Re: [Jack-Devel] what could cause: "impossible sample width (1) discovered!"

PrevNext  Index
DateSun, 13 Mar 2011 11:09:53 -0400
From Paul Davis <[hidden] at linuxaudiosystems dot com>
ToFelix Pfeifer <[hidden] at googlemail dot com>
Cc[hidden] at lists dot jackaudio dot org
In-Reply-ToFelix Pfeifer [Jack-Devel] what could cause: "impossible sample width (1) discovered!"
Follow-UpFelix Pfeifer Re: [Jack-Devel] what could cause: "impossible sample width (1) discovered!"
On Sun, Mar 13, 2011 at 11:03 AM, Felix Pfeifer
<[hidden]> wrote:

> If i add inputs
> or use inputs only i get this error message:
> "impossible sample width (1) discovered!"

i don't recall.

> Does anybody have an idea what can cause
> this error? What exactly means width?

bits per sample

a quick perusal of the source code for the alsa backend should give
you some deep clues about what is going on.

>Is it
> maybe an uncorrect samplerate in my hardware?
> When starting jack, does jack provide the masterclock
> for playback and record? Is it necessary that my
> soundcard acts as a samplerate slave or is it
> possible to use it as the mastersamplingclock?

JACK doesn't provide a sample clock to anything (directly). it does
rely on the backend's wait() function, which in your case means ALSA
waiting for the device to indicate that its ready for input and/or
output. jack doesn't do anythng on a per-sample basis - it waits for
the next block ("period") of data/space to be ready, runs the process
callback cycle to process/generate the data, and then waits for the
next block.

if you're using the normal ALSA USB1 class driver, the error is likely
to be in your system, because JACK and its ALSA backend works OK on
just about all the implementations that we know of (OK, not great,
thanks to issues with Linux USB support, but that's another long
story).
PrevNext  Index

1300029018.16392_0.ltw:2,a <AANLkTimEmTmDqHECTQ_bGp42mRfnUOwzBgxZyEHLMzwU at mail dot gmail dot com>