[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Q!Re: [linrad] Linrd installation...a few comments
- Subject: Q!Re: [linrad] Linrd installation...a few comments
- From: "Rein A. Smit" <rein0zn@xxxxxxxxxxxxxxxxx
- Date: Tue, 20 May 2003 06:44:08 -0700
For other readers:
Arie and I are talking about the use of a program, Krecord in
my case version 1.13, to check the workings of my, by now famous,
not working, AC97 sound section.
Krecord can be easily found and downloaded. It needs a complete
working compiler to install. It was here where differences showed
up between Arie's distribution and Mandrake 9.1. I did not have the
Kernel "source code" installed. I now got it what Chris was talking
about that if you want to recompile the Kernel ( as in windows
re-compiling, one needs the Kernel source code. As if you wanted
to recompile widows2000 or NT, you might need a lot of stuff but the
windows 2000 source code for sure as a minimum!
It is here where the wheels are heating the rubber. Biil gates does
not let you compile his windowsXX, The linux world can compile
the kernel it is in the public domain and it you feel changing,
it can do it as long as you go through the process and comply with
the procedures and rules.
Assuming that the kernel and the compiler are closely linked in
Linux I believe that the krecord compile was using .h(header files)
that are needed when you compile the kernel. I have compiled without
a doubt poorly written C or Java programs and never had any idea
that I should or was using header files that were part of the windows
Once it was clear that in compiling Krecord a certain file
was missing, it followed that the kernel code had to be absent or
I, however, never intended to recompile the kernel! What was needed
was to compile Krecord in order to install it. Lesson as Leif's
tells us make sure everything should be installed if you do not
quite know what you need or not!
BTW it is also possible that Arie knew this and loaded the kernel
source when he installed his distribution in the knowledge
or with the experience that this is the smart thing to do.
See Leif's statement about loading everything.
This Kernel source was just sitting there ready to be clinked on.
I did just did know this. My fault! Now when I install, I do also
this kernel-souce as well as nasm. It is the same procedure and
does not get installed either in Linrad if you use the standard
install. I figure that there is a fair amount assembly code
in Linrad and that is why you need nasm as per instructions.
It also seems that the compilation of Linrad does not require
anything Of the kernel source code As I could compile
linrad jut fine with missing source code.
I selected the "development" section in my innocence.
I thought that that would also take care of nasm, libs,
includes, etc. but it appears not to do this. I really think
that is where the experienced differences are, between lets say
Mandrake and RH. Arie and Kohjin have files that are needed
if you have to fool with things like this sound card, stored
in other directories or in sub directories then where I find them
Do not really understand why that is and if these things are
supposed to be in places, who's distribution is at what?
It appears that with Krecord the sound section behaves different
from what it should do in Linrad.
Arie Dogterom wrote:
> You should mute the input channel and activate capture. Then you
> will see at krecord level indicator that there is a recording
> signal while you do not hear it
It might really sound as if I can't read but all I got is 2 mutes/
un-mutes, the MIC and Line plus "capture" and PCB
I read from this that the spectrum display and the storing of data
in a .wave file, is going on with a muted input channel in your case?
Quite frankly I did not pay much attention to the "level indicator"
other than that it jumps between something like "-70 dB"
to at least -30 db or so when I unmet the MIC channel.
Have been looking more to the freq. display. I believe it also
jumps when I turn/off the PCM port.
I DO hear it and if I mute the mic, everything stops, the audio AND
the SA display. Again do you have anything other then the "MIC"
mute / slider that effects the input? I do not, all I have is the
> Then running krecord to play it back( goes via the pcm channel) you
> should hear the recording. Is that ok, stop krecord and run linrad.
It that was true in my case, I should hear both the pre processed audio
and the Linrad processed audio, being shifted by the linrad
processing time. I hear only the raw input and if I turn off the mic
the SA display in Linrad AND the raw audio stops!
So It appears there are 2 functions not working
1 -- decoupling my one input channel ( The mic/line ) from
whatever the mic-audio brings to the speaker output.
I am sure the audio runs through the sound card block function.
2 -- the processed output from Linrad does not connect to the
the the back end input of the sound card. I think that
is via the PCM channel.
Again, if that path was there I should here linrad processed audio.
in the presence of unprocessed audio from ( 1 )
> sure you then only will hear linrad processed sounds
> ----- Original Message -----
> From: Rein A. Smit <rein0zn@xxxxxxxxxxxxx>
> To: <linrad@xxxxxxxxxxxxxxxxxxxxxx>
> Sent: Tuesday, May 20, 2003 6:01 AM
> Subject: Re: [linrad] Linrd installation...a few comments
> > Hi John
> > What is it, you mean by cheaper onboard sound cards?
> > PCI stuff in particular. There have been issues with sound
> > cards all the way back to FFTDSP by Af9Y and plenty of DSP
> > programs being sensitive to the type of sound cards being
> > Some of them wanted 8 bit cards. I type this on a WIN95
> > machine with a 8 bit card installed. FFTDSP, DSPBlaster
> > run fine under DOS.
> > I suppose you are talking about the price of the mother board
> > to a certain extent.
> > Though it would be hard to detect any correlation in the final
> > price of the mother board and the sound section.
> > I have quite a few questions about this sound thing here.
> > Just for the fun, I will research the expensive boards and what they
> > use for onboard audio systems. ( ASUS et al )
> > I have questioned myself whether there is actually a problem with
> > that sound section. However it does everything in WIN98SE and as
> > a matter of fact in Linux as well .
> > It is also working with this krecord program using .wav files
> > In that mode however the mic/line inputs are not being muted.
> > it is not a matter of cross talk, the mic signal passes through
> > un-attenuated This krecord is neat, you can listen to signals
> > record them in memory and play them back, in my case I have to
> > turn down the audio from the receiver.
> > PA0SE has the same audio section running on a laptop and running
> > Linrad. The real strange point is that the alsamixer that he is
> > using from alsa_utils-0.9.3 has the ability to select channels and
> > allows for path switching/selection and I have nothing of that
> > in my installation, as if the mixer adjusts itself to who knows
> > what! The usual volume controls, MIC,LINE, Capture, PCM and
> > MASTER work OK.
> > I have been trying to find useful info and there are hundreds
> > of messages of people asking/reporting about that section but
> > it are always problems about no sound at all, distorted sound,
> > low volume etc. I have nothing of that as long as I do not try
> > to get Linrads output. Also Linrad reports the card properly
> > detected and working when I go to the set-up.
> > Do you have Linrad running? If so perhaps you could run through
> > http://www.nitehawk.com/linrad_dat/
> > Nothing BEATS information from users with a working setup.
> > 73 Rein, PA0ZN
> > "John Harrison, NI1B" wrote:
> > >
> > > Hi All,
> > >
> > > A couple of comments about getting a useful Linrad setup.
> > >
> > > MANY of the recent on-board and cheaper internal PCI cards have GROSS
> > > problems.
> > >
> > > In two cases (one on board and one add on card) the card had NO
> > > anti-aliasing filter (at all). The card would pick up high frequency
> > > signals and noise and fold it over into the lower freqs with NO
> > > attenuation !!
> > >
> > > In the other PCI case, the card was apparently very slow in taking
> > > samples so virtually NO TIME was left for Linrad :(
> > >
> > > ALL of the ISA cards I have tried have had neither of these problems,
> > > but try to find a recent motherboard with an ISA slot!!
> > >
> > > Anyways just another two technical details to check for when you are
> > > selecting a sound card.
> > >
> > > warm regards,
> > > john, ni1b