January 6, 2009 
< | 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 | >
January 6, 2009 [00:02:25] *** dvoid_ has quit IRC[00:03:02] *** dvoid_ has joined ##OpenGL[00:03:58] *** rutski_ has quit IRC[00:04:48] *** dusted has quit IRC[00:06:57] *** rutski_ has joined ##OpenGL[00:09:29] *** cplusplus has joined ##OpenGL[00:12:39] *** jfroy|work has quit IRC[00:14:33] *** forrestv has joined ##OpenGL[00:14:52] <forrestv> is there anything i can put in a display list that can translate to (0, 0, 0) ?[00:15:49] <b0000> glLoadIdentity();[00:15:56] *** pist0l-fish has quit IRC[00:16:58] <forrestv> b0000, i mean while keeping the current rotation[00:17:08] *** itewsh has quit IRC[00:18:47] *** Suprano has quit IRC[00:18:54] <b0000> cant you keep the rotation in the display list and call glLoadIdentity() before you draw it?[00:23:36] <forrestv> b0000, the rotation changes so it's outside ..[00:23:49] <forrestv> i'm trying to draw a skybox[00:24:00] <forrestv> with a display list[00:24:15] <forrestv> and the object doesn't have access to the camera direction[00:25:32] <b0000> ive not used display lists in years, but surely you can just keep all matrix manipulation outside the dl?[00:25:45] <yates> why are my triangles being reflected about the x axis? http://galois.digitalsignallabs.com/wxwaterfall/[00:26:19] *** cplusplus3 has quit IRC[00:28:16] <yates> actually they're being reflected about the line y = -1, and the reflections are all black (the non-reflected triangles are texture-mapped to a 1d texture)[00:28:41] <yates> s/line/plane/[00:30:50] <yates> see a snapshot at http://galois.digitalsignallabs.com/wxwaterfall/snapshot.jpg[00:30:59] <yates> test[00:31:04] *** Morphous has quit IRC[00:31:56] *** Morphous has joined ##opengl[00:32:09] *** androoid has quit IRC[00:34:11] *** calav3ra has joined ##opengl[00:38:04] *** Kasu has joined ##OpenGL[00:38:35] *** calav3ra has quit IRC[00:39:00] *** jfroy|work has joined ##OpenGL[00:41:02] *** calav3ra has joined ##opengl[00:41:36] *** vade has joined ##OpenGL[00:47:33] *** pist0l-fish has joined ##OpenGL[00:47:47] <pist0l-fish> hello. Are calls to glPush/PopMatrix() expensive?[00:48:32] *** calav3ra_ has quit IRC[00:50:03] <TheFlash> No, in my opinion (and in the implementation i am using).[00:51:52] <yates> MatthiasM: I need your help[00:51:58] <MatthiasM> ?[00:52:06] <yates> why are my triangles being reflected about the x axis? http://galois.digitalsignallabs.com/wxwaterfall/[00:52:13] <yates> see a snapshot at http://galois.digitalsignallabs.com/wxwaterfall/snapshot.jpg[00:52:37] <yates> actually they're being reflected about the plane y = -1, and the reflections are all black (the non-reflected triangles are texture-mapped to a 1d texture)[00:52:52] <yates> any ideas?[00:54:10] <yates> i wasn't seeing it at first because i originally had my background color set to black[00:55:11] <yates> i've disabled lighting (the call to InitMaterials) and don't transfer norms[00:55:43] <MatthiasM> you are most likly creating out of range texture coords - either <0 or >1[00:55:59] <MatthiasM> and you set wrap mode != repeat[00:56:27] <yates> ok, i'll check. don't think so, but i'll check.[00:56:28] <yates> thanks[00:56:33] *** neoneye has quit IRC[00:56:46] <MatthiasM> your code is too complex from looking over it[00:56:53] *** yosafbridge has joined ##opengl[00:56:58] <MatthiasM> you you have to step through it and inspect the values[00:58:52] <yates> right[00:59:17] <yates> Star Trek Enterprise computer: "Working..."[01:01:09] *** AlastairLynn has joined ##opengl[01:01:09] *** AlastairLynn has quit IRC[01:01:38] <yates> shit[01:03:14] <yates> oh man, stupid mistake.[01:04:51] <yates> yes, i was generating data out-of-range. thanks MatthiasM.[01:05:00] <MatthiasM> :)[01:05:01] <yates> all is right with the world, after all...[01:06:28] <b0000> MatthiasM: whats that file type? me and my system do not recognise the extension[01:07:00] <MatthiasM> java webstart - download it and start it with "javaws <jnlp-file>"[01:07:04] <yates> i too often get into a bad mindset that i've done something really, really convoluted and/or the system has a bug, and it turns out to be the obvious.[01:07:24] *** AlastairLynn has joined ##opengl[01:07:28] <MatthiasM> yates: it's mostly save to assume that the system and or libs are ok[01:08:32] <b0000> ...unless ati are involved[01:08:35] <yates> MatthiasM: Yah. I've been developing embedded code based on poor tools / new silicon too much.[01:08:49] <MatthiasM> heh :DD[01:09:02] <MatthiasM> yates: you don't need to tell me about ES silicium :([01:09:40] <yates> TI should give me royalties on the TMS320C55xx for filing so many bug reports.[01:10:08] <yates> I (Ericsson) was using the 55x for at least a year before they announced the part...[01:10:40] <yates> javaws isn't available in my fedora yum repos - got a source for it?[01:11:16] <MatthiasM> hmm - I only worked with the TMS320C6701[01:11:29] <yates> that's a complex machine[01:11:39] <yates> i worked with the dm642 - very very complex[01:12:00] <yates> no freaking stack and no call / return instructions! yowie![01:12:01] <MatthiasM> javaws is part of Sun's JRE[01:12:53] <yates> does the 6701 have the EDMA controller too?[01:13:30] <MatthiasM> I think yes - but it's already over 5 years ago[01:14:27] <yates> so i have to install by hand the Sun JRE?[01:14:36] <yates> i can't find any of this stuff in a yum repo[01:14:41] <MatthiasM> yates: no - they have an installer[01:15:07] <MatthiasM> you can also install it as user in your home folder[01:15:33] <yates> is the first item here what i need? http://java.sun.com/javase/downloads/index.jsp[01:15:50] <MatthiasM> yes[01:15:56] <yates> for linux?[01:16:11] <MatthiasM> or download the JDK if you also want the compiler[01:16:13] <MatthiasM> yes[01:16:29] <MatthiasM> but select x64 or without[01:17:16] *** pist0l-fish has quit IRC[01:19:42] *** dvoid_ has quit IRC[01:29:08] *** rocketmagnet has joined ##OpenGL[01:29:27] <rocketmagnet> hi, does someone know where i can find information on how to use opengl in an client/server application ??[01:29:58] *** x0rx0r has quit IRC[01:29:59] <rocketmagnet> because i've read in the docs that some parts can be stored on the server or client side for example[01:30:12] <rocketmagnet> like vertex buffer objects are stored on the server side[01:30:25] <rocketmagnet> but how to tell who is the server and who is client ??[01:30:51] <MatthiasM> rocketmagnet: the client is your app - and the server is what executes your GL calls[01:31:07] *** Spell_ has joined ##OpenGL[01:31:30] <rocketmagnet> hm ? what does execute the opengl calls ? my app or ?[01:31:52] <MatthiasM> the gl driver[01:32:06] <rocketmagnet> oh, so this has nothing to do with network client/server ?[01:32:09] <Rolenun> rocketmagnet, take a look at a couple open source projects. you can go as basic as the graphics engine like Ogre3D or Crystalspace, or as advanced as their derivative projects[01:32:32] <MatthiasM> when you call glRoatte - then some other part of software is executing this call somewhere[01:32:39] *** duzt has quit IRC[01:33:53] <MatthiasM> rocketmagnet: when using X windows and GLX you can redirect your display to another system - and if this can handle glX then you get a network between client/server[01:34:35] *** Spell_ is now known as Spell[01:34:47] *** dv_ has joined ##opengl[01:35:57] *** HuntsMan has joined ##opengl[01:36:48] <Spell> it seams my glversion is to new for my code... but wasnt OGL going to hold the backwardcompatibly?[01:37:15] <MatthiasM> Spell: http://mikeash.com/getting_answers.html[01:37:33] *** wcstok has joined ##OpenGL[01:42:13] *** Spell has left ##OpenGL[01:45:03] *** Aintaer has joined ##OpenGL[01:58:24] *** rutski_ has quit IRC[01:58:49] *** AlastairLynn has quit IRC[01:59:13] *** AlastairLynn has joined ##opengl[01:59:54] *** AlastairLynn has quit IRC[02:00:24] *** HuntsMan has quit IRC[02:09:05] <b0000> MatthiasM: do you have much experience with tangents/bitangents?[02:09:27] <MatthiasM> a bit - I did a displacement mapped torus :)[02:09:43] <b0000> MatthiasM: i seem to be getting some seams diagonally across a normal mapped cube...i dont think its coming from the shader[02:10:16] <b0000> each face has a seam across the diagonal... the parity for the bitangent is always 1... could this be the cause?[02:10:27] <MatthiasM> a cube normaly is considers as hard edges[02:10:34] <MatthiasM> so you would split all vertices[02:10:55] <b0000> ofcourse![02:11:03] <b0000> that explains the seams![02:11:20] <b0000> and the dodgy tangent values[02:11:58] <b0000> guess i just need to rexport with unique verts and they will disappear - thanks a lot dude[02:13:25] *** johndoe has quit IRC[02:14:09] *** e-roder has joined ##OpenGL[02:14:14] <e-roder> hey[02:15:29] <e-roder> i'm trying to use the mouse to look around[02:16:25] *** Kasu has quit IRC[02:16:34] <e-roder> i'm using the windows api to get the mouse position and i'm noticing a bit of natural mouse movement...i'm not touching anything and i get a change in movement of about 10[02:17:56] <MatthiasM> turn the music volume down :)[02:18:07] <Rolenun> lol[02:18:09] <e-roder> lol[02:18:45] <Rolenun> nah, that gives me movement of about 35 :D[02:19:04] <e-roder> but really[02:19:14] <b0000> e-roder: smells like a bug to me[02:19:17] <e-roder> have you guys noticed any calibration issues with the mouse?[02:19:21] <e-roder> think so?[02:19:22] <e-roder> ok[02:19:28] <b0000> e-roder: never :D[02:19:36] <Rolenun> start by picking the mouse up off the mousepad and holding it in the air[02:19:41] <MatthiasM> e-roder: what mouse do you use ?[02:19:41] <e-roder> i can unplug it[02:19:49] <e-roder> and i still have this issue[02:19:49] <Rolenun> and wave it around like you just dont care :D[02:19:54] <Rolenun> you killed my joke :D[02:20:02] <Rolenun> \[02:20:06] <e-roder> i'm using a laptop so there's the trackpad too[02:20:20] <e-roder> but...it's really odd i think[02:20:25] <MatthiasM> disable the touchpad - it's horrible anyway[02:20:39] <b0000> e-roder: that sounds more like the cause to me, try not touching it[02:21:56] *** Xmas| has quit IRC[02:23:05] *** Ademan has joined ##OpenGL[02:28:01] *** HideousNashimoto has quit IRC[02:33:01] *** braden has quit IRC[02:33:24] *** reprore_ has joined ##OpenGL[02:34:22] *** pietia has quit IRC[02:38:11] <e-roder> im not touching it[02:38:12] <e-roder> or anything[02:38:17] <e-roder> it could be a bug in my code[02:38:33] <e-roder> i'm reading the coordinates off a window message[02:38:45] <e-roder> and then i'm using setcursorpos to recenter it[02:38:57] <e-roder> anything about that sound like it could cause a bug?[02:39:07] <b0000> e-roder: setcursorpos will trigger a mouse move message[02:39:20] <MatthiasM> well - you could still have messages in the queue when you call setCursorPos[02:39:20] <e-roder> hm[02:39:58] <e-roder> so both of these are possible issues[02:40:10] <b0000> e-roder: im certain thats ur problem, you calling setcursorpos is raising another message[02:40:16] <e-roder> but if i'm not even moving the mouse[02:40:16] <e-roder> ok[02:40:36] <e-roder> but even when i commented out the setcursorpos call, i'm pretty sure it was still spinning[02:40:37] <MatthiasM> does the mouse cursor also move ?[02:40:44] <b0000> e-roder: you need to somehow ignore that message[02:40:49] <e-roder> i've hidden the cursor[02:40:54] <MatthiasM> show it[02:41:01] <b0000> MatthiasM: well hes moving it back to the centre of the window[02:41:21] <e-roder> yeah[02:41:27] <MatthiasM> but you would see it flickering[02:41:28] <b0000> MatthiasM: which is causing the window to recieve a mouse move message, from the setcursorpos[02:41:43] <b0000> i had this exact problem just the other week[02:41:43] <e-roder> b0000 i think you're right[02:41:46] <MatthiasM> if it does not flicker there should be no events[02:41:52] <e-roder> i had an initial call to center hte mouse[02:42:02] <MatthiasM> but there are better ways to get delta movements[02:42:10] <b0000> e-roder: i know im right[02:42:11] <e-roder> so without any calls to setcursorpos it doesn't do anything[02:42:21] <b0000> MatthiasM: such as?[02:42:24] <e-roder> b0000: haha alright[02:42:49] <MatthiasM> I know that LWJGL is working without such issues when you grab the mouse[02:42:56] <e-roder> LWJGL?[02:42:58] <MatthiasM> so you could look at it's native code[02:42:59] <b0000> just as a hack ignore the next mouse move message after your call to setcursorpos[02:43:04] <MatthiasM> www.lejgl.org[02:43:10] <b0000> you will see it magically work[02:43:14] <MatthiasM> www.lwjgl.org[02:43:16] <MatthiasM> grr[02:44:01] <b0000> then you can think of a better way to ignore the message once you know its the cause[02:44:22] <e-roder> ok[02:45:16] <b0000> bool g_ignoreMouseMove = false; void mySetCursorPos( int x, int y ) { SetCursorPos( x, y ); g_ignoreMouseMove = true; }[02:45:39] <e-roder> yup[02:45:49] <e-roder> i think you're reading my source[02:46:31] *** druggy_ has joined ##opengl[02:47:13] <b0000> then in WM_MOUSE_MOVE:... if( !g_ignoreMouseMove ) { //process mouse here } else { g_ignoreMouseMove = false }[02:47:22] <e-roder> yeah[02:47:27] <b0000> that will work around the problem nicely enough to begin with[02:47:30] <e-roder> but i don't think that's working[02:47:38] *** pietia has joined ##OpenGL[02:49:08] *** Spkka has quit IRC[02:49:51] <e-roder> any other ideas?[02:50:25] <MatthiasM> e-roder: did you look at the lwjgl code ?[02:50:49] <e-roder> i don't really wanna download a whole framework just to look through their code for one thing[02:51:01] <e-roder> which is written in java anyway, so probably doesn't use the window's api[02:51:47] <b0000> e-roder: why not printf to x,y values you get each wm mouse move[02:52:03] <b0000> and also printf when you set cursor pos, you will see im correct with your problem[02:52:44] <e-roder> hmmmm[02:52:51] <e-roder> i'll try[02:53:04] <e-roder> i'm using opengl so i dunno if i can printf[02:53:09] <MatthiasM> e-roder: it uses windows API on windows[02:53:09] <e-roder> not easily anyway[02:53:50] <e-roder> MatthiasM well java does underneath, but are you certain this framework directly accesses the api?[02:54:01] <MatthiasM> yes[02:55:20] <b0000> e-roder: change your config to console app[02:55:26] <b0000> and change WinMain to main[02:55:37] <MatthiasM> or just call AllocConsole[02:55:44] <b0000> or use OutputDebugString() to print to visual studio console[02:55:56] <b0000> MatthiasM: he will need to redirect stdio in that case[02:56:06] <b0000> MatthiasM: and thats like 20 lines of code[02:56:56] <cplusplus> does someone know about head tracking?[02:57:13] <e-roder> cplusplus? hmm?[02:57:39] <mlucassmith> MatthiasM: I'm running my opengl program on my MacBook Pro and on a Vista machine. The vista machine has an Nvidia 9800 card in it. I gave the program to a friend who has a windows xp machine with an ombility x300 and the darn thing doesn't render anything. He changed the code to check, opengl is definitely running: eg he can clear the screen to a specific color, but nothing going through the shaders appears on the screen.. yet[02:57:39] <cplusplus> http://www.youtube.com/watch?v=Jd3-eiid-Uw[02:57:44] <e-roder> b0000: well it's kind of tricky because the mouse handler is set up in the winmain call[02:57:58] <e-roder> cplusplus: i don't follow youtube links, sorry[02:58:03] <b0000> e-roder: just rename it[02:58:05] <mlucassmith> MatthiasM: I'm using only generic attributes and I vaguely recall reading somewhere that there are certain attribute ids you're not meant to bind to, but I thought it was with nvidia, not ati.. my question is: how the heck am I meant to debug it?[02:58:21] <cplusplus> a 3d image on a tft looks like a 2d[02:58:56] <e-roder> cplusplus: like a sprite?[02:59:54] <e-roder> b0000: i'm not sure how to hook up a mouse listener to a console app[03:00:30] *** synchris has quit IRC[03:00:39] <cplusplus> the 3d image is rendered dependig on the position of your head[03:01:36] <MatthiasM> mlucassmith: can't help you - glVertexPointer is the only fixed attribute that I still use - but I don't bind the custom ones - I let GL do it[03:02:16] <mlucassmith> I'm letting GL do it too. May be the X300 requires at least a vertex pointer.. if so, that kinda sucks.[03:02:38] *** druggy has quit IRC[03:02:46] <MatthiasM> well - if you don't use the fixed gl_Vertex attribute - then you need to bind one to index 0[03:04:53] <mlucassmith> okay[03:05:08] <mlucassmith> it's probable that's happening anyway, but I'll add in some debug info to check[03:05:44] <MatthiasM> make sure the the right one is bound to index 0[03:05:50] <mlucassmith> indeed :)[03:06:18] <mlucassmith> the first example program only has vertex information, nothing else, so I'll ask him to check that it gets bound to index 0[03:06:32] <cplusplus> e-roder: The display properly reacts to head and body movement as if it were a real window creating a realistic illusion of depth and space[03:06:33] *** yates has quit IRC[03:06:53] <e-roder> cplusplus: i realize that but[03:07:03] <cplusplus> but?[03:07:04] <e-roder> that's similar to just using a mouse to look around[03:07:12] <e-roder> im not sure what your question is again[03:07:53] <cplusplus> it look more realistic[03:08:12] <cplusplus> Using the infrared camera in the Wii remote and a head mounted sensor bar (two IR LEDs), you can accurately track the location of your head and render view dependent images on the screen. This effectively transforms your display into a portal to a virtual environment.[03:08:13] <MatthiasM> cplusplus: just buy one of these and interface it to your app[03:08:38] <MatthiasM> cplusplus: you don't need to quote wikipedia[03:08:45] <e-roder> wait, using opengl, usually you should call translate and then call rotate right? because of the stack?[03:08:50] <cplusplus> its not wiki;)[03:09:05] <MatthiasM> e-roder: you call them in the order you want it to happen[03:09:08] <e-roder> then the rotation will actually be done first, correct?[03:09:09] <e-roder> o[03:09:11] <e-roder> really?[03:09:25] <MatthiasM> the last call is "executed" first[03:09:31] <e-roder> that's what i thought[03:09:33] <e-roder> ok[03:09:37] <MatthiasM> that's why you do the camera transforms first[03:09:43] <e-roder> i'm looking at code that is rotating before translations[03:09:57] <MatthiasM> then it moves the origin[03:11:25] <e-roder> wow gamedev.net has a lot of old iD source[03:11:30] <e-roder> thanks john carmack[03:11:55] <MatthiasM> remember to keep safety distance to old id code[03:12:04] <MatthiasM> and other old code[03:12:09] <e-roder> heh[03:12:17] <e-roder> i love carmack[03:12:24] <e-roder> it's hard to keep my hands off[03:12:34] <MatthiasM> he did a lot of optimisation which are no longer useful[03:13:15] *** cplusplus has quit IRC[03:13:19] <MatthiasM> eg in the early days the rendering itself was expensive - nowadays the batchcount is important[03:13:56] <e-roder> MatthiasM: ummm im looking at the documentation for the jgl and they have a mouse call: getDX()...i dont think you directly access the api[03:14:47] <MatthiasM> e-roder: LWJGL abstract all native details - as it run on 4 OSes with 3 (4?) kinds of instruction sets[03:15:01] <e-roder> right[03:15:08] <e-roder> i'm not abstracting[03:15:21] <MatthiasM> but once you called Mouse.setGrabbed(true) you get deltas[03:15:37] <MatthiasM> eg Mouse.getEventDX()[03:16:01] <e-roder> but i'm writing in c++[03:16:06] <e-roder> using opengl[03:16:08] <MatthiasM> and ?[03:17:10] <tmccrary> opengl doesn't know anything about mice[03:17:24] <e-roder> right[03:17:27] <e-roder> windows api[03:17:34] <e-roder> not java abstractions[03:17:50] <MatthiasM> lol - what do you think is LWJGL using on windows to get this data ?[03:17:50] <tmccrary> well any sane developer is going to have abstract code anyway[03:17:55] <tmccrary> regardless of the language[03:18:23] <MatthiasM> or do you think the data appears magicly in it's variables ?[03:18:24] <tmccrary> dynamic binding isn't a huge bottleneck to worry about most of the time[03:21:49] <e-roder> b0000: how about calling getCursorPos()?[03:21:59] <e-roder> instead of using wm_mousemove[03:24:43] <b0000> e-roder: good thinking...that should work actually[03:24:58] <b0000> just call get() set() once per frame, should work just fine[03:25:02] <e-roder> yeah[03:27:37] <e-roder> seems to be solid[03:27:57] <e-roder> now i just need to figure out why my camera rotation is bad[03:28:24] <e-roder> i'm rotating around the y axis, then taking a cross product of the resultant vector and UP and rotating around that[03:28:39] <e-roder> doesn't that seem like it should work?[03:30:13] *** paul424 has quit IRC[03:30:31] *** HuntsMan has joined ##opengl[03:30:49] <MatthiasM> why not use a yaw and pitch rotation ?[03:31:17] <e-roder> i'm not 100% on those[03:31:27] <b0000> e-roder: maths is not my strong point, i tend to use more trial and error[03:31:29] <e-roder> but i think that's what im doing[03:31:36] <e-roder> alright[03:32:09] <b0000> infact my camera code has a bug currently...it cant look through 360 degrees[03:32:20] <e-roder> what happens if you go past 360?[03:32:24] <b0000> it appears to lock at 90 up or down[03:32:29] <e-roder> hmmm[03:32:34] <b0000> e-roder: nothing...it just wraps around[03:32:36] <e-roder> could be cuz of c++[03:32:39] <tmccrary> lol[03:32:46] <MatthiasM> using yaw/pitch you would be limited to +-90°C for pitch and 360° for yaw[03:32:47] <e-roder> 's trig functions[03:32:54] <tmccrary> rofl[03:33:09] <b0000> MatthiasM: is that so? i though i could pitch > 90[03:33:11] <MatthiasM> +-90°[03:33:29] <b0000> > +90 and < -90[03:33:34] <MatthiasM> b0000: assuming 0° be horrizontal and 90° up and -90° down[03:33:58] <MatthiasM> your head can't roll back :)[03:34:02] <b0000> yeah...i dont understand the locking... i think the sign changes or something...cos it kind of judders?[03:34:16] <b0000> thats cos its attached to my neck[03:34:23] <MatthiasM> yep[03:34:23] <b0000> my ogl camera is not :p[03:34:34] <MatthiasM> and yaw normaly implies turning the player[03:34:53] <e-roder> well even if you look straight up in modern shooters like halflife2 or sometthing you can see a bit of strange behavior[03:35:00] <b0000> sure but i thought the maths held up for 360 pitching[03:35:11] <b0000> not in my case, but in a general case[03:35:24] <b0000> cos as i say a lot of my maths is a bit sucky[03:35:47] *** MatthiasM has quit IRC[03:35:53] *** MatthiasM has joined ##opengl[03:36:03] <e-roder> b0000: for the sake of this discussion (considering it could involve trig functions) could you refrain from using 'cos' to mean because?[03:36:13] <e-roder> i'm getting really confused[03:37:02] <MatthiasM> http://www.matthiasmann.de/worldscape <-- this demo uses yaw & pitch to walk around[03:37:27] <MatthiasM> and to simulate a person looking around you dont' want to allow even a 90° pitch[03:37:37] <MatthiasM> maybe only 80 or 85°[03:38:16] <MatthiasM> as a pitch of 180° would make you look backwards with the world flipped upside down[03:38:19] <e-roder> i can look straight up[03:38:24] <e-roder> i just catch flies in my mouth[03:38:41] * MatthiasM throws something at e-roder[03:38:52] <e-roder> i can catch that in my mouth too[03:45:03] <e-roder> MatthiasM: so how do i employ pitch/yaw?[03:45:13] <e-roder> first i yaw, then i pitch, right?[03:45:16] <MatthiasM> you read about it[03:45:19] <e-roder> :([03:45:38] <MatthiasM> you want to pitch first[03:46:11] <e-roder> really?[03:46:16] <MatthiasM> wiki has pictures[03:46:25] <e-roder> ok[03:46:28] <MatthiasM> again - the order depends on what you want to do[03:46:43] <e-roder> you appear to be right[03:46:47] <MatthiasM> if you don't know what you want - then you can't code it[03:46:47] <e-roder> but i'm not sure i understand why[03:47:16] <MatthiasM> well - I'm not going to teach math to you[03:47:45] <e-roder> wiki under pitch-yaw-roll?[03:48:06] <MatthiasM> check for yourself[03:48:30] <e-roder> this pertainst o airplanes[03:49:10] <MatthiasM> and?[03:50:23] <e-roder> airplanes don't care about order of operations[03:50:56] <MatthiasM> they do[03:53:24] *** tokemonstah has quit IRC[04:14:14] <forrestv> e-roder, just randomly try rotations until it works[04:14:43] <mlucassmith> lol[04:14:46] <MatthiasM> forrestv: that's the sledge-hammer variant of bruteforce :)[04:14:53] <mlucassmith> do that in a plane :)[04:15:31] <MatthiasM> if your only tool is a hammer then every problem looks like a nail[04:16:26] <mlucassmith> if your only tool is a nail gun, you probably love lethal weapon 3[04:16:39] <MatthiasM> heh[04:16:48] <MatthiasM> n1 @ mlucassmith[04:17:31] * mlucassmith tries to think of other great nail gun movies[04:19:00] <b0000> nothing wrong with brute force...i find it faster sometimes[04:19:26] *** jfroy|work has quit IRC[04:20:48] <mlucassmith> I dunno.. knowing what the heck you're doing often helps :)[04:21:54] <MatthiasM> I suggest to wait with coding until you have an idea/plan in your head of what you want to do and how[04:22:27] <mlucassmith> a computer alone is not enough to inspire you, you also need alcohol and some stupid youtube videos :)[04:22:45] <MatthiasM> lol[04:23:37] <mlucassmith> i got stuck on caramelldansen for a few days last month[04:23:50] <mlucassmith> i implemented the camera from that in opengl.. see, inspiration :)[04:23:58] <MatthiasM> heh[04:29:13] *** yates has joined ##OpenGL[04:29:25] <yates> as i suspected, my app's loading the cpu quite heavily[04:29:46] <yates> i was thinking of using vertex arrays , but..[04:30:30] <b0000> use VBO[04:30:35] <yates> since resizing the window reduces cpu time drastically, i'm thinking the majority of the time is spent just draying, not transferring[04:30:46] <b0000> immediate mode is sooooo 90s dude[04:30:50] <yates> s/draying/drawing/[04:31:03] <yates> b0000: and it's my very first opengl program too[04:31:12] <yates> consider the level of the programmer, dude[04:31:35] <b0000> yates: ah then immediate mode is fine...the next step is to optimise to vertex array and then VBO[04:31:50] <b0000> when i dev i still initially use immediate mode[04:31:56] <MatthiasM> yates: I suggest to get VAs working first[04:32:03] <yates> VAs?[04:32:05] *** phrosty has quit IRC[04:32:07] <yates> vertex arrays[04:32:07] <mlucassmith> if resizing your window is making a difference, make sure you're actually using hardware accelerated drivers[04:32:17] <b0000> then i got to VBO directly...but id recommend VA first if ur new to gl[04:32:18] <yates> yes, i am, nvidia[04:32:41] <mlucassmith> i dunno, resizing your window makes a difference sounds -really- suspect to me yates[04:32:56] <b0000> resizing window improving performance suggests you are fill rate limited[04:32:58] <MatthiasM> are you running with or without vsync ?[04:33:17] <yates> MatthiasM: i don't know - how do i check?[04:33:27] <MatthiasM> what's your framerate you get ?[04:33:39] <yates> i haven't mreasured that yet[04:33:45] <yates> i'm just looking at cpu time[04:33:57] <MatthiasM> measure the FPS[04:34:16] <MatthiasM> if it's higher then what your monitors shows then you run without vsync[04:34:16] <yates> i've got a timer set so that, if it's running real time, it should be 50 Hz[04:34:37] <b0000> yates: http://pastebin.com/d9747966[04:34:46] <MatthiasM> and cpu load without vsync is not meaningful[04:34:54] <b0000> a useful code snippet... pass false to disable vsync[04:35:25] <MatthiasM> b0000: don't use strstr to check for extensions ![04:35:38] *** phrosty has joined ##opengl[04:35:48] <b0000> MatthiasM: it hardly matters to call it once at startup[04:36:16] <mlucassmith> isn't GL_EXTENSIONS deprecated?[04:36:35] <b0000> have you ever heard of the 80/20 rule or whatever they call it these days[04:36:41] <MatthiasM> b0000: there are extension which start with the same string as other extensions - eg ARB_vertex_program and ARB_vertex_program2 etc[04:36:50] <b0000> fair point[04:36:52] <MatthiasM> so using strstr can give a false result[04:37:01] <b0000> yes agreed...ill change that now[04:37:03] <MatthiasM> you need to check atleast for a space before/after[04:37:17] <MatthiasM> and if the check fails check the rest of the string[04:37:25] <RTFM_FTW> meh just use gluCheckExtension[04:37:29] <b0000> if( strstr( extensions, "WGL_EXT_swap_control " ) == 0 ) ?[04:37:34] <RTFM_FTW> see above[04:37:49] <MatthiasM> b0000: if this extension is reported last then this won't work[04:38:00] <b0000> ok i understand[04:38:07] <MatthiasM> better to use glew[04:38:17] <b0000> i use glew in my engine[04:38:22] <b0000> what do you suggest?[04:38:28] <MatthiasM> then ask glew for this information[04:38:31] <RTFM_FTW> GLboolean exiswts = gluCheckExtension( glGetString( GL_EXTENSIONS ), "GL_ARB_vertex_buffer_object" );[04:38:33] <yates> when should vsync be called? along with loading matrix modes and what not (i.e., the begining)?[04:38:37] <RTFM_FTW> there ya go[04:38:48] <b0000> yates: call it once at startup[04:38:57] <b0000> yates: certainly not every frame :p[04:39:10] <mlucassmith> RTFM_FTW: what is GL_EXTENSIONS replaced with post 3.0 deprecation?[04:39:51] <b0000> mlucassmith: im using opengl 2.1[04:39:55] <RTFM_FTW> short of VERSION there isn't one actually :P[04:40:04] <b0000> RTFM_FTW: thanks dude[04:40:13] <mlucassmith> RTFM_FTW: okay, so you just sort of, attempt to bind the function and if it works, yay, if not, tough luck.. heh[04:40:13] <RTFM_FTW> and of course VERSION doesn't give you everything you need[04:41:24] <MatthiasM> mlucassmith: no - there is a new extension to enumerate the extensions - eg getCount - getExtension[04:41:39] <mlucassmith> is that part of 3.0 or will it be a 3.1 thing?[04:41:58] <MatthiasM> mlucassmith: read the OpenGL extension registry[04:42:03] <b0000> 1>c:\users\andy\opengame\include\og_graphics.h(221) : error C3861: 'gluCheckExtension': identifier not found[04:42:05] *** jfroy has joined ##OpenGL[04:42:16] <b0000> and yes im including GL/glu.h[04:42:20] <RTFM_FTW> link with the GLU library[04:42:22] <mlucassmith> it's a nice catch 22, you have to check GL_EXTENSIONS to check for the extensions extension... :)[04:42:38] <b0000> RTFM_FTW: its a compile error, and i do link with glu[04:43:05] <b0000> does win32 glu support this function?[04:43:12] <MatthiasM> b0000: look into your glu header - maybe you have an old one[04:43:23] <b0000> dude win32 glu is ancient[04:43:27] *** qeed has quit IRC[04:43:29] <MatthiasM> heh[04:43:43] <MatthiasM> b0000: then the easiest approch is to walk from space to space :)[04:43:50] <MatthiasM> eg use strchr(str, ' ')[04:43:59] *** pietia has quit IRC[04:44:04] <RTFM_FTW> LOL just install a newer GLU[04:44:08] <RTFM_FTW> and be done w/ it :P[04:44:11] <MatthiasM> and don't forget to check the last one :)[04:44:25] <b0000> RTFM_FTW: im looking to ditch glu rather than upgrade it[04:44:28] <MatthiasM> or read the docu of glew on how to check for extensions[04:44:46] <b0000> the only function im currently using is gluPerspective that is easily replaced[04:45:06] <b0000> glu is bloated crap[04:45:09] <RTFM_FTW> so? ... using a few minor GLU routines isn't going to kill anyone[04:45:20] <tmccrary> Down with GLU![04:45:23] <b0000> its another dependency[04:45:23] <yates> i too don't like the idea of glu..[04:45:30] <yates> i've been avoiding it[04:45:32] <b0000> and its written like shit, by idiots[04:45:34] <tmccrary> It's infecting our youth[04:45:38] <RTFM_FTW> and modern dynamic library design solves the latter issue[04:45:51] <yates> glu is glue[04:46:28] <b0000> RTFM_FTW: glu = suckage[04:46:37] <MatthiasM> b0000: ??[04:46:46] <MatthiasM> b0000: it's just a lib - use it or not[04:47:21] <mlucassmith> glut == suckage, glu == utilities[04:47:32] <RTFM_FTW> and any OS worth half a shit isn't going to page the entire dll, dylib, ... into memory when you access one or two API methods :P[04:47:32] <b0000> MatthiasM: glu is shit, its good for prototyping, much like glut is[04:47:39] <RTFM_FTW> so one again don't fucking worry about it[04:47:43] <b0000> but not good in the long run[04:47:50] <RTFM_FTW> err once again[04:48:06] <yates> wxwaterfall.cpp:429: error: ~wglGetProcAddress~ was not declared in this scope[04:48:16] <yates> where is wglGetProcAddress supposed toc ome from?[04:48:32] <b0000> RTFM_FTW: you write as crap code as you like, ill stick to the real code (very rare to see code as efficient as i tend to write)[04:48:37] <mlucassmith> windows opengl32.dll[04:48:46] <yates> that's nice[04:48:50] <b0000> have you ever looked at the source for gluPerspective?[04:48:51] <yates> i'm on linux[04:49:12] <mlucassmith> yates, use the glx equivalent[04:49:17] <RTFM_FTW> "you write as crap code as you like, ill stick to the real code (very rare to see code as efficient as i tend to write)" ...bwahahahahahah[04:49:24] <RTFM_FTW> "dude I'[04:49:43] <RTFM_FTW> err "dude I'm 3337 bitches... I don't use gluCheckExtension :P"[04:49:55] <b0000> RTFM that was a sample off the net[04:50:06] <yates> 3337?[04:50:08] <yates> '[04:50:15] <mlucassmith> MatthiasM: There's always a space at the end of the GL_EXTENSIONS string, so you can suffix a space to the spec you're checking[04:50:38] <MatthiasM> are you sure ? I would not count on it[04:50:39] <RTFM_FTW> I prefer to solve useful problems... not attempt to optimize extension loading code which is *only* queried at initialization time... once :P[04:50:47] <tmccrary> lol[04:51:16] <mlucassmith> MatthiasM: well if you're owrried about it, stick a space on the end of the string you get back from GL_EXTENSIONS ;P[04:51:24] <MatthiasM> but this extension checking is already done by glew - and if one would read the docu it would be clear[04:51:26] <b0000> RTFM_FTW: same here now get over it. i wont ask what you do for a day job...but i suspect its not optimising 360 and ps3 for one of the biggest developers on the planet[04:51:35] <tmccrary> rofl[04:51:40] <mlucassmith> lol[04:51:50] <RTFM_FTW> hehe this makes me think of the guy we had in here a few months ago who was writing his extension loading code in fairly cryptic C with the idiotic idea of saving a few nanoseconds during init time :P[04:52:08] <RTFM_FTW> hell he might have been using assembly :P[04:52:12] <b0000> RTFM_FTW: are you upset because i dont want to use glu...get over it[04:52:19] <yates> RTFM_FTW: now don't exaggerate - it was microseconds...[04:52:20] <mlucassmith> omg[04:52:25] <mlucassmith> i'm in tears with laughter right now[04:52:44] <mlucassmith> b0000: this is funny as shit because you don't know who RTFM_FTW works for or what he does[04:52:51] <RTFM_FTW> honestly I don't give a fuck what you do... OTOH I *do* find it amusing that you use this act as a justification for your OMG awesome111 coding skills :P[04:53:07] <b0000> RTFM_FTW: you started this...now will we let it drop now?[04:53:09] <yates> ok, here's my stupid: i don't know glx cause i'm using wxwidget's wxglcanvas class, and i don't see this function ANYWHERE[04:53:11] <RTFM_FTW> because honestly we aren't impressed :P[04:53:21] <RTFM_FTW> or at least I'm not :P[04:53:23] <RTFM_FTW> heheh[04:53:34] <MatthiasM> yates: which function ?[04:53:39] <mlucassmith> yates: it's in the X11.so and Xext.so and libGL[04:53:40] <b0000> im not here to impress anyone. im here to discuss opengl in a civilised manner...ill ask one more time will we let it drop?[04:53:49] <tmccrary> I love how a guy who can't handle basic 3d math is employed "optimizing optimizing 360 and ps3 code for one of the biggest developers on the planet"[04:53:51] <yates> wglGetProcAddress[04:54:02] <MatthiasM> yates: wgl == windows only[04:54:03] <RTFM_FTW> "same here now get over it. i wont ask what you do for a day job...but i suspect its not optimising 360 and ps3 for one of the biggest developers on the planet" ...no actually I write the drivers you use which make that possible :P[04:54:06] <RTFM_FTW> so there[04:54:11] <mlucassmith> yates: glxGetProcAddress[04:54:18] <b0000> nv drivers?[04:54:36] <RTFM_FTW> since when does the 360 use "nv drivers"?[04:54:47] <MatthiasM> lol[04:54:53] <RTFM_FTW> you mentioned multiple consoles :P[04:54:58] <tmccrary> RTFM_FTW: don't try to correct him, you know he must be right[04:55:01] <RTFM_FTW> from multiple vendors[04:55:03] <mlucassmith> yates: each platform has their own interface for setting up OpenGL. On windows it's called WGL, on X11 it's called GLX, etc..[04:55:10] <tmccrary> He works for one of the biggest developers on the planet[04:55:22] <RTFM_FTW> LOL[04:55:24] <mlucassmith> tmccrary: Andre the Giant[04:55:34] <mlucassmith> that dude is crazy huge[04:55:52] <MatthiasM> mlucassmith: larger then the collos of rodos ?[04:56:05] <RTFM_FTW> damn that is pretty large[04:56:07] <mlucassmith> MatthiasM: I'd have to check, but since i'm on opengl, I'll just tell you you're wrong and I'm right[04:56:16] <MatthiasM> heh[04:56:32] <b0000> yeah well im not on here to name drop, all i gotta say is this is getting pretty pathetic im gonna shut up now. no disrespect RTFM_FTW[04:56:55] <tmccrary> ah yeah dogg, I heard you like to profile your 360, so I puts a PROFILER in yo PROFILER[04:57:05] <b0000> lets just leave the petty arguements...take it offline its just lowering the tone of the channel[04:57:36] <RTFM_FTW> you have won the GODLIKE CODER badge :D[04:57:46] <mlucassmith> it's an achievement you can unlock[04:57:54] <tmccrary> We all know you work for Blizzard[04:58:02] <tmccrary> On their numerous xbox 360 and ps3 titles[04:58:08] <RTFM_FTW> heheh[04:58:11] <mlucassmith> because blizzard really pushes the GPU hard in WoW :)[04:59:59] <yates> what's an easier way to turn vsync off?[05:00:11] <mlucassmith> yates: no idea on linux[05:00:17] <RTFM_FTW> that depends upon the OS[05:00:28] <MatthiasM> yates: just use a lib which has support for this[05:00:32] <yates> linux[05:00:32] <RTFM_FTW> since the *only* methods used to control vsync are OS dependent[05:00:35] <yates> fedora 8[05:00:48] <yates> oh, never mind, -it's glX. Ahh![05:03:10] <yates> and i'm supposed to turn vsync OFF?[05:03:17] <MatthiasM> on[05:03:25] <RTFM_FTW> indeed[05:03:57] <yates> what is the extension name to check for glx? my strstr is not finding[05:04:08] <mlucassmith> no, glx is always there on linux[05:04:32] <yates> so the check is not necessary?[05:04:55] <mlucassmith> not only is it not necessary, it's not possible[05:05:18] <MatthiasM> you can check the glx version[05:05:44] <MatthiasM> without glx you can't create an opengl context[05:07:02] <yates> version 1.3[05:08:05] *** |t4bz| has joined ##OpenGL[05:08:21] <yates> http://pastebin.com/m1493803f[05:10:16] <yates> ok, so now what?[05:10:30] <yates> it's still maxing the cpu[05:10:46] <yates> i'm not at all sure the "enabling" worked, though[05:10:49] <MatthiasM> either you use a lib which provides what you want - or you go and read about glx[05:11:07] <MatthiasM> well - check your framerate[05:13:11] <yates> when you enable vsync, you're doing it for the monitor/card, i.e., for the whole video display subsystem, right?[05:13:23] <mlucassmith> no[05:13:27] <mlucassmith> just your program[05:17:17] <yates> perhaps i should comment on my environment[05:18:00] <yates> i'm shortsighted, so i use a single-head system, sit about 8 inches from the screen, and run in 2048x1536 resolution[05:18:45] <yates> my initial gl window size was 500x500. should i be getting 50 Hz frame rate at that size?[05:19:33] <mlucassmith> what card do you have?[05:19:36] <yates> whoa. wait. it's only 20 Hz.[05:19:47] <mlucassmith> what are you rendering to the scene?[05:19:59] <yates> GeForce FX 5200/AGP[05:20:21] <mlucassmith> isn't that super old?[05:20:52] <yates> a triangle mesh (using GL_TRIANGLE_STRIP) of about 64k triangles[05:21:10] <yates> the card is about 2 1/2 years old[05:21:24] <mlucassmith> so you're filling a fifth of your window.. meh, it should be able to do that at full screen[05:21:25] <yates> i thought the technology was moderate at the time[05:21:51] <yates> so is it the vertex arrays?[05:21:55] <yates> (or lack thereof?)[05:22:01] <mlucassmith> how are you rendering the scene now?[05:22:08] <tmccrary> 5200 is like a geforce 2 mx or something[05:22:09] <mlucassmith> call lists?[05:22:20] <yates> no, the old testament way[05:22:28] <mlucassmith> a bunch of begin/end ?[05:22:32] <yates> glBegin/glEnd[05:22:34] <yates> yes[05:22:39] *** calav3ra has quit IRC[05:22:39] <mlucassmith> wrap them in call lists for an immediate speed up[05:23:15] <yates> MatthiasM: do you agree? what will be the biggest performance bang for my development time buck?[05:23:28] *** t4bz has quit IRC[05:23:49] <yates> the code is here: http://galois.digitalsignallabs.com/wxwaterfall/[05:26:17] <yates> getting late - going to bed - leving the channel open in case of any comments. thanks everyone![05:27:26] *** reprore_ has quit IRC[05:28:19] *** nywles has quit IRC[05:33:04] *** MissGeek has joined ##OpenGL[05:34:28] <b0000> MatthiasM: cheers dude much better, if( wglewIsSupported( "WGL_EXT_swap_control" ) ) { wglSwapIntervalEXT( enable ); }[05:35:07] *** davidc__ has quit IRC[05:37:02] *** heap_leak has joined ##OpenGL[05:42:34] *** heap_leak has quit IRC[05:46:45] *** amz has joined ##opengl[06:04:13] *** wcstok has quit IRC[06:11:01] *** e-roder has quit IRC[06:11:52] *** rocketmagnet has quit IRC[06:16:22] *** Entelin has quit IRC[06:17:33] *** Entelin has joined ##OpenGL[06:21:41] *** tttt has joined ##OpenGL[06:28:17] *** davidc__ has joined ##opengl[06:30:28] *** davidc__ has quit IRC[06:34:23] *** tttt has left ##OpenGL[06:47:25] *** MissGeek has quit IRC[06:53:26] *** walt has quit IRC[07:19:32] *** korff_home has joined ##OpenGL[07:19:58] *** walt_ has quit IRC[07:37:26] *** scai has joined ##opengl[07:40:51] *** pietia has joined ##OpenGL[07:46:10] *** xp_prg has quit IRC[07:51:12] *** electrostatic has joined ##OpenGL[07:51:26] <electrostatic> blarhg[07:56:12] <electrostatic> Question... http://opengl.org/registry/specs/ARB/pixel_buffer_object.txt does not seem to give any way to ensure that a packed buffer is done reading without mapping the buffer to client address space. (As they say in the comments of their example code, // Process partial images. Mapping the buffer waits for[07:56:12] <electrostatic> // outstanding DMA transfers into the buffer to finish. ) Is there any way to ensure that a read is finished without mapping the entire sotre to client address space? Or is mapping the store somehow a smaller overhead operation than what I am thinking?[07:56:50] *** scai has left ##opengl[08:02:25] *** pist0l-fish has joined ##OpenGL[08:04:10] <pist0l-fish> hello all. I have encountered strange behavior where I can blend one texture over another texture, but cannot blend an translucent color over a texture (results in the opaque color instead). Can anyone help? Thanks in advance.[08:08:07] <electrostatic> How are you trying to blend the color over the top?[08:08:56] <pist0l-fish> electrostatic: I just created a glColor4f(0,0,0,.25) quad on top of a quad that is textured[08:10:06] <electrostatic> And you are using glBlendFunc?[08:14:22] <pist0l-fish> yes[08:14:34] <pist0l-fish> glBlendFunc(GL_SRC_ALPHA, GL_ONE_MINUS_SRC_ALPHA)[08:16:23] <electrostatic> huh, and you said it is the opaque polygon showing, not the textured one? (Which would mean it isn't a depth issue)?[08:17:06] <electrostatic> An you would have to have glEnable(GL_BLEND);, or else the texturing part wouldn't have worked...[08:17:40] <pist0l-fish> electrostatic: I think I have found my problem; it looks like things are getting drawn in the other I am specifying them, not with regards to their depth values[08:17:43] <pist0l-fish> but...why[08:18:24] <electrostatic> You could try enableing depth tes, but it should be on by default, I believe[08:18:31] <electrostatic> *test[08:18:47] <pist0l-fish> I have depth test enabled, and I have glDepthFunc(GL_LEQUAL)[08:18:53] <electrostatic> Are you drawing into an FBO by any chance?[08:19:00] <pist0l-fish> what is an FBO?[08:19:08] <electrostatic> Well, most likely no then[08:19:13] <electrostatic> (Frame Buffer Object)[08:19:25] <electrostatic> Off sceen rendering[08:20:04] <electrostatic> You could try pasting the code to the link in the topic and hope someone who knows more than me takes a look, that's about all I can think off[08:20:14] <electrostatic> http://cpp.sf.net/[08:20:24] *** rnx has left ##opengl[08:20:25] <pist0l-fish> yeah. There's just so much code and it's everywhere. I'm just going to try and figure this out/give up. lol[08:20:40] <pist0l-fish> could it have anything to do with me using glOrtho?[08:21:15] <electrostatic> Possibly, but I'm not sure...[08:21:39] <electrostatic> You could try giving the top layer an exaggerated Z value and see if that helps...[08:22:26] <electrostatic> (IIRC Ortho just kind of "zooms back" till everything "might as well be normal")[08:23:22] <pist0l-fish> one last question. When you translate points, they change right? If I have a vertex at (x,y,z), translating it will also affect depth test, right?[08:24:17] <electrostatic> yes[08:27:47] <pist0l-fish> electrostatic: sorry to keep bothering you, but maybe you can help me by looking at this one chunk of code: http://codepad.org/OhdtdUbG - if you switch the order of the two blocks of code, you get different results. Is this normal?[08:28:16] *** synchris has joined ##opengl[08:29:05] <electrostatic> Do you have an unseen glLoadIdentity(); before the first block of code?[08:30:19] <pist0l-fish> electrostatic: I do[08:30:49] <electrostatic> I wouldn't think so then, but I am by no means an expert[08:32:00] <pist0l-fish> okay. Thanks for your help. I'm sure it's something trivial, and I'm also sure it's going to drive me insane[08:32:01] <pist0l-fish> haha[08:32:04] <electrostatic> wait, maybe it has something to do with your active texture?[08:32:23] <pist0l-fish> perhaps- you can only bind one texture at a time though, right?[08:33:40] *** davidc__ has joined ##opengl[08:33:55] <electrostatic> Yes, but you have to make sure to set the glTexEnvi(GL_TEXTURE_ENV,GL_TEXTURE_ENV_MODE,GL_BLEND); of both, I think[08:35:07] <electrostatic> (any other tex params too, for that matter...)[08:35:47] <pist0l-fish> oh okay[08:35:48] <pist0l-fish> i will try that[08:36:31] <pist0l-fish> ah it certainly did something[08:37:10] <pist0l-fish> electrostatic: thank you :)[08:37:15] <electrostatic> np =)[08:37:24] *** rabbit- has joined ##OpenGL[08:38:53] <pist0l-fish> ...wait, that wasn't it >_<[08:40:38] *** boghog has quit IRC[08:41:12] <electrostatic> meh[08:41:43] <pist0l-fish> are there any restrictions on near/far values for calls to glOrtho?[08:44:26] <electrostatic> dunno... What is the Z value in your overlay compared to your base?[08:44:31] *** UUncia has joined ##OpenGL[08:44:35] *** braden has joined ##OpenGL[08:45:14] <pist0l-fish> overlay = -9.0f, base = -10.0f[08:46:27] <electrostatic> and what are the last two arguments to glOrtho (near clip and far clip)?[08:48:31] <electrostatic> ok, one last question.. what is your depth buffer size (in your pixel format)?[08:49:07] <pist0l-fish> sorry, I don't follow[08:49:59] <electrostatic> I was just wondering how many bits your depth buffer was[08:50:05] *** Renderwahn has joined ##OpenGL[08:50:25] <pist0l-fish> i think the library that i use does not allow me to specify that[08:52:04] <electrostatic> one thing you could try really quick is changing the ortho from -10.0f, 10.0f to -1.0f, 1.0f and changing your test layers to overlay = .25f, base = -.25f[08:52:29] <electrostatic> Don't know if that will do anything though[08:57:38] <pist0l-fish> electrostatic: I don't know why I always have such trouble using opengl + 2d... something always happens. I won't bother you anymore though, I'm going to call it a night. Thanks for your help[08:57:54] *** pist0l-fish has quit IRC[08:58:46] *** electrostatic has quit IRC[09:04:25] *** amz has quit IRC[09:08:24] *** oss has joined ##OpenGL[09:08:39] <oss> Hello, where to download GLU for Linux?[09:09:48] *** Bollinger has joined ##OpenGL[09:14:51] *** walt has joined ##OpenGL[09:26:12] *** walt has quit IRC[09:29:41] *** walt has joined ##OpenGL[09:30:00] *** rutski_ has joined ##OpenGL[09:33:36] *** marenz_ has joined ##OpenGL[09:35:53] *** kbotnen has joined ##OpenGL[09:39:38] *** predaeus has joined ##opengl[09:40:07] *** groton has joined ##OpenGL[09:52:04] *** sohail has quit IRC[09:54:29] *** Nescafe has joined ##OpenGL[09:56:57] *** walt has quit IRC[09:57:18] *** rutski_ has quit IRC[09:58:01] *** boghog has joined ##opengl[09:59:07] *** oss has quit IRC[10:01:03] *** walt has joined ##OpenGL[10:01:40] *** tokemonstah has joined ##OpenGL[10:04:21] *** groton has quit IRC[10:04:43] *** pietia has quit IRC[10:05:50] *** walt has quit IRC[10:08:50] *** fargiolas has joined ##OpenGL[10:10:19] *** belou has joined ##OpenGL[10:13:42] *** walt has joined ##OpenGL[10:14:48] *** dvoid_ has joined ##OpenGL[10:15:37] *** walt has quit IRC[10:17:18] *** UUncia has quit IRC[10:17:49] *** UUncia has joined ##OpenGL[10:18:01] *** UUncia has quit IRC[10:24:09] *** walt has joined ##OpenGL[10:30:11] *** UUncia has joined ##OpenGL[10:30:16] *** walt has quit IRC[10:44:27] *** LordHavoc has quit IRC[10:46:47] *** yates has quit IRC[10:53:30] *** HuntsMan has quit IRC[10:53:30] *** groton has joined ##OpenGL[10:54:20] *** doub has joined ##opengl[10:54:45] *** dusted has joined ##OpenGL[11:03:26] *** Ingenu has joined ##OpenGL[11:04:49] *** dusted_ has joined ##OpenGL[11:06:17] *** dusted has quit IRC[11:11:48] *** maxton has joined ##OpenGL[11:25:20] *** dolphin has joined ##OpenGL[11:26:56] *** Spkka has joined ##OpenGL[11:30:54] *** LordHavoc has joined ##OpenGL[11:31:55] *** Morphous is now known as Amorphous[11:33:15] *** cplusplus has joined ##OpenGL[11:43:58] *** _JSC415 has joined ##opengl[11:44:20] <_JSC415> hm..[11:44:32] <_JSC415> where can i find a cool code editor ?[11:44:51] *** hogbog has joined ##opengl[11:44:55] *** boghog has quit IRC[11:45:24] *** LordMetroid has joined ##OpenGL[11:45:48] <MatthiasM> _JSC415: google[11:45:49] *** hogbog is now known as boghog[11:58:18] *** neoneye has joined ##OpenGL[11:58:27] *** marenz_ has quit IRC[11:58:44] <_JSC415> http://www.softpedia.com/get/Programming/Other-Programming-Files/SourceCode-to-Flowchart.shtml[11:58:47] <_JSC415> i found .[11:58:56] <_JSC415> but i need some more features.[11:59:00] *** rutski has joined ##OpenGL[11:59:04] <_JSC415> or just free one . ;P[12:01:18] *** maxton has quit IRC[12:01:21] *** geocalc has quit IRC[12:02:07] *** geocalc has joined ##opengl[12:02:29] *** paul424 has joined ##OpenGL[12:08:07] *** Suprano has joined ##OpenGL[12:13:03] *** paul424 has quit IRC[12:14:04] *** _JSC415 has quit IRC[12:23:45] *** walt has joined ##OpenGL[12:24:11] *** Plagman has quit IRC[12:25:29] *** Plagman has joined ##OpenGL[12:26:39] *** ol1veira has joined ##OpenGL[12:26:53] *** Plagman has quit IRC[12:27:38] *** Plagman has joined ##OpenGL[12:38:05] *** Gonzo has quit IRC[12:45:19] *** mouserSVK has joined ##OpenGL[12:46:24] *** groton has quit IRC[12:52:15] *** Gonzo has joined ##OpenGL[13:17:56] *** maxton has joined ##OpenGL[13:29:11] *** batbrat has joined ##opengl[13:29:50] <batbrat> I would like to thank all the members of the forum who helped me with understanding the redbook polarview example.[13:30:00] <batbrat> And all the other things of course.[13:30:32] <batbrat> Especially : aep,sparky,b0000,MatthiasM and others.[13:30:40] <MatthiasM> :)[13:31:44] <batbrat> I have an answer to the polar example.The reason why the redbook calls glRotated() with a positive azimuth is because it is specified as 360-conventional_azimuth => -azimuth :)[13:36:33] *** rabbit- has quit IRC[14:20:03] *** JernejL has joined ##OpenGL[14:21:18] *** sohail has joined ##OpenGL[14:33:42] *** Jernej has quit IRC[14:52:17] *** doub has quit IRC[14:55:20] *** sohail has quit IRC[14:57:49] *** eXtronuS has quit IRC[15:02:30] *** doub has joined ##opengl[15:02:43] *** Suprano has quit IRC[15:04:57] *** UUncia has quit IRC[15:08:23] *** johndoe has joined ##opengl[15:12:56] *** rutski has quit IRC[15:14:05] *** autonomy_ is now known as autonomy[15:18:11] *** jcouture has joined ##OpenGL[15:19:42] *** rutski has joined ##OpenGL[15:20:06] *** rutski has quit IRC[15:35:24] *** Suprano has joined ##OpenGL[15:35:58] *** jm|earth has joined ##OpenGL[15:40:57] *** jm|afk has quit IRC[15:45:11] *** wojtek has quit IRC[15:50:48] *** Kasu- has joined ##OpenGL[15:55:29] *** calav3ra has joined ##opengl[15:59:40] *** b0000 has quit IRC[16:07:15] *** b0000 has joined ##opengl[16:12:55] *** HuntsMan has joined ##opengl[16:14:44] *** NightVisio has joined ##OpenGL[16:15:14] *** NightVisio has quit IRC[16:29:32] *** rnx has joined ##opengl[16:36:45] *** pietia has joined ##OpenGL[16:42:14] *** k3t^ has quit IRC[16:58:38] *** bijoo_appdev has quit IRC[16:58:38] *** bijoo_osdev has quit IRC[17:00:38] *** mattn2|home has joined ##OpenGL[17:11:51] *** Jernej has joined ##OpenGL[17:12:14] *** JernejL has quit IRC[17:23:49] *** HuntsMan has quit IRC[17:24:52] *** zigonick has joined ##OpenGL[17:25:33] *** tokemonstah^ has joined ##OpenGL[17:26:41] *** reprore_ has joined ##OpenGL[17:27:00] *** Ankou has joined ##OpenGL[17:27:29] <Ankou> hi, how can I disable GL_ARB_texture_non_power_of_two?[17:29:57] *** x0rx0r has joined ##OpenGL[17:31:18] <x0rx0r> hello[17:32:48] *** Jernej has quit IRC[17:33:38] *** SilentAssassin has joined ##OpenGL[17:33:48] *** jfroy has quit IRC[17:33:52] <SilentAssassin> anyone knows how to find out the colision between a line segment and ellipsoid?[17:36:08] <dindinx> SilentAssassin: change the coordinate system so the ellipsoid becomes a centered unit sphere, and it will be a piece of cake afterwards.[17:36:44] <MatthiasM> Ankou: you don't use it[17:37:02] <zigonick> okay, im currenlty drawing points based on RL lat/Long vaules, what is the function to set the current display vector. Where i can set the top bottom left right of screen min/max vector#[17:37:15] <MatthiasM> http://www.realtimerendering.com/intersections.html[17:38:21] <Ankou> well I want to use it if it exists and I don't want to use it if it doesn't exists but I need to disable it in order to test the code which is executed if NPOT textures aren't allowed[17:38:39] *** jcouture has quit IRC[17:39:16] <MatthiasM> Ankou: lol - then add a flag to your extension checking code[17:39:19] <quicksilver> you can't disable extensions.[17:39:21] <quicksilver> Just lie about it.[17:40:22] *** jcouture has joined ##OpenGL[17:40:32] <Ankou> mhm but the problem is that it's not so easy to lie about it because I'm working with a library..[17:41:34] *** Yustme has joined ##OpenGL[17:42:17] *** tokemonstah has quit IRC[17:50:47] *** batbrat has left ##opengl[17:52:22] *** HuntsMan has joined ##opengl[17:53:28] *** NightVisio has joined ##OpenGL[17:57:20] *** synchris has quit IRC[17:57:30] *** synchris has joined ##opengl[17:59:43] *** sohail has joined ##OpenGL[18:01:17] *** JernejL has joined ##OpenGL[18:03:04] *** UUncia has joined ##OpenGL[18:03:50] *** Goosey has quit IRC[18:07:52] *** johndoe has quit IRC[18:09:34] *** Goosey has joined ##OpenGL[18:12:34] *** XT95_ has joined ##OpenGL[18:13:18] *** Goosey has quit IRC[18:16:44] *** mm765^away has left ##OpenGL[18:17:23] *** mm765^away has joined ##opengl[18:18:49] *** x0rx0r has quit IRC[18:19:10] *** x0rx0r has joined ##OpenGL[18:22:32] *** vade has quit IRC[18:26:34] *** belou has quit IRC[18:31:34] *** k3t has joined ##OpenGL[18:31:42] *** GuShH has quit IRC[18:36:27] *** NightVisio has quit IRC[18:38:45] *** scai has joined ##opengl[18:40:12] *** Jernej has joined ##OpenGL[18:40:14] <SilentAssassin> so i create a first identity matrix then translate this matrix to the elipsoid center then scale this matrix so it is 1 unit and then multiply this matrix with my line start and end?[18:43:01] *** chriswright has quit IRC[18:49:14] *** itewsh has joined ##OpenGL[18:49:35] *** JernejL_ has joined ##OpenGL[18:54:34] *** calav3ra_ has joined ##opengl[18:58:06] *** JernejL has quit IRC[19:03:36] *** Jernej has quit IRC[19:05:34] *** groton has joined ##OpenGL[19:07:04] *** Shel2476 has joined ##OpenGL[19:07:13] *** Shel2476 has left ##OpenGL[19:12:43] *** calav3ra has quit IRC[19:15:53] *** JernejL_ has quit IRC[19:26:05] *** doub has quit IRC[19:26:41] *** Nescafe has quit IRC[19:32:54] *** JernejL has joined ##OpenGL[19:36:14] *** jfroy|work has joined ##OpenGL[19:38:46] *** Goosey has joined ##OpenGL[19:42:04] *** UUncia has quit IRC[19:46:24] <SilentAssassin> when do you think is the best way in a game to store the mouse data... because i want to player to get hit at proper time... because if i render at one frame and check the hit in another ... the object may be gone already[19:47:46] *** suppahasrv has joined ##OpenGL[19:48:41] *** jlongster has joined ##OpenGL[19:50:25] *** x0rx0r has quit IRC[19:52:08] *** Xmas| has joined ##OpenGL[19:52:35] *** suppahsrv has quit IRC[19:52:41] *** dv_ has quit IRC[20:07:12] *** reprore_ has quit IRC[20:11:41] *** Ankou has left ##OpenGL[20:20:41] *** LordMetroid has quit IRC[20:24:28] *** vade has joined ##OpenGL[20:35:12] *** johndoe has joined ##opengl[20:48:42] *** pfo has quit IRC[20:49:11] *** pfo has joined ##OpenGL[20:51:01] *** Madsy has joined ##OpenGL[20:55:40] *** predaeus has quit IRC[20:56:01] *** stevosICBM has joined ##OpenGL[21:00:26] *** yates has joined ##OpenGL[21:00:50] *** suppahasrv is now known as suppahsrv[21:01:49] *** HuntsMan has quit IRC[21:05:38] <yates> MatthiasM: you available?[21:05:53] <MatthiasM> maybe :)[21:06:03] <yates> vsync[21:06:06] <yates> why?[21:06:12] <MatthiasM> because[21:06:16] <yates> oh[21:06:17] <yates> ok[21:06:25] <yates> :)[21:06:41] <MatthiasM> without vsync you get tearing artefacts[21:07:03] <MatthiasM> eg you may see the top half of the old image and the bottom half of the new image[21:07:03] *** Nescafe has joined ##OpenGL[21:07:08] <yates> ok, seriously, if my monitor refresh rate is 50 Hz, and I'm not even getting 20 Hz in my app, then why is messing with vsync important?[21:07:37] <yates> at this stage i don't care about artifacts - i just want to see what's taking all the cpu time[21:07:48] <MatthiasM> to synchronize the time when you switch to the backbuffer[21:08:04] <MatthiasM> use a profiler[21:08:06] <yates> why do i care whether it's synchronized or not?[21:08:16] <yates> for the tearing?[21:09:14] <yates> do you know of a linux-based open source profiler that's good for this kind of thing (C/C++)?[21:09:19] <MatthiasM> if you want to benchmark then turn vsync off - if you want to provide a clean image to the user and not waste power by computing more frames then displayable the turn vsync on[21:09:32] <MatthiasM> gcc has a profiler[21:10:02] <yates> but turning vsync off is only necessary when your frame rate is being limited by the monitor's vsync rate, right?[21:10:12] <yates> it isn't.[21:10:31] <MatthiasM> no - vsync will always add a delay until the monitor starts a new image[21:11:10] *** rorokimdim__ has joined ##OpenGL[21:11:15] <MatthiasM> eg if you can render 21 frames without vsync and you run with 50hz refresh then you get 50/3 ==> 16.6 fps[21:11:47] <yates> where did 3 come from?[21:12:10] <MatthiasM> 50/21 == 2.3[21:12:32] <MatthiasM> so it wait until the next frame starts - which means every 3 frames it shows a new image[21:12:35] <SilentAssassin> ^^[21:13:10] *** rorokimdim__ has quit IRC[21:13:15] *** rorokimdim___ has joined ##OpenGL[21:13:40] *** rorokimdim___ has quit IRC[21:13:44] *** rorokimdim___ has joined ##opengl[21:14:52] *** rorokimdim___ has quit IRC[21:14:56] *** rorokimdim___ has joined ##opengl[21:19:16] <pietia> anyone use jogl ?[21:19:28] <MatthiasM> no LWJGL[21:20:01] *** rorokimdim___ has left ##opengl[21:21:18] <pietia> i'm looking for some tutorial of not complicated arcade game (galaga or sth ;)[21:21:58] <MatthiasM> for LWJGL kevin has a asteroids tutorial[21:21:59] <tmccrary1> http://www.cokeandcode.com/asteroidstutorial[21:22:06] <MatthiasM> yep - that one[21:22:16] <pietia> ok, i will check that[21:22:20] <pietia> thanks[21:22:24] <MatthiasM> it has sound graphics and gameplay[21:24:26] *** rorokimdim___ has joined ##opengl[21:32:31] <pietia> the tutorial covers many interesting topics for beginer :) i have to make some kind of easy game in qt-opengl[21:32:44] <pietia> i'm thinking of importing something from blender[21:33:06] <MatthiasM> why qt-opengl ?[21:33:57] <pietia> requirments from my teacher[21:34:12] <MatthiasM> ohh - show him the tutorial ;)[21:34:34] <pietia> i tried to ask if i can make it in java but he said no : )[21:34:55] <MatthiasM> probably he thought that you can't use OGL from Java :DD[21:35:17] <MatthiasM> what class is this ?[21:36:00] <stevosICBM> just use opengl in C, imho. LWJGL is good but the C API is easy[21:36:18] <MatthiasM> it's basicly the same API[21:37:03] *** jcouture is now known as jcouture|away[21:37:50] <pietia> i dont know how to translate the name into english :) hm.... something like: " utilisation of computers" - stupid name. we learn mainly QT4[21:38:06] *** kbotnen has quit IRC[21:38:12] <MatthiasM> hmm[21:38:19] *** HuntsMan has joined ##opengl[21:38:58] <pietia> and it must be 3D game project[21:39:07] *** KU0N has joined ##opengl[21:39:08] *** jcouture|away is now known as jcouture[21:39:09] <pietia> in .... QMainWindow ;)[21:39:19] *** eXtronuS has joined ##OpenGL[21:40:52] <yates> MatthiasM: is gprof what you were referring to?[21:40:59] <MatthiasM> yep[21:43:34] *** Spkka has quit IRC[21:43:38] *** nettikkcah has joined ##opengl[21:44:31] *** hackkitten has quit IRC[21:44:33] *** nettikkcah is now known as hackkitten[21:44:37] *** Gorgoroth has joined ##OpenGL[21:49:04] *** mm765^away is now known as mm765[21:52:06] *** Kasu| has joined ##OpenGL[21:55:36] *** Kasu- has quit IRC[21:56:59] *** dolphin has quit IRC[21:58:47] *** itsmonktastic has joined ##opengl[21:58:59] *** stevosICBM has quit IRC[22:01:42] *** itsmonktastic has left ##opengl[22:01:49] *** cplusplus2 has joined ##OpenGL[22:02:07] *** emzic has joined ##OpenGL[22:06:08] *** jcouture is now known as jcouture|away[22:11:39] *** stevosICBM has joined ##OpenGL[22:12:34] *** rorokimdim___ has quit IRC[22:13:02] *** rorokimdim___ has joined ##opengl[22:17:15] *** pfo has quit IRC[22:19:00] *** A2intaer has joined ##OpenGL[22:19:22] *** cplusplus has quit IRC[22:19:52] *** e-roder has joined ##OpenGL[22:19:57] <e-roder> hey[22:21:21] *** mouserSVK has quit IRC[22:23:30] <hackkitten> hi :)[22:24:00] <e-roder> do i need to manually transform light positions to match camera transformations?[22:26:25] <Madsy> e-roder: The preferred way of rendering nowadays is to use shaders.[22:26:38] <e-roder> cg?[22:26:48] <Madsy> Then you can compute lighting however you want.[22:27:15] <e-roder> avoid using glBegin() and glEnd()?[22:27:26] <Madsy> I assume your question was biased towards the fixed function pipeline[22:27:36] <e-roder> i guess?[22:27:42] <Madsy> In that case, glLight* with GL_POSITION is treated just as a vertex[22:27:53] <Madsy> It is transformed by the modelview matrix[22:28:01] <e-roder> automatically?[22:28:44] <Madsy> Automatically doesn't make sense. You don't need to do the multiplication yourself, but you have to call glLight with GL_POSITION[22:29:07] <e-roder> hmmmm[22:29:15] <Madsy> The position is multiplied with the modelview matrix as it's seen just before the call.[22:29:25] *** pfo has joined ##OpenGL[22:29:25] <e-roder> so inside my render function, after the apropriate transforms[22:29:46] <Madsy> Also, if you want to treat the 4D vector for glLight with GL_POSITION as a position, make sure that the w component is set to 1[22:30:00] <Madsy> With w = 0, it is treated as a directional vector.[22:30:10] <e-roder> aren't opengl lights directional?[22:30:29] <e-roder> meaning i would want to have w=0[22:30:53] <Madsy> Rather ask yourself, what's the difference between a purely directional light and one with a position?[22:31:14] <e-roder> a directional one always points from the origin?[22:31:41] <Madsy> No, it doesn't use that space.[22:31:59] <e-roder> what are the implications of not using the space?[22:32:00] *** AlastairLynn has joined ##opengl[22:32:12] <Madsy> brb[22:32:15] <e-roder> ok[22:33:59] *** androoid has joined ##OpenGL[22:34:58] <Madsy> In the equation for phong lighting, a constant directional vector can be used instead of a vector from the current vertex and to the light position.[22:35:18] *** Ingenu has quit IRC[22:35:24] <e-roder> why?[22:35:27] <Madsy> So directional lights have a constant angle[22:35:40] <e-roder> phong is per-pixel? per-vertex?[22:35:50] <Plagman> per-pixel phong is per-pixel[22:35:53] <Plagman> et cetera[22:35:58] <e-roder> ah ok[22:36:03] <e-roder> so to set the light up[22:36:08] <e-roder> inside my render call[22:36:15] <Plagman> fixed function lighting is per-vertex phong, or something analog to it[22:36:40] <Plagman> just set it after you made your transformations[22:36:44] <Plagman> in the same space as your vertex data[22:37:18] *** Aintaer has quit IRC[22:37:23] <e-roder> glLight(glLightfv(GL_LIGHT0 + light_number, GL_POSITION, &lights[light_number]);[22:37:37] *** XT95_ has quit IRC[22:38:30] *** rorokimdim___ has quit IRC[22:38:38] *** rorokimdim___ has joined ##opengl[22:40:23] *** qeed has joined ##opengl[22:41:28] *** rorokimdim___ has quit IRC[22:42:00] <e-roder> Madsy: can you tell me more about using the shader for rendering?[22:42:20] <Madsy> http://www.opengl.org/registry[22:43:17] <Madsy> Either read the OpenGL 2.1 specifcation, or read about the following extensions: GL_ARB_shader_objects, GL_ARB_vertex_shader and GL_ARB_fragment_shader[22:43:33] <Madsy> And get the v1.20 or later version of the GLSL specification[22:43:43] <e-roder> ok[22:43:56] <Madsy> Both the GLSL and OpenGL language specifications are found on opengl.org[22:43:56] *** dusted_ has quit IRC[22:44:05] <e-roder> hmmm[22:44:24] <Madsy> opengl.org/registry list extensions which are not a part of the core spec yet.[22:44:37] <e-roder> alright[22:45:02] <Madsy> Those who are added to the core, aren't removed though.[22:45:12] <Madsy> They're still listed for completeness.[22:45:18] <e-roder> can i do stuff like write to the stencil buffer and such with the shaders?[22:45:42] <Madsy> If there is a way, I haven't heard about it.[22:45:57] <e-roder> so that stuff still has to be done by rendering in ogl?[22:46:04] <Madsy> Stencil operations aren't exactly like depth buffers or framebuffers[22:46:08] <e-roder> i'm working on dynamic lighting with shadow volumes right now[22:46:28] <Madsy> Stencil shadows isn't the end-all-be-all for shadowing[22:46:41] <Madsy> You can use depth textures instead.[22:47:01] <Madsy> I don't see a single reason for using stencil buffers anymore.[22:47:02] <e-roder> depth textures?[22:47:08] <e-roder> i'm learning i guess[22:54:30] *** Nescafe has quit IRC[22:54:40] *** KU0N has quit IRC[22:54:56] *** GX has joined ##OpenGL[22:57:49] *** Yustme has quit IRC[23:00:50] *** dusted_ has joined ##OpenGL[23:04:16] *** Bollinger has quit IRC[23:04:37] <e-roder> Madsy: how can i use depth textures?[23:05:06] <Madsy> I could give an unsatisfying answer, and you could keep on asking new question.[23:05:16] <Madsy> Except, we would never get done[23:05:28] <e-roder> fair enough[23:05:30] <Madsy> I recommend you read the OpenGL 2.1 specification[23:06:13] *** mouserSVK has joined ##opengl[23:06:59] <emzic> hello, i think i just got my implementation of tangent calculation ready[23:07:16] <emzic> now how do i pass them on to the shader? with a vertex attribute?[23:07:34] *** stevosICBM has quit IRC[23:13:37] *** essoe- has quit IRC[23:21:41] *** GX has quit IRC[23:27:39] <AlastairLynn> emzic: yes[23:27:53] *** groton has quit IRC[23:29:21] <emzic> ok, but how do i get the id for the tangent-variable?[23:29:43] <emzic> i am trying with glGetAttribLocation(shaderid, "tangent"); but this gives me -1[23:34:03] <emzic> and in the vertexshader i have "attribute vec3 tangent;"[23:42:55] <mlucassmith> MatthiasM: it was index 0 that was the vertex problem, it was being bound to index 1. Nice little gotcha there.[23:43:06] <MatthiasM> heh[23:48:36] *** twist has joined ##OpenGL[23:49:48] <emzic> does anyone know why my glGetAttribLocation returns -1?[23:50:56] *** dusted_ has quit IRC[23:51:24] <MatthiasM> emzic: because the attribute was optimized away[23:51:46] <emzic> aha, because i didnt use it yet?[23:51:52] <MatthiasM> yep[23:52:12] *** itewsh has quit IRC[23:53:10] <emzic> no, even if i use it, it is still -1[23:58:07] *** JernejL has quit IRC