[linrad] Re: 40 kHz of bandwidth with NO extra PCI-cards; adapting output sampling (CPU-load); optimum frequencies; Mini-ITX

[<leif@xxxxxxxxxx>]

> Regarding the CPU-load I would like to ask Leif/SM5BSZ, to comment
> whether a slight modification of the Lagrange-interpolation could
> work.
The problem is that someone (myself sometimes) wants the true bandwidth
of a 48kHz output sampling speed. Doing something that is equivalent
to a bandwidth limiting filter would destroy the generality of Linrad
as a tool box for general processing. Obviously it will be possible
to check the baseband bandwidth and select something that is fast if
the output sampling speed is much higher than required and also much
higher than 8 kHz (my ears do not like distortion below 4 kHz).

Much better would be to do it properly. I will think about it.
Presumably the interpolation can be used to step up the sampling
speed by a non-integer factor in the range 1.00 to 1.9999999 and
then a 4 bin fft can be back-transformed after being zero-padded
to the size giving the desired output rate. This should be pretty
efficient and accurate (I think)

In the meantime I think the OSS drivers can solve the problem.
Late OSS drivers contain a resampling routine. I have disabled
it on the input because it is not a very good idea to use it
(but it does allow 192kHz from my Delta44). On the output side
it should be a good idea to use it.

You may edit the file par_userint and specify 192kHz as the A/D
sampling rate for a Delta44. This works fine:-)

You may also edit par_userint for "min da speed" much lower than
the hardware speed. Then you can open the input at 48kHz and the
output at 8kHz on a card opened RDWR. Unfortunately this does not
work because the input data rate becomes the same as the data rate
set for the output - the OSS driver does not use its builtin
rate converter in this case.

How bad is it? Please set a modest bandwidth (100Hz or below) and
use 48kHz for input and output. What is the CPU load when
no signal is selected and what is it when a signal is clicked
upon? (Use 'Z' to clear the averaging for CPU load)

73

Leif / SM5BSZ



#############################################################
This message is sent to you because you are subscribed to
  the mailing list <linrad@xxxxxxxxxxxxxxxxxxxxx>.
To unsubscribe, E-mail to: <linrad-off@xxxxxxxxxxxxxxxxxxxxx>
To switch to the DIGEST mode, E-mail to <linrad-digest@xxxxxxxxxxxxxxxxxxxxx>
To switch to the INDEX mode, E-mail to <linrad-index@xxxxxxxxxxxxxxxxxxxxx>
Send administrative queries to  <linrad-request@xxxxxxxxxxxxxxxxxxxxx>


LINRADDARNIL