<HTML><BODY style="word-wrap: break-word; -khtml-nbsp-mode: space; -khtml-line-break: after-white-space; "><BR><DIV><DIV>On Mar 29, 2006, at 9:45 AM, Warren wrote:</DIV><BR class="Apple-interchange-newline"><BLOCKQUOTE type="cite"> Craig Partin wrote: <BLOCKQUOTE cite="mid71203aac0603290825x2a518050lc989939abfc59ec3@mail.gmail.com" type="cite"> <PRE wrap="">On 3/29/06, Mudit Wahal <A class="moz-txt-link-rfc2396E" href="mailto:mwahal@gmail.com"><mwahal@gmail.com></A> wrote:
</PRE> <BLOCKQUOTE type="cite"> <PRE wrap="">On 3/29/06, Craig Partin <A class="moz-txt-link-rfc2396E" href="mailto:cpartin@gmail.com"><cpartin@gmail.com></A> wrote:
</PRE> <BLOCKQUOTE type="cite"> <BLOCKQUOTE type="cite"> <BLOCKQUOTE type="cite"> <PRE wrap="">On 3/29/06, Craig Partin <A class="moz-txt-link-rfc2396E" href="mailto:cpartin@gmail.com"><cpartin@gmail.com></A> wrote:
</PRE> <BLOCKQUOTE type="cite"> <PRE wrap="">This is the first time I have been stumped in over two years of mythtv
use. I got my HD-3000 working just fine tuning QAM stations over
cable in myth. The system works flawlessly and reliably except that
QAM recordings get trashed when anything else is happening on the
system. It records beautifully when nothing else is happening. This
happens on all channels, HD and SD. I can playback 1080i content with
XVMC just fine, even with bob.
Hardware
MSI K7N Delta-2 lsr
Avermedia m179
MX4000 8X AGP
512MB memory
Sempron 2200
250GB sata HD
using onboard sata - also had IDE drive previously. no change
I have searched here and pchdtv.com and am afraid the answer is going
to be a new mobo. I have tried swapping power supplies. The card has
its own irq. I have switched PCI slots, it is currently in the orange
slot. I have upgraded to udev and kernel 2.6.15-gentoo-sources-r1.
</PRE> </BLOCKQUOTE> </BLOCKQUOTE> </BLOCKQUOTE> <BLOCKQUOTE type="cite"> <PRE wrap="">On 3/29/06, Mudit Wahal <A class="moz-txt-link-rfc2396E" href="mailto:mwahal@gmail.com"><mwahal@gmail.com></A> wrote:
</PRE> <BLOCKQUOTE type="cite"> <PRE wrap="">Do you see any messages in the logfile such as IOBuffer etc ?
</PRE> </BLOCKQUOTE> </BLOCKQUOTE> <PRE wrap="">On 3/29/06, Mudit Wahal <A class="moz-txt-link-rfc2396E" href="mailto:mwahal@gmail.com"><mwahal@gmail.com></A> wrote:
</PRE> <BLOCKQUOTE type="cite"> <PRE wrap="">Sorry I meat IOBound messages in the logfile.
</PRE> </BLOCKQUOTE> <PRE wrap="">No IOBound errors in the backend log. The frontend just has the ac
tex damaged, invalid mb errors
</PRE> </BLOCKQUOTE> <PRE wrap="">Are you running 0.18 or 0.19 ? Also I read somewhere that the SATA
drivers for the older chipset had some issues which was causing file
corruption.
You can use azap to tune to a channel and dump the HD output from the
card to a file directly with and without load. This will eliminate
myth from the middle. You can do vmstat or top and check the iowait%
etc.
I am running 0.18.1 and had IOBound issues while recording and doing
something else (commercial detection/file deletion/etc). But the
IOBound messages were getting logged in the logfile. Had to change
some buffersize etc to get around the messages.
</PRE> </BLOCKQUOTE> <PRE wrap="">I am running 0.19-fixes, whatever version is in the Gentoo portage
tree as of the past weekend. The issue existed under 0.18.1 also. It
happens with or without myth. The problem is not really myth related
but this is the most knowledgable group around for these issues.
There are several similar threads at the pchdtv forums but no answers
</PRE></BLOCKQUOTE> I do know that is the board has a Silicon Imaging (Images?) controller on it there are known problems with dumping an HDTV stream to the drives while anything else is happening. I have that problem with my pchdtv 3000 card.<BR></BLOCKQUOTE><BR></DIV><DIV>Take a look at PCI Latency on the WiKi.</DIV><DIV><BR class="khtml-block-placeholder"></DIV><DIV><A href="http://www.mythtv.org/wiki/index.php/PCI_Latency">http://www.mythtv.org/wiki/index.php/PCI_Latency</A></DIV><DIV><BR class="khtml-block-placeholder"></DIV><DIV>In particular look at this link in the "credits" area.</DIV><DIV><BR class="khtml-block-placeholder"></DIV><DIV><A href="http://www-128.ibm.com/developerworks/library/l-hw2.html">http://www-128.ibm.com/developerworks/library/l-hw2.html</A></DIV><DIV><BR class="khtml-block-placeholder"></DIV><DIV>This sounds like a classic case of feeding data to the PCI bus faster than the system can write it to the drives. You'll probably want to increase the latency of your IDE subsystem if you can.</DIV><DIV><BR class="khtml-block-placeholder"></DIV></BODY></HTML>