[pvrusb2] New driver snapshot: pvrusb2-mci-20060101

[Mike Isely]

On Wed, 4 Jan 2006, Frans Meulenbroeks wrote:

> Hi Mike,
>
> Thanks for your answer.
> I did not manually set it to 288 or to interlaced.
> Basically what I did is rewrite the PAL-BG (actually
> this was in the program I use to capture my
> recordings), and I ended up with this.

The source code for the driver has one place where it an gratuitously 
change the resolution and interlace (near the top of 
pvr2_hdw_commit_ctl_internal() in pvrusb2-hdw.c) - and in that spot it can 
only set the vertical resolution to either 480 or 576 and the interlace is 
always cleared.  Yet the log info you printed here clearly shows an 
attempt to set the vertical resolution to 288 and the interlace being set. 
Given that this can't come from within the driver then some outside entity 
must be commanding the driver to that, and at that point I'm puzzled. 
That's why I asked if you had tried to do that.  If you hadn't then I 
really have no idea what is happening there.

However the 20060101 snapshot definitely picks the wrong vertical 
resolution (480 instead of 576 as it should) when initializing to PAL-BG, 
and that is definitely going to be a problem for you.  So let's not worry 
about the strange symptom for the moment and instead try the 20060103 
snapshot and let's see where things stand before trying to dig deeper.


>
> Do you have any comments on the differences that I get
> with file on the new and old files (see my previous
> message).
> Did I for some reason get mpeg1 ?
> And what about the program vs system format?

Sorry, I missed this question before.  A lot has been flying around lately 
:-)

I can't offer any real explanation for that.  The code which programs the 
cx23416 has not been changed in quite some time - basically it has been 
unchanged since last July's stability fix there.  So the result should be 
the same.  Perhaps with that wrong resolution and interlace setting the 
encoder produced something screwed up that fooled other stuff down the 
line.  I think that's really a stretch, but it's the only reasoning I can 
come up with.

   -Mike

-- 
                         |         Mike Isely          |     PGP fingerprint
      Spammers Die!!     |                             | 03 54 43 4D 75 E5 CC 92
                         |   isely @ pobox (dot) com   | 71 16 01 E2 B5 F5 C1 E8
                         |                             |