<div class="gmail_quote">On Mon, Aug 29, 2011 at 8:43 AM, Andrew Stadt <span dir="ltr"><<a href="mailto:acstadt@stadt.ca">acstadt@stadt.ca</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
<div class="im">On 29/08/2011 12:56 AM, Jerry Rubinow wrote:<br>
> On Sat, Aug 27, 2011 at 12:28 AM, Jerry Rubinow <<a href="mailto:jerrymr@gmail.com">jerrymr@gmail.com</a><br>
</div><div class="im">> <mailto:<a href="mailto:jerrymr@gmail.com">jerrymr@gmail.com</a>>> wrote:<br>
><br>
> On Fri, Aug 26, 2011 at 11:28 PM, Jerry Rubinow <<a href="mailto:jerrymr@gmail.com">jerrymr@gmail.com</a><br>
</div><div><div></div><div class="h5">> <mailto:<a href="mailto:jerrymr@gmail.com">jerrymr@gmail.com</a>>> wrote:<br>
><br>
> I just upgraded the firmware on my HD-PVR from version 0x12<br>
> May 7, 2009 to the latest from the Hauppauge web site, 0x17<br>
> July 27, 2011. Now all recordings from the HD-PVR are way<br>
> oversaturated. Catting /dev/video0 and playing the file with<br>
> vlc yields the same result. According to v4l2-ctl -l, the<br>
> settings are all at defaults. Any ideas what the problem<br>
> could be?<br>
><br>
> Kern.log says "untested firmware, the driver might not work".<br>
> That's a little scary. Any way to obtain and install a<br>
> previous firmware version?<br>
><br>
> -Jerry<br>
><br>
><br>
> I tried adjusting the brightness and contrast in the channel<br>
> change script, via:<br>
><br>
> /usr/bin/v4l2-ctl --device=/dev/video0 --set-ctrl=brightness=100<br>
><br>
> But it seems to have no effect. I can use the same command during<br>
> a recording from the command line and see it effect the<br>
> brightness. It appears that when the recording starts, these<br>
> HD-PVR settings get reset to defaults.<br>
><br>
> I'm running 0.24.1-fixes, btw.<br>
><br>
><br>
> And a 2.6.38-10 kernel.<br>
><br>
> It seems like Myth must be setting the hd-pvr to a default state after<br>
> running the channel change script. Or for some reason the v4l2-ctl<br>
> lines aren't having an effect when run inside the script. Is there<br>
> any other point when a recording starts, besides the channel change<br>
> script, to run a script? If so, and if there is another script point,<br>
> at least I could have a workaround to adjust the picture adequately<br>
> until I found a real fix.<br>
><br>
> -Jerry<br>
><br>
</div></div>An ugly gludge I used once for a PVR150 with audio issues simply<br>
involved spawning another script from inside the channel change script.<br>
The 2nd script would sleep for a couple secs, while myth did its post<br>
channel change processing, then reset the audio channel via v4l2-ctl.<br>
Might work here, YMMV.<br></blockquote><div><br></div><div>That's an excellent kludge, thanks for the idea. I'll try that out tonight and hopefully that'll last until I'm able to find older firmware or otherwise figure out why this started happening. </div>
<div><br></div><div>Janne, Jarod, or anyone who's familiar with the driver, any ideas? Is there a potential kernel incompatibility with the firmware? I wanted to save doing a full system update as a last resort since things are otherwise pretty stable.</div>
<div><br></div><div>-Jerry</div></div>