[00:06:37] *** DobosCake has joined ##OpenGL
[00:15:23] *** renaatier has joined ##OpenGL
[00:17:23] *** LiQuiDninja has joined ##OpenGL
[00:17:29] *** wrenaatier has joined ##OpenGl
[00:17:43] <wrenaatier> Hi
[00:18:17] <wrenaatier> anyone here?
[00:18:47] <Makegho> yup
[00:19:13] <wrenaatier> Is this a good place to ask for help about my opengl assignment :)
[00:20:45] <Makegho> You may try
[00:23:27] *** smiler has joined ##OpenGL
[00:23:40] <wrenaatier> I'm new to texture binding... they gave me the code to read in a bmp file.. but I need to map it to a 70x70 2d grid of polygons--where each tiny square get a corresponding portion of the image
[00:24:02] <wrenaatier> I think I missed the class on how to do it heh
[00:24:34] <Makegho> In theory opengl isn't interested on the pixels, but you can of course force it yourself
[00:25:55] <Makegho> (That is not completely true though, but usually the view is kept something like [-1..1]x[-1..1] instead of something like [0..799]x[0..599])
[00:26:18] <Makegho> So, can you draw a polygon of the size that you want?
[00:26:59] <wrenaatier> I know that each polygon is .1 x .1... and I have a grid of 3600 of them making a 6 x 6 square on the XZ plane
[00:27:45] <Makegho> ah, sorry, I misunderstood
[00:27:58] <wrenaatier> yeah I didn't explain it well at first
[00:29:12] <Makegho> Well you have to create a list of vertices and assign them all the position in the scene, and the corresponding vertex coordinate
[00:29:47] <wrenaatier> So what I'm trying to do is like decal a 500 pixel x 480 pixel (or something) bmp onto the little squares... so each portion gets a part of the image, and the entire grid makes up the full picture
[00:29:49] <Makegho> What language, btw?
[00:30:06] <wrenaatier> using visual studio 2003 C++ lol
[00:31:33] <Makegho> The texture will be loaded to size 512 x 512 (the texture has to be of that size or you have to add borders)
[00:31:46] <Makegho> (or any other 2^n x 2^m)
[00:32:21] <wrenaatier> I'm thinking I have to somehow find the right scale for glTexCoord2f(x, z) for each glVertex3f(xx, 0, zz)... but the man page kind of assumes I know wtf I'm doing lol
[00:32:33] *** Yustme has quit IRC
[00:32:48] <Makegho> yeah, you kind of have to know what you're doing :)
[00:33:05] <Makegho> First step would be to take a pencil and some paper, and draw a grid
[00:33:41] <Makegho> and think.. What do you want? To draw a textured polygon, you have to know the coordinates of the polygon, and the coordinates of the texture where every corner of the polygon maps to
[00:34:11] <Makegho> Mm.. I was a bit unclear
[00:34:59] <wrenaatier> I have that down.  It starts at (far left) -3.0, 0, -3.0    and ends at (near right)  3.0, 0, 3.0    each tile being .1x.1  so it's 60x60.
[00:35:07] <Makegho> You assign every corner of the polygon 1) one point in space 2) one point in texture. The texture is an area where x and y both belong to [0, 1]
[00:35:32] <Makegho> 1 meaning 100 % of width or height
[00:36:35] <wrenaatier> Hmm ok...
[00:37:38] <wrenaatier> 100% of width OR height?
[00:38:55] <Makegho> for x 100% of width, for y 100% of height
[00:40:16] <wrenaatier> so what would [0, .1] do?
[00:41:40] <LiQuiDninja> if i might add a tip, to divise and check the positions of vectors before you go ahead and connect them with triangles strips or quads use glBegin(GL_LINES) and two calls to glVertex3f to render lines
[00:42:10] <Makegho> (0.0, 0.1) would point to (512*0.0, 512*0.1) = (0, 51.2) of the texture
[00:42:39] <wrenaatier> liquid you just confused me further :)
[00:43:01] <LiQuiDninja> oh shush your competant
[00:43:20] <Makegho> wrenaatier: he said that you should draw the grid's edges first, to be sure that you have placed them right
[00:43:30] <wrenaatier> ahhh gotcha...
[00:43:34] <Makegho> and only after it draw the textures
[00:44:08] *** LtJax has quit IRC
[00:45:01] <Makegho> you draw a line by first going to line drawing mode: glBegin(GL_LINES); and then adding the lines: /* line start point */ glVertex3f(x0, y0, z0); /* line end point */ glVertex3f(x1, y1, z1); and then add more of those here if you want to draw more lines
[00:45:20] <Makegho> then in the end you return from the line drawing mode with glEnd();
[00:46:37] <wrenaatier> Hmmm.. I think I got it, maybe.   so if I have my first vertice at -3, 0, -3  (far left)... That would correspond to [0, 0].  Then, for example... for the near right of that tile I'd have vertice -2.9, 0, -2.9.. I'd put [1/60.0, 1/60.0] ?
[00:46:51] <wrenaatier> (if I had 60x60 tiles)
[00:47:26] <Makegho> yup
[00:47:42] <wrenaatier> rawk!
[00:50:31] <wrenaatier> I think I can in some private data members for my tile class to calculate it automatically... then simply put inthe texcoords[S, T] in my draw function
[00:52:23] <wrenaatier> thanks makegho, and liquidninja
[00:53:36] <wrenaatier> I'm going to go try it.  cyu later.
[00:53:38] <Makegho> no problem, and happy coding :)
[00:53:48] *** wrenaatier has quit IRC
[00:54:04] *** renaatier has quit IRC
[00:56:32] *** dv_ has quit IRC
[00:57:16] *** dv_ has joined ##OpenGL
[01:00:49] *** niteon has quit IRC
[01:03:53] *** JernejL_ has joined ##OpenGL
[01:03:58] *** ced117 has quit IRC
[01:04:53] *** TheLorax has joined ##opengl
[01:05:45] *** prophile has quit IRC
[01:07:01] <meteors> Hey, really stupid question here
[01:07:33] <meteors> ah, never mind
[01:07:35] <Makegho> ask it to get an answer :-)
[01:07:46] <meteors> I figured it out
[01:07:49] <Makegho> good
[01:08:33] *** Burga has quit IRC
[01:14:38] *** TheChuckster has joined ##OpenGL
[01:14:54] <TheChuckster> hey what might cause a model to cause a stencil shadow test to fail?
[01:14:59] *** magnet_ has joined ##OpenGL
[01:15:11] <TheChuckster> when i move around if the model is in front of the shadow, there's holes in the shadow
[01:15:16] <TheChuckster> i'm not sure why either
[01:15:24] <TheChuckster> i think it has to do with the depth buffer
[01:15:39] <TheChuckster> for some reason when i'm rendering the shadow volume, it thinks it is behind the 3d model even though its not
[01:22:36] *** Jernej has quit IRC
[01:23:50] *** darka has quit IRC
[01:32:55] *** LordMetroid has quit IRC
[01:34:55] *** MatthiasM has quit IRC
[01:35:02] *** MatthiasM has joined ##opengl
[01:42:23] *** amz has joined ##opengl
[01:42:43] <TheChuckster> the weird thing is this only happens with z fail
[01:42:46] <TheChuckster> z pass works perfect
[01:43:05] *** tmccrary has joined ##OpenGL
[01:45:57] *** Walt_ has quit IRC
[01:47:24] *** Walt has joined ##opengl
[01:48:18] *** TheLorax has quit IRC
[01:49:31] *** noteventime has quit IRC
[01:49:58] *** HuntsMan has quit IRC
[01:51:49] *** braden has quit IRC
[01:58:09] *** replor has quit IRC
[02:03:58] *** amalon has quit IRC
[02:08:36] *** mm765 is now known as mm765^away
[02:09:52] *** Xmas| has joined ##OpenGL
[02:10:50] *** TheChuckster has left ##OpenGL
[02:10:58] *** braden has joined ##OpenGL
[02:13:41] *** Burga has joined ##OpenGL
[02:21:14] *** tmccrary has left ##OpenGL
[02:28:52] *** juanmabc has quit IRC
[02:31:36] *** hibread has joined ##opengl
[02:41:02] *** reduz has quit IRC
[02:45:49] *** amalon has joined ##opengl
[02:48:28] <hibread> g'day knuckle heads!
[02:53:02] *** morgajel has left ##OpenGL
[03:02:51] *** rutski has joined ##OpenGL
[03:03:10] *** StephaneLenclud has joined ##OpenGL
[03:03:14] <StephaneLenclud> Hi
[03:03:49] <StephaneLenclud> If Lighting is enabled glColor won't work right?
[03:04:01] <StephaneLenclud> I should use glMaterial instead?
[03:06:42] *** Shel2476 has joined ##OpenGL
[03:07:50] *** Shel2476 has left ##OpenGL
[03:07:50] <hibread> glColorMaterial()
[03:08:11] <hibread> "glColorMaterial specifies which material parameters track the current color"
[03:08:47] *** TheLorax has joined ##opengl
[03:10:23] <StephaneLenclud> mmmh, no such a thing as glColorMaterial in OpenGL|ES 1.1
[03:10:30] *** Shel2476 has joined ##OpenGL
[03:10:31] *** Shel2476 has left ##OpenGL
[03:13:28] <hibread> StephaneLenclud: sorry i can't be of anymore help
[03:13:41] <Makegho> You can still enable GL_COLOR_MATERIAL I think
[03:14:02] *** Shel2476 has joined ##OpenGL
[03:14:03] *** Shel2476 has left ##OpenGL
[03:14:04] <Makegho> just can't change it from GL_AMBIENT_AND_DIFFUSE
[03:14:23] <Makegho> http://www.khronos.org/registry/gles/specs/1.1/es_cm_spec_1.1.10.pdf page 16
[03:15:46] <Satan_Inside> page 11 actually :)
[03:16:00] <Satan_Inside> specifically subsection 2.14
[03:16:24] <Makegho> ah yes, I looked at the pdf numbering
[03:16:34] <Makegho> but yes it's page 11 :)
[03:16:52] *** gaffa has quit IRC
[03:18:33] <StephaneLenclud> let me look at that
[03:19:16] *** tmccrary has joined ##OpenGL
[03:19:27] <tmccrary> damn, bugle is pretty nice
[03:20:24] *** IJs\laptop has quit IRC
[03:21:30] <Satan_Inside> mmm in regards to an earlier comment I saw mentioning display list versus VBO performance -- on (near) current generation renderers display lists are effectively layered through static VBOs on the server side therefore the key advantages one has with the direct VBO path over using DLs is reduced subroutine call overhead and the ability to directly feed the server "friendly" data which can be a key win if your vendor's DLs optim
[03:23:06] <Satan_Inside> many DL implementations will end up falling off of the fast path (i.e. passing primitive data off as IM to the server) if they can't reorganize the data-sets they are presented with into a optimal format for static VA / VBO
[03:23:21] *** speedy1 has joined ##OpenGL
[03:23:31] <Satan_Inside> those edge cases are ultimately renderer and vendor specific though
[03:24:08] <Satan_Inside> and I've never seen this openly documented
[03:24:21] <tmccrary> I'm not sure if you're talking about the discussion I had
[03:24:32] <tmccrary> Where the guy said vertex arrays are faster than display lists
[03:24:48] *** inoxor has quit IRC
[03:25:03] <StephaneLenclud> Thanks all. Good reading COLOR_MATERIAL fixed my issue
[03:25:18] <tmccrary> but it makes sense to use the static VBO functionality to do display lists, since there so similar
[03:25:27] <tmccrary> *they're so similar
[03:25:56] <Satan_Inside> and like I said above if the vendor can do it for the data you give them then they will
[03:26:40] <Satan_Inside> OTOH this is only possible (in the most optimal case) for a small sub-set of possible DL applications
[03:26:53] <tmccrary> the applet opengl profiler tool is pretty nice
[03:27:10] <Satan_Inside> since people frequently tend to throw all sorts of stuff in a DL expecting it to be "fast"
[03:29:32] *** Lucine2 has quit IRC
[03:30:07] *** Lucine2 has joined ##OpenGL
[03:37:32] <speedy1> if I can choose between:  discard; and letting fragment through, what is more performant?
[03:38:04] <speedy1> letting fragment through hit's the memory, but does discard have some hidden costs?
[03:38:16] <speedy1> *hits
[03:44:01] <Satan_Inside> that depends upon what you are doing with the discard operation
[03:44:17] <Satan_Inside> i.e. is this shader based alpha testing? something else?
[03:44:45] <Satan_Inside> since it might be possible to reject fragments via dedicated HW rather than pushing this off to the programmable pipeline
[03:46:04] <Satan_Inside> generally speaking shader KILL does have a cost (how substantial will depend upon a number of factors) associated with it
[03:46:33] <speedy1> concretely, i'm getting logarithmic luminance average from float16 RGB surface via 16 step approximation
[03:46:59] *** tmccrary has quit IRC
[03:47:17] <speedy1> there is no real depth or color output - perhaps I could set outgoing alpha to 0 and cull the fragment there
[03:47:29] *** Xmas| has quit IRC
[03:48:29] *** HuntsMan has joined ##opengl
[03:48:56] <speedy1> I'm counting fragments which passed via occlusion query
[03:53:02] *** rutski has quit IRC
[03:56:04] *** StephaneLenclud has quit IRC
[04:19:50] *** ajww has quit IRC
[04:35:59] *** Ademan has joined ##OpenGL
[04:44:12] *** kaotrix has quit IRC
[04:51:31] *** TheLorax has quit IRC
[05:00:14] *** blight_ has quit IRC
[05:06:48] *** peda__ has joined ##OpenGL
[05:08:09] *** speedy1 has quit IRC
[05:09:10] *** ecks has joined ##OpenGL
[05:09:27] <ecks> hello, how would i be able to print out a GLfloat variable?
[05:11:14] <HuntsMan> the same way you print a float variable
[05:11:50] <ecks> HuntsMan : well, i cant print it using printf
[05:12:15] <HuntsMan> why not?
[05:12:34] <ecks>  error: cannot convert 'GLfloat' to 'const char*' for argument '1' to 'int printf(const char*, ...)
[05:12:45] <HuntsMan> and the code is?
[05:13:26] <ecks> printf(windowWidth);
[05:13:52] <HuntsMan> that's crearly wrong
[05:14:01] <HuntsMan> should be printf("%f", windowWidth);
[05:14:29] <HuntsMan> as printf always takes the format string then the variable argument list
[05:14:56] <ecks> oh thanx, its been a while since i used C
[05:15:17] *** TheLorax has joined ##opengl
[05:16:51] <ecks> oh, it seems that opengl's default output is not the terminal
[05:17:32] <HuntsMan> default output of what?
[05:17:48] <ecks> standard output i mean
[05:17:56] <ecks> stdout
[05:18:34] <HuntsMan> opengl doesn't control stdout
[05:20:51] *** m4ggus has joined ##opengl
[05:23:03] *** braden has quit IRC
[05:23:45] *** mm765^away is now known as mm765
[05:24:33] *** peda_ has quit IRC
[05:32:46] *** braden has joined ##OpenGL
[05:38:41] *** odietsch has quit IRC
[05:44:01] *** adamp has joined ##opengl
[05:58:29] *** ahelon has quit IRC
[06:03:09] *** amz has quit IRC
[06:06:25] *** rnx has left ##opengl
[06:10:27] *** Lemml has joined ##OpenGL
[06:15:08] *** amalon has quit IRC
[06:37:33] *** Maerz has quit IRC
[06:38:16] *** Maerz has joined ##OpenGL
[06:45:19] *** BahamutZERO has quit IRC
[06:57:32] *** Lemml has quit IRC
[06:58:32] *** Lemml has joined ##OpenGL
[07:00:22] *** mm^away has joined ##opengl
[07:15:25] *** Killari has quit IRC
[07:17:34] *** mm765 has quit IRC
[07:19:57] *** Killari has joined ##OpenGL
[07:46:55] *** charlie_zzz is now known as charlie5
[07:47:38] *** ALPSINC has joined ##OpenGL
[07:47:56] <ALPSINC> hi all... i need a tad bit of help
[07:48:26] <TheLorax> just ask
[07:48:35] <ALPSINC> heheh it's kind of long
[07:48:54] <TheLorax> just ASK! :P
[07:49:28] <ALPSINC> this map: gtkradient isn't rendiering a 'map preview' the whole way... as it stops at a certan point... I had a diffrent graphics card before, and it was working fine, but now... well... ya
[07:49:39] <ALPSINC> s/map/program
[07:49:51] <ALPSINC> and on windows ftw
[07:50:00] <TheLorax> that's not an opengl problem
[07:50:17] <TheLorax> no idea
[07:50:49] <ALPSINC> ahh
[07:50:50] <ALPSINC> ok..
[07:51:08] <TheLorax> it's open source, just take a look
[07:51:14] <ALPSINC> thanks for the help though :)
[07:51:30] <ALPSINC> yea... just looking into that...
[07:51:33] <ALPSINC> :) thanks
[07:51:36] *** ALPSINC has left ##OpenGL
[08:15:48] *** TheLorax has quit IRC
[08:35:44] *** fargiolas has joined ##OpenGL
[08:39:40] *** fargiolas has quit IRC
[08:44:24] *** neoneye has joined ##OpenGL
[08:45:27] *** sohail has quit IRC
[08:53:47] *** blbmp has left ##OpenGL
[08:54:30] *** Yustme has joined ##opengl
[08:56:06] *** fargiolas has joined ##OpenGL
[08:56:51] *** dv_ has quit IRC
[08:57:34] *** dv_ has joined ##OpenGL
[09:08:51] *** Justme has joined ##opengl
[09:17:58] *** Yustme has quit IRC
[09:23:01] *** Yustme has joined ##opengl
[09:38:22] *** Justme has quit IRC
[09:40:18] *** Suprano has joined ##OpenGL
[09:43:45] *** DobosCake has quit IRC
[09:54:19] *** Walt has quit IRC
[10:31:02] *** adamp has quit IRC
[10:32:05] *** adamp has joined ##opengl
[10:57:36] *** fargiolas has quit IRC
[10:58:00] *** carlopalma has joined ##OpenGL
[11:00:15] *** BahamutZERO has joined ##OpenGL
[11:00:26] *** kandinski has quit IRC
[11:00:28] *** kandinski has joined ##OpenGL
[11:04:21] *** carlopalma has quit IRC
[11:10:19] *** ced117 has joined ##OpenGL
[11:14:38] *** fargiolas has joined ##OpenGL
[11:22:30] *** madsy_ has joined ##opengl
[11:32:22] *** Madsy has quit IRC
[11:35:50] *** m4ggus_ has joined ##opengl
[11:36:03] *** karabash has joined ##OpenGL
[11:39:07] *** ajww has joined ##OpenGL
[11:51:46] *** m4ggus has quit IRC
[11:54:18] *** Walt has joined ##opengl
[12:02:42] *** amalon has joined ##opengl
[12:06:16] *** darka has joined ##opengl
[12:09:45] *** snoffler has joined ##opengl
[12:26:51] *** adamp has quit IRC
[12:27:15] *** adamp has joined ##opengl
[12:36:20] *** Suprano has quit IRC
[12:36:24] *** marenz_ has joined ##OpenGL
[12:36:54] *** marenz__ has joined ##OpenGL
[12:42:40] *** sanjin has joined ##OpenGL
[12:42:59] <sanjin> hi
[12:46:36] *** Burga has quit IRC
[12:47:23] *** groton has joined ##OpenGL
[12:47:34] *** doggliewr has joined ##opengl
[12:48:18] <sanjin> is there something like glReadPixels for textures? (i don't want glGetTexImage because i only need a specific rectangle of the texture)
[12:56:05] *** doggliewr has quit IRC
[12:59:06] *** peda_ has joined ##OpenGL
[13:08:06] *** snoffler has quit IRC
[13:08:31] *** Ademan has quit IRC
[13:12:06] *** marenz__ has quit IRC
[13:12:08] *** karabash has quit IRC
[13:16:53] *** peda__ has quit IRC
[13:19:21] *** fargiolas_ has joined ##OpenGL
[13:23:34] *** hypernewbie has joined ##OpenGL
[13:32:27] *** blight_ has joined ##opengl
[13:40:16] *** fargiolas has quit IRC
[13:41:49] *** boghog has joined ##OpenGL
[13:43:21] *** LordMetroid has joined ##OpenGL
[13:52:04] *** fargiolas_ has quit IRC
[13:53:56] *** fargiolas has joined ##OpenGL
[13:58:29] *** m4ggus_ has quit IRC
[14:00:36] *** Watermelon2 has joined ##OpenGL
[14:03:00] *** Gudy has quit IRC
[14:09:28] *** IJs\laptop has joined ##OpenGL
[14:14:01] *** amalon_ has joined ##opengl
[14:16:44] *** sanjin has left ##OpenGL
[14:17:35] *** Lemml has quit IRC
[14:21:18] *** jh559 has joined ##opengl
[14:21:34] *** [this] has joined ##opengl
[14:22:46] *** Lemml has joined ##OpenGL
[14:24:41] *** amalon has quit IRC
[14:31:43] *** amalon_ has quit IRC
[14:32:29] *** adamp has quit IRC
[14:33:11] *** adamp has joined ##opengl
[14:34:12] *** Xmas| has joined ##OpenGL
[14:40:09] *** rnx has joined ##opengl
[14:51:23] *** hypernewbie has quit IRC
[14:54:39] *** stuckie has joined ##opengl
[14:54:48] *** [this] has quit IRC
[14:59:03] *** prophile has joined ##opengl
[14:59:54] *** KU0N has joined ##OpenGL
[15:00:58] <KU0N> hello
[15:04:53] *** bsod2_ has quit IRC
[15:11:33] *** karabash has joined ##OpenGL
[15:12:01] <boghog> hi KU0N
[15:16:46] *** rutski has joined ##OpenGL
[15:26:42] *** mapreduce has quit IRC
[15:26:50] *** mapreduce has joined ##OpenGL
[15:38:35] *** Walt_ has joined ##opengl
[15:39:41] *** chondrite has joined ##OpenGL
[15:43:21] *** BadChoice has joined ##OpenGL
[15:43:31] <BadChoice> hy everyone,
[15:43:43] <BadChoice> does anyone know how to load jpg textures into opengl?
[15:44:36] <Watermelon2> ...
[15:44:37] <Makegho> the same way you load any other textures into opengl
[15:44:57] *** marenz_ has quit IRC
[15:45:25] <BadChoice> I load bmp and I have a variable Image whre I load bmp
[15:45:44] <Watermelon2> ...
[15:45:55] <Watermelon2> i dont understand that sentence..
[15:46:01] <KU0N> you must use a jpg loader
[15:46:04] <BadChoice> sorry
[15:46:08] <KU0N> you can't load a jpg as a bmp
[15:46:10] <Watermelon2> 'I load bmp and I have a variable Image whre I load bmp'...
[15:46:34] <BadChoice> I found a function to load BMP
[15:46:43] <BadChoice> but it can't load jpg
[15:46:47] <KU0N> yea but you CAN'T use it to load jpg
[15:46:48] <Makegho> (don't be hostile :)
[15:46:49] <Watermelon2> is it part of winapi?
[15:47:07] <BadChoice> no, i'm using linux
[15:47:11] <Watermelon2> hmmm
[15:47:16] <Watermelon2> SDL?
[15:47:20] <BadChoice> glut
[15:47:33] <BadChoice> is there any jpg loading library or function compatible with opengl?
[15:47:41] <Watermelon2> libjpeg...
[15:47:46] <KU0N> BadChoice: there is no "compatible with opengl"
[15:47:56] <KU0N> it will works as long as you can load the data in CPU memory
[15:48:41] *** Walt has quit IRC
[15:48:51] <BadChoice> but doesn't it must have some specifiyc format ?
[15:49:05] <Watermelon2> char*
[15:49:58] <Watermelon2> images are identical after loaded into memory
[15:50:11] <Watermelon2> widht/height and bpp might vary
[15:50:15] <Watermelon2> but still same stuff
[15:50:55] <BadChoice> its an tri-dimensional array with each rgb values and the widht and heihg
[15:51:04] <BadChoice> mmm
[15:51:37] <Watermelon2> ...
[15:53:02] <quicksilver> BadChoice: there are a few choices in in memory image format, and opengl supports the obvious ones, as parameters when you call TexImage and friends.
[15:53:14] <quicksilver> most image loaders these days load into 32bit RGBA, it's pretty standard.
[15:53:36] <BadChoice> wuo, really thanks
[15:53:55] <BadChoice> I need a image loadre like libjpeg
[15:54:15] <BadChoice> then load the image and make it as a texture with loadtexture funcion ?
[15:54:44] <Watermelon2> http://www.ijg.org
[16:02:34] *** Suprano has joined ##OpenGL
[16:05:18] *** joakim_ has joined ##OpenGL
[16:14:39] <boghog> BadChoice, I use DevIL myself, but I haven't used it for anything other than opengl test programs
[16:15:05] <boghog> http://openil.sourceforge.net/
[16:15:22] <boghog> one nice thing about it is that it's API is very opengl-like
[16:16:42] <BadChoice> boghog, I'll take a look, now I'm dealing with ligjpeg
[16:18:56] *** Xmas| has quit IRC
[16:21:26] *** LiQuiDninja has quit IRC
[16:23:24] *** joakim__ has quit IRC
[16:36:47] *** odietsch has joined ##OpenGL
[16:37:48] <prophile> i tend to use SDL_image on windows/linux and ImageIO on os x
[16:47:43] *** charlie5 is now known as charlie_zzz
[16:48:53] <Watermelon2> he is not using SDL...
[16:49:06] <BadChoice> no.. I'm using glut
[16:49:25] <BadChoice> but if sdl it's better I cane give it a try
[16:50:05] <Watermelon2> not better but more 'complete'
[16:50:47] <BadChoice> mmm
[16:51:02] <BadChoice> and performance?
[16:51:06] *** kaotrix has joined ##OpenGL
[16:51:25] <Watermelon2> peformance of what?
[16:51:30] <Watermelon2> performance rather
[16:52:51] <BadChoice> of 3D
[16:52:56] <BadChoice> I mean wich is fast?
[16:53:00] <Watermelon2> ...
[16:53:06] <Watermelon2> both are using opengl
[16:53:13] <Watermelon2> there is no difference
[16:54:12] *** juanmabc has joined ##opengl
[16:55:44] *** dv_ has quit IRC
[16:56:26] *** dv_ has joined ##OpenGL
[16:57:02] <BadChoice> ah
[16:57:52] *** ecks has quit IRC
[17:00:30] *** CNU has quit IRC
[17:01:02] <BadChoice> then when use glut and when use sdl?
[17:01:08] *** NightBird has quit IRC
[17:01:23] <quicksilver> lots of people will say 'never use glut'
[17:01:31] <odietsch> both serve the same purpose
[17:01:38] <quicksilver> glut is quite a simple way to get started and is bundled as standard on macos and linux
[17:01:44] <quicksilver> sdl is considerably more capable
[17:01:58] *** chondrite has quit IRC
[17:02:02] <quicksilver> (better keyboard / joystick handles, more utilities like loading images, drawing fonts, etc)
[17:02:13] <BadChoice> uhmm...
[17:02:26] <BadChoice> I'll try using sld then
[17:02:37] <BadChoice> it will make life easier
[17:10:18] <rnx> sdl doesnt draw fonts
[17:11:23] <groton> while GLUT does
[17:13:35] <quicksilver> rnx: well, sdl_ttf does
[17:14:13] <quicksilver> GLUT has some simple font capabilities but they're not very useful to be honest.
[17:16:40] <rnx> there being an optional extension library that creates text that you can then load as an texture by yourself and draw via opengl doesnt exactly count as "sdl draws fonts"
[17:17:48] <quicksilver> well, I was simplifying :)
[17:18:10] <quicksilver> but the ability to render decent fonts is certainly a common request.
[17:19:33] <rnx> using something like ftgl from glut is less work
[17:19:56] <rnx> not that there is anything wrong with sdl ... i use it myself
[17:21:32] *** juanmabc_ has joined ##opengl
[17:22:03] *** juanmabc has quit IRC
[17:22:14] *** juanmabc_ is now known as juanmabc
[17:25:13] *** ahelon has joined ##opengl
[17:45:50] *** snoffler has joined ##opengl
[17:54:46] *** TheLorax has joined ##opengl
[17:56:17] *** snoffler has quit IRC
[18:00:53] *** BadChoice has left ##OpenGL
[18:12:43] *** juanmabc has quit IRC
[18:17:52] *** fargiolas_ has joined ##OpenGL
[18:19:20] *** fargiolas has quit IRC
[18:21:18] *** rnx has quit IRC
[18:22:39] *** rnx has joined ##opengl
[18:23:52] *** madsy_ has quit IRC
[18:26:46] *** Madsy has joined ##opengl
[18:28:19] *** DobosCake has joined ##OpenGL
[18:32:43] *** amz has joined ##opengl
[18:37:24] *** Amorphous has joined ##opengl
[18:48:31] *** Jup4 has joined ##OpenGL
[18:48:49] *** Jupp3 has quit IRC
[18:48:50] *** Jup4 is now known as Jupp3
[18:51:51] *** LordMetroid has quit IRC
[18:53:08] *** neoneye has quit IRC
[18:57:07] *** DobosCake has quit IRC
[18:57:25] *** DobosCake has joined ##OpenGL
[18:58:49] *** m4ggus has joined ##opengl
[19:00:26] *** DobosCake has quit IRC
[19:01:28] *** DobosCake has joined ##OpenGL
[19:01:29] *** Walt_ has quit IRC
[19:01:56] *** amz has quit IRC
[19:02:23] *** amz has joined ##opengl
[19:14:48] *** juanmabc has joined ##opengl
[19:17:25] *** DobosCake has quit IRC
[19:17:30] *** Lemml has quit IRC
[19:20:19] *** DobosCake has joined ##OpenGL
[19:24:11] *** Suprano has quit IRC
[19:24:35] *** DobosCake has quit IRC
[19:25:10] *** DobosCake has joined ##OpenGL
[19:29:16] *** DobosCake has quit IRC
[19:29:33] *** karabash has quit IRC
[19:31:45] *** sohail has joined ##OpenGL
[19:32:48] *** Lucine2 has quit IRC
[19:37:49] *** kiras has joined ##OpenGL
[19:45:01] *** m4ggus has quit IRC
[19:45:25] *** Lucine2 has joined ##OpenGL
[19:50:29] *** charlie_zzz has quit IRC
[19:58:54] *** Lekane has quit IRC
[20:17:55] *** chondrite has joined ##OpenGL
[20:20:53] <rutski> I know that glGetError() should be call in a loop to check all error flags
[20:21:09] <rutski> but what if I call glGetError() after every single OpenGL call?
[20:21:20] <rutski> is it then guaranteed that it'll only ever return 1 error flag?
[20:22:12] <Madsy> Nope. It can return several
[20:22:19] *** sanjin has joined ##OpenGL
[20:22:22] <rutski> hmm
[20:22:28] <Madsy> That's why you call it recursively until it returns GL_NO_ERROR
[20:22:50] <rutski> right, but did you get the point about it returning GL_NO_ERROR before function call A
[20:23:04] <Madsy> Say, a function could return both GL_INVALID_ENUM, GL_INVALID_OPERATION and GL_OUT_OF_MEMORY
[20:23:15] <rutski> in other words, if it returns GL_NO_ERROR before gl call A, then can it possibly return multiply flags right after call A?
[20:23:20] <rutski> ah, right; ok
[20:23:22] <rutski> gotcha
[20:23:34] <sanjin> is there any nice way to save only a part ( a rectangle ) of a texture in a local array? (like glGetTexImage does for the whole texture)
[20:23:41] <Madsy> The next API call resets the error state IIRC.
[20:23:42] <Satan_Inside> always call it within a loop as Madsy stated -- http://paste.lisp.org/display/23737
[20:23:54] *** charlie5 has joined ##OpenGL
[20:24:52] <Satan_Inside> sanjin simply render the sub-rectangle and capture via glGetTexImage( ... )
[20:25:03] *** Suprano has joined ##OpenGL
[20:25:23] <sanjin> well, my problem is that i want it not to change any render states...
[20:25:38] <sanjin> im actually writing a c++ stream for gpu textures used as buffer for gpgpu ;p
[20:27:29] <Satan_Inside> so? I fail to see why changing a select number of render state(s) might be a problem here
[20:28:02] <Satan_Inside> in any case if you only want to capture a sub-rectangle what I described above is your only option
[20:28:48] *** neoneye has joined ##OpenGL
[20:28:56] *** karabash has joined ##OpenGL
[20:30:21] <sanjin> well, i will need to change matrix transformation states, set up a vertex array and will have to change the bound texture .. only to get some bytes of my texture ... all this whole gpgpu stuff will propably run side by side with the rendering engine so this might be costly :<
[20:30:24] <Satan_Inside> another option is simply masking a sub-rectangle via glReadPixels( ... )
[20:30:40] <Satan_Inside> the PBO specification illustrates how to do this
[20:30:52] <Satan_Inside> (using the fast path for FB captures)
[20:31:03] <sanjin> so i will need a framebuffer to use glReadPixels to read from my texture?
[20:31:24] <Satan_Inside> correct
[20:31:49] <Satan_Inside> glReadPixels( ... ) reads pixels directly from the framebuffer
[20:34:06] <sanjin> thats nice... so i can just use a framebuffer and access my texture via it ... do you know if it is costly to have a framebuffer for every texture-buffer ( i want to use them for things like force/acceleration/velocity or other GPGPU stuff) ?
[20:36:46] <Satan_Inside> well you have a linear increase in the number of mutable surface textures the driver now has to allocate, manage and ultimately page on or off as needed
[20:37:11] <Satan_Inside> so its not necessarily cheap from that perspective
[20:37:52] *** charlie5 has quit IRC
[20:37:54] <sanjin> yes but i think i will need a framebuffer anyway for rendering into the buffers... i will propably provide a stream for both ... a framebuffer and a texture itself
[20:38:16] <sanjin> thanks for you help Satan_Inside !
[20:38:18] <Satan_Inside> due to this you end up using more VRAM... how much depends upon the texture types / formats, the texture / surface dimensions and the number of textures
[20:38:37] <Satan_Inside> no problem
[20:40:54] *** blbmp has joined ##OpenGL
[20:45:05] *** TheLorax has quit IRC
[20:47:12] *** amz has quit IRC
[20:47:41] *** charlie5 has joined ##OpenGL
[20:50:58] *** _boto has joined ##opengl
[20:51:20] *** amz has joined ##opengl
[20:53:06] *** charlie5 has quit IRC
[20:56:35] *** DobosCake has joined ##OpenGL
[20:58:17] *** Nate75Sanders has joined ##OpenGL
[20:59:18] *** charlie5 has joined ##OpenGL
[21:00:35] *** DobosCake has quit IRC
[21:03:02] *** Nate75Sanders has quit IRC
[21:12:17] *** DobosCake has joined ##OpenGL
[21:15:18] *** DobosCake has quit IRC
[21:17:15] *** Vixus has joined ##OpenGL
[21:17:54] <Vixus> Hey does anyone have links to any good masking+texturing tutorials or are NeHe the best ones?
[21:22:24] <Madsy> Why do you need "tutorials" for that?
[21:22:43] <Madsy> It should be pretty obvious how it works.
[21:23:17] <Madsy> It works just like alpha blending would.
[21:24:19] <Madsy> You render a grayscale image in the first pass, so that dst contains black and white. Then you enable blending wih GL_DST_COLOR, GL_ZERO
[21:24:49] <Madsy> with*
[21:26:54] *** LiQuiDninja has joined ##OpenGL
[21:27:12] <Vixus> ok, so say I have a texture, then an image mask that defines the shape of that texture in black and white.. how would that work?
[21:28:34] *** neoneurone has joined ##OpenGL
[21:30:42] <Madsy> Like I said.
[21:31:26] *** DMINATOR has joined ##OpenGL
[21:31:31] <Madsy> If *that's* too difficult for you, I recommend start with the basics first. (read: software rendering, NOT OpenGL)
[21:32:16] <Watermelon2> ...
[21:32:51] *** chondrite has quit IRC
[21:33:34] *** chondrite has joined ##OpenGL
[21:35:44] *** DobosCake has joined ##OpenGL
[21:44:19] *** Watermelon2 has quit IRC
[21:45:01] *** NeoThermic has quit IRC
[21:46:17] <Vixus> well manipulating the images beforehand is easy enough
[21:46:21] *** Vixus has quit IRC
[21:48:23] *** Walt has joined ##opengl
[21:51:19] *** Jup4 has joined ##OpenGL
[21:51:30] *** Jupp3 has quit IRC
[21:51:31] *** Jup4 is now known as Jupp3
[21:54:26] *** karabash has quit IRC
[21:54:57] *** chondrite has quit IRC
[22:06:13] *** magnet__ has joined ##OpenGL
[22:06:42] *** chrisjw has joined ##opengl
[22:06:57] *** NeoThermic has joined ##OpenGL
[22:11:49] *** JernejL_ has quit IRC
[22:11:55] *** JernejL has joined ##OpenGL
[22:15:01] *** Suprano has quit IRC
[22:15:11] *** marenz_ has joined ##OpenGL
[22:17:41] *** kiras has quit IRC
[22:20:34] *** magnet_ has quit IRC
[22:22:46] *** marenz_ has quit IRC
[22:23:12] *** marenz_ has joined ##OpenGL
[22:23:26] *** speedy1 has joined ##OpenGL
[22:28:03] *** karabash has joined ##OpenGL
[22:31:48] *** fargiolas_ has quit IRC
[22:35:28] *** neoneurone has quit IRC
[22:40:01] *** mm^away is now known as mm765
[22:41:10] *** thesquib has joined ##OpenGL
[22:42:23] *** Lekane has joined ##OpenGL
[22:43:32] *** adamp has quit IRC
[22:44:15] *** adamp has joined ##opengl
[22:48:30] *** charlie55 has joined ##OpenGL
[22:48:58] *** charlie5 has quit IRC
[22:49:03] *** charlie55 is now known as charlie5
[22:54:42] *** adamp has quit IRC
[22:55:06] *** adamp has joined ##opengl
[22:55:25] *** Walt_ has joined ##opengl
[23:01:15] *** Lacerta-2 has joined ##OpenGL
[23:01:33] <Lacerta-2> evening
[23:02:16] <Lacerta-2> can someone help me with my FBO problem?
[23:02:35] *** kiras has joined ##OpenGL
[23:02:50] <Lacerta-2> i can't seem to find a working texture format combination
[23:04:18] <Lacerta-2> a moment ago i made a function that tests all possible combinations (citation needed)
[23:05:15] <Lacerta-2> i got about 60k combination and every single one give me "missing attachment" FBO status
[23:05:55] <Lacerta-2> now i bet the error is somewhere else entirely but i've run out of ideas :/
[23:06:35] <Lacerta-2> i'm using GF6600 GT on Arch Linux, repository drivers
[23:09:55] <speedy1> if a simple FBO with one pow2 RGBA 32bit texture attached does not work, you probably have a driver or system problem
[23:10:54] <speedy1> try if any of the FBO samples from the internet work
[23:10:59] <Lacerta-2> nope
[23:11:00] *** Walt has quit IRC
[23:11:05] <Lacerta-2> haven't got anything to work
[23:11:43] <speedy1> then it's most likely a driver / system problem
[23:12:38] *** LordMetroid has joined ##OpenGL
[23:12:54] <Lacerta-2> internal/format/type -> RGBA8 / RGBA / UNSIGNED_BYTE is supposed to work anywhere?
[23:13:27] <Lacerta-2> i'd like to know some combination that works on most implementations
[23:13:39] <Lacerta-2> so i could just focus on getting that to work
[23:20:37] *** _boto has quit IRC
[23:21:40] <sanjin> can FRAMEBUFFER_INCOMPLETE_ATTACHMENT_EXT be caused by using unsupported formats (like LUMINANCE8 ) or not?
[23:22:08] <sanjin> ( i'd say only GL_FRAMEBUFFER_INCOMPLETE_FORMATS_EXT would be returned then )
[23:22:30] <Lacerta-2> that will give "unsupported" status
[23:23:04] <Lacerta-2> come to think of it i didn't get any "unsupported" status.. even though i should have
[23:23:17] <Lacerta-2> i'll double check my code
[23:23:49] <sanjin> so what can cause GL_FRAMEBUFFER_INCOMPLETE_FORMATS_EXT ?
[23:24:01] <sanjin> sorry, typo: i meant GL_FRAMEBUFFER_INCOMPLETE_ATTACHMENT_EXT
[23:24:59] <sanjin> maybe its because of my GL_LUMINANCE8 internal format?
[23:28:48] <sanjin> bye
[23:29:44] *** sanjin has left ##OpenGL
[23:48:38] *** Lemml has joined ##OpenGL
[23:50:21] *** odietsch has quit IRC

   May 10, 2008  
< | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | 31 | >