27 May
2003
27 May
'03
6:07 p.m.
Mark Knecht wrote:
that should be divided by three to get bytes as there are 8 bits in
every byte the quattro uses 24 with 3 periods (not packed in four like
other cards).
24 * 3 bytes * 48000 = 3072000 bytes
=> 3072 Kb
=> 3.072 Mb/S
Another thing
about the quattro is that even the windows drivers will
not allow you to record and playback on all 4 channels at one time.
You can use
4 in, 2 out 48khz 24 bit
2 in, 4 out
2 in, 2 out 94 khz 24 bit
I think I am running up against this with my latest attempts. It may be
that the quattro cannot do all 8 channels because of a hardware switch
maybe for power consumption or something. They do run awfully hot and I
can imagine that it would be possible to melt something if their is too
much power running through them.
USB 1.0/1.1 is built on a 12Mb/Sec physical layer.
96000 * 24-bits * 4-channels = 9.216Mb/S if I did the calculations
correctly.
I think that's probably more than USB can really handle. When I worked for a
company that did USB hardware, we found in many PCs (at that time -
1996-1999) that most USB implementations would really do much more than
about 2Mb/Sec reliably. The newer implementations are much better, but these
numbers bring you up to 75% bus efficiency. Even 1394 doesn't get much
beyond that and it uses much larger packet sizes to get some of that.
That comes as no surprise to me that the usb protocol cannot handle
effectively it's full bandwidth.
But it should be easily able to handle this:
44100 * 2 bytes * 8 channels (4 in, 4 out) = 705600 bytes
=> 705.6 Kb/S
--
Patrick Shirkey - Boost Hardware Ltd.
Http://www.boosthardware.com
Http://www.djcj.org - The Linux Audio Users guide
========================================
Being on stage with the band in front of crowds shouting, "Get off! No!
We want normal music!", I think that was more like acting than anything
I've ever done.
Goldie, 8 Nov, 2002
The Scotsman