That looks like a core image is in a surface subdivision mode. Its
actively trying to optimize the kernel by sub-dividin the work due to
GPU texture size limitations. Im going to bet the GMA950 has texture
limitations that require splitting up the workload like this, so Core
Image is actually running your kernel twice (or 4 times) and splititng
the workload between sub images.
Ive seen issues like this when working on my own filters, and it
turned out I was not handling bounds correctly on my end. Adjusting
the strength and scale filters also adjusts the 'artifacts' showing up
on the 'extra' lines, so I think this is an issue with your kernel
bumping in to how Core Image handles images larger than the GPU can
natively handle well.
On Sep 15, 2009, at 6:06 PM, Christopher Wright wrote:
Bad link sorry (thanks for spotting it Vade!), file is here:
http://web.me.com/psonice/CIglitch.qtz
Chris, can you supply an example composition with a filter that
exhibits said problem, so others can test against it?
Here are my results from 10.5.8/GMA950 --
http://softpixel.com/~cwright/programming/qcciglitch/
I went through and found the boundary cases where it creates a
stripe on each axis, as well as a few intermediate widths just to
flesh out the page.
--
[ 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/doktorp%40mac.com
This email sent to [email protected]
_______________________________________________
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]
