May 29, 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 | >
May 29, 2008 [00:02:04] *** TehLorax has joined ##opengl[00:02:17] *** TheLorax has quit IRC[00:04:23] *** ajww_ has quit IRC[00:07:07] *** ajww__ has quit IRC[00:09:50] *** tmccrary has quit IRC[00:17:55] *** elite01 has quit IRC[00:23:05] *** pfo_atruediger is now known as pfo[00:25:25] *** aalex_ has joined ##OpenGL[00:25:44] *** jparishy has joined ##OpenGL[00:27:04] *** KU0N has quit IRC[00:29:30] *** TehLorax has quit IRC[00:29:35] *** RTFM_FTW has joined ##OpenGL[00:29:41] *** TheLorax has joined ##opengl[00:34:16] *** danieljames_ has quit IRC[00:34:46] *** groton has quit IRC[00:36:03] *** Walt has quit IRC[00:40:22] *** aalex_ has quit IRC[00:40:47] *** rutski has quit IRC[00:49:10] *** bbeausej has quit IRC[00:55:25] *** rutski has joined ##OpenGL[00:56:37] *** mm765 is now known as mm765^away[01:02:50] *** WhitAngl has joined ##OpenGL[01:04:14] *** scy has left ##opengl[01:05:56] <WhitAngl> Hi! I was wondering: when rendering a GL_LINE with a large width, how are the attributes of the point interpolated? I mean, if my first vertex has an attribute with value 0 and the other vertex 1, where is the iso value 0.5 in my fragment program ? 2 circles around my vertices ? the mediator ? anything else ?[01:06:53] <WhitAngl> In fact, in my fragment shader I'm looking how far away from the line I am[01:08:07] *** prophile has quit IRC[01:11:13] <Plagman_> it's just linear[01:13:11] <WhitAngl> linear along the line.. but is it also linear along the perpendicular of the line ?[01:13:42] <WhitAngl> I mean my line has a width, it's not a single pixel line[01:13:44] <RTFM_FTW> yep its all linearly interpolated[01:14:24] <WhitAngl> then perpendicular to the line, at my first vertex, what does it interpolate ?[01:14:35] <WhitAngl> I'll do a drawing[01:15:36] *** bbeausej has joined ##OpenGL[01:18:23] <Plagman_> yes, it's linear along the perdendicular[01:18:42] <Plagman_> you have to think of wide lines like quads, the two edge points having duplicated attributes[01:19:58] <WhitAngl> www.bonneel.com/interp.png[01:20:11] <WhitAngl> something is disturbing me[01:20:49] <WhitAngl> right at the first vertex my value is 0.. then with which value can the shader interpole perpendicularly to the line ?[01:20:56] *** prophile has joined ##opengl[01:21:29] <Plagman_> no I wasn't clear[01:21:34] <WhitAngl> you need 2 values to interpolate linearly.. and if the two values are the same (duplicated 0 for example), then it turns out to be a constant interpolation[01:21:41] <Plagman_> it doesn't interpolate anything along the perdendicular[01:21:49] <Plagman_> I wasn't meaning that[01:22:06] <Plagman_> I meant that it was interpolating linearly along the two points of the line with the same values along the perpendicular[01:22:30] <Plagman_> as I said you have to think of wide lines like quads, with two vertices of the same attributes on each edge[01:22:59] <WhitAngl> oh ok.. so I have no way to get an idea on how far I am from the line ? I would like my lines to be shaded like thin cylinders[01:23:30] <Plagman_> no[01:23:34] <Plagman_> use quads[01:23:43] <Plagman_> two quads if you want cylinders[01:23:48] <Plagman_> with a seam along the center[01:24:43] <WhitAngl> ok, thanks :)[01:30:02] <WhitAngl> mmm.. nope.. when my camera will look perpendicularly to the cylinder, it will vanish[01:30:16] <WhitAngl> I'll need 2 quads at least[01:30:24] <WhitAngl> oops, that what you said ;)[01:32:33] <Plagman_> exactly[01:32:46] <Plagman_> but I meant two quads for each axix[01:32:48] <Plagman_> axis*[01:33:24] <WhitAngl> for each axis?[01:33:26] *** prophile has quit IRC[01:33:31] *** m4ggus has quit IRC[01:33:35] *** m4ggus_ has joined ##OpenGL[01:33:37] *** m4ggus_ is now known as m4ggus[01:33:39] <Plagman_> because you won't be able to shade a single quad to look like a "cylinder" using only fixed function colors[01:33:50] <Plagman_> you need to quads for each "side"[01:33:56] <WhitAngl> oh no, I shade my quads myself ;)[01:34:03] <Plagman_> ah, okay then[01:36:16] *** Dew420 has quit IRC[01:36:39] *** MatthiasM has quit IRC[01:36:47] *** MatthiasM has joined ##opengl[01:39:49] *** korff has quit IRC[01:40:48] *** Walt has joined ##opengl[01:40:57] *** jparishy has quit IRC[01:42:53] *** speedy1 has quit IRC[01:43:40] *** amz has joined ##opengl[01:43:52] *** thomash_ has quit IRC[01:44:05] *** thomash_ has joined ##OpenGL[01:44:21] *** speedy1 has joined ##OpenGL[01:44:59] <speedy1> glYoda, good news, ATI released new extensions for OGL in the latest drivers :) now I can't flame you that much ;)[01:47:00] <speedy1> uh? I don't think you do that part - t'was just very good news and a joke :)[01:49:23] <speedy1> aye[01:50:54] <speedy1> very cool of you for lingering on IRC if you really work for ATI[01:51:14] *** Dew420 has joined ##OpenGL[01:52:10] <speedy1> kk :)[01:52:21] <RTFM_FTW> since it always manages to turn into a complete flame-war[01:53:02] <speedy1> yeah, just skip it - doesn't really matter that much[01:53:31] *** Dew420 has quit IRC[01:55:36] *** juanmabc has quit IRC[01:55:54] *** Dew420 has joined ##OpenGL[01:59:23] *** RTFM_FTW_ has joined ##OpenGL[02:04:23] *** jparishy has joined ##OpenGL[02:08:13] *** Andos has joined ##OpenGL[02:08:44] <Andos> Hi[02:09:11] *** RTFM_FTW_ has quit IRC[02:10:46] <Andos> Can anyone sketch out for me what the practical difference between attaching a color Renderbuffer to a FBO and attaching a normal texture to the FBO ?[02:17:07] *** RTFM_FTW has quit IRC[02:17:12] *** jparishy has quit IRC[02:19:08] *** elektr1k|afk has joined ##opengl[02:19:08] *** elektr1k has quit IRC[02:26:04] *** adhv has quit IRC[02:27:15] *** RTFM_FTW has joined ##OpenGL[02:34:11] *** Xmas| has quit IRC[02:36:09] *** jparishy has joined ##OpenGL[02:37:42] *** jparishy has quit IRC[02:47:30] *** TheLorax has quit IRC[03:01:15] *** Maerz has quit IRC[03:01:49] *** dvoid has quit IRC[03:01:59] *** thomash_ has quit IRC[03:02:05] *** thomash_ has joined ##OpenGL[03:03:51] *** Tenac has joined ##OpenGL[03:03:53] <Tenac> What is a better name for my class that controls all the the interface objects, the cursor, the GUI, and HUD... UIElement or UIComponent, or maybe something else[03:04:07] <speedy1> ui_manager[03:05:00] <speedy1> or ui_core[03:05:08] *** Morphous has joined ##opengl[03:06:10] *** paragon has quit IRC[03:07:12] <Tenac> well, I'd like to think of a gui button and cursor as a Component of the interface, so that just seems better to me but idk if it makes sense or not[03:07:44] *** huperniketes has quit IRC[03:08:43] *** tschona has joined ##OpenGL[03:10:51] *** speedy1 has quit IRC[03:20:47] *** Amorphous has quit IRC[03:30:33] <WhitAngl> if I pass the vertices positions as attributes of my vertices, should their value be exactly the same as gl_Vertex in my vertex shader, or are they transformed somewhere before ?[03:30:41] *** rnx has left ##opengl[03:33:25] *** huperniketes has joined ##OpenGL[03:34:48] *** Dew420 has quit IRC[03:38:33] *** Dew420 has joined ##OpenGL[03:43:07] *** huperniketes has quit IRC[03:43:13] *** thomash_ has quit IRC[03:43:28] *** thomash_ has joined ##OpenGL[03:44:39] *** huperniketes has joined ##OpenGL[03:58:30] *** Ragnarok has joined ##OpenGL[03:59:25] <Ragnarok> would u recommand the GLSL shader files embedded in headers etc?[03:59:41] <Ragnarok> even though u would need to recompile each time a change[04:00:22] *** TheChuckster has joined ##OpenGL[04:03:33] *** mm765^away is now known as mm765[04:06:12] *** NeoThermic has quit IRC[04:07:07] *** jerkface03 has joined ##OpenGL[04:09:19] *** neunon has quit IRC[04:09:21] *** neunon_ has joined ##OpenGL[04:09:26] *** m4ggus has quit IRC[04:10:41] *** neunon_ is now known as neunon[04:11:22] *** m4ggus has joined ##opengl[04:20:41] *** blbmp has quit IRC[04:22:26] *** WhitAngl2 has joined ##OpenGL[04:24:06] *** tschona has left ##OpenGL[04:32:27] *** ginoman has joined ##OpenGL[04:33:10] <ginoman> http://3dcodingtutorial.com/Working-with-3D-models/Exporting-the-model.html[04:38:05] *** neunon has quit IRC[04:38:22] *** neunon_ has joined ##OpenGL[04:39:11] *** WhitAngl has quit IRC[04:39:42] *** cit has joined ##OpenGL[04:48:54] *** brutopia has quit IRC[04:48:56] *** Killari has quit IRC[04:49:02] *** RTFM_FTW has quit IRC[04:49:02] *** Andos has quit IRC[04:49:04] *** enoex_ has quit IRC[04:49:07] *** hibread has quit IRC[04:49:09] *** bagu has quit IRC[04:49:11] *** Andon has quit IRC[04:49:11] *** druggy_ has quit IRC[04:49:11] *** smiler has quit IRC[04:49:15] *** Lucine has quit IRC[04:49:15] *** bobbens has quit IRC[04:49:31] <Tenac> This probabaly isn't important, but when I load a texture, which way should I say the filepath: "Textures/Image.png", "Textures\Image.png", or "Textures\\Image.png"[04:52:26] *** hibread has joined ##opengl[04:57:19] *** WhitAngl has joined ##OpenGL[04:57:25] *** RTFM_FTW has joined ##OpenGL[04:57:25] *** Andos has joined ##OpenGL[04:57:25] *** brutopia has joined ##OpenGL[04:57:25] *** enoex_ has joined ##OpenGL[04:57:25] *** bagu has joined ##OpenGL[04:57:25] *** Andon has joined ##OpenGL[04:57:25] *** druggy_ has joined ##OpenGL[04:57:25] *** Lucine has joined ##OpenGL[04:57:25] *** bobbens has joined ##OpenGL[04:57:25] *** smiler has joined ##OpenGL[04:57:43] *** Killari has joined ##OpenGL[05:01:06] <Ragnarok> Tenac, /[05:01:43] <Ragnarok> only use // when escaping / unless ur string class in ur language does it for you automatically[05:01:45] *** neunon_ is now known as neunon[05:01:59] <Ragnarok> i know C# does[05:05:07] *** WhatAHam has joined ##OpenGL[05:05:14] <WhatAHam> so how do i query the refresh rate?[05:05:37] *** cit has quit IRC[05:07:05] *** cit has joined ##OpenGL[05:07:50] *** cit has quit IRC[05:07:57] *** Inside has joined ##OpenGL[05:07:59] <Inside> Hey guys \o/[05:08:12] <Inside> Can you explain to me how exactly setting vertex colors in a textured mesh help?[05:08:17] *** cit has joined ##OpenGL[05:15:47] *** WhitAngl2 has quit IRC[05:21:50] <Tenac> ok Ragnarok, ty[05:21:57] <Ragnarok> np[05:27:20] *** juanmabc has joined ##opengl[05:42:04] *** ginoman has left ##OpenGL[05:42:30] *** ginoman has joined ##OpenGL[05:45:21] *** bfrog has joined ##opengl[05:45:28] *** Andos has quit IRC[05:45:36] <bfrog> wow, I really had no clue how graphics hardware worked before... I atleast have a vague idea now[05:45:41] <bfrog> very cool shite[05:47:46] *** Ragnarok has quit IRC[05:52:44] *** Walt has quit IRC[06:04:07] *** Ragnarok has joined ##OpenGL[06:04:12] *** m4ggus has quit IRC[06:04:18] <Ragnarok> what would cause a glClearColor to not work?[06:10:09] <cit> how can i speed up glTextSubImage2D?[06:18:26] <RTFM_FTW> use native alignment, use a native internalFormat, format and type and use a optimal upload / submission path[06:18:55] <RTFM_FTW> GL_ARB_pixel_buffer_object or GL_APPLE_client_storage, GL_APPLE_texture_range for example[06:24:36] *** mastro has quit IRC[06:24:58] *** bbeausej has quit IRC[06:43:48] *** mediogre has joined ##OpenGL[06:48:08] <cit> i didn't find pixel buffer on the iphone, tnx =)[06:48:19] *** WhatAHam has quit IRC[06:50:13] *** mm765 is now known as mm765^sleep[06:50:30] *** servus_ has joined ##opengl[06:51:25] *** servus has quit IRC[06:51:36] *** servus_ is now known as servus[06:55:28] *** blbmp has joined ##OpenGL[07:00:30] *** mm^away has joined ##opengl[07:02:04] *** Dew420 has quit IRC[07:09:20] *** scrav has quit IRC[07:09:45] *** aalex__ has joined ##OpenGL[07:10:11] *** Dew420 has joined ##OpenGL[07:11:55] *** neunon has quit IRC[07:17:38] *** mm765^sleep has quit IRC[07:21:34] *** juanmabc has quit IRC[07:31:10] *** Dew420 has quit IRC[07:33:53] *** bfrog has quit IRC[07:40:57] *** Scutt has joined ##OpenGL[07:43:43] *** amz has quit IRC[07:45:04] *** aalex__ has quit IRC[07:47:51] *** KU0N has joined ##OpenGL[07:48:21] <KU0N> hello[07:48:38] <KU0N> what's up?[07:59:44] <Jupp3> everything that isn't down[08:00:13] <KU0N> hehe[08:01:18] *** servus_ has joined ##opengl[08:02:13] <KU0N> I wonder, I have some spaceships, and I wonder which shadow method I should use.[08:02:59] <KU0N> Each ship will cast shadow on itself (sun being light source) other ships (close ones) will cast shadow as well[08:03:15] *** scy has joined ##opengl[08:04:32] *** Inside has quit IRC[08:17:10] *** servus has quit IRC[08:27:46] *** Jupp3 has quit IRC[08:32:04] *** KU0N has quit IRC[08:32:11] *** cit_ has joined ##OpenGL[08:36:38] *** Roderic has joined ##OpenGL[08:36:40] *** Roderic is now known as Ingenu[08:40:20] *** blbmp has quit IRC[08:46:28] *** cit has quit IRC[08:46:49] *** scy has left ##opengl[08:53:20] *** Tenac has quit IRC[08:53:36] *** Tenac has joined ##OpenGL[08:58:50] *** Sisu has quit IRC[09:02:44] *** Sisu has joined ##opengl[09:05:21] *** blbmp has joined ##OpenGL[09:10:39] *** michele_m has joined ##OpenGL[09:14:41] *** servus_ is now known as servus[09:17:57] *** t4bz has joined ##OpenGL[09:18:02] *** [AD]Turbo has joined ##OpenGL[09:18:43] *** predaeus has joined ##opengl[09:19:44] <[AD]Turbo> yo[09:21:14] *** sohail has quit IRC[09:37:01] *** belou has joined ##OpenGL[09:39:27] *** cit_ has quit IRC[09:39:32] *** cit has joined ##OpenGL[09:44:02] *** avacore^ has joined ##OpenGL[09:46:54] *** avacore^ has quit IRC[09:47:20] *** avacore^ has joined ##OpenGL[09:56:28] *** avacore has quit IRC[09:58:12] *** avacore^ is now known as avacore[10:14:46] *** hd_ has joined ##OpenGL[10:18:51] *** hd_ has left ##OpenGL[10:19:22] *** gotan666 has joined ##OpenGL[10:25:41] *** blbmp has quit IRC[10:25:43] *** kenws has joined ##OpenGL[10:26:48] *** andikr has joined ##OpenGL[10:29:51] *** [this] has joined ##opengl[10:40:37] <belou> aloha !![10:43:37] *** Obfuscate has quit IRC[10:44:15] *** adhv has joined ##OpenGL[10:45:48] <michele_m> I have a c# gui that passes the window handle to a c++ dll which contains my opengl engine[10:45:56] <michele_m> the handle is correct[10:46:10] <michele_m> but getDC generates an GL_INVALID_OPERATION[10:46:55] <michele_m> the fun thing is that getDC does NOT return a NULL value[10:46:55] <michele_m> :S[10:49:50] <michele_m> any ideas?[10:52:03] <Ingenu> no clue[10:55:43] *** groton has joined ##OpenGL[10:55:45] <predaeus> michele_m, how can getDC generate a GL error if it is totally unrelated to GL?[10:56:21] <michele_m> then why gl error is set after calling getdc? O_o[10:56:35] <quicksilver> maybe it was set before, too...[10:57:00] <michele_m> nope, I flush all the error before doing anything[10:57:19] <predaeus> michele_m, do you setup the context?[10:57:46] <predaeus> sorry if asking obvious things, sometimes helps[10:58:22] <michele_m> you mean choosing pfd and such?[10:58:26] <predaeus> yes[10:58:37] <michele_m> right after getting the dc, so yes[10:58:55] <michele_m> however it's failing[10:59:44] <michele_m> what I'm afraid is that it's failing because I'm using windows as a virtualized os[11:00:08] <michele_m> gotta restart in order to be sure of that[11:02:56] <michele_m> mm[11:03:15] <michele_m> I think gl_invalid_operation is set because I don't even have a context[11:04:04] *** dvoid has joined ##OpenGL[11:25:12] *** WhitAngl2 has joined ##OpenGL[11:35:59] *** brutopia has quit IRC[11:36:04] *** brutopia has joined ##opengl[11:42:31] *** brutopia has quit IRC[11:43:56] *** WhitAngl has quit IRC[11:44:29] *** Suprano has joined ##OpenGL[11:51:16] *** Ademan_ has joined ##OpenGL[11:51:19] *** elite01 has joined ##opengl[11:51:47] *** Ademan has quit IRC[11:51:52] *** Ademan_ is now known as Ademan[11:57:58] *** brutopia has joined ##opengl[12:00:27] *** brutopia has quit IRC[12:06:18] *** brutopia has joined ##opengl[12:21:45] *** brutopia has quit IRC[12:26:06] *** groton has quit IRC[12:26:06] *** gotan666 has quit IRC[12:26:13] *** nytejade has quit IRC[12:26:15] *** Wagoo has quit IRC[12:26:19] *** keitsi has quit IRC[12:27:05] *** WhitAnglAtWork has quit IRC[12:27:05] *** groton has joined ##OpenGL[12:27:05] *** gotan666 has joined ##OpenGL[12:27:07] *** nytejade has joined ##OpenGL[12:27:07] *** Wagoo has joined ##OpenGL[12:27:07] *** keitsi has joined ##OpenGL[12:27:28] *** brutopia has joined ##opengl[12:36:37] *** LordMetroid has joined ##OpenGL[12:36:57] *** brutopia has quit IRC[12:42:37] *** brutopia has joined ##opengl[12:44:57] *** rsaltini has joined ##OpenGL[12:50:51] *** peda_ has quit IRC[12:50:59] *** peda_ has joined ##OpenGL[12:55:28] *** WimLeers has joined ##OpenGL[12:57:42] *** LordMetroid has quit IRC[12:58:20] *** LordMetroid has joined ##OpenGL[13:01:18] *** Suprano has quit IRC[13:02:20] *** TheChuckster has quit IRC[13:02:45] *** brutopia has quit IRC[13:07:22] *** michele_m has quit IRC[13:08:09] *** brutopia has joined ##opengl[13:09:26] *** neoneye has joined ##OpenGL[13:14:41] *** brutopia has quit IRC[13:15:02] *** Suprano has joined ##OpenGL[13:20:20] *** brutopia has joined ##opengl[13:21:53] *** lordmetroid_ has joined ##OpenGL[13:24:31] *** lordmetroid_ has quit IRC[13:30:54] *** Walt has joined ##opengl[13:31:45] *** XzN`HydrAv|1x has quit IRC[13:38:33] *** LordMetroid has quit IRC[13:39:57] *** arikel has joined ##OpenGL[13:41:08] <arikel> Hello, does anyone compile c++/opengl programs with gcc on linux and could tell a real noob what's the line to use to compile? I get " error: GL/glut.h: No such file or directory " errors...[13:41:10] *** ginoman has quit IRC[13:41:32] <arikel> i tried -lGL within the gcc line, but it didn't work[13:43:39] <ville> arikel: look up -I, -L and -l flags from the GCC man page.[13:44:29] <arikel> okay, thanks ville[13:45:37] *** prophile has joined ##opengl[13:47:34] *** fargiolas has joined ##OpenGL[13:50:03] *** WimLeers has quit IRC[13:50:08] *** WimLeers_ has joined ##OpenGL[13:54:43] <quicksilver> arikel: that error has nothing to do with link flags.[13:54:52] <quicksilver> arikel: it means you don't have glut.h installed[13:54:57] <quicksilver> (or installed in the wrong place[13:55:08] *** WhitAnglAtWork has joined ##OpenGL[13:55:42] <quicksilver> you want to install what debian calls freeglut3-dev[13:56:06] <quicksilver> your distribution may call it something different[13:56:23] <WhitAnglAtWork> Hi! If I abruptly close my program without releasing any GPU ressources, does it create a memory leak on my graphic card ? I mean, I just sent several millions of vertices on my GPU, killed my program and now I cannot run it again normally[13:58:03] <Weiss> WhitAnglAtWork: i've often wondered that myself - i think it depends on how good the driver is[13:58:26] *** predaeus has quit IRC[13:58:26] <WhitAnglAtWork> last time I had to reboot my computer to get my program work again[14:00:39] *** Walt has quit IRC[14:00:44] <Weiss> that would make sense to me[14:00:58] <Weiss> but there are other people in here who can talk better about driver architecture...[14:01:19] <WhitAnglAtWork> they remain quiet ;)[14:02:51] <WhitAnglAtWork> brb, I need to reboot[14:03:14] *** WhitAnglAtWork has quit IRC[14:05:00] *** brutopia has quit IRC[14:05:57] *** brutopia has joined ##opengl[14:06:23] <WimLeers_> gluBuild2DMipmaps() works for me, but my entire scene gets very dark. If I comment it out (and ONLY that line), textures stop working, but the scene goes back to normal lighting. Any ideas?[14:08:26] *** rnx has joined ##opengl[14:09:19] *** WhitAnglAtWork has joined ##OpenGL[14:12:18] *** brutopia has quit IRC[14:13:56] <arikel> quicksilver: thanks a lot sir! On fedora it's called "freeglut-devel", but you got it, thank you.[14:14:08] *** Alexa has joined ##OpenGL[14:14:59] *** Dark_Angel has joined ##opengl[14:15:04] <quicksilver> godo![14:15:09] <Dark_Angel> Hello[14:15:25] *** Rangar has quit IRC[14:15:30] <Dark_Angel> I need some help in opengl..can anyone help me?[14:15:54] <Dark_Angel> i only want to learn how to make a cup of tea[14:16:23] <quicksilver> 1) boil kettle; 2) put teabag in cup; 3) pour water on teabag[14:16:34] <quicksilver> that should get you started. we can cover advanced techniques later.[14:18:03] *** brutopia has joined ##opengl[14:18:34] *** PainBank has joined ##OpenGL[14:18:48] <Dark_Angel> i only want to darw it..to make it like i made a teapot[14:19:33] <quicksilver> I was joking.[14:19:52] <quicksilver> there is a lot of opengl to learn before you can draw a cup, anyhow.[14:20:00] <Dark_Angel> i know[14:20:02] <quicksilver> you will want to start with some basic tutorials.[14:20:07] <Dark_Angel> i have learned smth so far[14:20:27] <Dark_Angel> i know that u can make it with call List..[14:20:27] <quicksilver> simplest way to make a complex object like a cup is probaly to design it in a modelling program.[14:20:32] <quicksilver> I use blender.[14:20:40] <Dark_Angel> or blender[14:20:42] <quicksilver> then you need to pick a model format, save it and load it into your program.[14:21:02] <quicksilver> you can try to draw it by hand if you prefer.[14:21:04] <Dark_Angel> and where can i find that model?[14:21:20] <quicksilver> most people find the maths of calculating vertex coords and normals by hand boring or hard or both.[14:21:31] *** neunon has joined ##OpenGL[14:21:35] <Dark_Angel> :([14:22:21] <Dark_Angel> i can only imagine[14:22:29] *** freespace has quit IRC[14:22:58] <Dark_Angel> do u have some tutorials that could explain how to draw this cup?[14:24:44] <Dark_Angel> are u still there?[14:26:50] <quicksilver> have you done the nehe tutorials?[14:26:52] <quicksilver> they are quite popular.[14:27:31] <Dark_Angel> i know about that site[14:29:53] <quicksilver> well, have you done the them?[14:31:25] <Dark_Angel> i havent had time to done all that stuf..and i dont understan some of them....ive tried..believe me[14:31:48] *** arikel has left ##OpenGL[14:34:37] <quicksilver> if you don't understand that stuff you are not going to be able to draw a teacup[14:34:42] <quicksilver> I'm sorry about that :([14:34:49] <quicksilver> but you cannot run before you can walk.[14:35:58] *** WimLeers_ has quit IRC[14:36:13] *** WimLeers has joined ##OpenGL[14:36:18] <Dark_Angel> i know..[14:36:26] <Dark_Angel> but i found smth else from another source[14:36:29] *** Vixus has joined ##OpenGL[14:36:37] <Dark_Angel> and it was easier than on nehe[14:37:06] <Vixus> Has anyone here used SOIL for loading textures?[14:38:07] <WhitAnglAtWork> I use mkOpenGLJPEGImage SDK, it's fairly easy to use[14:39:29] <Vixus> Well it's a more a case of not being able to use the actual library because of compile issues.[14:39:54] <Vixus> I don't want to switch to another library just go come up against the same wall.[14:40:00] <Vixus> *to[14:40:13] <WhitAnglAtWork> mkOpenGLJPEGImage provides prebuilt binaries ;)[14:40:29] <Vixus> so does this unfortunately.[14:41:12] <Vixus> and I don't want to go near JPEGs.[14:41:29] <WhitAnglAtWork> it also loads bmp[14:41:39] <WhitAnglAtWork> and 5 minutes integration into my code :p[14:42:58] <Vixus> I think I'm forgetting to tell g++ to link stuff..[14:45:32] <quicksilver> unless you show us the error we can only guess :P[14:45:37] <quicksilver> pastebin preferred.[14:45:54] <quicksilver> Dark_Angel: the opengl books are quite good, in my opinion.[14:46:09] <quicksilver> Dark_Angel: they're available online, see topic.[14:48:15] <Vixus> Sure[14:49:43] <quicksilver> Dark_Angel: as I said, see topic.[14:51:05] <Dark_Angel> ok[14:51:58] <Vixus> http://pastebin.com/d2927ea80[14:52:59] <quicksilver> Vixus: are you correctly bracketing the SOIL headers in extern "C" ?[14:53:17] <Vixus> probably not[14:53:19] *** eduard has joined ##OpenGL[14:53:31] <eduard> hello there![14:53:49] <Vixus> is it because I'm using a C++ compiler?[14:53:57] <quicksilver> that is my first guess, at least.[14:54:09] <quicksilver> when compiling C and C++ together, the C++ parts need to know that the C is, well, C[14:54:18] <quicksilver> otherwise it tries to link it using C++ calling conventions.[14:54:24] <Vixus> ah ok[14:54:42] <Vixus> so do I need to put it in SOIL.h?[14:55:38] *** bolides has joined ##OpenGL[14:55:48] <Vixus> the file already has #ifdef __cplusplus // extern "C" { // #endifdef[14:55:56] <Vixus> *#endif[14:56:13] <Vixus> so I need to define __cplusplus Xp[14:57:51] <Vixus> quicksilver, Ok, none of that changed anything.[14:58:29] <quicksilver> not that then[14:58:34] <quicksilver> odd.[14:59:45] <Vixus> does undefined reference just mean it's not able to find the library/link it?[14:59:59] <quicksilver> no[15:00:08] <quicksilver> it means it can't find that particular symbol[15:01:17] <Vixus> hmm[15:02:47] <Vixus> no idea.[15:03:10] *** eduard has quit IRC[15:03:17] <Weiss> quicksilver: which format do you export to from Blender for your programs? i tried using Wavefront OBJ format, but Blender's export seems to have broken normals and inconsistent polygon winding[15:04:11] <quicksilver> I'm using OBJ[15:04:16] <quicksilver> seems fine to me :)[15:05:23] <Weiss> hmm.. interesting[15:05:25] <quicksilver> of course, it can't do animation so I'll have to investigate something else.[15:05:33] <quicksilver> but for straight quads, obj is working fine.[15:07:11] <Weiss> hmm.. which versioN?[15:08:36] <quicksilver> http://roobarb.jellybean.co.uk/~jules/Picture%2039.png[15:08:44] <quicksilver> Weiss: those models were all made in blender and exported as objs[15:09:14] <quicksilver> 2.45, apparently[15:09:52] <Weiss> quicksilver: very nice..[15:10:03] <Weiss> could you let me know which version of the OBJ exporter you have as well?[15:10:10] <quicksilver> I wonder how I can tell[15:10:42] <Weiss> find the script (somewhere in /usr/share i think..) and look at the comments[15:10:48] <Weiss> hang on.. i'll let you know the location[15:12:04] <quicksilver> Blender.app//Contents/MacOS/.blender/scripts/export_obj.py[15:12:08] <quicksilver> in my case :)[15:12:24] <quicksilver> version "1.1" by "Campbell Barton, Jiri Hnidek"[15:13:34] <Weiss> hmm, same here[15:14:06] <quicksilver> I think your normal loading code may be at fault then, rather than the exporter :)[15:14:16] <quicksilver> maybe you were looking normals up in the vertex list[15:14:18] <quicksilver> or vice versa[15:14:37] <Vixus> quicksilver, hehe made it work..[15:15:09] <quicksilver> Vixus: cool, what did you have to do?[15:15:34] <Vixus> make... make install.... ugh.[15:15:38] <Weiss> quicksilver: i came across a forum post a while ago with someone else complaining about the same problem.. i need to investigate more thoroghly[15:16:00] <Vixus> now I just need to learn how to use textures properly so that I don't get this massive white quad.[15:18:34] <Vixus> right, so say I've loaded a texture ID into an int..[15:19:01] <Vixus> I would use glBindTexture(GL_TEXTURE_2D, texid) to begin working with that texture?[15:19:22] *** freespace has joined ##opengl[15:19:26] <quicksilver> right.[15:20:05] <Vixus> if the texture image is not a power of 2, would opengl just display plain white? :p[15:20:17] <Vixus> image dimensions I mean[15:22:27] <quicksilver> most modern cards support non power of 2 textures.[15:22:39] <quicksilver> well, I dunno about most actually.[15:22:41] <quicksilver> mine does :)[15:22:52] *** mediogre has left ##OpenGL[15:23:01] <quicksilver> plain white can be a sign your mipmapping when you don't have mipmaps[15:23:08] <Vixus> ah[15:23:11] <Vixus> that's probably it[15:23:34] <Vixus> how do I not mipmap?[15:24:14] *** lexs has joined ##OpenGL[15:24:50] *** [this] has quit IRC[15:25:03] <lexs> if i want to store a color for every vertex in a vbo, is it good to use GL_BYTE instead of GL_FLOAT?[15:27:31] *** Alexa has quit IRC[15:27:54] <quicksilver> Vixus: when you call glTexParameter[15:28:02] <quicksilver> Vixus: don't choose the _MIPMAP_ options :)[15:28:18] <Vixus> heh, are they enabled by default?[15:28:33] <quicksilver> I'm not sure, I always make explicit calls.[15:28:47] <quicksilver> normally GL_LINEAR for both GL_TEXTURE_MIN_FILTER and MAG[15:28:51] <prophile> no u[15:28:59] <Vixus> yep[15:29:14] <prophile> lexs: GL_UNSIGNED_BYTE preferably[15:29:21] <Vixus> I found it on some strange SGI IRIX docs.[15:29:30] <prophile> for all other types of array use floats though[15:30:08] <lexs> prophile: yeah, unsigned was what i meant :)[15:30:57] <Vixus> quicksilver, hmm I still get the plain white.. maybe I'm not loading my textures properly.[15:33:12] <quicksilver> did you enable TEXTURE_2D ?[15:33:27] <quicksilver> it's worth checking out a tutorial just to see all the boilerplate[15:33:37] <quicksilver> there's lots of boring things to remember in opengl :)[15:34:29] <Vixus> would one of the nehe ones do?[15:34:31] *** hibread has quit IRC[15:36:00] *** Suprano has quit IRC[15:36:21] <quicksilver> yeah[15:37:20] *** hibread has joined ##opengl[15:38:00] <Weiss> Vixus: check the power of two thing - my card doesn't support it, and it wasn't bought too long ago (though it isn't great)[15:38:02] *** bolides has quit IRC[15:38:30] <Vixus> Haha, it works![15:38:44] <Vixus> Scales properly and everything[15:40:06] *** hibread has quit IRC[15:40:22] *** hibread has joined ##opengl[15:40:44] *** servus_ has joined ##opengl[15:41:04] <quicksilver> gluBuild2DMipmaps is a good way around the power of two thing if you need nice scaling[15:41:34] <Vixus> SOIL does that for me. : )[15:41:44] <quicksilver> you get all the mipmaps build and solves Po2[15:41:52] <quicksilver> I imagine soil solves Po2 itself.[15:41:57] <Vixus> yeah[15:42:21] <Vixus> i'd need a non-supporting graphics card to test however[15:44:03] <Vixus> Anyone played liero?[15:44:11] *** WimLeers has quit IRC[15:44:28] <hibread> So whats happening guys?[15:46:36] <prophile> well, we're breathing and our hearts are beating[15:48:00] <hibread> sounds pretty raunchy. Some women came past?[15:50:17] *** Dark_Angel has left ##opengl[15:50:32] <prophile> men.[15:50:38] *** bbeausej has joined ##OpenGL[15:52:26] *** Suprano has joined ##OpenGL[15:53:06] <Vixus> heh, I just noticed it draws my texture upside down[15:53:13] <Vixus> because bottom left is 0,0.[15:55:07] *** servus has quit IRC[15:56:07] <quicksilver> yes, that happens to most people[15:56:24] <quicksilver> just reverse the y coords in your glTexCoord calls[15:56:37] <quicksilver> althouhg soil may have a v-flip flag[15:56:39] *** elite01 has quit IRC[15:57:12] <Vixus> nah, i got it.. it would be nice if the texture and quad coords were synced though[15:59:38] *** frobnitz has joined ##opengl[15:59:40] *** elite01 has joined ##opengl[16:00:43] <elite01> are "exotic" glBlendFunc and glBlendEquation really slow? like having a few quads at 1280x800 making fps drop to just a few tens on this gf7400?[16:01:34] *** Jupp3 has joined ##OpenGL[16:02:13] <Ingenu> they aren't exotic AFAIR[16:02:22] <Ingenu> it shouldn't[16:02:35] <Ingenu> well of course blending is expensive[16:02:35] <elite01> it's GL_ONE, GL_ONE and GL_FUNC_ADD[16:02:40] <Ingenu> read modify write[16:02:41] <elite01> but *that* expensive?[16:02:50] <Ingenu> fullscreen quads ?[16:02:52] <elite01> yeah[16:03:03] <elite01> what i want is a blur effect[16:03:26] <quicksilver> using double buffers?[16:03:26] <Ingenu> just image the bandwidth required[16:03:34] <quicksilver> (front buffer blending is more expensive)[16:03:39] <Ingenu> read every pixel, modify it and write it[16:03:48] <quicksilver> but yes full screen image operations are always fairly expensive.[16:03:50] <elite01> using glCopyTexImage2D and drawing that texture a few times[16:03:55] <Ingenu> that's a lot of bandwidth, on a high-end card it should be fine though[16:04:01] <quicksilver> glCopy is fairly expensive too[16:04:08] <elite01> hm, alternatives?[16:04:10] <quicksilver> FBOs[16:04:17] <quicksilver> if your card supports them[16:04:27] <elite01> and what about drawing that many quads?[16:05:17] <elite01> a shader would be even slower, wouldn't it?[16:05:28] <Ingenu> define "that many" and "quad" and how you draw them[16:05:34] <quicksilver> shader might be faster[16:05:48] <quicksilver> shader might make it one pass instead of many passes[16:05:50] <Ingenu> prolly not, post processing is done by shaders using the scene as a texture[16:05:52] <Ingenu> most oftne[16:05:59] <quicksilver> depending if I'm understanding what you're saying.[16:06:17] <elite01> ok, at the moment, i'm drawing 9 full screen quads, using additive blending[16:06:43] *** sohail has joined ##OpenGL[16:06:43] <elite01> but a larger blur radius would be cool[16:07:21] <Ingenu> you would rather use a filter in a shader to achieve the same effect[16:07:33] <Ingenu> would ease your poor gfx card memory bus[16:08:24] <elite01> so, 9 texture2D calls for each pixel?[16:08:32] <Ingenu> yes[16:08:44] *** XzN`HydrAv|1x has joined ##OpenGL[16:08:44] <Ingenu> but since you'll be hitting the cache most of the time it won't be that expensive[16:08:50] <elite01> cool[16:08:55] <Ingenu> and you'll be saving a lot of writes[16:09:01] <Ingenu> (from 9 to 1)[16:09:15] <quicksilver> that's what I meant by one pass instead of many )[16:09:30] <Ingenu> (I got it)[16:09:32] <hibread> elite01: so you'll still have to read 9 different textures?[16:09:35] <quicksilver> one pass algorithms have better cache coherency and perhaps even better pipelining.[16:09:37] <elite01> 9 times the same texture[16:09:50] <quicksilver> although I don't understand much about pipelining on modern GPUs ;[16:09:55] <Ingenu> I assumed he was reading 9 times the same given his previous explanation[16:10:22] <elite01> so, out of interest, is every combination of glBlendFunc and glBlendEquation typically equally fast or are there differences?[16:10:37] <Ingenu> same speed[16:10:48] <hibread> elite01: so 9 times the same texture equals one time if you process everything you need to in one hit[16:10:53] <Ingenu> but as I said you don't want to use blending unless you are FORCED to[16:11:13] <elite01> well, for a 3x3 blur, every output pixel depends on 9 input pixels, doesn't it?[16:11:17] <Ingenu> yeah he's cutting his bandwidth requirement by 9[16:11:19] <RTFM_FTW> actually multi pass algorithms aren't a problem WRT pipelining and cache coherency[16:11:36] <Ingenu> depends what you want to do exactly[16:11:58] <Ingenu> yep only texkill are evil[16:12:02] <Ingenu> and alpha test[16:12:04] <RTFM_FTW> not necessarily[16:12:04] <Ingenu> and depth out[16:12:06] <Ingenu> and... ;p[16:12:11] <RTFM_FTW> again not necessarily[16:12:37] <Ingenu> yeah we've been toying with low resolution particle buffer blended to the screen last using depth writes[16:12:43] <Ingenu> it's a huge perf win[16:12:58] <RTFM_FTW> depth writes can definitely be considered "evil" in the average case but A test and T kill are easily circumvented[16:13:03] <RTFM_FTW> via the compiler[16:13:05] <Ingenu> but the brute force/simple approach result is as you expect not so good looking[16:13:07] <RTFM_FTW> and a good optimizer[16:13:46] <Ingenu> well asking a programmer to trust a compiler is a bit much ;p[16:13:56] <Ingenu> better make its job easy[16:14:41] <RTFM_FTW> well you would be wise to trust the compiler since none of you are coding directly to the underlying HW arch. :P[16:15:15] <RTFM_FTW> OTOH you shouldn't write idiotic code... that we can all agree with :P[16:15:29] <frobnitz> I have a 2D projection, and I want to draw two textured rectangles (one slightly rotated and/or translated) on top of each other. Trick is, I need to attach an "intensity" value to each texel somehow, and only draw the "stronger" texel at each overlap point.[16:16:33] <frobnitz> The intensity isn't related to the texel color. Is there a simple way to do this? I've thought about using the depth buffer to hold the intensity, but I'm not sure what the best approach is.[16:16:38] <elite01> you want a z buffer[16:16:54] <elite01> (or depth buffer)[16:17:17] <frobnitz> Is there a way to fill the z buffer without a glVertex for each point?[16:17:39] *** PainBank has left ##OpenGL[16:17:47] <RTFM_FTW> actually glVertex doesn't fill the Z buffer at all[16:17:49] <elite01> hm, just set the z coordinates of the textured rectangles[16:18:16] *** belou has quit IRC[16:19:11] <frobnitz> elite01: each point in the texture (probably 2048x2048) has its own intensity value. I was hoping to only draw one rectangle, and somehow apply a depth offset.[16:19:32] *** belou has joined ##OpenGL[16:19:32] <elite01> oh[16:19:50] *** Baba has joined ##OpenGL[16:20:08] <elite01> sounds like using a pixel shader and/or abusing the alpha component[16:20:53] <frobnitz> elite01: That's what I'm affraid of. I can do it, but I was hoping there was a simpler way.[16:21:14] <elite01> i'm sorry, i can't think of anything else either[16:21:25] <Vixus> anyone know of any guides to working with RGB colour? ie. to get the ratios of the constituents of a certain colour?[16:21:58] <elite01> Vixus, experiment with a color chooser component[16:22:04] <elite01> as found in paint, gimp, photoshop[16:22:16] <Vixus> alright[16:22:51] <Vixus> i want to get the blend of pink and white (antialiasing) to contribute to an alpha value[16:24:18] *** BahamutZERO has quit IRC[16:25:39] *** m4ggus has joined ##opengl[16:26:56] *** Baba__ has quit IRC[16:29:07] <elite01> http://pastebin.org/39457 - can you see anything wrong with that shader (3x3 blur)? it looks pretty, but i think it's much too slow[16:30:43] <Vixus> has anyone ever had a statement has no effect warning?[16:31:18] <elite01> sure[16:31:23] <Vixus> what does it mean?[16:31:35] <elite01> it's a warning that a statement has no effect[16:32:04] <Vixus> that's worrying in my case[16:32:51] <elite01> what statement is it?[16:34:19] <Vixus> http://pastebin.com/d468538e2 .. the three if statements in the for loop.[16:35:03] <Vixus> oh crap[16:35:05] <Vixus> i used ==[16:35:07] <Vixus> instead of =[16:35:12] <elite01> hehe yeah[16:37:39] <Vixus> now I get an interesting effect[16:38:06] *** higgsfett has quit IRC[16:38:19] *** higgsfett has joined ##OpenGL[16:38:30] *** nplus has joined ##OpenGL[16:43:42] <elite01> come on, make http://pastebin.org/39457 faster so http://img365.imageshack.us/img365/3950/coolnessud8.jpg doesn't run at 40fps (:[16:45:53] <Weiss> elite01: perhaps try manually unwinding those loops?[16:46:01] <elite01> thanks for the suggestion[16:46:37] <Weiss> however as RTFM_FTW says the compiler should know what to do... (?)[16:49:16] <elite01> hmm, it seems to, still jerky :/[16:50:20] <Weiss> what size of framebuffer?[16:51:04] <Weiss> my low-end card struggles a bit doing fullscreen (1680x1050) post-processing (one step) at 10fps[16:51:06] <elite01> 1280x800[16:51:17] <elite01> yeah, it's just a gf7400 for me[16:51:39] <Weiss> mine's worse (mobility radeon x600)[16:52:26] <elite01> mhm, laptops and opengl[16:52:44] <elite01> so i guess i won't get more than 40fps out of this[16:54:22] *** kaotrix has joined ##OpenGL[16:54:31] *** Vixus has quit IRC[17:02:41] *** Suprano has quit IRC[17:03:48] *** speedy1 has joined ##OpenGL[17:07:26] *** stringfellow has joined ##opengl[17:07:35] <Ragnarok> anyone tell me why my clearcolor won't work. I can't figure it out :([17:07:37] <Ragnarok> http://rafb.net/p/ejfVNI39.html[17:10:40] <hibread> Ragnarok: tried just glClearColor( 0.0f, 0.0f, 1.0f, 1.0f); ?[17:10:52] <Ragnarok> yes[17:10:54] <Ragnarok> no luck[17:11:12] <hibread> so whats happening? Yellow not coming up?[17:12:09] <Ragnarok> yeah[17:12:11] <Ragnarok> only black[17:12:25] <Ragnarok> but colouring the polygon works[17:13:22] <quicksilver> tried setting the clear color immediately before the call to clear ?[17:13:54] *** LiQuiDninja has joined ##OpenGL[17:14:30] <speedy1> ragnarok: also try disabling Qt from overwriting the background of the window - there used to be a parameter for that[17:15:12] <hibread> speedy1: i use Qt, and i dont think i ever had to do someting like that[17:16:17] <Ragnarok> somehow now it works if I use it before glClear. But I've done it like this before and its always worked[17:16:20] <Ragnarok> hmmm[17:17:03] <hibread> the context is somehow not setup during the initializeGL() ?[17:17:07] <quicksilver> someting in your framework is resetting the clear colour[17:17:19] <quicksilver> or the initialise GL is applying to a different context, yes[17:17:52] <Ragnarok> it looks exactly almost like my old code and it works perfectly there I think lemme try[17:19:13] <Ragnarok> my old code[17:19:13] <Ragnarok> http://rafb.net/p/RQY6IG46.html[17:20:02] *** aalex__ has joined ##OpenGL[17:21:34] *** mm^away is now known as mm765[17:23:38] *** Jupp3 has quit IRC[17:25:17] *** ginoman has joined ##OpenGL[17:33:42] *** aalex__ has quit IRC[17:42:56] *** neunon has quit IRC[17:43:20] *** RTFM_FTW has quit IRC[17:46:25] *** Obfuscate has joined ##opengl[17:48:09] *** HuntsMan has joined ##opengl[17:48:21] *** higgsfett has quit IRC[17:49:07] *** elite01 has quit IRC[17:50:45] *** Jupp3 has joined ##OpenGL[17:51:42] *** WimLeers has joined ##OpenGL[17:51:53] *** Jupp3 has quit IRC[17:51:53] *** Jup4 has joined ##OpenGL[17:52:48] *** Walt has joined ##opengl[17:53:10] *** gotan666 has quit IRC[17:53:27] *** speedy1 has quit IRC[17:56:50] *** gotan666 has joined ##OpenGL[17:57:49] *** bbeausej has quit IRC[17:57:51] *** bbeausej_ has joined ##OpenGL[17:58:52] *** gotan666 has quit IRC[18:05:27] *** aalex__ has joined ##OpenGL[18:07:28] *** aalex__ has quit IRC[18:07:33] *** bbeausej_ has quit IRC[18:07:35] *** bbeausej has joined ##OpenGL[18:08:06] *** WhatAHam has joined ##opengl[18:08:37] <WhatAHam> anyone remeber the GLX extension to get a framecount from the hardware? i need this to detect skipped frames..[18:09:04] *** groton has quit IRC[18:09:09] *** higgsfett has joined ##OpenGL[18:12:41] *** LordMetroid has joined ##OpenGL[18:12:43] *** lordmetroid_ has joined ##OpenGL[18:13:25] *** lordmetroid_ has quit IRC[18:13:45] *** LordMetroid has quit IRC[18:13:50] *** LordMetroid has joined ##OpenGL[18:14:49] *** WimLeers has left ##OpenGL[18:17:00] *** Yustme has joined ##opengl[18:19:07] *** belou has quit IRC[18:19:11] *** peda_ has quit IRC[18:20:07] *** [AD]Turbo has quit IRC[18:24:13] *** lordmetroid_ has joined ##OpenGL[18:26:25] *** lordmetroid_ has quit IRC[18:39:10] *** nplus has quit IRC[18:39:42] *** nplus has joined ##OpenGL[18:39:44] *** kenws has quit IRC[18:41:05] *** LordMetroid has quit IRC[18:42:28] *** TheLorax has joined ##opengl[18:45:05] *** scy has joined ##opengl[18:47:46] *** Xmas| has joined ##OpenGL[19:00:17] *** Grizzly` has joined ##opengl[19:05:25] *** kaotrix has quit IRC[19:08:15] *** pfo has quit IRC[19:08:34] *** andikr has quit IRC[19:10:07] *** Jup4 is now known as Jupp3[19:19:44] *** mccbaka has quit IRC[19:24:12] *** kiras has joined ##OpenGL[19:28:34] *** WhitAngl2 has quit IRC[19:31:19] *** nplus has quit IRC[19:31:43] *** nplus has joined ##OpenGL[19:32:24] *** fargiolas has quit IRC[19:35:34] *** BahamutZERO has joined ##OpenGL[19:45:44] *** white_gecko has joined ##OpenGL[19:46:36] <WhatAHam> anyone know the most efficient way to do this: generate an 800 x 600 checkerboard of 2x2 squares, each square having a random color? i am thinking of doing it as a series of quads specified as arrays of veretex and color data but that might be slow (lots of vertices)![19:46:47] *** futurestack has joined ##OpenGL[19:48:36] *** m4ggus has quit IRC[19:50:35] *** LordMetroid has joined ##OpenGL[19:53:22] *** WhitAngl has joined ##OpenGL[19:53:24] *** fargiolas has joined ##OpenGL[19:53:59] <TheLorax> WhatAHam, could put it in a texture[19:54:04] <TheLorax> trade verts for space[19:54:07] <WhatAHam> yeah was thinking that[19:54:20] <TheLorax> WhatAHam, do you know about vertex buffer objects?[19:54:25] *** lordmetroid_ has joined ##OpenGL[19:54:27] <WhatAHam> TheLorax: no.. i'm a n00b[19:54:49] <TheLorax> WhatAHam, they are about 1000 times faster than glVertex(..)[19:55:10] <WhatAHam> what do i look at in the big red book?[19:55:31] <WhatAHam> you mean vertex arrays?[19:55:39] <TheLorax> basically, it lets you load the model INTO the video card, then you just say glDrawArray(vbo); so that may take less space than a texture and be even faster since you dont' have to sample the texture[19:56:11] <TheLorax> not vertex arrays, and i don't own the red book[19:56:24] <prophile> I'd do that as a texture[19:56:31] <WhatAHam> hmm.. well my little models (checkerboard squares) each change color 1 per frame.. that would work?[19:56:35] <prophile> VBOs kinda work on top of vertex arrays[19:56:54] <TheLorax> you can't use a texture if they change color then[19:57:05] <WhatAHam> prophile: can you modify textures that are already defined? or do i need to do a texture subimmage? i am essentially generating a new texture once per frame..[19:57:23] <prophile> so glTexSubImage2D it into place once you regenerate it[19:57:25] *** Xmas| has quit IRC[19:57:42] <prophile> or you could use a FBO and a shader and regenerate it on the GPU[19:57:50] <WhatAHam> FBO?[19:57:55] <prophile> frame buffer object[19:57:57] <WhatAHam> shader?[19:57:59] <WhatAHam> gah[19:58:02] <WhatAHam> i ned to read up on this stuff[19:58:02] <WhatAHam> haha[19:58:07] <WhatAHam> what's a shader?[19:58:09] <WhatAHam> hah[19:58:14] <prophile> um[19:58:28] <prophile> www.google.co.uk/search?q=shaders[19:58:33] <prophile> www.google.co.uk/search?q=shaders+OpenGL[19:58:40] <WhatAHam> hehe[19:58:56] *** amalon has joined ##opengl[19:58:58] *** HuntsMan has quit IRC[19:59:23] *** m4ggus has joined ##opengl[19:59:47] <prophile> WhatAHam: also I want to introduce you to GL_TEXTURE_RECTANGLE_ARB[19:59:50] <prophile> which removes the POT requirement[20:00:03] <WhatAHam> POT? power of two?[20:00:05] <WhatAHam> hehe[20:00:07] <WhatAHam> man[20:00:10] <WhatAHam> i am SUCH a n00b, but thanks[20:00:36] <prophile> yes, POT = power of two[20:00:36] <WhatAHam> woah so i can get this to run ON the graphics card?[20:00:38] <WhatAHam> n33t[20:00:40] <prophile> yes[20:02:56] <WhatAHam> so like i have to create a Cg program or something?[20:11:04] *** LordMetroid has quit IRC[20:13:12] *** Grizzly` has quit IRC[20:16:56] *** lordmetroid__ has joined ##OpenGL[20:17:09] *** Grizzly` has joined ##opengl[20:17:17] * Grizzly` bows to the room[20:19:12] <Makegho> greetings.[20:19:53] <Grizzly`> With lighting disabled, if i draw a polygon with, for example, glColor3f(1.0,0.0,0.0) or glColor3b(127,0,0), why will the face not be collored 255/0/0, b ut 246/0/0? similar things happen to the other two color channels...[20:20:26] <Grizzly`> white isn't white :/[20:21:18] <Grizzly`> if i clear the ColorBuffer with white (1.0,1.0,1.0) and draw a white polygon witgh glColor 1.0,1.0,1.0 i can see a distinct light-rose polygon[20:21:43] <Grizzly`> Is there anything i can do to fix that?[20:24:45] *** rutski has quit IRC[20:27:50] *** ViRUS has joined ##opengl[20:28:08] <Grizzly`> No ideas...[20:28:36] *** lordmetroid__ has quit IRC[20:29:07] <Jupp3> Grizzly`: Becouse OpenGL doesn't guarantee how the resulting image will look on various hardware and software combinations[20:30:03] <WhatAHam> read about opengl correctness.. i am a noob so i dunno but i remember seeing that in that big red book[20:30:29] <Grizzly`> that is a nuisance when trying to do picking though :/[20:31:12] <WhatAHam> oh yuo wanna figure out where the mouse is?[20:31:17] <WhatAHam> like what object?[20:31:42] *** TheLorax has quit IRC[20:33:15] <Grizzly`> I have up to 6 classes of objects with up to 15 objects each, my idea was to render them into the picking buffer with the class index as red value and the number within the class as green value (or similar) and then use the color value that glReadPixels returns under the mouse button directly to index the model i clicked...[20:33:34] *** lordmetroid_ has quit IRC[20:33:48] <Grizzly`> but seems like i have to make a lookup tabel with color values or something instead of using the channel values as indexes directly[20:34:03] <Jupp3> OpenGL is relatively old standard. If it did guarantee same look on all hardware, it would obviously have to be "worst possible"[20:35:46] <Grizzly`> I don't care about the actual "look", usually i wouldn't swap to the pickign buffer anyway, i'm only looking at it for debuging purposes right now... but i thought with lighting switched off i would get the rgb values i set for glColor[20:37:09] <Grizzly`> doesn't matter, i'll somehow manage. there are other problems :) for example why do my textured faces sometimes only reflect ambient but no diffuse light (and sometimes both, as they should)? material (non textured) objects always work fine...[20:39:43] *** replor_ has quit IRC[20:39:46] <Jupp3> No, you most definitely will not get "rgb value you set for glColor"[20:39:51] *** pfo has joined ##OpenGL[20:40:22] <Jupp3> You get a color that hardware/software decides to use with given values[20:40:32] <Jupp3> Which might be more or less what you specified[20:40:32] <Grizzly`> actually it's the other way round, it always reflects ambient, but sometimes only diffuse[20:40:34] *** rooly has quit IRC[20:41:11] <Grizzly`> Jupp3, okay. I got you.[20:44:34] *** blbmp has joined ##OpenGL[20:45:52] <Jupp3> Grizzly`: I guess most hardware uses floats internally for colors[20:47:02] <Grizzly`> probably. Really, I got the point about the correctness about the colors. my mind has already moved on to the next completely unrelated problem. I'll find a solution for the picking stuff anyway[20:47:16] <Grizzly`> it just would have been the lazy way if it worked.[20:51:39] *** mastro has joined ##OpenGL[20:55:09] *** neoneye has quit IRC[21:04:25] *** dv_ has joined ##OpenGL[21:06:17] <jerkface03> can someone quickly give me a brief outline of what an orthographic view is?[21:10:47] <rnx> gives you a parallel projection ... no perspective[21:12:38] <jerkface03> rnx i'm not sure what that means, do you know if there's a screenshot comparing a normal view iwth an orthographic view?[21:16:16] <rnx> quite likely[21:16:33] *** fargiolas has quit IRC[21:16:40] *** groton has joined ##OpenGL[21:24:45] *** WhitAngl has quit IRC[21:29:17] *** rsaltini has quit IRC[21:30:12] *** rsaltini has joined ##OpenGL[21:30:12] *** Jernej has joined ##OpenGL[21:30:23] <lexs> anyone have a good reference or something to implement a matrix class?[21:30:32] *** Xmas| has joined ##OpenGL[21:41:05] *** nplus has quit IRC[21:48:06] *** JernejL has quit IRC[21:49:34] *** WhitAngl has joined ##OpenGL[21:55:59] *** mm765 has quit IRC[21:57:15] *** kiras has quit IRC[21:57:30] *** gotan666 has joined ##OpenGL[21:57:52] *** mm765 has joined ##opengl[22:01:10] *** scy has left ##opengl[22:01:33] *** SmokingRope has joined ##OpenGL[22:03:37] <prophile> lexs: you have to be careful with those[22:03:42] <prophile> when your values escape...[22:04:02] <prophile> they then start to invade[22:04:04] <prophile> steal away other values[22:04:26] <prophile> and then there's The One who prematurely frees the entire array[22:05:19] *** speedy1 has joined ##OpenGL[22:05:19] *** rsaltini has quit IRC[22:05:59] *** rsaltini has joined ##OpenGL[22:12:23] *** groton has quit IRC[22:15:34] *** ViRUS has quit IRC[22:25:32] *** Ingenu has quit IRC[22:32:56] *** Yustme has quit IRC[22:38:45] *** juanmabc has joined ##opengl[22:40:14] *** prophile has quit IRC[22:46:59] *** prophile has joined ##opengl[22:48:01] *** elite01 has joined ##opengl[23:24:02] *** WhitAngl2 has joined ##OpenGL[23:25:10] *** gotan666 has quit IRC[23:27:14] *** white_gecko has quit IRC[23:28:14] *** Jupp3 has quit IRC[23:42:34] *** WhitAngl has quit IRC[23:43:34] *** mm765 is now known as mm765^away[23:44:25] *** cit has quit IRC[23:44:29] *** cit has joined ##OpenGL[23:48:24] *** BahamutZERO has quit IRC[23:50:30] *** BahamutZERO has joined ##OpenGL[23:51:24] *** ginoman has quit IRC[23:54:46] *** deniz_ has left ##OpenGL