Throughput Speeds for Firewwire and USB2

[Guest]

Archived from groups: rec.audio.pro (More info?)

A while back there was some good discussion here about the speed of
these interfaces relative to the number of audio channels at a given
sample rate that could go through the pipleline. I know that the
theoretical limits aren't even close to being achieved in practice,
but I'm looking for both what the specification allows and what kind
of practical implementations we have now.

I haven't been able to find the right key words to find that thread in
the Google archive so if someone better at this than I can point me to
it (or if you want to start discussing it all over again - I'm sure
things have changed on both ends in the past few months) I'd
appreciate it.

This is basic research for an article on interfacing, so you might
read it back elsewhere some time in the future.



--
I'm really Mike Rivers (mrivers@d-and-d.com)
However, until the spam goes away or Hell freezes over,
lots of IP addresses are blocked from this system. If
you e-mail me and it bounces, use your secret decoder ring
and reach me here: double-m-eleven-double-zero at yahoo

 

[Guest]

Archived from groups: rec.audio.pro (More info?)

> A while back there was some good discussion here about the speed of
> these interfaces relative to the number of audio channels at a given
> sample rate that could go through the pipleline. I know that the
> theoretical limits aren't even close to being achieved in practice,
> but I'm looking for both what the specification allows and what kind
> of practical implementations we have now.
>
> I haven't been able to find the right key words to find that thread in
> the Google archive so if someone better at this than I can point me to
> it (or if you want to start discussing it all over again - I'm sure
> things have changed on both ends in the past few months) I'd
> appreciate it.
>
> This is basic research for an article on interfacing, so you might
> read it back elsewhere some time in the future.

Both formats have overhead which eats bandwidth, but the significant
difference is that USB has processor latency overhead. Firewire is for the
mostpart autonomous, in fact there is hardware support for transfers between
iPods without a host computer, but naturally it has been disabled in the
firmware for the sake of iTunes licensing. USB on the otherhand is slave,
governed entirely by the CPU, and bandwidth is lost while waiting for the
CPU to confirm packet transfers and decide what data is to go where.

The same is true of IDE/ATA compared to SCSI, which is why SCSI can achieve
much faster access speeds and is prefered for high-traffic servers. When
you have an ATA drive on a USB interface the problem is compounded. Audio
editing requires very frequent access to multiple files on a timely basis,
so USB is not recommended, especially not for external hard drives.
Thankfully the price difference between USB2 and Firewire enclosures is very
reasonable compared to ATA vs SCSI.

I don't have anything more specific, hopefully that will help direct your
inquiries.

 

[Guest]

Archived from groups: rec.audio.pro (More info?)

In article <PSLFd.16810$0y4.3002@read1.cgocable.net> nobody@home.com writes:

> Both formats have overhead which eats bandwidth, but the significant
> difference is that USB has processor latency overhead. Firewire is for the
> mostpart autonomous

I knew that, but thanks for reminding me. It's a good point. I've
noticed the CPU-o-meter spiking whenever I have a USB device
connected, even if it's something that isn't full time busy.


--
I'm really Mike Rivers (mrivers@d-and-d.com)
However, until the spam goes away or Hell freezes over,
lots of IP addresses are blocked from this system. If
you e-mail me and it bounces, use your secret decoder ring
and reach me here: double-m-eleven-double-zero at yahoo

 

[Guest]

Archived from groups: rec.audio.pro (More info?)

On Fri, 14 Jan 2005 04:04:19 -0500, "Sugarite" <nobody@home.com>
wrote:

> Audio
>editing requires very frequent access to multiple files on a timely basis,
>so USB is not recommended, especially not for external hard drives.
>Thankfully the price difference between USB2 and Firewire enclosures is very
>reasonable compared to ATA vs SCSI.

I assumed Mike was asking about audio interfaces. You, that he was
asking about external drives. Was it either, or both?


CubaseFAQ www.laurencepayne.co.uk/CubaseFAQ.htm
"Possibly the world's least impressive web site": George Perfect

 

[Guest]

Archived from groups: rec.audio.pro (More info?)

In article <9hpku05ia9o8dglsqou40ug0rs5pi89hvs@4ax.com> l@laurenceDELETEpayne.freeserve.co.uk writes:

> I assumed Mike was asking about audio interfaces. You, that he was
> asking about external drives. Was it either, or both?

Yes, and I'm surprised that nobody has piped up with the (possibly
valid) data or even pointed me to a Google URL for the fairly recent
discussion of the topic that I couldn't find myself. Or maybe they have
and there was just less of a discussion than I remembered. I've gathered
some useful information though.

--
I'm really Mike Rivers (mrivers@d-and-d.com)
However, until the spam goes away or Hell freezes over,
lots of IP addresses are blocked from this system. If
you e-mail me and it bounces, use your secret decoder ring
and reach me here: double-m-eleven-double-zero at yahoo