May 4, 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 4, 2008 [00:00:58] *** LordMetroid has quit IRC[00:05:51] *** LordMetroid has joined ##OpenGL[00:39:04] *** Burga_ has quit IRC[00:39:14] *** [this] has quit IRC[00:43:19] *** HuntsMan has joined ##opengl[00:46:08] *** dv_ has quit IRC[00:51:15] *** Yustme has quit IRC[00:54:34] *** scy has left ##opengl[00:55:56] <ezeki3l> can someone tell me what might cause texture rendering problem like this[00:55:58] <ezeki3l> http://cloudmotion.com/~pavlenko/tmp/spheremap.png[00:56:31] <ezeki3l> i'm using gluSphere(q, 0.5f, 32, 16);[01:00:48] *** dv_ has joined ##OpenGL[01:16:10] *** rutski has quit IRC[01:18:25] *** rutski has joined ##OpenGL[01:19:22] *** x0rx0r_ has quit IRC[01:21:15] *** rsp has left ##opengl[01:22:03] *** LordMetroid has quit IRC[01:25:45] *** Jernej has joined ##OpenGL[01:33:11] <LiQuiDninja> ezeki3l: a request like that will attract only guesses like "hmm maybe the texture coords are reversed" im not saying dont ask but read the topic and paste code samples www.pastebin.com[01:37:40] *** Xmas| has joined ##OpenGL[01:43:14] *** DMINATOR has quit IRC[01:44:17] *** JernejL_ has quit IRC[01:46:13] *** Jorachim has quit IRC[01:58:17] <ezeki3l> ok[01:58:43] <ezeki3l> http://pastebin.com/m62d14ead[01:58:53] <ezeki3l> here is the file i compile[01:59:34] <ezeki3l> just wierd texture[01:59:37] <ezeki3l> i don't know why[02:13:10] <Andon> You sure that's the right URL? It's 404 for me...[02:13:35] <ezeki3l> http://pastebin.com/m62d14ead[02:13:37] <ezeki3l> yea[02:13:46] *** kaotrix has quit IRC[02:14:34] <ezeki3l> http://cloudmotion.com/~pavlenko/tmp/test.c[02:20:06] <Andon> Well, for one thing, you never enable tex coord generation. I don't remember if gluSphere needs that, but you set it to GL_SPHERE_MAP at one point, but leave it disabled (default state).[02:22:12] *** Jernej has quit IRC[02:23:55] <ezeki3l> Oo[02:26:13] <ezeki3l> you mean enable it with glTexGeni?[02:27:04] <Andon> No, enable GL_TEXGEN_S and GL_TEXGEN_T with glEnable (...);[02:28:00] <Andon> You have those two lines commented out, which makes setting the texgen mode to GL_SPHERE_MAP in your init function amount to nothing.[02:28:06] *** Tibor__ has quit IRC[02:28:56] <ezeki3l> um[02:29:19] <ezeki3l> yea, they're comented out, i don't know why but with them the texture doesn't move at all[02:29:36] <ezeki3l> it's as if it is a plane no matter how you look at it[02:31:08] <ezeki3l> ok[02:31:10] <ezeki3l> i fixed[02:31:14] <ezeki3l> i didn't have[02:31:15] <ezeki3l> glEnable(GL_CULL_FACE);[02:31:16] <ezeki3l> glCullFace(GL_BACK);[02:32:07] *** Ragnarok has quit IRC[02:36:30] <Andon> That doesn't sound right though... You shouldn't be running into backface issues drawing a single sphere.[02:36:52] <Andon> Also, you need to clear the depth buffer every frame.[02:38:46] <ezeki3l> Oo[02:38:48] <ezeki3l> ok[02:39:28] *** JohnP789 has joined ##OpenGL[02:39:38] <ezeki3l> what's the command for clearing depth buffer?[02:40:01] <Andon> Add | GL_DEPTH_BUFFER_BIT to your glClear call.[02:42:00] <ezeki3l> thanks[02:42:19] <Andon> Do you still need to enable backface culling after that change?[02:42:31] <Andon> That might have been your problem all along.[02:42:54] *** juanmabc has quit IRC[02:46:27] <ezeki3l> no[02:46:33] <ezeki3l> still works[02:46:42] <ezeki3l> without re enabling backface calling[02:48:06] <Andon> Ah, good. You should probably go and enable it anyway, doesn't hurt to leave it on.[02:50:35] <ezeki3l> ok[02:50:40] *** blight_ has quit IRC[02:51:03] <ezeki3l> Andon: do you know how to disable fragment programs?[02:51:08] <ezeki3l> is it just the unbind call?[02:51:18] *** servus_ has joined ##opengl[02:54:35] <Andon> That'd probably work... but I think the proper thing to do is call glDisable (GL_FRAGMENT_PROGRAM_ARB);[03:00:59] *** aalex has joined ##OpenGL[03:02:43] <ezeki3l> hm[03:02:44] *** blight_ has joined ##opengl[03:03:04] <ezeki3l> how would you enable/disable two different fragment programs?[03:04:08] <Andon> You can't. To do that, you'd need to use multiple passes.[03:04:56] <ezeki3l> hmm[03:05:00] <RTFM_FTW> RTT and glBindProgramARB(...)[03:06:03] <RTFM_FTW> i.e. [bind program zero] ... render into a texture attachment ... [bind program one ] ... render into texture attachment ... etc.[03:06:30] <ezeki3l> Oo, interesting[03:06:32] *** servus has quit IRC[03:06:39] <RTFM_FTW> then on the final display pass bind to the default window buffer and render a quad (or another primitive)[03:06:58] <ezeki3l> i wonder tho[03:06:58] *** dvoid_ has quit IRC[03:07:16] <ezeki3l> if i can pass a parameter between certain glVertex calls into fragment program[03:07:24] <ezeki3l> and then just branch in the program it self[03:07:28] <ezeki3l> i only need a small change[03:07:35] <RTFM_FTW> yes you can[03:07:45] <ezeki3l> that would make it atonz slower[03:07:51] <ezeki3l> heh[03:08:37] <RTFM_FTW> you can pass per vertex "varying" in GLSL speak parameters into the programmable fragment pipeline (i.e. whatever fragment shader you happen to be using)[03:09:31] *** JohnP789 has quit IRC[03:09:39] <RTFM_FTW> those "varying" variables are calculated per vertex (i.e. written to) and read from per fragment[03:11:11] <ezeki3l> Oh[03:11:24] <ezeki3l> man, i should have looked into useing GLSL[03:11:32] <ezeki3l> i'm using GC right now[03:11:38] <Andon> Cg you mean? :)[03:11:41] <ezeki3l> er[03:11:42] <ezeki3l> yea[03:11:43] <RTFM_FTW> yes that would be wise since it is the shading path going forward[03:11:44] <ezeki3l> heh[03:12:48] *** dv_ has quit IRC[03:13:46] <Andon> Cg's not a bad place to start - it's almost identical to HLSL (DirectX's language) and all of the concepts exist in GLSL.[03:14:14] <RTFM_FTW> not necessarily[03:14:31] <RTFM_FTW> they exist on a single vendor specific implementation[03:14:43] <RTFM_FTW> they do *not* exist on all other implementations[03:14:45] <Andon> Well, if you avoid the nv30/40/xxx stuff in Cg.[03:14:46] <RTFM_FTW> unlike GLSL[03:15:29] <RTFM_FTW> GLSL (on implementations which offer hardware support for it) support the full (required) breadth of the implementation which cannot be said for Cg[03:16:06] <RTFM_FTW> furthermore its not adequately supported or platforms beyond Linux or Windows[03:16:48] <RTFM_FTW> OTOH GLSL is (for the most part) adequately supported on any OpenGL accessible OS[03:19:56] *** Jorachim has joined ##openGL[03:21:36] <Andon> That still doesn't make it a bad place to _start_. It's a lot simpler to implement , with quite a bit more reading material.[03:21:41] <Andon> (on the platforms that support it)[03:21:56] <Andon> It's certainly not a good place to remain...[03:22:11] <RTFM_FTW> actually everything I said makes it a "bad place to start" AFAIC[03:23:28] <RTFM_FTW> and its honestly no easier to implement (on the driver side) nor is it any easier to develop with on the client side[03:23:37] <RTFM_FTW> versus GLSL[03:23:59] <RTFM_FTW> if anything I'd say that the opposite happens to be true[03:24:24] <Andon> Maybe not on OS X, but other platforms make managing the GLSL set of extensions quite tedious.[03:24:39] <Andon> For a beginner.[03:24:40] <RTFM_FTW> being that you don't have to be worried about whether or not your specific Cg on vendor XYZ implementation (assuming its not necessarily a Nvidia GPU) happens to fully support the runtime[03:24:47] <HuntsMan> Andon: just use core :)[03:24:48] <RTFM_FTW> and who mentioned Mac OS X?[03:24:53] <RTFM_FTW> I know I didn't[03:25:34] <RTFM_FTW> and as HuntsMan stated assuming you have a GLSL core GL API implementation (i.e. anything GL 2.0 or newer) that also isn't much of a issue[03:25:52] <HuntsMan> yeah[03:26:10] <HuntsMan> RTFM_FTW: BTW what GLSL version does the ARB version implement?[03:26:13] <HuntsMan> 110?[03:27:05] <RTFM_FTW> GL 2.0 happens to be 110 on several implementations... although there are implementations where it happens to be 120[03:27:32] <RTFM_FTW> so that might vary depending upon the specific implementation(s) being looked at[03:27:35] <HuntsMan> yeah, but using ARB versions does provide version 120?[03:27:50] <RTFM_FTW> as of 2.1 yes[03:28:09] <HuntsMan> ok[03:28:17] <RTFM_FTW> one of the things GL 2.1 did is promote the GLSL 1.20 specification[03:28:58] <RTFM_FTW> as noted by this page: http://opengl.org/documentation/current_version/[03:29:03] *** blight_ has quit IRC[03:29:12] *** Xmas| has quit IRC[03:30:21] *** blight_ has joined ##opengl[03:39:23] *** cyates has quit IRC[03:41:42] *** rnx has left ##opengl[03:44:08] *** Osah has joined ##OpenGL[03:46:47] *** blight_ has quit IRC[04:20:37] *** thesquib has joined ##OpenGL[04:24:30] *** Suprano has quit IRC[04:26:56] <speedy1> which GLSL version (if any) enables defining constant arrays?[04:27:31] <HuntsMan> 120[04:27:42] *** Burga has joined ##OpenGL[04:36:19] *** Jorachim has quit IRC[04:47:02] *** Burga has quit IRC[04:47:56] *** Latheesan has joined ##opengl[04:48:11] <Latheesan> hi everyone[04:48:20] <Latheesan> in C++ OpenGL & GLUT, number 27 == Esc key, what number corresponds to Up, Down, Left and Right?[04:50:51] <Latheesan> ah, nvm, found it[04:54:56] *** apple_ide has joined ##OpenGL[04:55:41] <hibread> Latheesan: next time, ask your question out aloud to yourself. You're bound to find the answer! :)[04:58:02] <Latheesan> lol, i asked and got impatient for a reply, so i googled random queries and finally arrived at a solution[04:58:32] <Latheesan> tho i did do some searching before asking, i just got lucky this time[04:58:38] *** aalex has quit IRC[04:59:05] <hibread> what, waiting for 2 mins?? :)[05:00:29] *** thesquib has quit IRC[05:01:21] <apple_ide> [Newbie Q, 2D:] You know how glTexCoord2f(); takes parameters which are coordinates for mapping a texture? if you plug in normal numbers, (eg 36,36), then the origin is at the centre of the (say... QUAD) being mapped. Is there any way to have the origin at top left corner of the QUAD? offsetting the parameters ,(which I couldn't get to work)?[05:03:47] *** deathstar550 has joined ##OpenGL[05:03:50] <apple_ide> currently, the texture mapping is *fine*... but the texture starts at the centre of the quad and wraps around the quad[05:05:58] <apple_ide> .. which is not fine[05:09:03] <apple_ide> All the tutorials uses (1.0f, 0.0f,0.0f) etc.... but I couldn't get an exact floating point number correlating to the height and width of the rectangle within the texture which I want to use[05:09:17] <apple_ide> that seems to screw up the texture mapping[05:10:55] <Latheesan> can someone take a quick look at this : http://pastebin.ca/1006599[05:11:06] <Latheesan> in the game, if i press "esc" key, the game window exists[05:11:22] <Latheesan> however, if i press up or down, the cannon in the game does not rotate as expected[05:11:34] <NightBird> if I wanted to know if a point would end up lieing on the screen, what's the easiest way to do so(irregardless of if it's perspective/orthographic projections)?[05:11:52] <Latheesan> have i got the wrong key value for UP and DOWN ?[05:13:43] <NightBird> for some reason I suspect I could via multiplying the projection and model view matrix, then multiplying the point against that, and the resulting data could tell me if it's on screen or not... is that right?[05:15:04] *** chrisjw has joined ##opengl[05:20:07] *** Plagman has joined ##OpenGL[05:22:34] *** speedy1 has quit IRC[05:26:40] *** amalon has quit IRC[05:37:10] <apple_ide> worked around it..... yay[05:47:44] *** Ademan_ has quit IRC[05:48:30] *** Ademan_ has joined ##OpenGL[05:55:00] *** thesquib has joined ##OpenGL[05:55:44] <Latheesan> what does this error means?[05:55:44] <Latheesan> http://img76.imageshack.us/img76/999/79304030qc2.png[05:56:16] <HuntsMan> well you're trying to vsprintf into a too small buffer[05:57:13] <Latheesan> hmm today is my first day attempting to do simple game with c++ and openGL[05:57:24] <Latheesan> i started to get the error after i added these 4 lines[05:57:25] <Latheesan> 1 sec[05:57:46] <Latheesan> char angleString[20];[05:57:47] <Latheesan> int angle = ceil(cannon.GetAngleDegrees());[05:57:47] <Latheesan> sprintf_s(timeString, "Angle: %is", angle);[05:57:47] <Latheesan> renderBitmapString(320,350, GLUT_BITMAP_HELVETICA_18, timeString);[05:58:12] <HuntsMan> char angleString[128];[05:58:17] <Latheesan> ahh[05:58:18] *** BahamutZERO has quit IRC[05:58:39] <Latheesan> im still getting the same error[05:58:59] <Latheesan> could it be that the angle is a really long float number like 1.4939234234923912312312323....[05:59:10] <NightBird> er... you're having it print an integer[05:59:11] <Latheesan> i thought using ceil() function would round it to a single digit[05:59:16] <NightBird> the angle is -360 to 360..[05:59:18] <HuntsMan> you realize that you're printing into timeString, and what's it's size?[05:59:33] <Latheesan> haha my bad[05:59:49] <Latheesan> timeString was 10[05:59:52] <NightBird> ah yeah...[05:59:55] <NightBird> there you go then[05:59:56] <Latheesan> i forgot to change that to angleString[06:00:01] <HuntsMan> lol[06:00:18] <NightBird> copy and pasteing code should only done if you're paying very close attention[06:00:52] <Latheesan> lol i made quite a few bad mistake, bah im never copiing and pasting[06:00:54] <Latheesan> thanks guys[06:01:40] <Latheesan> you dont suppose you know how to print that little "o" that represent the number as a angle do you?[06:01:47] <Latheesan> e.g. 45 o[06:04:06] *** Plagman has quit IRC[06:04:15] <NightBird> the ascii value is 186[06:04:45] <NightBird> so you could do something like printf("%i%c",degrees,186);[06:05:03] <NightBird> actually, you may want to do (char)186 rather than just a straight 186[06:05:18] <Latheesan> ah excellent, thanks :)[06:05:21] <NightBird> there's probably an easier way to do it instring, but I don't know it off the top of my head[06:05:30] <NightBird> maybe /186 or something.. *shrugs*[06:05:32] *** Plagman has joined ##OpenGL[06:07:26] <Latheesan> hmm[06:07:27] <Latheesan> i tried this[06:07:34] <Latheesan> sprintf_s(angleString, "Cannon Angle: %i%c", angle, (char)186);[06:07:40] <NightBird> okay[06:07:41] <Latheesan> it didnt work[06:07:46] <NightBird> hm..[06:08:30] <NightBird> that may be, it depends on the format it's outputting, and if it supports characters outside the normal range[06:10:19] *** servus has joined ##opengl[06:10:27] <Latheesan> hmm[06:10:56] <Latheesan> in html, it's °[06:12:58] <NightBird> ...yeah..[06:13:52] <Latheesan> this website http://www.goascii.com/[06:13:54] <Latheesan> says[06:14:11] <Latheesan> degree symbol ° == decimal 176[06:14:45] <Latheesan> so, how about this : sprintf_s(angleString, "Cannon Angle: %i%c", angle, char(176));[06:15:02] <Latheesan> woops, i mean sprintf_s(angleString, "Cannon Angle: %i%c", angle, (char)176);[06:16:28] <NightBird> okay[06:16:37] <NightBird> I must have read it wrong or something[06:16:37] <Latheesan> :( still doenst work[06:17:02] * NightBird shrugs[06:17:13] <NightBird> can you put a lower case o up and scaled down by any chance?[06:20:17] <Latheesan> dunno[06:23:52] *** DR0ID_ has joined ##OpenGL[06:24:18] *** nerdzyboy has quit IRC[06:24:59] *** servus_ has quit IRC[06:27:01] *** charlie_zzz is now known as charlie5[06:38:29] *** rutski_ has joined ##OpenGL[06:40:31] *** Sp0tter|tf2 has joined ##OpenGL[06:41:15] <Sp0tter|tf2> What is a good way to control frames per second displayed? I created a little timer class, and in my main loop put a if(fps_timer->Read()){ display();} but after profiling I found that it takes more time to check the timer then to display[06:41:18] *** Sp0tter|tf2 is now known as Sp0tter[06:44:13] *** chrisjw has quit IRC[06:48:54] <Latheesan> i had a similar question, how do you display the fps on the game window?[06:49:29] <Sp0tter> thats simple[06:49:35] <Latheesan> oh?[06:49:38] <Sp0tter> just record the clock time over 10 frames[06:49:48] <Latheesan> hmm[06:49:53] *** thesquib has quit IRC[06:49:53] <Sp0tter> so you snapshot the clock, wait 10 frames, snap it again[06:49:54] <Sp0tter> divide by 10[06:50:28] <Latheesan> im new to this, can you explain how this works?[06:50:49] <Sp0tter> what exactly are you using, and what exactly are you trying to do[06:51:01] <Latheesan> im using C++ OpenGL and GLUT[06:51:06] <Latheesan> and what i am trying to do is[06:51:14] <Latheesan> show what the fps on my simple game[06:51:21] <Sp0tter> http://www.lighthouse3d.com/opengl/glut/index.php?fps[06:51:26] <Sp0tter> google is amazing :)[06:51:31] *** rutski has quit IRC[06:51:32] <Latheesan> bah[06:51:36] <Latheesan> its amazing to everyone else except me[06:51:46] <Latheesan> i been searching for something like this long time now[06:52:03] <Sp0tter> doubtful, since i just searched for 'opengl frames per second'[06:52:21] <Latheesan> i been searching "C++ opengl glut count fps"[06:52:27] <Latheesan> and similar variant :p[06:52:42] <Sp0tter> heh[06:53:20] <Sp0tter> I'm trying to restrict my frames per second, my display funciton goes loops through a lot of vectors and really no need to ever display more than 30 fps[06:53:39] <Sp0tter> need every scrap of cpu power for my other threads[06:54:55] <hibread> just put vsync on. That'll restrict frame rate to 60hz (i assume you have LCD?)[06:55:25] <Latheesan> i downloaded the glutfps demo project that demonstrates how the fps is worked out[06:55:31] <Latheesan> when i try to compile it, i get this error :[06:55:34] <Latheesan> Error 1 error C2381: 'exit' : redefinition; __declspec(noreturn) differs c:\program files\microsoft visual studio 8\vc\include\stdlib.h 406[06:55:41] <Latheesan> Error 2 error C3861: 'exit': identifier not found c:\users\latheesan\desktop\glutfps\glutfps.cpp 228[06:55:51] <Latheesan> so im guessing it's going to be difficult to get the fps running :([06:55:54] <Sp0tter> hibread i need to be able to cut it down to 15 fps actually[06:56:10] <Latheesan> download link : http://www.lighthouse3d.com/opengl/glut/examples/glutfps.zip[06:56:19] <Sp0tter> I'm running a genetic algorithm in multiple threads also[06:56:23] <LordHavoc> so I've confirmed the nature of the Apple GL bug that affects Nexuiz - glMatrixMode(GL_TEXTURE) operations seem to manipulate the texture matrix on the glClientActiveTexture unit rather than the glActiveTexture unit as the spec states, a direct violation of spec[06:56:46] <LordHavoc> all other implementations get it right[06:57:22] <LordHavoc> I was able to make the bug go away by adding glClientActiveTexture calls next to every glActiveTexture call (since they're cheap operations this isn't a big deal)[06:57:24] <hibread> LordHavoc: what do you expect from the Apply community? They want to left field, and do so quite convincingly most of the time[06:58:09] <LordHavoc> I had previously tried adding such calls in my texmatrix function[06:58:22] <LordHavoc> but that didn't fix it[06:58:50] <LordHavoc> because there was some other code that directly called glActiveTexture, once I changed it to keep the two in sync that code started working[06:59:14] <LordHavoc> basically I had set a texture matrix on a given unit in the previous frame, but my reset at the start of the new frame failed to reset the texture matrix on the right unit because of this driver bug[06:59:23] <LordHavoc> but my code still flagged it as reset[06:59:28] <LordHavoc> since it didn't know that hadn't taken hold[06:59:45] *** Lemml has joined ##OpenGL[07:00:14] <hibread> LordHavoc: how many hours do you reckon you've spend debugging code for apple as opposed to other major os's (linux and windows)?[07:00:38] <LordHavoc> hibread: no idea, I don't know the relative staff sizes[07:00:47] *** mm^away has joined ##opengl[07:07:43] <Latheesan> thanks Sp0tter, i got the FPS to work, it was indeed very simple to do :)[07:08:05] *** thesquib has joined ##OpenGL[07:08:44] <Sp0tter> cool[07:11:31] <Latheesan> hey can u help me workout some logic[07:11:42] <Latheesan> ?[07:12:09] <Sp0tter> heh, maybe[07:12:16] <Sp0tter> kinda workin on my own stuff though[07:12:22] <Latheesan> ah :([07:12:36] <Latheesan> well this might be real easy for you[07:14:36] <Sp0tter> can never hurt to ask[07:14:53] <Latheesan> ok lemme take a screenshot quick[07:15:27] <Latheesan> http://img297.imageshack.us/img297/1751/76169030gf2.png[07:15:43] <Latheesan> this is the simle game im working on to learn c++ an opengl as well as glut alltogther[07:15:52] <Latheesan> what im stuck on is[07:16:04] <Latheesan> when i press up, the cannon rotates upwards[07:16:11] <Latheesan> and opposite for the down key[07:16:24] <Latheesan> what i am trying to work out is how to map the same rotation onto the cannon ball[07:16:46] <Latheesan> so the ball stays inside the mouth of the canon[07:16:49] <Latheesan> cannon*[07:17:08] <Latheesan> and moves up when the cannon moves up, moves down when the cannon moves down[07:17:21] <hibread> do you know the position of the centre of the canon?[07:17:27] *** mm765^away has quit IRC[07:18:14] <Latheesan> do you mean the x,y position of cannon ?[07:19:23] <RTFM_FTW> LordHavoc as I believe I've told you before if you really feel this to be in violation of the GL specification and ultimately a driver side issue you would be wise to file a Radar against this with Apple[07:19:47] <RTFM_FTW> bugs will never get fixed without doing this.[07:20:07] <Latheesan> http://pastebin.ca/1006704 this what makes up the cannon[07:25:52] <Latheesan> hairbread, after some reading, yes, i know the center of the cannon[07:26:21] <Latheesan> woops, typo, hibread*[07:27:41] <ezeki3l> is there a rand() function in vertex shader?[07:33:22] <ezeki3l> oh man[07:33:28] <ezeki3l> it realy looks like it doesn't have one[07:42:20] *** Sp0tter has quit IRC[07:44:58] <NightBird> ezeki3l: nope[07:50:18] <NightBird> most random number generators operate off of ints, which most gpu's don't have support for[07:50:28] <NightBird> though, more modern ones do[07:51:30] <ezeki3l> yea[07:51:44] <ezeki3l> i looke up a pretty cool hack for random numbers in vertex shader[07:52:29] <NightBird> hm.... I'm guessing it will use the standard inputs(vertex position, color, texture coordinates, normals, etc), and combine it in a way to create a psudo random number?[07:52:43] <ezeki3l> you need like, to generate random colors per pixel texture, pass it to the vertex shader[07:52:46] <ezeki3l> etc[07:52:53] <NightBird> maybe including a parameter that gets changed once a frame...[07:53:07] <ezeki3l> is it possible to alter vertex actual coordinates in vertex shader?[07:53:18] <NightBird> yeah, but it's expensive to generate the texture..[07:53:26] <NightBird> (not so much that it's prohibitive, just not ideal)[07:53:33] <ezeki3l> ye[07:53:41] <NightBird> ezeki3l: most vertex shaders do modify the output vertex in some way[07:53:46] <ezeki3l> ok[07:54:06] <ezeki3l> so let's say i have a sphere, and then i multiply every vertex of that object by 1.5, would it result in a bigger sphere?[07:54:13] <ezeki3l> asuming it's centered at <0,0,0>[07:54:24] <NightBird> like changing the position, colors, normals, textures, etc... plus communicating important info to the fragment shader that may not be passed otherwise[07:54:33] <ezeki3l> ok[07:54:39] <ezeki3l> hmm[07:54:41] <NightBird> ezeki3l: sure, but that could easily be done with a glScale3f(1.5f,1.5f,1.5f);[07:55:12] <ezeki3l> ok[07:55:20] <ezeki3l> i just want to see how vertex shader works[07:55:21] * NightBird personally finds fragment shaders to be more interesting than vertex shaders, but I'm just not as imaginative, I'm sure[07:55:38] <Latheesan> is there a good forums for openGL discussion? where u can post ur problems and hopefully get some pointers?[07:55:55] <NightBird> I didn't even care much for geomtery shaders, though I started reading through my 3rd gpu gems and saw some really good uses for it, so I may have to revise my opinion on them..[07:56:21] <NightBird> I'm sure there are several, I just don't know of any off the top of my head to point you too[07:56:35] <Latheesan> hmm[07:57:28] <ezeki3l> NightBird: will this make the object 50 times bigger?[07:57:29] <ezeki3l> http://pastebin.com/m67799d00[07:57:58] <NightBird> oh... you're using the nvidia shader language rather than glsl..[07:58:31] <NightBird> you'll want to multiply the vertex before you multiply it by the modelviewproj[07:58:55] <NightBird> multiplying it by modelviewproj will put it into screen cordinates[07:59:27] <ezeki3l> oh[07:59:28] <ezeki3l> right[07:59:29] <ezeki3l> hm[07:59:32] <ezeki3l> still won't get bigger[08:00:16] <Latheesan> hmm this is how my game looks like at the moment, http://img381.imageshack.us/img381/2393/39016017gn0.png[08:00:33] <Latheesan> when i move the cannon, the ball moves along with it, however, it moves faster and gets out of the range[08:00:42] <Latheesan> this is what i use :[08:00:43] <Latheesan> cannonBall.SetPos(cannonBall.GetXPos() + cos(cannon.GetAngleRadians()), cannonBall.GetYPos() + sin(cannon.GetAngleRadians()));[08:00:52] <Latheesan> when moving cannon up[08:00:57] <Latheesan> and this is for moving cannon down[08:01:01] <Latheesan> cannonBall.SetPos(cannonBall.GetXPos() - cos(cannon.GetAngleRadians()), cannonBall.GetYPos() - sin(cannon.GetAngleRadians()));[08:01:19] <Latheesan> anyone might be able to shed some light on where might be the problem?[08:02:28] <NightBird> why are you doing it like that?[08:03:19] <Latheesan> im not exactly math savvy, my friend is and he helped me work it out like that[08:04:42] <NightBird> so wait, you can control the cannon angle while a ball is flying, right?[08:04:56] <Latheesan> the ball doesnt fly at the moment[08:05:07] <Latheesan> what i am trying to achieve is, when the cannon moves up or down[08:05:10] <NightBird> ah[08:05:13] <Latheesan> i wanted the ball to stay in the cannon's mouth[08:05:14] <NightBird> and the ball to move with it[08:05:15] <NightBird> gotcha[08:05:18] <Latheesan> yup[08:05:25] <Latheesan> at the moment, the ball moves too fast[08:05:35] <Latheesan> and gets out of the mouth into some random are[08:05:37] <Latheesan> area*[08:05:38] <NightBird> I'm guessing it looks fine for a bit, then goes crazy[08:05:52] <Latheesan> yup[08:05:55] <NightBird> yeah[08:05:56] <Latheesan> http://img258.imageshack.us/img258/3212/95325995mc5.png[08:06:04] <Latheesan> this waht happens if i move up too much[08:06:15] *** LiQuiDninja has quit IRC[08:06:23] <Latheesan> woops, wrong screenshot[08:06:58] <Latheesan> http://img380.imageshack.us/img380/9159/98738468ww1.png[08:06:59] <Latheesan> there[08:07:02] *** enoex_ has quit IRC[08:07:07] <NightBird> yeap[08:07:26] <NightBird> the problem is that you're modifying the position with the new angle, and not the original position with the new angle[08:07:45] <Latheesan> so instead of[08:07:50] <Latheesan> cannonBall.SetPos(cannonBall.GetXPos() + cos(cannon.GetAngleRadians()), cannonBall.GetYPos() + sin(cannon.GetAngleRadians()));[08:07:56] <Latheesan> i do like this :[08:08:12] <Latheesan> cannonBall.SetPos(cannonBall.OrignalXPos() + cos(cannon.GetAngleRadians()), cannonBall.OrignalYPos() + sin(cannon.GetAngleRadians()));[08:08:47] <NightBird> yeap[08:08:55] <NightBird> when doing adjustments such as rotation and scaling, the original data gets destorted over time, resulting in a loss of data.[08:08:59] <Latheesan> ok awsome, i'll try[08:09:48] <Latheesan> ok now, the ball moved 1 place[08:09:52] <Latheesan> and then it just stood there[08:10:23] <Latheesan> cannonBall.SetPos(75 + cos(cannon.GetAngleRadians()), 333 + sin(cannon.GetAngleRadians())); (this is what i used, 75 is the orignal x pos of the ball, 333 is the orignal y pos of the ball)[08:11:03] <NightBird> yeah, try adding a multiplier to the sin and cos functions[08:11:19] <Latheesan> multiply by what?[08:11:28] <NightBird> according to that, your cannonball will range from 74 to 76 for the x, and 332 and 334 for the y[08:11:31] <NightBird> probably not what you want[08:11:32] <NightBird> dunno[08:11:41] <NightBird> I would suggest... 8[08:11:50] <NightBird> tweak the number till it looks right[08:12:22] <NightBird> of course, you could be doing this via the opengl rotations and translations so it handles all of this math for you, but there you go[08:13:10] <Latheesan> hmm tried[08:13:12] <Latheesan> it didnt work[08:13:20] <NightBird> then you probably didn't do it right[08:13:23] <Latheesan> jumped to some random position and slowly rotated[08:14:04] <Latheesan> cannon.SetAngleDegrees(cannon.GetAngleDegrees() + 1);[08:14:04] <Latheesan> cannonBall.SetPos(75 + (cos(cannon.GetAngleRadians() * 8)), 333 + (sin(cannon.GetAngleRadians()) * 8));[08:14:13] <Latheesan> this is what i have, when u press up key[08:14:16] <NightBird> pushmatrix, translate to rotation point of the cannon, apply the cannon rotation, then translate out to where you want the ball to be. draw the circle that represents your ball, popmatrix[08:15:47] <Latheesan> hmm[08:17:51] <NightBird> okay, I'm heading to bed now[08:19:09] <Latheesan> nighto[08:21:20] <LordHavoc> RTFM_FTW: I am filing a report, but my theory is flawed, still testing out more specifics[08:21:37] <LordHavoc> RTFM_FTW: what I have confirmed is that it is absolutely a confusion between clientactivetexture/activetexture[08:21:52] <LordHavoc> RTFM_FTW: calling both in all cases makes the bug go away[08:22:03] <LordHavoc> RTFM_FTW: calling them in all cases to do with texture matrix setting, however, does not[08:22:08] <LordHavoc> RTFM_FTW: so there's something more to it[08:22:31] <LordHavoc> RTFM_FTW: and this is known to only affect GLSL, so I still think there is something funky about uniforms[08:22:41] <LordHavoc> RTFM_FTW: (built in uniforms, rather)[08:22:48] <LordHavoc> RTFM_FTW: but anyway, still working on the details[08:27:28] *** Coke has joined ##OpenGL[08:28:38] *** rutski_ has quit IRC[08:36:30] <ezeki3l> hmm[08:36:49] <ezeki3l> if i pass a texture to vertex shader, how can i access the color elevent of a pixel in it?[08:42:40] <Latheesan> if anyone is free to help, can you check this out please? http://www.opengl.org/discussion_boards/ubbthreads.php?ubb=showflat&Number=237876[08:43:38] <LordHavoc> ezeki3l: you are aware that only works on GF6/7/8/9/Radeon HD?[08:43:57] <ezeki3l> heh[08:44:06] <ezeki3l> yea, i just found ot from google[08:44:50] <ezeki3l> tex2Dlod( textureSampler, float4( uv, 0, 0) );[08:45:05] <ezeki3l> i duno what 'uv' is in this cas, and is it provided automaticaly[08:45:18] <LordHavoc> ezeki3l: ATI (annoyingly) doesn't have that capability in their older cards[08:45:30] <LordHavoc> ezeki3l: and encourages use of FBO rendering to vertex buffers instead[08:46:16] <LordHavoc> ezeki3l: I recommend looking at glsl_quickref.pdf[08:46:18] <ezeki3l> darn[08:46:20] <LordHavoc> ezeki3l: for parameter types[08:46:23] <ezeki3l> ok[08:48:12] <ezeki3l> so is there any way to access color info on texture with a given S, T values in Vertex Shader?[08:49:56] <LordHavoc> yes[08:50:00] <LordHavoc> uv are the s,t values[08:50:13] <LordHavoc> I don't know why some people call them uv[08:50:13] <LordHavoc> OpenGL calls them s,t[08:50:14] <ezeki3l> i mean on the older cards[08:50:17] <LordHavoc> nope[08:50:25] <LordHavoc> absolutely no way to read a texture on the older cards in a vertex shader[08:50:41] <ezeki3l> hm[08:50:42] <ezeki3l> ok[08:50:42] <LordHavoc> a real shame too, textures would be ideal for skeletal animation[08:50:44] <LordHavoc> (bone tables)[08:51:16] <ezeki3l> yeh[08:51:55] <adampetrone> u and v is what they're called in most 3d modeling apps[08:52:14] *** Killari has quit IRC[08:52:19] <LordHavoc> displacement terrain is lots of fun too[08:52:31] <ezeki3l> do you know, in CG, if the prfile is just the last ingeter in "ARBVP1"?[08:52:32] <LordHavoc> so easy to just have a fixed mesh that follows the camera, with a mipmapped heightmap texture[08:56:09] <ezeki3l> yay[08:56:13] <ezeki3l> VP40 works[08:56:17] <ezeki3l> i must have the "newer" card[08:56:18] <ezeki3l> heh[08:56:33] *** Killari has joined ##OpenGL[08:57:11] *** scy has joined ##opengl[09:05:43] *** prophile_ has joined ##opengl[09:13:04] *** prophile has quit IRC[09:18:37] *** enoex has joined ##OpenGL[09:30:07] *** darka has joined ##opengl[09:31:00] *** groton has joined ##OpenGL[09:33:28] <ezeki3l> is there an access to rotation matrix in Vertex Shader?[09:33:36] <ezeki3l> or i must build my own[09:38:22] <LordHavoc> RTFM_FTW: hmm... much more interesting result - the problem seems to be cured by simply moving the glMatrixMode(GL_TEXTURE) call after the glActiveTexture call[09:38:32] <LordHavoc> RTFM_FTW: nothing to do with glClientActiveTexture[09:42:47] *** LiQuiDninja has joined ##OpenGL[09:52:41] *** Burga has joined ##OpenGL[09:55:56] <ezeki3l> man[09:56:01] <ezeki3l> why do i get error[09:56:03] <ezeki3l> error C1020: invalid operands to "*" mat3[09:56:10] <ezeki3l> vec3 k = normalize(modelviewproj * normal);[09:56:12] <ezeki3l> on this line[09:56:23] <ezeki3l> modelviewproj is 3x3[10:03:12] <ezeki3l> even these lines[10:03:13] <ezeki3l> mat3 rot_m = mat3(vec3(1,0,0), vec3(0,1,0), vec3(0,0,1));[10:03:14] <ezeki3l> vec3 k = rot_m * vec3(1,0,0);[10:03:17] <ezeki3l> produce same error[10:04:44] <Muzzzle> have you done the right overloading?[10:04:53] <ezeki3l> what do you mean?[10:05:02] <ezeki3l> mat3 is 3x3 matrix right?[10:05:02] <Muzzzle> can you paste the code?[10:05:08] <ezeki3l> what's the overloading[10:05:30] <Muzzzle> isn't mat3 your own class?[10:05:36] <ezeki3l> the cgc compiler gives error error C1020: invalid operands to "*" mat3[10:05:42] <Muzzzle> and vec3[10:05:47] <ezeki3l> on the line vec3 k = rot_m * vec3(1,0,0);[10:06:00] <ezeki3l> it's the same error even if i use float3[10:06:19] <ezeki3l> i thought these types are default[10:06:26] <ezeki3l> shouldn't they be overloaded automagicaly?[10:06:27] <Muzzzle> what lib. are you using?[10:06:34] <ezeki3l> it's vertex shader[10:06:39] <Muzzzle> oh[10:06:55] <Muzzzle> weird then[10:07:01] <ezeki3l> yea[10:07:34] <Muzzzle> it's probably something about the overloading of * though[10:07:38] <Muzzzle> what else could it be[10:08:07] <ezeki3l> it's mat3 and vec3 are built in types[10:08:16] *** thesquib has quit IRC[10:08:22] <ezeki3l> i don't know how to begin the overloading even[10:10:11] <Muzzzle> you are sure you should use the * and not another command for multiplying?[10:10:25] <LordHavoc> ezeki3l: try vec3 * mat3 order instead[10:10:42] <LordHavoc> ezeki3l: (remember matrix multiply is not commutive)[10:10:44] <Muzzzle> yea order can matter yes[10:12:13] <ezeki3l> mat3 rot_m = mat3(vec3(1,0,0), vec3(0,1,0), vec3(0,0,1));[10:12:14] <ezeki3l> vec3 k = vec3(1,1,1)*rot_m;[10:12:16] <ezeki3l> same error[10:12:29] <ezeki3l> error C1020: invalid operands to "*"[10:13:07] <LordHavoc> ezeki3l: I don't know myself, because I simply don't use Cg or mat3[10:13:30] <ezeki3l> ok[10:13:33] <ezeki3l> thanks for your help[10:13:40] <LordHavoc> I currently only use the texture matrix builtins[10:13:42] <ezeki3l> i'm going to hardcode matrix operation[10:13:43] <ezeki3l> lol[10:13:51] <LordHavoc> what is your purpose?[10:14:23] <LordHavoc> usually the only times people use smaller than mat4 types are cases of skeletal animation[10:14:30] <LordHavoc> (which only need 3 of the vectors, not all 4)[10:15:29] <LordHavoc> so those tend to be fed as a vec4 array uniform[10:15:49] <LordHavoc> using dot() calls to do the multiply[10:15:57] <LordHavoc> (dot is slow on SSE in CPUs, but fast on GPUs)[10:16:34] <LordHavoc> it's even more ideal to feed the vec4's as a GL_FLOAT texture, but due to limited hardware support that's only an optimized case, still need the fallback[10:17:25] <LordHavoc> even if the fallback only covers Intel GMA950, GeForce FX, and ATI Radeon 9500 through X1850[10:18:09] <LordHavoc> of which the first is generally too bad for any serious graphics, the second is adequate only for the highest models (which are relatively rare), horrible otherwise, and the last ranges from adequate to good[10:18:31] <LordHavoc> so basically it's a fallback case for Radeon 9500-X1850[10:18:51] <LordHavoc> (some models in that range also have horrible performance, but that's getting into too many details)[10:19:23] <LordHavoc> and err, sorry for ranting about vertex shader texture fetches[10:21:43] *** rsp has joined ##opengl[10:28:46] *** deathstar550 has quit IRC[10:34:10] *** deathstar550 has joined ##OpenGL[10:35:04] *** Tibor__ has joined ##OpenGL[10:37:54] *** TurboAWAY is now known as [AD]Turbo[10:39:45] *** MiniuM has joined ##OpenGL[10:40:11] <MiniuM> hello world[10:40:35] <MiniuM> plz i've between my hands an old P3 computer[10:40:57] <MiniuM> i don't know is it supports openGL or no[10:41:18] <MiniuM> 'cause Wolfenstein didn't work :([10:42:03] <MiniuM> there is a way to know if my graphic card support opengl?[10:43:42] *** enoex_ has joined ##OpenGL[10:43:43] <MiniuM> anyone alive here? :D[10:44:20] <LiQuiDninja> MiniuM: first download the drivers from your cards website, there are auto detectors on nvidias site http://www.nvidia.com/Download/index.aspx?lang=en-us[10:44:43] *** enoex has quit IRC[10:44:45] <LiQuiDninja> select option 2 :P[10:45:08] <MiniuM> but i have an intel integrated g card[10:45:21] *** Latheesan_ has joined ##opengl[10:45:28] <MiniuM> the nvidia stuff will work?[10:46:13] <LiQuiDninja> well if your sure youve downloaded the latest drivers then execute a benchmarking program, just dont try crysis[10:46:36] <MiniuM> :D[10:46:45] <MiniuM> or COD4[10:48:11] <LiQuiDninja> OGL will only scare you with blank screens when your over confident with wild function calls, dont worry about damaging your new mean machine[10:48:53] *** karabash has joined ##OpenGL[10:53:27] *** apple_ide has left ##OpenGL[10:58:47] *** replor has joined ##OpenGL[10:59:56] *** thesquib has joined ##OpenGL[11:00:12] *** BahamutZERO has joined ##OpenGL[11:01:24] *** MiniuM has quit IRC[11:07:33] *** dvoid_ has joined ##OpenGL[11:08:44] *** amalon has joined ##opengl[11:08:59] *** Coke has quit IRC[11:11:28] *** Yustme has joined ##opengl[11:12:10] *** m4ggus has joined ##opengl[11:12:44] *** MatthiasM has quit IRC[11:12:51] *** MatthiasM has joined ##opengl[11:22:22] *** groton has quit IRC[11:24:08] *** karabash has quit IRC[11:24:09] *** prophile_ has quit IRC[11:26:03] *** groton has joined ##OpenGL[11:28:26] *** IJs\laptop has joined ##OpenGL[11:53:02] *** mm^away is now known as mm765[12:18:24] *** dv_ has joined ##OpenGL[12:21:13] *** LtJax has joined ##opengl[12:28:03] *** thesquib has quit IRC[12:32:39] *** prophile has joined ##opengl[12:40:28] <rsp> What will it be this time competition.linuxgamepublishing.com/comp.jpg[12:44:25] <bobbens> i'm pretty sure it's a space game[12:44:32] <bobbens> but at first i thought the lgp logo was a sun :P[12:44:41] <rsp> xD[12:44:57] <bobbens> bg looks like nebulae[12:45:44] <rsp> bg looks like feh3%58j2€£@[38q0fq9€£@%53[12:50:34] <Weiss> very slightly OT, but has anyone found that Wavefront OBJ files exported from Blender have inconsistent winding order?[12:50:36] *** devilsadvocate has joined ##OpenGL[12:52:02] *** deathstar550 has quit IRC[13:06:52] *** Raj has joined ##OpenGL[13:08:28] <LordHavoc> rsp: is that one of those random dot stereograms?[13:09:18] *** amalon has quit IRC[13:09:49] *** blight_ has joined ##opengl[13:10:31] *** Suprano has joined ##OpenGL[13:10:59] <LordHavoc> rsp: or just a very noisy image?[13:12:01] *** Raj has left ##OpenGL[13:13:00] <LordHavoc> rsp: ahh I see[13:13:08] <LordHavoc> rsp: I remember this site[13:13:20] <LordHavoc> rsp: guessing what it is as it's very slowly revealed, heh[13:13:37] *** Raj has joined ##OpenGL[13:16:18] *** Raj has left ##OpenGL[13:20:16] <rsp> LordHavoc: Heh :)[13:28:35] *** JernejL has joined ##OpenGL[13:30:15] *** amalon has joined ##opengl[13:31:38] <LordHavoc> rsp: I'd be half tempted to check each day and download the image, then remove all identical pixels between an early (very noisy) image and a later (less noisy) image, then fill in the gaps by interpolation[13:31:52] <LordHavoc> rsp: but that would be cheating[13:32:03] <LordHavoc> rsp: (and I'm not sure it would help much)[13:34:15] <rsp> Hehe ;)[13:37:08] *** MrPrise has joined ##OpenGL[13:37:11] <MrPrise> hello[13:38:46] <MrPrise> I'm using codeblocks with mingw on windows and I got glActiveTextureARB was not declared error. however I see there is glActiveTextureARB in the GL/glext.h[13:40:46] <rsp> Maybe you have to load the function[13:41:13] <rsp> You can do it manually or use glew[13:42:17] <MrPrise> I found this: http://www.3ddrome.com/articles/dynamiclightmaps.php I'm trying to use that technic in my app[13:43:00] <MrPrise> it does not mention loading of function[13:43:24] <MrPrise> how can I do this?[13:46:17] <MrPrise> ok, I found it's site. thanks[13:46:54] <LordHavoc> MrPrise: ahem, the topic says where glew is[13:49:25] <MrPrise> ah, ok sorry.[13:52:16] *** dolphin has joined ##OpenGL[13:53:13] *** Xmas| has joined ##OpenGL[13:57:38] *** scy has left ##opengl[14:11:13] *** devilsadvocate has quit IRC[14:16:56] *** Lemml has quit IRC[14:19:05] *** dolphin has quit IRC[14:21:09] *** rnx has joined ##opengl[14:22:22] *** dolphin has joined ##OpenGL[14:22:32] *** dolphin has quit IRC[14:26:25] *** FordCortina has joined ##OpenGL[14:32:13] *** m4ggus has quit IRC[14:34:32] *** amalon has quit IRC[14:35:07] *** m4ggus has joined ##OpenGL[14:56:59] *** amalon has joined ##opengl[14:58:49] *** rnx has quit IRC[15:00:47] *** rnx has joined ##opengl[15:06:36] *** Walt has quit IRC[15:11:35] *** aalex has joined ##OpenGL[15:11:38] *** aalex has quit IRC[15:17:14] *** Trollinator has joined ##OpenGL[15:17:34] <Trollinator> Does anybody know when OpenGL 3.0 will be released?[15:20:29] *** __abel__ has joined ##OpenGL[15:21:12] <groton> Trollinator, lol[15:21:26] <groton> Trollinator, nice joke :)[15:21:51] <Trollinator> groton: it wasn't meant to be a joke[15:22:10] <hibread> Trollinator: i hear ya[15:23:47] *** Rangar has joined ##OpenGL[15:24:09] *** Jupp3 has joined ##OpenGL[15:29:27] <Trollinator> i mean, it was supposed to be done in Sept. 2007, and it still isn't[15:29:27] *** Plagman has quit IRC[15:31:31] *** rnx has quit IRC[15:31:37] *** joakim__ has joined ##OpenGL[15:32:05] <hibread> Trollinator: im more frustrated by the lack of progress updates[15:32:34] <Trollinator> yes, it seems to be very hard to find information about what's going on.[15:34:20] <rsp> :/[15:35:42] *** rutski has joined ##OpenGL[15:36:23] <NightBird> didn't they release a statement somewhat recently basically stating that they're having some problems deciding on a portion of the spec?[15:36:34] <hibread> NightBird: link?[15:37:00] <NightBird> hibread: I don't have it, but for some reason I'm under the impression that something like that was implied by them...[15:37:12] <NightBird> which could of course mean I'm completely off...[15:37:44] <hibread> do you recall it being an official report or some speculation in a forum?[15:37:56] <NightBird> probably speculation[15:37:59] <NightBird> I think it was in here[15:39:04] <hibread> maybe the whole groups on long service?[15:40:32] * NightBird shrugs[15:51:47] *** Plagman has joined ##OpenGL[15:51:53] *** Walt has joined ##opengl[16:00:12] *** Jorachim has joined ##openGL[16:02:28] *** BahamutZERO has quit IRC[16:03:18] *** WimLeers|school has joined ##OpenGL[16:04:39] *** wcstok has joined ##OpenGL[16:08:23] *** joakim_ has joined ##OpenGL[16:09:10] *** scy has joined ##opengl[16:10:30] *** phrosty has quit IRC[16:13:23] *** magnet has quit IRC[16:13:33] *** magnet_ has joined ##OpenGL[16:16:54] *** MrPrise has left ##OpenGL[16:22:31] *** phrosty has joined ##opengl[16:22:54] *** joakim__ has quit IRC[16:26:15] *** Jupp3 has quit IRC[16:27:05] *** joakim_ has quit IRC[16:32:46] *** kaotrix has joined ##OpenGL[16:34:55] *** __abel__ has left ##OpenGL[16:35:31] *** rnx has joined ##opengl[16:39:51] *** BahamutZERO has joined ##OpenGL[16:48:15] *** replor has joined ##OpenGL[16:48:42] *** amalon has quit IRC[17:00:53] *** Eltran has joined ##OpenGL[17:01:10] <Eltran> hey, how do you call the way that we can use our mouse to interact on a screen ?[17:01:17] <Eltran> like in Quake 4, a GUI inside a texture[17:01:37] <Eltran> and you can click on buttons in it, edit text,...[17:02:05] <Xmas|> well, GUI?[17:02:22] <Eltran> yeah but err... another gui in a shader[17:02:31] <Killari> double gui? ;>[17:03:20] *** amalon has joined ##opengl[17:03:23] <Xmas|> why would you call a GUI differently based on how it is rendered?[17:04:07] <Eltran> because I can't find it in google and I think it has a special name[17:05:00] <Eltran> it's just a button on a screen (shader) and when you put your crosshair on it, you see a new cursor on the shader and you can click on the button inside it[17:05:28] <groton> mouse hovering?[17:06:07] <hibread> Eltran: shader?[17:07:03] <Muzzzle> glutSolidSphere for spheres, but what about ellipsoids?[17:08:02] <Eltran> http://img.clubic.com/photo/00151748.jpg it's like that, yknow the guy can click on the button, how to do something similiar ? (just tell me the basics pls if ya know em)[17:08:34] <hibread> Eltran: i guess you'd need some picking function[17:09:02] <groton> hibread won a prize :)[17:09:16] <hibread> and if you happen to "pick" the relevant object, do() something special[17:09:31] <groton> of course, it is in the Red Book, even version 1.1[17:09:44] *** Trollinator has quit IRC[17:09:46] <hibread> yay, what price did i win? :)[17:09:52] <hibread> prize[17:09:57] *** amz has joined ##opengl[17:09:59] <Eltran> ok thanks[17:10:06] *** RamboZa has quit IRC[17:10:32] <groton> hibread, you guessed it right, just glory, or proudness[17:10:57] <hibread> excellent. I do feel proud! Thank you all[17:11:01] <groton> np[17:11:17] <groton> :[17:11:21] <groton> ;)[17:15:18] * rsp np[17:17:18] *** Catfish_ has joined ##OpenGL[17:17:39] <Catfish_> Heya[17:18:14] <Catfish_> Anyone familiar with marching cubes?[17:18:25] *** blight_ has joined ##opengl[17:18:29] <Catfish_> A simple sphere under marching cubes shouldn't generate overlapping faces, right?[17:18:51] <LtJax> nothing should[17:18:53] <LtJax> so yea[17:18:58] *** LtJax has quit IRC[17:19:01] <Catfish_> Hmph[17:19:49] *** Lemml has joined ##OpenGL[17:24:23] <Muzzzle> is there an easy way to draw if (this->sphere.v.GetZ() < 0.0001)[17:24:24] <Muzzzle> this->sphere.v.SetZ(0);[17:24:25] <Muzzzle> woops[17:24:48] <Muzzzle> is there an easy way to draw ellipsoids in opengl?[17:24:49] <garou> Hi once again, I've got another noobish question lined up: As far as I see it so far, there are three matrix stacks: projection, modelview and texture. What the texture matrix does, seems (so far) obvious to me, but what do modelview and projection do?[17:25:01] <Muzzzle> like glutSolidSphere for spheres[17:25:38] <groton> garou, have you read the relevant part about matrix stacks in the RedBook?[17:25:48] <Xmas|> garou, they transform the vertex positions from model space to view space and from view space to clip space[17:28:08] <garou> groton: I have to admit, no. Formal texts in this matter are normally either quite dense or depend on knowledge that is scattered over the rest of the text and thus are good reference material, but bad for learning. Maybe that's just a prejudice, but I'm working on a very specific and relatively small problem ATM.[17:28:47] <groton> garou, well, the Xmas| answer has been far smarter than my one :)[17:29:17] <garou> Thanks, Xmas| :)[17:30:03] <hibread> garou: if you plan to spend a relatively large amount of time with opengl, you may as well learn it properly which those text will allow you to do[17:30:24] <garou> Is there a current redbook online? I've only found ones that are on the state of OpenGL 1.1 and I'm a little too short on money for ordering technical literature.[17:30:59] <groton> garou, 1.1 is perfect[17:31:04] <hibread> and if you wish to accelerate your learning, i suggest maybe OpenGL Distilled?[17:31:11] *** Catfish_ is now known as Catfish[17:31:18] <garou> hibread: Actually, I don't. ^^ I want to cram learning the basics in a few weeks, then move on.[17:31:30] <hibread> move on to what?? :)[17:32:46] <garou> Right now, non-realtime image generation and mathematical models of geometry and topology; after that, neural networks and fuzzy set theory.[17:32:49] *** juanmabc has joined ##opengl[17:32:54] <groton> move on better things i suppose (heresy!)[17:32:55] *** rnx has quit IRC[17:33:42] <garou> To other topics in which I need cource credits.[17:34:03] <hibread> garou: fair enough. Well, i dunno... skim over the NeHe tutorials?[17:34:04] <garou> s/cource/course/[17:34:27] <Catfish> This is definitely wrong for a marching cube sphere, right? http://img395.imageshack.us/my.php?image=picture2za0.png[17:35:54] <garou> Did so, they're quite... ad hoc, exploring specific topics as they come up, always in part or at least seeming so. I think the subtle hint in the redbooks direction may have been a good idea, though I'm not really happy with reading a text that soon will be two major releases behind the current standard.[17:37:04] *** Madsy has joined ##opengl[17:37:11] <hibread> garou: the basics of pre-opengl3.0 have not changed. Changes have been made through extensions[17:38:28] <garou> Okay, but... As far as I heard, big things are coming. ^^[17:38:35] <hibread> *Additions, rather, have been made through extensions[17:38:57] <hibread> garou: hahaha... yes we are all hoping a similar thing[17:39:51] *** Seeb_ is now known as Seeb[17:40:07] *** Baba has joined ##OpenGL[17:47:04] *** charlie5 is now known as charlie_zzz[17:51:47] *** baballama has quit IRC[17:52:16] *** HuntsMan has quit IRC[17:55:02] <garou> Hmmm... I just saw an out-of-focus effect on the teapots in the Redbook. A flatmate of mine, well, obsesses over effects like that. Is it relatively easy to implement or more one of the PITA things?[17:55:43] *** Catfish has quit IRC[18:02:39] <bobbens> garou: not that hard with shaders I think, but I'm really hoping my driver gets shaders soon to play around with that sort of thingy :)[18:03:29] *** ajww has quit IRC[18:05:07] <garou> Okaaay, at the far end of the development then. ^^[18:05:40] <bobbens> well nowadays people do everything with shaders[18:06:03] <bobbens> look at gallium3d, they want to just make cards use shaders and emulate older functions :)[18:06:16] <bobbens> but by doing this they reduce code for drivers and increase performance :)[18:07:26] <garou> Yup, I've even heard of physics being outsourced to IIRC vertex shaders, and I've seen clouds being simulated on 3D textures.[18:08:43] <bobbens> yeah, shaders are all the rage[18:10:02] <garou> Maybe I should keep that in mind... Actually all I really want to do is mostly putting some triangles on the screen comfortably with a mini-3D-engine, but heck, why not keep it open-ended... ^^[18:10:20] <bobbens> only issue with shaders is making them obligatory atm[18:10:26] <bobbens> since not everyone has them yet[18:10:34] <bobbens> (like me :()[18:10:40] <prophile> upgrade yer GPU[18:10:44] <bobbens> not gpu[18:10:46] <bobbens> but drivers :)[18:11:03] <bobbens> radeon open source drivers should get shaders soonish[18:11:23] <prophile> god[18:11:33] <garou> Yes, my son?[18:11:34] <prophile> i wish my radeon 9600 had GLSL support[18:11:49] <prophile> garou: you deserve to be chopped up into little pieces for that dreadfully poor joke[18:12:06] <prophile> i now have a geforce fx 5200[18:12:07] <prophile> it's not great[18:12:13] <prophile> but at least it can do shaders[18:12:18] <bobbens> my old box has a 9600[18:12:28] <bobbens> my new one has an X550 (pci-e made me upgrade)[18:12:35] <garou> Meeep...[18:12:38] <bobbens> much better performance, but still no shaders[18:12:47] <bobbens> altohugh I got clipping planes recently :)[18:14:25] *** joakim_ has joined ##OpenGL[18:16:44] *** Madsy has quit IRC[18:18:10] <garou> bbl, I'll verticalize myself for a while.[18:18:55] *** rnx has joined ##opengl[18:18:58] *** mm765 is now known as mm765^away[18:21:10] *** gaffa has joined ##opengl[18:33:51] <hibread> bobbens: still no shaders??[18:34:43] <hibread> the 9xxx series radeons had shader support...[18:34:51] <RTFM_FTW> *has*[18:35:20] <RTFM_FTW> all R3xx series ASICs offer GLSL support[18:37:50] <hibread> prophile: the 5200 would very unlikely have better performance than the 9600. Especially when using shaders[18:38:09] <bobbens> card has shader, driver doesn't yet[18:38:12] <bobbens> but I'm not running git[18:38:14] <RTFM_FTW> in fact the opposite would be likely[18:38:24] *** amalon has quit IRC[18:38:32] <hibread> RTFM_FTW: the 5200 having more performance?[18:38:36] <RTFM_FTW> since the NV3x series -- specifically that ASIC is quite weak concerning the programmable pipeline[18:38:58] <hibread> i said unlikely have better support than...[18:39:42] *** dolphin has joined ##OpenGL[18:39:45] <hibread> errr *performance[18:45:42] <prophile> hibread: I beg to differ[18:45:57] <prophile> this 5200 has much better performance on the shaders[18:46:02] <prophile> for the simple reason that it supports shaders[18:46:10] <prophile> when I had the 9200 it had to fall back to a software renderer[18:46:28] <Xmas|> a 9200 is something quite different than a 9600...[18:46:32] <hibread> you said 9600.[18:47:14] <RTFM_FTW> and both ASICs (i.e. 5200 from Nvidia and 9600, 9800 from ATI) "support shaders"[18:47:48] <RTFM_FTW> OTOH the 9200 (effectively a R2xx ASIC) will support vertex shaders but it lacks the support for anything advanced concerning the programmable fragment pipeline[18:47:54] <hibread> 9200 uses the R200[18:48:09] <RTFM_FTW> it does support ATI_text_fragment_shader / ATI_fragtment_shader[18:48:12] <prophile> oh[18:48:14] <prophile> I meant 9200[18:48:21] <prophile> no wait[18:48:25] <prophile> I meant 9600[18:48:26] <Jorachim> Haha[18:48:27] <prophile> I tell a lie[18:48:28] <prophile> heh[18:48:32] <hibread> well then yep.. use your blazing 5200 :)[18:48:33] <prophile> I was right first time[18:48:36] <prophile> stop trying to confuse me damnit[18:48:43] <hibread> which would barely drag the skin of a rice pudding[18:48:59] <hibread> prophile: so which is it? :)[18:49:12] <prophile> oh I dunno[18:49:16] <prophile> feelgood, dear boy[18:49:27] <prophile> what would an old eMac have had?[18:49:28] <hibread> *off[18:49:52] <prophile> *on[18:49:53] <prophile> wut[18:50:09] <RTFM_FTW> assuming you are referring to me I'm not feelgood.[18:50:16] <RTFM_FTW> notes handle[18:50:17] <hibread> doesn't matter, i was correcting my typo[18:50:58] <prophile> it's a 9200[18:50:59] *** amalon has joined ##opengl[18:51:00] <prophile> there we are[18:51:05] <prophile> an ATi Radeon 9200[18:51:12] <hibread> yep. Dispose of that hardware[18:51:18] <prophile> I'm not using it any more[18:51:28] <RTFM_FTW> in any case the NV3x (specifically the 5200 series) was so slow that Apple never bothered exporting hardware support for CoreImage and friends for this ASIC... in a number of cases the SW rasterizer ended up being faster.[18:51:47] <prophile> replacement GPUs are so expensive :([18:52:02] <prophile> i hardly found anything for less than like 200 quid[18:52:31] <hibread> prophile: 200 pounds? You'd get near on the best card atm for that price[18:53:00] <prophile> then maybe i'm looking in the wrong places[18:53:08] <prophile> what should I be getting for my poor old G5?[18:53:28] <prophile> also I appear to be bleeding[18:53:32] <prophile> this is not a good sign[18:53:52] <RTFM_FTW> get yourself a X1900[18:53:52] <hibread> wtf...[18:53:59] <RTFM_FTW> its a wonderful card for that box[18:54:18] <RTFM_FTW> since it fully supports PPC and its still supported by Apple[18:54:32] <prophile> k[18:54:37] <prophile> how much will that set me back?[18:54:42] <RTFM_FTW> furthermore its actually viable for that particular machine[18:54:45] <RTFM_FTW> that depends[18:55:11] <RTFM_FTW> I'd recommend that its time to start looking around[18:55:45] <prophile> well, see the first place I find on the internet quotes about $340[18:56:03] <prophile> which equates to about £170, and then with VAT on top of that it's around the £200 mark[18:56:16] <prophile> :/[18:56:56] <RTFM_FTW> I'd probably look on eBay[18:57:42] <RTFM_FTW> and it wouldn't surprise me that its commanding a premium online since its not (that I'm aware of) being produced anymore[18:57:59] <prophile> £25[18:58:00] <hibread> prophile: in australia, you can get an X1950Pro for about 200$... which is around 100 pound[18:58:01] <prophile> now that's what I'm talking about[18:58:09] <RTFM_FTW> so there is probably a fairly tight stock of them in circulation[18:58:48] *** ghenriks has quit IRC[18:59:02] <prophile> so is this a scam, this 25 quid one?[18:59:10] <hibread> 2nd hand?[18:59:13] <prophile> aye[19:00:05] *** juanmabc has quit IRC[19:00:53] <hibread> RTFM_FTW: what about the HD3850 AGP... that would be pretty decent right?[19:01:11] <RTFM_FTW> that would be very decent[19:01:28] <hibread> thats not all that much more expensive than the X1900 or what ever[19:02:23] <RTFM_FTW> its also SM4 compatible which for someone doing GL w/ their desktop is probably worth noting[19:03:21] <RTFM_FTW> i.e. supports EXT_geometry_shader, EXT_gpu_shader4, EXT_transform_feedback, EXT_bindable_uniform etc.[19:04:39] <prophile> excellent[19:05:03] <prophile> what's your opinion of core animation?[19:05:18] <prophile> all I've really heard from anyone is basically "well, it does animations"[19:05:49] <RTFM_FTW> and that is exactly right[19:05:51] <RTFM_FTW> heh[19:07:12] <RTFM_FTW> honestly when I've needed to do 2D animation I've done it directly via OpenGL (i.e. shaders + RTT via FBO or pbuffers etc.)... all CA does is allow this sort of thing for those not actively involved in graphics[19:07:22] *** DobosCake has joined ##OpenGL[19:08:09] <RTFM_FTW> its undoubtedly easier to get up to speed I suppose with CoreAnimation than it is with generating your own scenegraph for whatever animation you need to perform[19:09:26] <prophile> k[19:11:30] *** Neo|Laptop has quit IRC[19:14:00] *** Neo|Laptop has joined ##OpenGL[19:18:50] *** LordMetroid has joined ##OpenGL[19:22:08] *** groton has quit IRC[19:24:52] *** ^gfx has joined ##OpenGL[19:33:54] *** magnet_ has quit IRC[19:35:56] *** Tibor__ has quit IRC[19:36:11] *** dv_ has quit IRC[19:38:08] *** m4ggus has quit IRC[19:40:05] *** IJs\laptop has quit IRC[19:41:29] *** Eltran has quit IRC[19:47:19] *** rutski has quit IRC[19:48:43] *** dv_ has joined ##OpenGL[19:55:25] *** DobosCake has quit IRC[19:55:33] *** DobosCake has joined ##OpenGL[19:56:06] *** Makegho has joined ##OpenGL[20:06:22] *** scy has left ##opengl[20:09:38] *** amalon has quit IRC[20:10:14] *** WimLeers has joined ##OpenGL[20:14:02] *** DobosCake has quit IRC[20:14:11] *** DobosCake has joined ##OpenGL[20:16:37] *** replor_ has joined ##OpenGL[20:19:34] *** WimLeers_ has joined ##OpenGL[20:23:43] *** DobosCake has quit IRC[20:24:19] *** DobosCake has joined ##OpenGL[20:27:19] *** WimLeers|school has quit IRC[20:27:20] *** DobosCake has quit IRC[20:27:32] *** DobosCake has joined ##OpenGL[20:29:41] *** tmccrary has joined ##OpenGL[20:32:40] *** replor has quit IRC[20:36:18] *** WimLeers has quit IRC[20:39:35] *** aalex has joined ##OpenGL[20:46:49] *** ajww has joined ##OpenGL[20:48:01] *** hibread has quit IRC[20:48:25] *** WimLeers has joined ##OpenGL[20:48:56] *** dolphin has quit IRC[21:00:53] *** DMINATOR has joined ##OpenGL[21:03:42] *** mastro has quit IRC[21:04:01] *** IMX_ has joined ##OpenGL[21:04:19] *** IMX_ has quit IRC[21:04:36] *** WimLeers_ has quit IRC[21:07:04] *** karabash has joined ##OpenGL[21:08:50] *** WimLeers is now known as WimLeers|school[21:15:07] *** replor has joined ##OpenGL[21:15:36] *** ^gfx has quit IRC[21:17:52] *** karabash has quit IRC[21:19:03] *** juanmabc has joined ##opengl[21:19:52] *** juanmabc has quit IRC[21:20:04] *** juanmabc has joined ##opengl[21:28:20] *** ville has joined ##opengl[21:29:45] *** Xmas| has quit IRC[21:30:00] *** dvoid_ has quit IRC[21:32:46] *** replor_ has quit IRC[21:41:28] *** ville has quit IRC[21:49:20] *** DobosCake has quit IRC[21:50:19] *** DobosCake has joined ##OpenGL[21:53:19] *** DobosCake has quit IRC[21:56:39] *** DobosCake has joined ##OpenGL[21:58:57] *** scy has joined ##opengl[22:03:10] *** ville has joined ##opengl[22:04:59] *** mm765^away is now known as mm765[22:10:10] *** darka has quit IRC[22:10:25] *** darka has joined ##opengl[22:28:55] *** wcstok has quit IRC[22:29:04] *** wcstok has joined ##OpenGL[22:29:30] *** aalex has quit IRC[22:31:30] *** joakim__ has joined ##OpenGL[22:39:53] *** wcstok has quit IRC[22:44:27] *** speedy1 has joined ##OpenGL[22:48:37] *** joakim_ has quit IRC[22:54:14] *** pfo has quit IRC[22:58:56] *** pfo has joined ##OpenGL[22:59:56] *** [this] has joined ##opengl[23:11:27] *** amalon has joined ##opengl[23:22:11] *** lolage0 has joined ##OpenGL[23:22:32] *** aalex has joined ##OpenGL[23:23:42] *** StephaneLenclud has joined ##OpenGL[23:25:51] <StephaneLenclud> I'm trying to render a cylinder within a cylinder for my laser beam on OpenGLES 1.1 but it's kind of broken[23:27:02] <StephaneLenclud> I have blending enabled and I'm using a color with 0.2 as alpha channel[23:27:16] <StephaneLenclud> Some beam render fine but most don't[23:27:22] *** RTFM_FTW is now known as Satan_Inside[23:28:17] <StephaneLenclud> Funny enough when my billboarded explosions are running the beam are completely broken[23:29:16] <StephaneLenclud> Maybe I should try having proper material instead of using glColor[23:29:29] *** Yustme has quit IRC[23:30:20] *** thesquib has joined ##OpenGL[23:33:31] *** Lemml has quit IRC[23:35:38] <rnx> depth testing doesnt work for transparent surfaces[23:36:14] <StephaneLenclud> I'm[23:36:17] <StephaneLenclud> mmmh[23:36:23] <tmccrary> yep, you have to sort them[23:36:29] <tmccrary> render back to front[23:36:39] <StephaneLenclud> shall I disable depth then[23:37:51] <StephaneLenclud> My rendering order is proper[23:41:40] *** Latheesan_ has quit IRC[23:42:20] *** WimLeers|school is now known as WimLeers[23:44:06] *** scy has left ##opengl[23:45:05] <StephaneLenclud> disabling depth ain't helping[23:45:57] *** neoneurone has joined ##OpenGL[23:47:23] <rnx> post code and screenshots[23:55:43] *** DR0ID_ has quit IRC[23:55:46] *** [this] has quit IRC[23:57:49] <StephaneLenclud> http://img337.imageshack.us/my.php?image=sshottj9.png[23:58:03] <StephaneLenclud> there goes the screenshot[23:58:30] <StephaneLenclud> It shows like that when my explosions are not running[23:58:53] <StephaneLenclud> When explosions are running you can hardly see the beam at all