I'm trying out a new computer running Ubuntu Linux with a Tascam US-122
usb audio interface.  I tried driving an orchestra using a midi keyboard.
I optimistically set -b to 32 and -B to 64.  Csound quit with an error
and said that ALSA couldn't use the 64 value and to try 90 instead.

Remembering that the buffers values are meant to be powers of two,
I tried the -B 90 anyway.  When I ran csound my screen went immediately
black and the machine went down hard.  I tried a second time with the
same results.  Then I tried -B 1024 and everything worked.

The computer is running an AMD Phenom X4 processor.

While I'm at it, -B 1024 is the best I can do running a single
sinewave oscilator off of the keyboard.  My older computer could
do hundreds of  voices  with < 10ms latency without skipping a beat.  I'm using
the --sched flag and running as root.  Could this be the USB interface?
I can imagine that my old PCI card was more available to the processor.
The new board has no PCI slots, so I will have to wait for the EMU 0404
PCIe version that is almost available.

Thanks,

Toby


Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Regarding the US 122, I have one of these, but I can't use in Windows  
because it somehow crashes the machine, just freezes it. It used to  
work on Linux (fedora), but now it doesn't. When it did, I remember  
that I could not get very low latency with it (even if I had an RT  
kernel), say with Jack (or anything really). Whereas the onboard  
soundcard allowed me to go down to 64/32 . I guess the alsa driver for  
it was not particularly great.
And since it does not work on OSX 10.5, I can't use this soundcard  
anymore.

Victor

On 8 Dec 2009, at 02:27, Tobiah wrote:

> I'm trying out a new computer running Ubuntu Linux with a Tascam  
> US-122
> usb audio interface.  I tried driving an orchestra using a midi  
> keyboard.
> I optimistically set -b to 32 and -B to 64.  Csound quit with an error
> and said that ALSA couldn't use the 64 value and to try 90 instead.
>
> Remembering that the buffers values are meant to be powers of two,
> I tried the -B 90 anyway.  When I ran csound my screen went  
> immediately
> black and the machine went down hard.  I tried a second time with the
> same results.  Then I tried -B 1024 and everything worked.
>
> The computer is running an AMD Phenom X4 processor.
>
> While I'm at it, -B 1024 is the best I can do running a single
> sinewave oscilator off of the keyboard.  My older computer could
> do hundreds of  voices  with < 10ms latency without skipping a  
> beat.  I'm using
> the --sched flag and running as root.  Could this be the USB  
> interface?
> I can imagine that my old PCI card was more available to the  
> processor.
> The new board has no PCI slots, so I will have to wait for the EMU  
> 0404
> PCIe version that is almost available.
>
> Thanks,
>
> Toby
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
> "unsubscribe csound"



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Just an idea...

Try 48000 sample rate. Many USB cards only handle this sample rate so
SRC from 44100 has to be done in software by the drivers.

Cheers,
Andrés

On Tue, Dec 8, 2009 at 2:27 AM, Tobiah  wrote:
> I'm trying out a new computer running Ubuntu Linux with a Tascam US-122
> usb audio interface.  I tried driving an orchestra using a midi keyboard.
> I optimistically set -b to 32 and -B to 64.  Csound quit with an error
> and said that ALSA couldn't use the 64 value and to try 90 instead.
>
> Remembering that the buffers values are meant to be powers of two,
> I tried the -B 90 anyway.  When I ran csound my screen went immediately
> black and the machine went down hard.  I tried a second time with the
> same results.  Then I tried -B 1024 and everything worked.
>
> The computer is running an AMD Phenom X4 processor.
>
> While I'm at it, -B 1024 is the best I can do running a single
> sinewave oscilator off of the keyboard.  My older computer could
> do hundreds of  voices  with < 10ms latency without skipping a beat.  I'm using
> the --sched flag and running as root.  Could this be the USB interface?
> I can imagine that my old PCI card was more available to the processor.
> The new board has no PCI slots, so I will have to wait for the EMU 0404
> PCIe version that is almost available.
>
> Thanks,
>
> Toby
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"
>



-- 


Andrés


Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"