One glitchy comp: http://web.me.com/CIglitch.qtz
Actually testing it a bit more revealed that this happens with 8 bit
mode too! Play with the 'width' setting, you can see the image getting
divided as the resolution increases (here at least, with 10.5.8/radeon
2600).
I'm not doing any ROI stuff at all, and it uses only one filter input
so I can't see it being that. It's a custom sharpen filter, nothing
too special.
Anyone else getting this glitch?
Chris
On 15 Sep 2009, at 20:39, Christopher Wright wrote:
Probably a good idea. I came across that on leopard btw - it'll
probably already be fixed in 10.6 leaving me with the headache of
supporting both ;)
Are these built in CI filters or ones which you have written?
Sounds like a missing ROI function.
This is also plausible. My only "driver bug" indicator was his
statement:
CI seems to glitch with 32bit colour + high resolution, tiling the
image might help. It seems that if I use an image > 1024 pixels and
run a CI Filter that samples at an offset, it can't sample to the
right of the 1024pixel boundary for a destination to the left. I
might have to implement some kind of tiling to get around that..
If the behaviour is different in 32bit mode from 8bit mode, or if
the image dimensions affect sampling behaviour, an ROI problem
should still present itself in those cases (in my opinion -- there
are of course implementation details that may negate that).
Chris, can you supply an example composition with a filter that
exhibits said problem, so others can test against it?
--
[ christopher wright ]
[email protected]
http://kineme.net/
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Quartzcomposer-dev mailing list ([email protected])
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/quartzcomposer-dev/archive%40mail-archive.com
This email sent to [email protected]
