Hey all
I'm looking for a method similar to what [pix_equal] does but supports more than 8 bit per channel. With [pix_image] - [pix_equal], it seems I get only 256 steps with a greyscale image, even if the mask loaded with [pix_image] has 16 bit. Is this a limitation by openGL or by [pix_image]? Would I have more luck when doing it with shaders instead of [pix_*] objects?
Thanks Roman
afaik, Gem only supports 8bit per pixel images so all pix_* objects have this limitation I was wondering the same when working on image capture plugin with 10 or 12bit camera I didn't find any solution to increase the depth without code refactoring
-- do it yourself http://antoine.villeret.free.fr
2017-04-28 23:52 GMT+02:00 Roman Haefeli reduzent@gmail.com:
Hey all
I'm looking for a method similar to what [pix_equal] does but supports more than 8 bit per channel. With [pix_image] - [pix_equal], it seems I get only 256 steps with a greyscale image, even if the mask loaded with [pix_image] has 16 bit. Is this a limitation by openGL or by [pix_image]? Would I have more luck when doing it with shaders instead of [pix_*] objects?
Thanks Roman _______________________________________________ Pd-list@lists.iem.at mailing list UNSUBSCRIBE and account-management -> https://lists.puredata.info/li stinfo/pd-list
On Sam, 2017-04-29 at 15:30 +0200, Antoine Villeret wrote:
afaik, Gem only supports 8bit per pixel images so all pix_* objects have this limitation I was wondering the same when working on image capture plugin with 10 or 12bit camera I didn't find any solution to increase the depth without code refactoring
I think I found a work-round for my case, since I want to have more bit depth with only one channel (greyscale). I swapped the header of a 24bit greyscale TIFF with the header of an RGB/8bit TIFF and thus was able to get more resolution. However, this doesn't work with the [pix_equal] -> [pix_mask] setup anymore and it seems its too slow anyway. I'm now trying to do the masking with a shader instead and hopefully be able to use the extra bit depth with the trick mentioned above (I'm still quite a GLSL noob, though... this will probably take some time).
Roman
2017-04-28 23:52 GMT+02:00 Roman Haefeli reduzent@gmail.com:
Hey all
I'm looking for a method similar to what [pix_equal] does but supports more than 8 bit per channel. With [pix_image] - [pix_equal], it seems I get only 256 steps with a greyscale image, even if the mask loaded with [pix_image] has 16 bit. Is this a limitation by openGL or by [pix_image]? Would I have more luck when doing it with shaders instead of [pix_*] objects?
Thanks Roman _______________________________________________ Pd-list@lists.iem.at mailing list UNSUBSCRIBE and account-management -> https://lists.puredata.info/l istinfo/pd-list
-
Antoine Villeret -
Roman Haefeli