RE: glitches, another story *g*

["Ralph Metzler <" <rjkm@netcologne.de>]

Plamen Ganev writes:
 > However, I noted some relation between CPU speed and glitches. Actually, it
 > seems that higher CPU speeds create more glitches than slow CPU + well-tuned
 > hd. My test case is a TX chipset with a K6-2 233 CPU. When the CPU is
 > running at full speed, I could see glitches even without recording (*any*
 > disk activity was enough). By lowering the cpu speed (making it work at a
 > mere 133 MHz) I could record without any glitches.
 > 
 > All I can think of is either the card is getting too much noise from the
 > cpu+ide cable, or, there're race conditions with the bus, so that the card
 > misses frames (maybe waiting to the bus to free to send the frame in the
 > main memory?).

Those glitches are heat related.
I had the same effects last February. Like in your case they mostly
occured during high disk and CPU activity, even when they were
unrelated to recording.
They also were much more frequent when the CI interface was connected.
I have a fan blowing directly on the card since then and never had that problem again. 

I always blamed my specific setup for this: dual PII 350, hard disks
close to the card (heat + interference), many cables for on-board SCSI
and IDE surrounding the card, etc. 
With other PCs (single CPU and better ventilation) the problem did not occur.
So, this problem might come up if the air flow over the DVB board is
not sufficient.


Ralph