec, and therefore Firewire delivers more bandwidth in real-world applications, despite the better spec of USB (relative to comparable generations of USB vs. Firewire).

 

 

(b)USB is designed for printers, mice, etc whilst firewire is not

 

Firewire was designed as replacement for SCSI, and in particular to meet the needs of audio and video transmission. USB was not. The importance of this distinction is debatable.

 

 

©Many professional devices use firewire, so it must be the best

 

Don't know that this would make it best, but true professional audio recording devices (as opposed to cheaper devices targeting the so-called prosumer recording market - aka hobbyists) are much more likely to have Firewire than USB interfaces.

 

 

(d)You need USB 3.0 to do 24/192

 

Don't believe this is true, but it is true that 24/192 requires special handling/capabilities for USB implementation. Indeed, it seems rather difficult to provide at all at present. Firewire protocol does not require anything special to support 24/192, to the best of my knowledge.

 

(e)the host->peripheral architecture is inferior to peer->peer

 

IMO, as I stated above, this is potentially a true statement, although evidence is circumstantial at best. I will say that, IMO, peer to peer is IN NO WAY inferior to the host-peripheral architecture.

 

(f)the physical interface is better in firewire ( isolation, clock, power etc) than USB.

 

Some innovative vendors (starting with Gordon Rankin) have worked hard to mostly eliminate the inherent advantage Firewire has (and still maintains, IMO) over typical (read adaptive) USB installations.

 

(g)Driver issues

 

Manufacturers are required to provide drivers for their products to operate in the manner in which computer audiophiles use them as DAC interfaces, i.e. asynchronously (or otherwise).

 

OTOH, Firewire bus interfaces for many Firewire devices do not require driver (installation), e.g. Firewire hard drives, video cameras, etc.

 

(h)Availability of interface

 

For my purposes as an individual (who can use & afford only so many devices), the availability of Firewire in high quality devices is more than sufficient to meet my needs, e.g. Metric Halo, Weiss and Prism (Orpheus) all make incredibly top notch Firewire interfaces, at various price ranges, albeit NOT in the lower price ranges.

 

In my experience, the ubiquity of something (i.e. products) has an INVERSE relationship to quality & performance. YMMV.

 

clay

 

 

[Paul R]

(a)Firewire400 > USB2.0 in terms of speed and/or latency

Absolutely, in speed, latency, reliability, and several more technical issues, such as allocated (guaranteed) bandwidth for devices.

 

(b)USB is designed for printers, mice, etc whilst firewire is not

It was originally designed for that limited intent, with the idea of a low cost highly reliable interface that would be universal- unlike parallel and serial ports. It is indeed, pretty much just a serial port with standard pins and signals. It was expanded to be much more capable over time, but the primary issue with USB is that its transfer rate (bandwidth) is highly dependent upon the CPU of the host.

 

(e)the host->peripheral architecture is inferior to peer->peer

I am not clear on what you mean by this. And if I guess correctly what you mean, the answer is clearly- "it depends." For audio, which is bandwidth and timing sensitive, Firewire is clearly superior. For driving DASD devices? Firewire. For driving a printer or mouse? USB is clearly superior in these respects.

 

 

(g)Driver issues

Depends again. On a PC, there are few driver issues with USB and lots of them with Firewire. That reverses somewhat on a Mac, but at a higher baseline. Support for USB is better on a Mac than on a PC, and support for Firewire on a Mac is better than support for USB. Both are still good - on macs. On Windows, the support is spotty, unreliable, and often frustrating. For both devices.

 

(h)Availability of interface

I do not see this an issue. The hardware is available on both Windows, Mac, or Linux PC's for both interfaces, and third party hardware cards for Firewire or USB are cheap and easily obtained.

 

-Paul

 

 

 

[bdiament]

Hi Paul,

 

"...For driving a printer or mouse? USB is clearly superior in these respects..."

 

While I might tend to agree, I've never tried (or seen) a Firewire printer or mouse. ;-}

 

Best regards,

Barry

www.soundkeeperrecordings.com

www.barrydiamentaudio.com

 

 

[Audio_ELF]

The way some people talk, you'd think that USB had been forced on Apple. In fact Apple was one of the pioneers of using USB to replace their equivalent of Serial and PS/2 ports. The original (G3) iMac is notable not just for not having a floppy, but also for having USB ports for it's major connections - those these were supplemented with FireWire for connection of high speed devices.

 

Eloise

 

PS Barry commented... "I've never tried (or seen) a Firewire printer or mouse." I've never seen a FireWire mouse, but I do recall some large format Canon printers having FireWire, and before that SCSI connections. Early "professional" level digital cameras (original Canon EOS1D for example) also often had FireWire as did some flatbed scanners.

 

[Paul R]

The larger format color printers, like HP DesignJets have Firewire connectivity built into them, and the smaller Epson and Canon photo printers also have Firewire connections. Mostly to connect to cameras I think.

 

I had to google for a firewire mouse, as I never heard of one. It would seem that Google had not either... :)

 

 

 

 

[idiot_savant]

for their input so far - I'm going to give it a couple of days to allow more input before I give mine( as I seem to have an unrivalled effect of killing threads ),

 

if there's any sentence liable to set the idiot off, it's the "X, as everyone knows is better than Y"... maybe X really is better than Y, but why?

 

your friendly neighbourhood idiot