hello,
i would like to access color data of a framebuffer. just like pix_data, but working with a framebuffer. i'm specially interested with framebuffer because i need to have more than 8 bit/color (i'd like to use RGB32 format).
pix_write use successfully the frambuffer, but i can't make pix_snap to work.
is there a solution, and if not, what would be the best to do in order to make this possible?
thx cyrille
Hello Cyrille, I start with example/Gem/pix/framebuffer_readback.pd. I'm not sure that is what you are looking for (Do you want tu use the right outlet of [gemframebuffer] ?). Here the patch with the texture 'like' the right outlet of [gemframebuffer] and [pix_data]. ++
Jack
Le 31 mai 08 à 23:12, cyrille henry a écrit :
hello,
i would like to access color data of a framebuffer. just like pix_data, but working with a framebuffer. i'm specially interested with framebuffer because i need to have more than 8 bit/color (i'd like to use RGB32 format).
pix_write use successfully the frambuffer, but i can't make pix_snap to work.
is there a solution, and if not, what would be the best to do in order to make this possible?
thx cyrille
GEM-dev mailing list GEM-dev@iem.at http://lists.puredata.info/listinfo/gem-dev
Oops ! Send my last mail too fast. Here the patch (simple) i wanted to attach. ++
Jack
Le 31 mai 08 à 23:12, cyrille henry a écrit :
hello,
i would like to access color data of a framebuffer. just like pix_data, but working with a framebuffer. i'm specially interested with framebuffer because i need to have more than 8 bit/color (i'd like to use RGB32 format).
pix_write use successfully the frambuffer, but i can't make pix_snap to work.
is there a solution, and if not, what would be the best to do in order to make this possible?
thx cyrille
GEM-dev mailing list GEM-dev@iem.at http://lists.puredata.info/listinfo/gem-dev
hello,
thanks for your answer, this patch did not work here : pix_data did not output any value.
could you try this one and tell me what does output pix_info?
anyway, i had an other try with pix_snap, and today it is working! (i don't really understand why it was not working yesterday).
value are still in 8 bits as it's hardcoded in pix_snap. so, i'll try to see if i could make pix_snap to work also with 16 or 32 bits buffer.
thanks cyrille
Jack a écrit :
Oops ! Send my last mail too fast. Here the patch (simple) i wanted to attach. ++
Jack
Le 31 mai 08 à 23:12, cyrille henry a écrit :
hello,
i would like to access color data of a framebuffer. just like pix_data, but working with a framebuffer. i'm specially interested with framebuffer because i need to have more than 8 bit/color (i'd like to use RGB32 format).
pix_write use successfully the frambuffer, but i can't make pix_snap to work.
is there a solution, and if not, what would be the best to do in order to make this possible?
thx cyrille
GEM-dev mailing list GEM-dev@iem.at http://lists.puredata.info/listinfo/gem-dev
GEM-dev mailing list GEM-dev@iem.at http://lists.puredata.info/listinfo/gem-dev
#N canvas 192 148 489 651 10; #X msg 112 86 create , 1; #X obj 112 131 gemwin; #X msg 128 108 lighting 1; #X obj 161 269 cnv 15 150 150 empty empty scene 20 12 0 14 -233017 -66577 0; #X obj 82 196 gemhead 20; #X obj 82 220 gemframebuffer; #X obj 167 331 rotateXYZ; #X obj 167 376 teapot; #X obj 82 266 t a a b; #X obj 187 293 i; #X obj 220 292 + 1; #X obj 250 291 % 360; #X obj 82 243 translateXYZ 0 0 -4; #X obj 167 353 color 1 0 1; #X floatatom 226 313 5 0 0 0 - - -; #X obj 163 175 loadbang; #X msg 194 133 0 , destroy; #X obj 67 515 pix_data; #X floatatom 97 495 5 0 0 0 - - -; #X floatatom 135 496 5 0 0 0 - - -; #X obj 89 538 unpack f f f; #X floatatom 89 591 5 0 0 0 - - -; #X floatatom 123 591 5 0 0 0 - - -; #X floatatom 158 591 5 0 0 0 - - -; #X obj 51 491 bng 15 250 50 0 empty empty empty 17 7 0 10 -262144 -1 -1; #X msg 97 472 0.5; #X text 131 471 <- send this message; #X obj 89 559 * 256; #X obj 127 560 * 256; #X obj 168 561 * 256; #X obj 82 424 pix_info ---; #X floatatom 111 447 5 0 0 0 - - -; #X msg 162 195 dimen 128 128 , format RGB32; #X connect 0 0 1 0; #X connect 2 0 1 0; #X connect 4 0 5 0; #X connect 5 0 12 0; #X connect 6 0 13 0; #X connect 8 0 30 0; #X connect 8 1 6 0; #X connect 8 2 9 0; #X connect 9 0 10 0; #X connect 9 0 6 1; #X connect 9 0 6 2; #X connect 10 0 11 0; #X connect 11 0 9 1; #X connect 12 0 8 0; #X connect 13 0 7 0; #X connect 14 0 6 3; #X connect 15 0 32 0; #X connect 16 0 1 0; #X connect 17 1 20 0; #X connect 18 0 17 2; #X connect 19 0 17 3; #X connect 20 0 27 0; #X connect 20 1 28 0; #X connect 20 2 29 0; #X connect 24 0 17 0; #X connect 25 0 18 0; #X connect 25 0 19 0; #X connect 27 0 21 0; #X connect 28 0 22 0; #X connect 29 0 23 0; #X connect 30 0 17 1; #X connect 30 3 31 0; #X connect 32 0 5 0;
I try my last patch i sent and (i don't know why) it doesn't work, sorry.
But the first works fine (with [pix_buffer]). With this patch, I put the message [dimen 128 128, format RGB32 ( under [loadbang] and nothing worked. But if i only put [dimen 128 128( under loadbang then press [0.5( and the bang, [pix_data] output 255 0 255. Then i send the message [format RGB32( (after the creation of gemwin) to [gemframebuffer] then bang, and [pix_data] also output 255 0 255. But before creation of gemwin, nothing happen. Here the two patchs.
Le 1 juin 08 à 17:13, cyrille henry a écrit :
hello,
thanks for your answer, this patch did not work here : pix_data did not output any value.
could you try this one and tell me what does output pix_info?
anyway, i had an other try with pix_snap, and today it is working!
Cool !
(i don't really understand why it was not working yesterday).
value are still in 8 bits as it's hardcoded in pix_snap. so, i'll try to see if i could make pix_snap to work also with 16 or 32 bits buffer.
Have you try with [pix_buffer] ? I don't know if it works with 8, 16 or 32 bits. ++
Jack
thanks cyrille
Jack a écrit :
Oops ! Send my last mail too fast. Here the patch (simple) i wanted to attach. ++ Jack Le 31 mai 08 à 23:12, cyrille henry a écrit :
hello,
i would like to access color data of a framebuffer. just like pix_data, but working with a framebuffer. i'm specially interested with framebuffer because i need to have more than 8 bit/color (i'd like to use RGB32 format).
pix_write use successfully the frambuffer, but i can't make pix_snap to work.
is there a solution, and if not, what would be the best to do in order to make this possible?
thx cyrille
GEM-dev mailing list GEM-dev@iem.at http://lists.puredata.info/listinfo/gem-dev
GEM-dev mailing list GEM-dev@iem.at http://lists.puredata.info/listinfo/gem-dev
#N canvas 192 148 489 651 10; #X msg 112 86 create , 1; #X obj 112 131 gemwin; #X msg 128 108 lighting 1; #X obj 161 269 cnv 15 150 150 empty empty scene 20 12 0 14 -233017 -66577 0; #X obj 82 196 gemhead 20; #X obj 82 220 gemframebuffer; #X obj 167 331 rotateXYZ; #X obj 167 376 teapot; #X obj 82 266 t a a b; #X obj 187 293 i; #X obj 220 292 + 1; #X obj 250 291 % 360; #X obj 82 243 translateXYZ 0 0 -4; #X obj 167 353 color 1 0 1; #X floatatom 226 313 5 0 0 0 - - -; #X obj 163 175 loadbang; #X msg 194 133 0 , destroy; #X obj 67 515 pix_data; #X floatatom 97 495 5 0 0 0 - - -; #X floatatom 135 496 5 0 0 0 - - -; #X obj 89 538 unpack f f f; #X floatatom 89 591 5 0 0 0 - - -; #X floatatom 123 591 5 0 0 0 - - -; #X floatatom 158 591 5 0 0 0 - - -; #X obj 51 491 bng 15 250 50 0 empty empty empty 17 7 0 10 -262144 -1 -1; #X msg 97 472 0.5; #X text 131 471 <- send this message; #X obj 89 559 * 256; #X obj 127 560 * 256; #X obj 168 561 * 256; #X obj 82 424 pix_info ---; #X floatatom 111 447 5 0 0 0 - - -; #X msg 162 195 dimen 128 128 , format RGB32; #X connect 0 0 1 0; #X connect 2 0 1 0; #X connect 4 0 5 0; #X connect 5 0 12 0; #X connect 6 0 13 0; #X connect 8 0 30 0; #X connect 8 1 6 0; #X connect 8 2 9 0; #X connect 9 0 10 0; #X connect 9 0 6 1; #X connect 9 0 6 2; #X connect 10 0 11 0; #X connect 11 0 9 1; #X connect 12 0 8 0; #X connect 13 0 7 0; #X connect 14 0 6 3; #X connect 15 0 32 0; #X connect 16 0 1 0; #X connect 17 1 20 0; #X connect 18 0 17 2; #X connect 19 0 17 3; #X connect 20 0 27 0; #X connect 20 1 28 0; #X connect 20 2 29 0; #X connect 24 0 17 0; #X connect 25 0 18 0; #X connect 25 0 19 0; #X connect 27 0 21 0; #X connect 28 0 22 0; #X connect 29 0 23 0; #X connect 30 0 17 1; #X connect 30 3 31 0; #X connect 32 0 5 0;
Jack a écrit :
I try my last patch i sent and (i don't know why) it doesn't work, sorry.
But the first works fine (with [pix_buffer]).
it work also here. that's why i try pix_snap again.
With this patch, I put the message [dimen 128 128, format RGB32( under [loadbang] and nothing worked. But if i only put [dimen 128 128( under loadbang then press [0.5( and the bang, [pix_data] output 255 0 255. Then i send the message [format RGB32( (after the creation of gemwin) to [gemframebuffer] then bang, and [pix_data] also output 255 0 255. But before creation of gemwin, nothing happen. Here the two patchs.
ok. that is good to konw. here, both works.
anyway, i realised that none of the Gem object works with more than 8 bits/color.
so i have to understand and write some code in order to adapt some object for 32 bits. i'm unsuccessful for now, but i'll make more test as this could lead to a lot's of applications.
cyrille
Le 1 juin 08 à 17:13, cyrille henry a écrit :
hello,
thanks for your answer, this patch did not work here : pix_data did not output any value.
could you try this one and tell me what does output pix_info?
anyway, i had an other try with pix_snap, and today it is working!
Cool !
(i don't really understand why it was not working yesterday).
value are still in 8 bits as it's hardcoded in pix_snap. so, i'll try to see if i could make pix_snap to work also with 16 or 32 bits buffer.
Have you try with [pix_buffer] ? I don't know if it works with 8, 16 or 32 bits. ++
Jack
thanks cyrille
Jack a écrit :
Oops ! Send my last mail too fast. Here the patch (simple) i wanted to attach. ++ Jack Le 31 mai 08 à 23:12, cyrille henry a écrit :
hello,
i would like to access color data of a framebuffer. just like pix_data, but working with a framebuffer. i'm specially interested with framebuffer because i need to have more than 8 bit/color (i'd like to use RGB32 format).
pix_write use successfully the frambuffer, but i can't make pix_snap to work.
is there a solution, and if not, what would be the best to do in order to make this possible?
thx cyrille
GEM-dev mailing list GEM-dev@iem.at http://lists.puredata.info/listinfo/gem-dev
GEM-dev mailing list GEM-dev@iem.at http://lists.puredata.info/listinfo/gem-dev
#N canvas 192 148 489 651 10; #X msg 112 86 create , 1; #X obj 112 131 gemwin; #X msg 128 108 lighting 1; #X obj 161 269 cnv 15 150 150 empty empty scene 20 12 0 14 -233017 -66577 0; #X obj 82 196 gemhead 20; #X obj 82 220 gemframebuffer; #X obj 167 331 rotateXYZ; #X obj 167 376 teapot; #X obj 82 266 t a a b; #X obj 187 293 i; #X obj 220 292 + 1; #X obj 250 291 % 360; #X obj 82 243 translateXYZ 0 0 -4; #X obj 167 353 color 1 0 1; #X floatatom 226 313 5 0 0 0 - - -; #X obj 163 175 loadbang; #X msg 194 133 0 , destroy; #X obj 67 515 pix_data; #X floatatom 97 495 5 0 0 0 - - -; #X floatatom 135 496 5 0 0 0 - - -; #X obj 89 538 unpack f f f; #X floatatom 89 591 5 0 0 0 - - -; #X floatatom 123 591 5 0 0 0 - - -; #X floatatom 158 591 5 0 0 0 - - -; #X obj 51 491 bng 15 250 50 0 empty empty empty 17 7 0 10 -262144 -1 -1; #X msg 97 472 0.5; #X text 131 471 <- send this message; #X obj 89 559 * 256; #X obj 127 560 * 256; #X obj 168 561 * 256; #X obj 82 424 pix_info ---; #X floatatom 111 447 5 0 0 0 - - -; #X msg 162 195 dimen 128 128 , format RGB32; #X connect 0 0 1 0; #X connect 2 0 1 0; #X connect 4 0 5 0; #X connect 5 0 12 0; #X connect 6 0 13 0; #X connect 8 0 30 0; #X connect 8 1 6 0; #X connect 8 2 9 0; #X connect 9 0 10 0; #X connect 9 0 6 1; #X connect 9 0 6 2; #X connect 10 0 11 0; #X connect 11 0 9 1; #X connect 12 0 8 0; #X connect 13 0 7 0; #X connect 14 0 6 3; #X connect 15 0 32 0; #X connect 16 0 1 0; #X connect 17 1 20 0; #X connect 18 0 17 2; #X connect 19 0 17 3; #X connect 20 0 27 0; #X connect 20 1 28 0; #X connect 20 2 29 0; #X connect 24 0 17 0; #X connect 25 0 18 0; #X connect 25 0 19 0; #X connect 27 0 21 0; #X connect 28 0 22 0; #X connect 29 0 23 0; #X connect 30 0 17 1; #X connect 30 3 31 0; #X connect 32 0 5 0;
GEM-dev mailing list GEM-dev@iem.at http://lists.puredata.info/listinfo/gem-dev
cyrille henry a écrit :
Jack a écrit :
I try my last patch i sent and (i don't know why) it doesn't work, sorry.
But the first works fine (with [pix_buffer]).
it work also here. that's why i try pix_snap again.
With this patch, I put the message [dimen 128 128, format RGB32( under [loadbang] and nothing worked. But if i only put [dimen 128 128( under loadbang then press [0.5( and the bang, [pix_data] output 255 0 255. Then i send the message [format RGB32( (after the creation of gemwin) to [gemframebuffer] then bang, and [pix_data] also output 255 0 255. But before creation of gemwin, nothing happen. Here the two patchs.
ok. that is good to konw. here, both works.
anyway, i realised that none of the Gem object works with more than 8 bits/color.
so i have to understand and write some code in order to adapt some object for 32 bits. i'm unsuccessful for now, but i'll make more test as this could lead to a lot's of applications.
well, look like it's hopeless: http://www.opengl.org/sdk/docs/man/xhtml/glReadPixels.xml glReadPixels does not work with 16 / 32 bits color.
++ c
cyrille
Le 1 juin 08 à 17:13, cyrille henry a écrit :
hello,
thanks for your answer, this patch did not work here : pix_data did not output any value.
could you try this one and tell me what does output pix_info?
anyway, i had an other try with pix_snap, and today it is working!
Cool !
(i don't really understand why it was not working yesterday).
value are still in 8 bits as it's hardcoded in pix_snap. so, i'll try to see if i could make pix_snap to work also with 16 or 32 bits buffer.
Have you try with [pix_buffer] ? I don't know if it works with 8, 16 or 32 bits. ++
Jack
thanks cyrille
Jack a écrit :
Oops ! Send my last mail too fast. Here the patch (simple) i wanted to attach. ++ Jack Le 31 mai 08 à 23:12, cyrille henry a écrit :
hello,
i would like to access color data of a framebuffer. just like pix_data, but working with a framebuffer. i'm specially interested with framebuffer because i need to have more than 8 bit/color (i'd like to use RGB32 format).
pix_write use successfully the frambuffer, but i can't make pix_snap to work.
is there a solution, and if not, what would be the best to do in order to make this possible?
thx cyrille
GEM-dev mailing list GEM-dev@iem.at http://lists.puredata.info/listinfo/gem-dev
GEM-dev mailing list GEM-dev@iem.at http://lists.puredata.info/listinfo/gem-dev
#N canvas 192 148 489 651 10; #X msg 112 86 create , 1; #X obj 112 131 gemwin; #X msg 128 108 lighting 1; #X obj 161 269 cnv 15 150 150 empty empty scene 20 12 0 14 -233017 -66577 0; #X obj 82 196 gemhead 20; #X obj 82 220 gemframebuffer; #X obj 167 331 rotateXYZ; #X obj 167 376 teapot; #X obj 82 266 t a a b; #X obj 187 293 i; #X obj 220 292 + 1; #X obj 250 291 % 360; #X obj 82 243 translateXYZ 0 0 -4; #X obj 167 353 color 1 0 1; #X floatatom 226 313 5 0 0 0 - - -; #X obj 163 175 loadbang; #X msg 194 133 0 , destroy; #X obj 67 515 pix_data; #X floatatom 97 495 5 0 0 0 - - -; #X floatatom 135 496 5 0 0 0 - - -; #X obj 89 538 unpack f f f; #X floatatom 89 591 5 0 0 0 - - -; #X floatatom 123 591 5 0 0 0 - - -; #X floatatom 158 591 5 0 0 0 - - -; #X obj 51 491 bng 15 250 50 0 empty empty empty 17 7 0 10 -262144 -1 -1; #X msg 97 472 0.5; #X text 131 471 <- send this message; #X obj 89 559 * 256; #X obj 127 560 * 256; #X obj 168 561 * 256; #X obj 82 424 pix_info ---; #X floatatom 111 447 5 0 0 0 - - -; #X msg 162 195 dimen 128 128 , format RGB32; #X connect 0 0 1 0; #X connect 2 0 1 0; #X connect 4 0 5 0; #X connect 5 0 12 0; #X connect 6 0 13 0; #X connect 8 0 30 0; #X connect 8 1 6 0; #X connect 8 2 9 0; #X connect 9 0 10 0; #X connect 9 0 6 1; #X connect 9 0 6 2; #X connect 10 0 11 0; #X connect 11 0 9 1; #X connect 12 0 8 0; #X connect 13 0 7 0; #X connect 14 0 6 3; #X connect 15 0 32 0; #X connect 16 0 1 0; #X connect 17 1 20 0; #X connect 18 0 17 2; #X connect 19 0 17 3; #X connect 20 0 27 0; #X connect 20 1 28 0; #X connect 20 2 29 0; #X connect 24 0 17 0; #X connect 25 0 18 0; #X connect 25 0 19 0; #X connect 27 0 21 0; #X connect 28 0 22 0; #X connect 29 0 23 0; #X connect 30 0 17 1; #X connect 30 3 31 0; #X connect 32 0 5 0;
GEM-dev mailing list GEM-dev@iem.at http://lists.puredata.info/listinfo/gem-dev
GEM-dev mailing list GEM-dev@iem.at http://lists.puredata.info/listinfo/gem-dev
well, look like it's hopeless: http://www.opengl.org/sdk/docs/man/xhtml/glReadPixels.xml glReadPixels does not work with 16 / 32 bits color.
What exactly do you mean by 16/32 bits color? Do you mean per-channel or combined RGBA values? glReadPixels works with any format that I've come across. What makes you think it won't work for your situation?
wes
hello,
Wesley Smith a écrit :
well, look like it's hopeless: http://www.opengl.org/sdk/docs/man/xhtml/glReadPixels.xml glReadPixels does not work with 16 / 32 bits color.
What exactly do you mean by 16/32 bits color? Do you mean per-channel or combined RGBA values?
i mean per channel.
glReadPixels works with any format that I've come across. What makes you think it won't work for your situation?
the doc i find there : http://www.opengl.org/sdk/docs/man/xhtml/glReadPixels.xml does not specify a format with 16/32 bit per channel. unless i'm misunderstand something...
did your really used glReadPixels with 16 or 32 bit per channel? and if yes, with what type and format?
thanks Cyrille
wes
GEM-dev mailing list GEM-dev@iem.at http://lists.puredata.info/listinfo/gem-dev
What exactly do you mean by 16/32 bits color? Do you mean per-channel or combined RGBA values?
i mean per channel.
glReadPixels works with any format that I've come across. What makes you think it won't work for your situation?
the doc i find there : http://www.opengl.org/sdk/docs/man/xhtml/glReadPixels.xml does not specify a format with 16/32 bit per channel. unless i'm misunderstand something...
did your really used glReadPixels with 16 or 32 bit per channel? and if yes, with what type and format?
The OpenGL docs like this one don't cross-reference other uses/abuses of the function. glReadPixels as documented in your link refers to its use on the standard OpenGL drawing buffers but does not mention anything about extensions like FBOs. Typical OpenGL draw buffers can only have integer types, so you don't see the higher bpp type enums in the docs. However, when an FBO is active, you can set the read buffer to the FBO as in glReadBuffer(GL_COLOR_ATTACHMENT0_EXT). Then when you call glReadPixels, it will read the pixels in the FBO and if the FBO happens to have a float texture target attached to it, you will be able to read floating point values via glReadPixels.
HTH, wes
ok, thanks for the answer. i still don't understand all of this, i need to have a deeper look.
now, i know where to look.
thanks cyrille
Wesley Smith a écrit :
What exactly do you mean by 16/32 bits color? Do you mean per-channel or combined RGBA values?
i mean per channel.
glReadPixels works with any format that I've come across. What makes you think it won't work for your situation?
the doc i find there : http://www.opengl.org/sdk/docs/man/xhtml/glReadPixels.xml does not specify a format with 16/32 bit per channel. unless i'm misunderstand something...
did your really used glReadPixels with 16 or 32 bit per channel? and if yes, with what type and format?
The OpenGL docs like this one don't cross-reference other uses/abuses of the function. glReadPixels as documented in your link refers to its use on the standard OpenGL drawing buffers but does not mention anything about extensions like FBOs. Typical OpenGL draw buffers can only have integer types, so you don't see the higher bpp type enums in the docs. However, when an FBO is active, you can set the read buffer to the FBO as in glReadBuffer(GL_COLOR_ATTACHMENT0_EXT). Then when you call glReadPixels, it will read the pixels in the FBO and if the FBO happens to have a float texture target attached to it, you will be able to read floating point values via glReadPixels.
HTH, wes
GEM-dev mailing list GEM-dev@iem.at http://lists.puredata.info/listinfo/gem-dev
GL_FLOAT is a 32 bit per component format. .
On Sun, Jun 1, 2008 at 3:06 PM, cyrille henry cyrille.henry@la-kitchen.fr wrote:
hello,
Wesley Smith a écrit :
well, look like it's hopeless: http://www.opengl.org/sdk/docs/man/xhtml/glReadPixels.xml glReadPixels does not work with 16 / 32 bits color.
What exactly do you mean by 16/32 bits color? Do you mean per-channel or combined RGBA values?
i mean per channel.
glReadPixels works with any format that I've come across. What makes you think it won't work for your situation?
the doc i find there : http://www.opengl.org/sdk/docs/man/xhtml/glReadPixels.xml does not specify a format with 16/32 bit per channel. unless i'm misunderstand something...
did your really used glReadPixels with 16 or 32 bit per channel? and if yes, with what type and format?
thanks Cyrille
wes
GEM-dev mailing list GEM-dev@iem.at http://lists.puredata.info/listinfo/gem-dev
GEM-dev mailing list GEM-dev@iem.at http://lists.puredata.info/listinfo/gem-dev
-
chris clepper -
cyrille henry -
Jack -
Wesley Smith