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

RE: [linrad] Sound: Recording overflow

Hi folks,

I opened up my RX2500 yesterday, what a surprise!
Seeing many pot cores and parts, I shut my hat off for Leif!
My very first fire-up was successful.

At 18:17 02/11/07 +0100, Leif SM5BSZ wrote:
>Yes. Linrad uses svgalib and if you switch away the console Linrad will
>fail setting pixels on the screen, It will just wait until setpixel
>becomes successful (returns from svgalib). As a consequence the audio 
>input overflows because data is never taken care of. 

A console switching problem depends on the video card in use.
Matrox Millennium G400 causes the problem while nVIDIA GeForce2
and ATI Radeon AGP don't.

An audio overflows etc also are caused by the system's (kernel's)
other job, cron etc meant depends on CPU load.

>You have to escape from Linrad before switching console - or perhaps,
>if you are lucky it may be possible to tell the sound system to not
>bother you with the error messages. My system does not accept console
>swithing safely. Sometime Linrad crashes. With Linrad00-48 I have been
>able to recover by killing Linrad from another console but in earlier
>versions the big reset button was the only escape. I think the problem
>on my system is outside Linrad and the improvement in 00-48 is that
>the idle loop (in wcw.c) now gives the kernel some time by going to 
>sleep for short intervals now and then. 

A sound board selection is also important as my Turtle Beach Santa Cruz
causes many unknown problems and crashes many times.
Replacing by Delta 44 and/or Sound Blaster PCI128, all problems were
solved and I have never met any crashes since then.
My system is Red Hat 7.3 w/OSS 500MHz Pentium III 384MB/Ram.
By the way, RHL 8.0 is impressive though it eats CPU load a lot.
I ended up resuming back to 7.3 :-(

Share and Enjoy!

Kohjin - JR1EDE