On Tue, Nov 30, 2004 at 01:05:55PM -0600, Chris Kennedy wrote:
> 
> 0x01024000-0x01026fa0 - YUV Buffer for Decoding?
> 0x01029000-0x0107d600 - YUV Buffer for Decoding?
> 0x0108e400-0x010b8700 - YUV Buffer for Decoding?
> ...
> 0x01240400-0x012949f0 - YUV Buffer for Decoding?
> 0x012a5800-0x012cfb00 - YUV Buffer for Decoding?
> 0x012d8200-0x0132c7f0 - YUV Buffer for Decoding?


I have calculated some, these seem right on size, exactly right ;-).

(I think the bytes being short a few bytes would be from my calculating the
offset a little off possibly, or the video data being 00's or something at
the end of it, now seems to actually have data evenly laid out to make them
all 345600 bytes)

0x01024000-0x01026fa0  - is 12192 bytes in size (??? PCM perhaps ???)

 = Space of 8288 bytes =

0x01029000-0x0107d600  - is 345600 bytes in size (Y buffer)

 = Space of 69120 bytes =

0x0108e400-0x010b8700  - is 172800 bytes in size (UV buffer)


 = Space of 1604864 bytes =


0x01240400-0x012949f0  - is 345584 bytes in size (Y buffer)

 = Space of 69136 bytes =

0x012a5800-0x012cfb00  - is 172800 bytes in size (UV buffer)

 = Space of 34560 bytes =

0x012d8200-0x0132c7f0  - is 345584 bytes in size (Y buffer)


So at least the first is not the Y buffer, and second is the Y buffer it seems,
and so that explains why we aren't working, at least one reason why it isn't.
We seem to need a split of the DMA offset after 345600 bytes, then add and jump
up to the UV buffer offset I guess.  Sounds kinda tricky, but this is for NTSC 
and for PAL I bet the buffers are just filled to the end possibly (actually may
be easier writing in PAL then (all space between, so like the frame buffer,
probably just fill them a bit farther in that case).  That first buffer seems
odd, maybe PCM I guess, or including timing information, control stuff???
The oddest thing is how the buffers go ...

 [PCM?? -> Y -> UV -> Y -> UV -> Y] 

But seems to be that way and if look forward, backward, seems this is the
entire buffers right there, nothing beyond or before that are like these
in size or data (or moving).  So not sure what that means, or how YUV totally
works on this level, so maybe this layout can make it easier for people to
analyze without much work, since I've done all the mapping and calculating
above :-).


I think if using the newest versions (I put them into the OLD/ivtv-0.2.0-rc
directories for now, timing stuff by the vsync interrupt seems a little
experimental, although guess working actually better, just can't do it on
the itvc16 and so kinda a concern), perhaps moving the YUV buffer offset 
to 0x01029000 may prove interesting (although don't know what that
first buffer is, if it has to be filled to make things work or not).  At least
in PAL I *think* then the Y and UV wouldn't have to be split, but in NTSC
I'm certain they do.  Also it may be we really have to fill at least the first
2 sets of Y-UV buffers, hopefully the last Y one is perhaps where the
hardware/firmware copies the current Y buffer to display or something, still
a mystery though where it's UV buffer is for that.


Thanks,
Chris

Also the second set is also the exact right sizes it seems, again same 
> 
> ---
>  Chris Kennedy / [EMAIL PROTECTED]
>   Engineer KMOS-TV/KTBG-FM
>   Broadcasting Services Department
>   Central Missouri State University
> 
> 
> -------------------------------------------------------
> SF email is sponsored by - The IT Product Guide
> Read honest & candid reviews on hundreds of IT Products from real users.
> Discover which products truly live up to the hype. Start reading now. 
> http://productguide.itmanagersjournal.com/
> _______________________________________________
> ivtv-devel mailing list
> [EMAIL PROTECTED]
> https://lists.sourceforge.net/lists/listinfo/ivtv-devel

-- 
---
 Chris Kennedy / [EMAIL PROTECTED]
  Engineer KMOS-TV/KTBG-FM
  Broadcasting Services Department
  Central Missouri State University


-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now. 
http://productguide.itmanagersjournal.com/
_______________________________________________
ivtv-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/ivtv-devel

Reply via email to