[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: [linrad] FFT performance

Hi Leif,

I did try RDONLY and RDRW configurations to no avail. I'm including the 
soundboard_init.log at the end of this message. It seems fine.

I just finished performing a test with a SoundBlaster Live card and the
problem is indeed in the sound driver, as I've expected. I have also
encountered a lot of problems with sound drivers. It is actually quite
common to find drivers that do not comply with the OSS programming
standards and there are drivers that don't even work with select().

I will perform a new test with the alsa drivers and see what happens. It
is a pity since the C-Media chip has a decent SNR and is very simple to
control. My SB-Live card is a nightmare. The mixer has twenty two thousand
controls. I have no idea of what they do since I am yet to find a diagram
for the mixer connections! But the driver API works flawlesly!

Lief, on a side subject, have you considered spliting the user interface
from the processing code for linrad? Another area of problems for many
folks is the svgalib configuration. In all my machines I had to tweak the
svga config to get it working. Although X11 consumes a good ammount of
resources, I believe it would be easier to deal than svgalib.


-- Edson

On Fri, 15 Nov 2002, Leif Asbrink wrote:

> Hi Edson,
> > I can't get to the screen where the values are displayed. This is what is 
> > happening with details:
> > 
> > $ ./linrad
> > 
> > ADC sampling is set to 8000, 2 channels (I/Q receiver). Parameters are 
> > saved (W pressed).
> > 
> > At the main screen I press "F" for the hardware test mode. The message
> > "You are prompted to the parameter selection screens for the following
> > reason:  par_test file missing". Then at the "Set parameters for mode:
> > HARDWARE TEST MODE RX CHannels: 1" screen, no matter what values I set,
> > when I press enter, linrad always tells that "Speed far too slow. Try less
> > demanding FFT!". That is it. The only way to get out is to press ESC, thus
> > quiting the application.
> Most probably something is wrong with your device driver. It was already
> tested in the setup process and I do not know why it does not work here.
> There are many problems with different device drivers not really 
> following the specifications. Some may give incorrect info about the
> buffer pointers for example.
> Did you get any choice at all in the A/D setup? Try a different device
> if there is one, otherwise try RDWR or RDONLY and possibly another input 
> speed.
> Problems like this are common with ALSA but OSS usually works well.
> With a sampling speed of 8000 and fft1 bandwidth 100Hz or above you
> should not see the "Speed far too slow" message even on a 486 computer
> so the speed is surely evaluated incorrectly. 
> Check the soundcard_init.log file. Perhaps it contains some 
> useful information.
> You may download a really old Linrad version and check if the problem
> is present. I have done some changes in order to work around some of
> the incompatibilities. 
> 73
> Leif  /  SM5BSZ 

/dev/dsp opened as RDONLY 
16bit format supported
Max no of channels = 2
Max input speed 48000 Hz
/dev/dsp opened as WRONLY 
16bit format supported
Max no of channels = 2
Max input speed 48000 Hz
/dev/dsp opened as RDWR 
16bit format supported
Max no of channels = 2
Max input speed 48000 Hz

 /dev/dsp selected by user for input.
User selected RDONLY
Min sampling speed = 8000 Hz
8000 Hz selected by user
8000 Hz reported by device
Input format: 16 bit
Read test OK
No output device found while input is open.
/dev/dsp opened as WRONLY with no input open.
Max output speed 48000
User selected /dev/dsp for output.
Normal end