January 3, 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 3, 2009 [00:01:33] *** korff_home has quit IRC[00:05:34] *** iion has quit IRC[00:06:26] *** iion has joined ##OpenGL[00:06:42] *** mattn2|home has quit IRC[00:08:50] *** ccfreak2k has joined ##opengl[00:09:14] *** caine607 has quit IRC[00:09:32] *** caine607 has joined ##OpenGL[00:13:12] *** LordHavoc has joined ##OpenGL[00:14:22] *** davidc__ has quit IRC[00:14:52] *** groton has quit IRC[00:15:34] *** sparky has quit IRC[00:15:48] *** sparky has joined ##OpenGL[00:23:51] *** scai has left ##opengl[00:26:25] *** HuntsMan has quit IRC[00:30:04] *** itewsh has quit IRC[00:30:41] *** k3t has quit IRC[00:39:10] *** eXtronuS has joined ##OpenGL[00:42:52] *** xp_prg has joined ##OpenGL[00:48:45] *** vade has quit IRC[00:51:56] *** grullizzle has joined ##OpenGL[00:52:42] *** rutski_ has quit IRC[00:54:45] *** Jernej has joined ##OpenGL[00:55:01] *** Suprano has joined ##OpenGL[00:55:22] *** rutski_ has joined ##OpenGL[00:56:45] *** eXtronuS has quit IRC[00:58:22] *** jcazevedo has quit IRC[01:01:50] *** gaffa has quit IRC[01:01:55] *** Yustme_ has quit IRC[01:01:59] *** _Rangar_ has joined ##OpenGL[01:05:59] *** XT95___ has quit IRC[01:07:48] *** JernejL has quit IRC[01:12:01] *** rutski_ has quit IRC[01:17:50] *** Rangar has quit IRC[01:17:52] *** ata2 has joined ##OpenGL[01:20:02] <ata2> Hello everybody. I have the following problem: I am working on skeletonization of point clouds and I need to display point cloud and skeleton in OpenGL. Point clouds are usually displayed as little disks oriented according to the estimated normal. Skeleton points can be represented as little spheres inside the shape. However, if the density of the cloud is high the disks of the cloud would hide the skeletal spheres, thus I need to u[01:20:04] <ata2> se transparency. How hard a task is this going to be?[01:21:35] <ata2> To give a helloworld example you could create two spheres with GLU, a big one outside and a small one inside. The one outside needs to be transparent otherwise you wouldn't see the one inside[01:22:00] <ata2> I remember reading that there are a bunch of considerations upon the rendering ordering to make if you want to use alpha blending[01:22:29] <ata2> thus, before starting to do anything I would like to ask you OpenGL experts[01:23:27] <HideousNashimoto> How to set wireframe in opengl, I forgot function name :)[01:23:44] <RTFM_FTW> glPolygonMode[01:23:47] <HideousNashimoto> thx[01:24:16] <RTFM_FTW> and yes order is important (i.e. sorting by Z) when it comes to transparent primitives[01:24:23] *** rabbit- has joined ##OpenGL[01:26:07] <MatthiasM> ata2: you could try point sprites and additive blending[01:30:06] <ata2> RTFM_FTW: I don't understand something though then.. how do they do it in games? You often see transparent stuff and I cannot imagine that everything get sorted before rendering....[01:30:11] <ata2> (triangle by triangle)[01:31:13] <tmccrary> yeah, that's what happens, you just do it efficiently and cache the sort order when possible[01:31:21] <MatthiasM> there are very fast sorting algorithms[01:31:38] <ata2> wow that's a heck of a job to do then...[01:31:47] <ata2> tmccrary: thanks for the suggestion[01:31:54] <MatthiasM> and often you can render the transparent stuff without sorting while you know that it will be (mostly) sorted[01:32:15] <ata2> yah but in my case I know that's not going to happen :)[01:32:39] <ata2> I guess I will have to sort everything... that's gonna take a while to code :([01:32:43] <tmccrary> its really not that bad or complex[01:33:13] <ata2> do you have a pointer to a tutorial on alpha blending for tranparency?[01:34:15] *** HuntsMan has joined ##opengl[01:34:49] <ata2> MatthiasM: Point sprites seems to be mainly concerned with speed.. I don't really care.. I just need to generate few renderings[01:34:59] <ata2> MatthiasM: (static)[01:35:26] <MatthiasM> but they are also easier to use - just one vertex and you can render an array with one call[01:35:35] *** Suprano has quit IRC[01:35:45] <MatthiasM> and are always screenaligned[01:35:50] <ata2> well, I use a small sphere or a glpoint[01:36:09] <MatthiasM> spheres are expensive[01:36:19] <ata2> my models are small :P[01:36:30] <ata2> anyway, thanks for the chat guys, very appreciated[01:43:58] *** ccfreak2k has quit IRC[01:49:22] *** m4ggus has quit IRC[01:49:24] *** m4ggus_ has joined ##opengl[01:49:34] *** m4ggus_ is now known as m4ggus[01:50:21] *** HideousNashimoto has quit IRC[01:52:13] *** pa has quit IRC[01:52:29] *** jberg has joined ##OpenGL[01:55:58] *** Bollinger has quit IRC[02:00:28] *** HuntsMan has quit IRC[02:01:07] *** eXtronuS has joined ##OpenGL[02:07:59] *** Kasu has quit IRC[02:12:33] *** k3t has joined ##OpenGL[02:21:37] *** Wagoo has quit IRC[02:25:12] *** calav3ra has joined ##opengl[02:27:38] *** amz has quit IRC[02:28:19] *** x0rx0r has quit IRC[02:30:42] *** Wagoo has joined ##opengl[02:32:26] *** grullizzle has quit IRC[02:32:56] *** nitrotrigger has quit IRC[02:35:10] *** Gorgoroth has quit IRC[02:39:21] *** Morphous has quit IRC[02:44:27] *** wisey has joined ##OpenGL[02:55:11] *** calav3ra has quit IRC[02:59:45] *** Morphous has joined ##opengl[03:03:32] *** nettikkcah has joined ##opengl[03:03:36] *** eXtronuS_ has joined ##OpenGL[03:05:52] *** eXtronuS has quit IRC[03:05:52] *** rabbit- has quit IRC[03:05:52] *** hackkitten has quit IRC[03:05:52] *** pfo has quit IRC[03:05:52] *** dindinx has quit IRC[03:05:52] *** Andon has quit IRC[03:06:46] *** pfo has joined ##OpenGL[03:07:54] *** dindinx has joined ##opengl[03:07:58] *** Andon has joined ##OpenGL[03:21:29] *** amz has joined ##opengl[03:22:49] *** AlastairLynn has quit IRC[03:23:01] *** AlastairLynn has joined ##opengl[03:29:20] *** dvoid_ has quit IRC[03:30:49] *** k3t has quit IRC[03:31:00] *** Tenac_ has joined ##opengl[03:31:02] *** k3t has joined ##OpenGL[03:32:00] *** Tenac_ has quit IRC[03:35:44] *** MatthiasM has quit IRC[03:35:50] *** MatthiasM has joined ##opengl[03:44:34] *** Tenac has quit IRC[03:46:09] *** Zenja_Solaja has joined ##OpenGL[03:50:26] *** cplusplus has quit IRC[03:52:00] <Zenja_Solaja> Hi everyone. I've just encountered something bizarre (in a good way). I've made the classic mistake of enabling GL_INDEX_ARRAYS with my vertex/texture/normal arrays in a new section of code (yep, accidently enabled index colours). I was stunned to see the frame rate go from 840-850fps to 1350-1400fps for my project (on NV8800GT hardware). I've removed it from the new section of code, and placed it my initialisation[03:52:00] <Zenja_Solaja> believe that I got more than 50% performance increase by enabling something I'm not even using. Anyone have any explanations as to why this is happening?[03:53:57] *** iion has quit IRC[03:56:30] *** grullizzle has joined ##OpenGL[03:57:19] <MatthiasM> Zenja_Solaja: there is no GL_INDEX_ARRAYS[03:57:43] <Zenja_Solaja> MatthiasM: glEnableClientState(GL_INDEX_ARRAY);[03:59:47] <MatthiasM> do you also set a glIndexPointer ?[03:59:53] <Zenja_Solaja> No.[04:00:48] <MatthiasM> good luck then - it might crash somewhere ;D[04:00:51] *** GuShH has joined ##OpenGL[04:01:00] <Zenja_Solaja> I'm still suprised. During the last few months, I've been adding scenes and watching the frame rate slowly decrement as the complexity increases. I was hovering around 840-850 for the last week. With the mistake I made today, I suddenly got 1350fps.[04:01:17] <Zenja_Solaja> You're right about the possibility of it crashing :)[04:01:27] <MatthiasM> are you sure that everything is still executed ?[04:01:38] <Zenja_Solaja> The scene counters seem right.[04:01:40] <MatthiasM> not that some draw calls create a GL error[04:01:46] <Zenja_Solaja> I cannot see anything weird.[04:01:50] <Zenja_Solaja> Lighting still looks good.[04:02:00] <MatthiasM> screenshot?[04:02:08] <Zenja_Solaja> before/after?[04:02:13] <GuShH> perhaps you're not in debug mode anymore[04:02:16] *** bijoo_appdev has joined ##OpenGL[04:02:17] * GuShH giggles[04:02:20] *** bijoo_osdev has joined ##opengl[04:02:36] <Zenja_Solaja> That was the first thing I checked. Recompiled everything, and have a toggle in release mode[04:02:57] <Zenja_Solaja> with the toggle key, i enable/disable GL_INDEX_ARRAY[04:03:08] <Zenja_Solaja> And observe the performance go up by over 50%[04:03:27] <Zenja_Solaja> I've tried this in another rendering engine (Irrlicht), and saw no difference in performance.[04:03:45] <Zenja_Solaja> Now I'm confused as to what is going on.[04:03:55] <GuShH> Perhaps a dead moth that was stuck in between your gpu fan got released and bam! sudden burst of power![04:04:02] <Zenja_Solaja> BTW - index colours are deprecated in OpenGLES and 3.0[04:04:17] <Zenja_Solaja> GuShH: -that is the most logical explanation yet :)[04:04:36] *** bijoo_appdev_ has joined ##OpenGL[04:04:36] *** bijoo_osdev_ has joined ##opengl[04:04:58] <GuShH> Unless it uses passive cooling and you eat moth for breakfast.[04:05:00] <tmccrary> oh o[04:05:16] <tmccrary> you have discovered the top secret GL_INDEX_ARRAYS enum[04:05:25] <tmccrary> expect black helicopters out your window any second[04:05:29] <tmccrary> run while you still have a chance![04:05:35] <Zenja_Solaja> tmccrary: :)[04:05:47] <GuShH> too late they're already watching him[04:06:25] <Zenja_Solaja> So that's what "Valued partners" get - advice to turn on the secret speed switch.[04:06:43] <Zenja_Solaja> The non-valued partners dont know about this secret switch.[04:06:53] <GuShH> "2 missed event reminders: take the pill" Now that explains things.[04:07:07] <Zenja_Solaja> I wish.[04:07:16] <GuShH> If only I could hear my cellphone while I'm asleep...[04:09:52] <Zenja_Solaja> Well I've tried enabling this for Irrlicht and WildMagic4, and none of them notice the speed difference. They also dont crash, so that's good.[04:11:06] <MatthiasM> well - expect it to crash on other vendors or OS[04:11:18] <Zenja_Solaja> Well, I can always leave a hidden preference setting called "Magic Frame Rate Increase - use at own risk", and no one will be the wiser :)[04:11:27] <Zenja_Solaja> Testing on Windows, OSX and iPhone.[04:11:36] <Zenja_Solaja> Wont work on iPhone due to OpenGLES[04:11:47] <Zenja_Solaja> Dont notice any speed difference on OSX.[04:12:01] <Zenja_Solaja> It's only with nVidia 8800GT on my Windows box where I notice it.[04:12:04] <Zenja_Solaja> Weird.[04:12:05] *** bijoo_appdev_ has quit IRC[04:12:05] *** bijoo_osdev_ has quit IRC[04:12:35] <MatthiasM> maybe this is triggering some hidden quality/perf trade for some game/benchmark ?[04:12:47] <Zenja_Solaja> MatthiasM: that's what I'm guessing.[04:13:00] <MatthiasM> and next update you get the opposite effect :P[04:13:41] <Zenja_Solaja> MatthiasM: I'm actually 100% positive that I've stumbled onto something silly which Futuremark uses, and nVidia optimises for.[04:13:58] <Zenja_Solaja> I'm just suprised I haven't crashed.[04:14:00] <MatthiasM> unlikly[04:14:54] <Zenja_Solaja> Anyhow, I've left a big note in the source code. The big google doesn't have anything on this issue. I might email nVidia and see what sort of reaction I'll get.[04:15:09] <MatthiasM> use the dev forum[04:15:33] <MatthiasM> but do you have a screenshot of your app?[04:15:37] <Zenja_Solaja> OK. But it's saturday, so I'll try sometime during the following week.[04:16:06] <Zenja_Solaja> http://www.users.on.net/~zenja/files/tennis.jpg[04:16:15] <Zenja_Solaja> I need to make a before/after set.[04:16:28] <Zenja_Solaja> I'll do that before contacting nVidia.[04:17:41] <Zenja_Solaja> However, when pressing the toggle key, I dont see any difference other than a frame rate increase.[04:18:18] <RTFM_FTW> vendor specific optimizations in common benches aren't at all uncommon :D[04:18:37] <RTFM_FTW> irrespective of what others might be telling you :P[04:18:59] <MatthiasM> but index_array is unlikly in any 3d benchmark code[04:20:03] <Zenja_Solaja> agreed[04:20:22] <tmccrary> What would the purpose be though, crippling your card during normal operation?[04:20:42] <tmccrary> Maybe automatically doing lower precision calculations, to get better performace[04:21:04] <RTFM_FTW> heh a number of "tricks" can be used[04:21:30] <Zenja_Solaja> Well, my program is in RGBA mode, so it cannot use indexed colours[04:21:41] <RTFM_FTW> in any case such an optimization would be completely pointless if it cripples performance of existing codes at the cost of boosting it in a specific bench :P[04:22:00] <Zenja_Solaja> However, by calling glEnableClientState(GL_INDEX_ARRAY); I seem to be getting the dramatic frame rate increase.[04:22:05] <RTFM_FTW> indexed color has been dead for several generations now[04:22:13] <Zenja_Solaja> RTFM_FTW: tell me about it.[04:22:20] <Zenja_Solaja> I accidently enabled it today[04:22:28] <Zenja_Solaja> Classic amatuer mistake :)[04:22:42] <Zenja_Solaja> Removed it, then noticed what it did to frame rate.[04:23:03] <Zenja_Solaja> I'm reading the red book, and it says that in RGBA colour modes, index arrays are not used.[04:23:30] *** bijoo_osdev has quit IRC[04:23:35] <Zenja_Solaja> sorry index colours[04:25:49] <Zenja_Solaja> I'm puzzled, since this only happens in my engine, and not in 2 other engines I've tried (Irrlicht and WildMagic4)[04:26:09] *** vrmad1 has joined ##OpenGL[04:26:10] <MatthiasM> maybe you also have a lot of other strange states enabled :P[04:26:34] *** bijoo_appdev has quit IRC[04:27:19] <Zenja_Solaja> I doubt it. Maybe the driver allocates memory differently, and I just happen to hit a CPU cached sweet spot.[04:28:26] <Zenja_Solaja> That would explain it as well.[04:28:39] *** rabbit- has joined ##OpenGL[04:28:59] *** _Rangar_ has quit IRC[04:29:04] *** vrmad1 has quit IRC[04:30:04] <Zenja_Solaja> I'm uploading before and after pictures.[04:30:07] <Zenja_Solaja> Give me a second.[04:32:24] <Zenja_Solaja> With glEnableClientArray(GL_INDEX_ARRAY) = http://www.users.on.net/~zenja/files/tennis_on.jpg[04:32:32] <tmccrary> maybe your cpu is faster than your gpu?[04:32:38] <tmccrary> so when it drops into software mode[04:32:43] <tmccrary> what are you using?[04:32:44] <MatthiasM> lol[04:32:55] <Zenja_Solaja> With glDisableClientArray(GL_INDEX_ARRAY) http://www.users.on.net/~zenja/files/tennis_off.jpg[04:33:07] <Zenja_Solaja> nVidia 8800GT[04:33:12] <Zenja_Solaja> 512Mb VRAM[04:33:12] <tmccrary> ah, probably not[04:33:28] <Zenja_Solaja> on an AMD x2 4200+, 2Gb, Windows XP SP3[04:33:34] <MatthiasM> Zenja_Solaja: do you have mipmaps for the fence ?[04:33:49] <MatthiasM> looks like alias[04:34:05] *** X-Scale has joined ##OpenGL[04:34:47] <Zenja_Solaja> Yes (mipmapping). (ignore the fence/stadium art, have a friend reworking that as we speak)[04:35:13] <Zenja_Solaja> Since I'm developing this for the iPhone, I'm running really small screen resolution (640x960), which is double the iPhone resolution.[04:35:24] <Zenja_Solaja> Dont tell Apple that I develop iPhone games under Windows XP :)[04:35:32] <MatthiasM> heh[04:36:04] <Zenja_Solaja> On top of both screenshots you can see the frame rate difference from that single line of code which is toggled.[04:36:35] <Zenja_Solaja> And plenty of jpeg artifacts :)[04:38:19] <Zenja_Solaja> Anyway, I'll send the screenshots and problem description to nVidia to see whether they will react. I dont see this effect on OSX with Radeon X1600.[04:39:11] <Zenja_Solaja> Emphasis on HOT[04:39:31] <RTFM_FTW> emphasis on fast actually[04:39:38] <RTFM_FTW> fast, stable and high quality :)[04:41:38] <tmccrary> I was working on it the other day, but it rebuffed my advances[04:42:26] *** dmh has joined ##opengl[04:42:50] <RTFM_FTW> you must not be good enough for it[04:43:11] <Zenja_Solaja> I'm scared to update the nVidia driver to the latest OpenGL 3.0 driver they've released 2 weeks ago.[04:43:27] <Zenja_Solaja> Though it would be interesting to see if this phenonemom disappears.[04:44:58] <Zenja_Solaja> RTFM_FTW: I cannot really use the MBP and X1600 for more than 2 minutes (high GPU usage) before I start getting blue squares on the screen. I think the laptop is overheating.[04:45:09] <Zenja_Solaja> When it's cool, then the graphics are fast and good.[04:45:29] <RTFM_FTW> it might be bad VRAM[04:45:41] <RTFM_FTW> in any case I'd recommend taking it in[04:45:53] <Zenja_Solaja> The laptop is over 2 years old.[04:46:00] <Zenja_Solaja> I thought that was normal :)[04:46:06] <RTFM_FTW> no[04:46:24] <Zenja_Solaja> Even in iPhoto I get these blue squares, but only with Leopard.[04:46:27] <RTFM_FTW> I've already managed to kill the GPU on this G8x based MBP[04:46:33] <Zenja_Solaja> Never had the blue squares with Tiger.[04:46:58] <Zenja_Solaja> RTFM_FTW: that's what you meant by not good enough for it :)[04:47:16] <Zenja_Solaja> RTFM_FTW - Who do you work now?[04:47:23] <Zenja_Solaja> Work for now?[04:47:34] <tmccrary> net code is probably the most boring thing to work on[04:47:58] <Zenja_Solaja> tmccrary: it is exciting to see 2 machines talk after you've written the net stack.[04:48:04] <RTFM_FTW> nah this: "I've already managed to kill the GPU on this G8x based MBP" is the "Charlie Problem" :P[04:48:14] <Zenja_Solaja> Charlie?[04:48:27] <RTFM_FTW> read the Inquirer[04:48:57] <Zenja_Solaja> Oh, now I get it.[04:50:15] <RTFM_FTW> in any case the issue I'm referring to is the one documented here: http://support.apple.com/kb/TS2377[04:50:42] <RTFM_FTW> this was the prior generation (G8x based) MBP[04:51:03] <tmccrary> aka the "Nvidia botched their entire line of hardware" problem[04:51:15] <RTFM_FTW> yep they really fucked up[04:51:40] <RTFM_FTW> lying about it is even worse :D[04:51:47] <Zenja_Solaja> Reading the article - the first thing that came into my mind is that an nVidia production manager got a wonderful bonuses for cost cutting.[04:51:58] <Zenja_Solaja> Geez, nVidia really screwed up,.[04:52:23] <Zenja_Solaja> I'd hate to be in that managers position now (assuming he still works for nVidia).[04:52:25] *** jberg has quit IRC[04:52:49] <RTFM_FTW> well there is a reason their stock took a ~20+% dive a few months ago :P[04:53:13] <RTFM_FTW> and still hasn't recovered[04:53:17] <Zenja_Solaja> Well, at least on the Apple link you supplied it says that Apple are taking in MBP even if they are out of warranty.[04:53:48] <RTFM_FTW> oh well I won't buy another Nvidia GPU again thanks to this bullshit :D[04:53:51] <Zenja_Solaja> That means that my partners laptop may be effected - lucky that she's not a gamer :)[04:54:14] <Zenja_Solaja> RTFM_FTW: you have to buy from someone - though this season, AMD is doing all right.[04:56:36] <tmccrary> Now he knows[04:57:22] <Zenja_Solaja> From the article: One last thing to toss into the mix, cost. The PI layer is expensive, it adds about $50 to the cost of a wafer. Wafers from TSMC on a high end process cost about $3,000 to $5,000 depending on a lot of details. Adding the PI layer increases the cost of silicon by a noticeable amount, and adds to the defect rate.[04:57:22] <Zenja_Solaja> For cards that sell to big OEMs for $30 or so, silicon can't be more than a few dollars of the total. Adding 25 cents to the cost of a chip is a big deal, it can mean the difference between profit and loss for the entire run. One engineer suggested that Nvidia might have shot down the PI layer on cost grounds, but we don't buy that. They weren't that desperate, were they?[04:57:53] <RTFM_FTW> in any case I *personally* stay far, far away from Nvidia product :PO[04:58:26] <Zenja_Solaja> Look at it from this point of view - they cannot afford to screw up again, so their next product cycle will actually be good.[04:58:33] <RTFM_FTW> LOLOLOL[04:58:46] <tmccrary> Their current product cycle is pretty good[04:58:59] <tmccrary> It was the 8 series that was screwed up[04:59:04] <RTFM_FTW> I'd say that the GT2xx disaster speaks volumes about "good" :P[04:59:27] <tmccrary> It's among the fastest out there, I have a GTX280 and it rocks ;)[04:59:53] <tmccrary> That said, I got screwed by nvidia on my laptop, so I'm not quite as fanboyish for them as I was[05:00:08] <tmccrary> and nvidia won't fix my laptop even though its about 1 1/2 weeks old[05:00:11] <tmccrary> fuckers[05:00:25] <RTFM_FTW> and they likely lose money on everyone sold :P so I stand by my earlier statement[05:00:56] <Zenja_Solaja> Reading the article, it makes me wonder who gave the researcher a tunnelling electron microscope. Guess one of the competitors (AMD/Intel/whoever) checked out the nVidia chip. he he[05:01:03] <RTFM_FTW> furthermore the GT2xx doesn't exist in the important mobile space[05:01:21] <RTFM_FTW> all they have there is the re-hashed G9x[05:01:23] <tmccrary> well, intel rules that arena anyway[05:03:26] *** amz has quit IRC[05:07:54] <Zenja_Solaja> c-ya guys - it's time for a 1 hour walk in the park.[05:08:44] *** Zenja_Solaja has left ##OpenGL[05:08:49] *** glcrazy has quit IRC[05:09:54] *** wisey has quit IRC[05:10:31] *** HuntsMan has joined ##opengl[05:15:36] *** ata2 has quit IRC[05:18:02] *** Tenac has joined ##opengl[05:26:54] *** Tenac has quit IRC[05:27:18] *** ata2 has joined ##OpenGL[05:27:45] <ata2> Suppose that I want to orient a GLU disk in 3D such that his normals points in a determined direction[05:27:50] *** X-Scale has quit IRC[05:27:55] <ata2> what do I need to do in the rendering primitive?[05:28:12] <ata2> is multiplying by a rotation matrix the only solution?[05:28:30] *** Tenac has joined ##opengl[05:32:54] *** joe-coder has joined ##OpenGL[05:34:06] <joe-coder> I'm using glBlendFunc(GL_SRC_ALPHA, GL_ONE) (additive blending) and also have global fog enabled. when i render translucent sprites, the complately transparent regions still seem to add to the fog color, so i see a white square around all of my sprites.[05:34:15] <joe-coder> anyone know what i'm doing wrong?[05:35:20] <HuntsMan> additive blending is GL_ONE and GL_ONE[05:36:44] *** theGreg101 has quit IRC[05:37:20] <joe-coder> I changed it (and will keep it that way), but no difference in rendering.[05:38:14] <MatthiasM> joe-coder: read in the GL spec how fog is applied[05:39:22] <joe-coder> i know it mixes a color based on the value written to the depth buffer. perhaps it's because my sprites aren't writing to the depth buffer?[05:42:00] <MatthiasM> fog is not related to depth buffer in any way[05:42:35] *** m4ggus has quit IRC[05:42:42] <MatthiasM> fog depends on the distance to the camera[05:43:04] <MatthiasM> but if you would read the spec it would be clear[05:43:56] *** Rolenun has joined ##OpenGL[05:47:07] <joe-coder> http://www.opengl.org/documentation/specs/version1.1/glspec1.1/node90.html[05:47:31] <joe-coder> I don't seeing anything that explains my issue[05:49:05] <joe-coder> when i render a quad that has a black texture and is completely transparent, that quad increases the brightness of the color behind it.[05:49:26] <joe-coder> but only when fog is enabled.[05:50:05] <joe-coder> so i guess the white fog is blending with the black of the texture before being added[05:50:16] <joe-coder> but how can I make it take the alpha into account when it blends with the fot?[05:50:33] <joe-coder> *fog[05:54:05] <joe-coder> BlendEquationSeparate ?[06:02:59] *** pierrelux has joined ##OpenGL[06:08:56] *** Tenac_ has joined ##opengl[06:12:35] <joe-coder> i'm trying different arguments to glBlendFunc and glBlendFuncSeparate with no luck.[06:17:10] *** iion has joined ##OpenGL[06:22:59] *** k3t has quit IRC[06:23:46] *** Tenac has quit IRC[06:28:25] *** k3t has joined ##OpenGL[06:29:26] *** stringfellow has joined ##opengl[06:44:07] <joe-coder> I still haven't figured out how to do standard opengl fog with additive blended polygons.[06:44:39] <tmccrary> That's odd, I could swear you did[06:45:42] <joe-coder> I couldn't find out anything to pass to glBlendFunc[Separate] that gave the effect I wanted.[06:46:57] <joe-coder> additionally, the 3rd and 4th arguments to glBlendFuncSeparate seem to be ingored for my purposes[06:47:12] <joe-coder> passing GL_ZERO to both renders the same as passing GL_ONE to both.[06:47:30] <joe-coder> I would've thought zero source alpha would make my quad transparent.[06:54:54] *** joe-coder has quit IRC[07:01:43] *** eXtronuS_ has quit IRC[07:04:59] *** batbrat has joined ##opengl[07:05:06] *** pa has joined ##OpenGL[07:06:24] *** rabbit- has quit IRC[07:08:32] *** korff_home has joined ##OpenGL[07:09:08] <batbrat> I've been going over the redbook edition2 -I am currently reading the third chapter(Transformations).There is an example of setting up a camera for a plane pilot in it, which I have been trying to understand.On reading it, I found that the axes chosen for rotation seemed to be different than the ones I would choose.I searched the internet and found this site-http://dremba.bol.ucla.edu/3dg_camera.htm which chooses the same axes as I do.Is this a misprint i[07:12:42] <batbrat> *If the errata can be found elsewhere // Not "if they can be found elsewhere"[07:17:53] *** iion has quit IRC[07:21:13] *** solifugus has joined ##OpenGL[07:21:43] <solifugus> on ubuntu, what package contains headers so I can compile opengl stuff?[07:21:59] <batbrat> Specifically, I found that the roll is about z, pitch about y and heading(yaw?) about x.Shouldn't this be roll - z pitch - x and heading - y? Because in the first instance, (imho)the planes orientation would be wrong at least in the case when roll=0 pitch=0 and heading >0.[07:22:09] <batbrat> I'll tell you the packages...[07:22:13] <batbrat> 1 moment please[07:22:22] <solifugus> ok.. eagerly waiting..[07:25:27] <batbrat> I use mesa-common-dev,mesa-utils,mesademos,freeglut3,glut-doc,libglu1-mesa,libglu1-mesa-dev,libgle3. However, if you install graphics card drivers(Nvidia definitely,the rest I'm not sure), the necessary OpenGL headers and libraries get installed so you need not install Mesa3D. Strictly speaking, Mesa isn't OpenGL.[07:25:35] <batbrat> I hope that helps[07:26:10] <batbrat> Wait, are you compiling C/C++ OpenGL stuff or in some other language?[07:28:17] <batbrat> solifugus: If you are compiling for another language, you will also need the language specific bindings. Ex: Python -> PyOpenGL package- python-opengl[07:30:09] <solifugus> I am using C[07:31:43] <solifugus> Also, I have nvidia GeForce Go 7600[07:32:04] <solifugus> drivers installed.. but there's no /usr/X11R6/include directory..[07:32:16] <solifugus> like the tutorial I am follows says..[07:32:24] <HuntsMan> install nvidia-glx-dev and use /usr/include/GL[07:34:00] <batbrat> One second solifugus... See if you have /usr/include/GL ?[07:34:06] *** tokemonstah has joined ##OpenGL[07:34:12] <solifugus> installing now..[07:34:37] <batbrat> I just saw your post.... Sorry.[07:34:56] <solifugus> yes.. it has this stuff in it: glext.h gl.h gl_mangle.h glu.h glu_mangle.h glxext.h glx.h glx_mangle.h[07:35:35] <HuntsMan> those are the GL includes[07:35:50] <batbrat> Thats good.... Thats all you need to start really.[07:36:53] <batbrat> HuntsMan: If drivers are installed, is nvidia-glx-new necessary? I am not sure it is.[07:37:09] <HuntsMan> probably not[07:37:23] <batbrat> HuntsMan: Ok thanks for clarifying.[07:38:15] <solifugus> batbrat: What about this line from my tutorial: cc -o 3d 3d.o -L/usr/X11/lib -lglut -lGLU -lGL -lX11-1Xext -1m[07:38:26] <solifugus> There is no /usr/X11/lib[07:38:46] <solifugus> there is /usr/X11R6/lib .. just point to that?[07:38:56] <batbrat> Ok try this:[07:39:59] <batbrat> cc -o3d 3d.o -lglut -lGLU -lGL -lX111-Xext ... I think that will work... I am not certain though.[07:40:13] <batbrat> *-lX11-ext[07:40:21] <batbrat> **-lX11-Xext[07:41:05] <batbrat> solifugus: I think you can skip -L/usr/X11/lib safely...[07:41:41] <batbrat> Or from what HuntsMan said, you could use -L/usr/include/GL/ instead[07:41:41] <solifugus> ok... I am also missing glut stuff.. so installing the glut dev package..[07:42:00] <batbrat> No dont do that install freeglut3 thats more up to date[07:43:10] <batbrat> There is no problem if you install glut-dev but freeglut3-dev is more recent afaik[07:43:19] <solifugus> that's what I got.. freeglut3-dev[07:43:49] <solifugus> now my errors are: cc: unrecognized option '-1m' and /usr/bin/ld: cannot find -lX11-1Xext[07:43:52] *** korff_home has quit IRC[07:43:57] <batbrat> Ok cool[07:44:09] <solifugus> don't even know what those are... just copied from tutorial in book[07:44:40] <batbrat> Ok can you post the code... I'll try compiling it for you and see what needs to be done...[07:46:47] <solifugus> batbrat: Here's the code: http://pastebin.com/m2e1f4eac[07:46:53] <batbrat> Thanks...[07:47:03] <batbrat> I'll give it a shot[07:47:30] <solifugus> batbrat: Here's how I'm trying to compile: cc -c -I/usr/include/GL my3d.c && cc -o my3d my3d.o -L/usr/X11iR6/lib -lglut -lGLU -lGL -lX11-1Xext -1m[07:48:34] *** gilgrissom- has joined ##OpenGL[07:48:56] *** pierrelux has quit IRC[07:50:24] <batbrat> cc Help.c -o Help -lGLU -lGL -lglut[07:50:48] <batbrat> That worked for me where Help.c is the file you posted.[07:51:01] <batbrat> I had to rename initcamera to initCamera though[07:51:12] *** Tenac_ is now known as Tenac[07:51:36] <batbrat> solifugus: Try making that change and redoing the compile with the options I posted[07:52:38] <solifugus> Wow.. ok, yes.. it works now!![07:52:39] <solifugus> Thanks![07:52:50] <batbrat> Cool! Glad that it worked[07:53:09] <solifugus> Now I just need to use the bathroom and then on to the next page in my book![07:53:24] <sparky> what book[07:53:27] <batbrat> Good.All the best. BTW which book are you using?[07:53:40] <solifugus> it's "Linux Game Programming"[07:53:57] <batbrat> Nice.I haven't read it though...[07:54:03] *** gilgrissom- has joined ##OpenGL[07:54:26] *** gilgrissom- has quit IRC[07:54:45] <solifugus> Yeah.. I'll be happy if I can just do what I want in wire frames, even..[07:56:02] <batbrat> I hope you succeed :)[07:59:26] *** grullizzle has quit IRC[08:00:50] *** xp_prg has quit IRC[08:00:50] *** dv_ has joined ##opengl[08:02:08] <batbrat> Does anyone know where the errata at http://www.sgi.com/Technology/openGL has been shifted?[08:03:10] <batbrat> Sorry to repeat myself.I had forgotten to metion the site in my post.[08:03:33] *** tokemonstah has quit IRC[08:06:50] *** korff_home has joined ##OpenGL[08:06:54] *** rabbit- has joined ##OpenGL[08:07:40] <ata2> why as soon as I insert a statement like: glMaterialfv(GL_FRONT, GL_AMBIENT_AND_DIFFUSE, gray); the shading of my primitives it turned off? (like if there is no lighting)[08:08:10] <hibread> ata2: the secret art of opengl programming has emerged[08:08:30] <ata2> I tried to use a glColor3f but it has no effect, my disks/spheres are still whitish and shaded[08:08:34] <hibread> ata2: just like everything, we need to realize the absolute truth that is our experiences[08:11:18] <ata2> what is that supposed to mean?[08:12:09] <hibread> it means that you use 'shaders' for everything[08:12:47] <ata2> no, I don't talk about any shaders.. I mean default opengl shading[08:13:19] <ata2> I had no call to glColor or glMaterial* and I got GLU disks shaded according to their orientation[08:13:25] <ata2> w.r.t. the light source[08:13:59] <ata2> Now I would like to change the color of those little disks, but inserting a glColor right before the primitive makes no differnece[08:14:50] <ata2> while inserting a glMaterial effectively colors the primitives but stops the shading and every primitive is colored with the same constant value color (flat)[08:17:17] <batbrat> ata2: I think what hibread meant was that you should use shaders to achieve whatever effect you want.I think all th glMaterialfv etc. are emulated by shaders now-a-days anyway.I am new to OpenGL, so I may be wrong.[08:17:47] <ata2> batbrat: I see, thank you[08:18:13] <ata2> batbrat: nevertheless I don't need shaders.. I just need something very basic[08:19:02] <batbrat> ata2: I am not sure about how to do that. Sorry.[08:19:17] <ata2> batbrat: it's ok[08:22:45] <hibread> ata2: been a while since i've used the fixed function system. Could you paste some example code?[08:22:55] *** X-Scale has joined ##OpenGL[08:23:09] <ata2> it's very possible it was just my eyes crossing after many hours of work[08:23:22] <ata2> take a look at this image... do you think shading is working?[08:24:00] <batbrat> Please link the image...[08:24:06] <ata2> http://img145.imageshack.us/img145/9347/picture8na5.png[08:24:10] *** dmh has quit IRC[08:24:12] <ata2> yah, I was :P[08:24:54] <batbrat> Ok...[08:25:13] <hibread> shanghi?[08:26:27] <hibread> ata2: and what does it look like before you use the glMaterial function?[08:26:39] *** solifugus has quit IRC[08:26:54] <ata2> A different shade of gray, i guess that is the default OpenGL color[08:27:21] <ata2> Anyway, I believe I was just not seeing well :) the following is more important[08:27:44] <ata2> You have seen the image with the point cloud displayed as little disk and a blown up version of a skeleton[08:28:08] <ata2> Now I would like to display those little disks with transparency using alpha blending[08:28:58] <ata2> Since the skeleton is naturally "behind" the cloud I should render it before the cloud (I don't display backfacing samples so there is nothing behind the skeleton)[08:29:05] <ata2> This is my drawing routine[08:29:09] <ata2> http://rafb.net/p/jBOmcG57.html[08:29:40] <ata2> The reason for which I was using materials is that simple calls to glColor were giving me no change in coloring[08:30:02] <hibread> im too drunk to help you, sorry ;)[08:30:18] <ata2> lol ok[08:30:41] <batbrat> :)[08:45:44] <ata2> well, i will try to ask the same question tomorrow[08:45:54] <ata2> bye for now[08:45:55] *** ata2 has quit IRC[08:47:06] *** rnx has left ##opengl[09:06:07] *** Suprano has joined ##OpenGL[09:09:57] <batbrat> I tried implementing the pilot view code (Code -http://pastebin.com/m3d22b85f). I am having trouble with selecting the right axes of rotation. With the code taken from the redbook, pitch and heading effects are interchanged.With my code, they aren't interchanged but I get "moved" away from my position when I rotate.Can you guys please help me decipher the right way to do things as well as the correct axes of rotation?[09:11:00] <batbrat> I have created a 20x20 Quad placed flat at the origin to simulate the ground, attached Up arrow, down arrow to pitch, A/D to heading and Right arrow/Left arrow to roll.[09:11:51] <batbrat> Quad coordinates (-10,0,-10) (10,0,-10) (10,0,10) (-10,0,10)[09:14:59] *** Suprano has quit IRC[09:28:30] *** Renderwahn has joined ##OpenGL[09:29:18] <batbrat> I have made some changes and the rotation seems to be working. The new code is at : http://pastebin.com/ma0eca38. Can anyone who is free please see if this is the right way to do things?[09:39:55] *** predaeus has joined ##opengl[09:45:52] *** SinnerG has quit IRC[09:49:51] *** walt has joined ##OpenGL[09:50:13] <batbrat> Ok I have re-checked the code.It is still wrong. Apparently, I have to consider what happens to the local axes after rotation... I am not sure, but that may need re-orienting the axes dynamically.[09:51:24] <batbrat> Currently, I am doing Roll - About z axis Pitch - About x axis Heading - About y axis. The redbook doesn't do that it does Pitch - y axis, Heading- x axis.[09:54:40] <batbrat> The ordering of transformations also varies. I have written - Roll-Heading-Pitch while the redbook says Roll-Pitch-Heading. There are now two degrees of confusion :). So If I were doing this type of camera for a flight simulator, can I solve this problem without a Quaternion camera?[10:04:03] *** glcrazy has joined ##OpenGL[10:05:23] *** scai has joined ##opengl[10:06:58] *** GuShH has quit IRC[10:09:52] <batbrat> Ok bye all. I have to go to windows for a bit...[10:09:55] *** batbrat has left ##opengl[10:09:56] *** GuShH has joined ##OpenGL[10:12:48] *** stefodnb has joined ##OpenGL[10:12:48] <stefodnb> Hi[10:13:53] <stefodnb> If I want a 20x32000 texture (to map to a quad), I should instead use 1000 textures of size 32x32, and map/draw them separately, right?[10:15:08] *** groton has joined ##OpenGL[10:15:30] <stefodnb> like with a quad strip[10:17:05] *** GuShH has quit IRC[10:27:16] *** Yustme has joined ##OpenGL[10:30:27] *** uster has joined ##opengl[10:30:43] *** Suprano has joined ##OpenGL[10:34:24] *** glcrazy has quit IRC[10:39:08] <uster> When incrementing and displaying a score for a game using GLUTs glutBitmapCharacter I see the following. http://img165.imageshack.us/my.php?image=39431064oz7.png. The left side shows the before, and the right shows after. The right draws 10 in red correctly but with a 0 in my glClearColor behind it. Why is this?[10:40:03] *** Bollinger has joined ##OpenGL[10:41:02] <uster> I'm just wondering if anyone has an idea about what it could be, keyword or somthing.[10:42:28] <stefodnb> i dont know but i really doubt that drawing the "10" has the side effect of drawing a black 0 first. maybe you somehow aren't glClearing properly. zbuffer etc. try painting the score at a different location everytime to get a feel whats going on[10:46:38] *** sohail has quit IRC[10:50:40] *** pietia has joined ##OpenGL[10:59:04] *** Rangar has joined ##OpenGL[10:59:06] *** UUncia has joined ##OpenGL[11:00:03] *** GuShH has joined ##OpenGL[11:01:06] *** batbrat has joined ##opengl[11:01:31] *** batbrat has left ##opengl[11:01:37] *** batbrat has joined ##opengl[11:04:45] *** uster has quit IRC[11:08:33] *** HideousNashimoto has joined ##OpenGL[11:08:53] <HideousNashimoto> Hey, I am thinking of buying OpenGL Distilled book is it good on?[11:08:55] <HideousNashimoto> *one[11:16:58] <MatthiasM> stefodnb: always make the textures as big as possible - eg try 32 textures of size 32x1024[11:18:00] <stefodnb> Ah I thought they need to be square. Thanks ;)[11:19:12] <MatthiasM> no - POT if you don't have certain extensions - but not square[11:20:01] <stefodnb> right[11:31:13] *** fargiola` has joined ##OpenGL[11:37:05] *** rabbit- has quit IRC[11:40:36] *** xp_prg has joined ##OpenGL[11:43:08] *** fargiolas has quit IRC[11:43:32] *** Suprano has quit IRC[11:44:25] *** urbanmonk has joined ##OpenGL[11:47:37] *** nettikkcah is now known as hackkitten[12:00:25] *** X-Scale has quit IRC[12:01:13] *** wisey has joined ##OpenGL[12:03:47] *** Mathematicz has joined ##OpenGL[12:04:23] <Mathematicz> Anyone here use processing?(.org)[12:04:47] <MatthiasM> no - I use Java & LWJGL directly :)[12:04:59] <Mathematicz> LWJ?[12:05:20] <Mathematicz> Just I was wondering what is was like for 3D games, and i'll be using C++ :D[12:05:24] <MatthiasM> http://lwjgl.org/ and #lwjgl[12:05:58] <Mathematicz> Is it the same as OpenGL?[12:06:12] <MatthiasM> it IS an OpenGL binding and also OpenAL[12:09:36] *** Spkka has joined ##OpenGL[12:12:26] *** dindinx has quit IRC[12:12:26] *** Tempoe has quit IRC[12:12:26] *** Burga has quit IRC[12:12:26] *** GNU\caust1c has quit IRC[12:12:26] *** ville has quit IRC[12:13:57] *** dindinx has joined ##OpenGL[12:13:57] *** Burga has joined ##OpenGL[12:13:57] *** ville has joined ##OpenGL[12:13:57] *** GNU\caust1c has joined ##OpenGL[12:13:57] *** Tempoe has joined ##OpenGL[12:14:34] *** wisey has quit IRC[12:15:01] *** Mathematicz has left ##OpenGL[12:16:17] <HideousNashimoto> Can I use SDL_LoadBMP("sometexture.bmp"); with OpenGL?[12:16:51] *** Xmas| has joined ##OpenGL[12:16:52] <stefodnb> yes of course[12:17:03] <stefodnb> OpenGL wants a pointer to a block of pixels[12:17:13] <stefodnb> SDL_Surface::pixels gives you that[12:17:53] <HideousNashimoto> SDL_Surface *Texture=SDL_LoadBMP("sometextreu.bmp"); Texture.pixels[12:18:14] <HideousNashimoto> Oh, well I guess I would managed to figure it out[12:20:47] *** NotWil is now known as Wil[12:21:10] <Wil> I feel like an ass when I don't use Wil for years and then decide to ghost whoever uses it. lol[12:23:29] *** UUncia has quit IRC[12:24:51] *** UUncia has joined ##OpenGL[12:27:59] *** reprore_ has joined ##OpenGL[12:34:56] *** tokemonstah has joined ##OpenGL[12:40:03] *** UUncia_ has joined ##OpenGL[12:42:46] *** dolphin has joined ##OpenGL[12:43:05] *** urbanmonk has quit IRC[12:44:33] *** dolphin has quit IRC[12:44:55] *** XT95_ has joined ##OpenGL[12:47:36] *** pa has quit IRC[12:47:41] *** stefodnb has quit IRC[12:48:23] *** cplusplus has joined ##OpenGL[12:48:35] *** cplusplus has quit IRC[12:51:58] *** walt has quit IRC[12:52:52] *** UUncia has quit IRC[12:54:21] *** walt has joined ##OpenGL[12:54:38] *** pa has joined ##OpenGL[12:56:32] *** GX has joined ##OpenGL[13:03:17] *** dvoid_ has joined ##OpenGL[13:04:53] *** reprore_ has quit IRC[13:05:05] <HideousNashimoto> OMG, is there any tutorial how to Texture cube on linux, using SDL and Opengl, I have been searching whole morning[13:05:20] *** reprore has joined ##OpenGL[13:06:08] *** GX has quit IRC[13:06:23] <exDM69> HideousNashimoto: no, but there should be numerous examples of it[13:06:41] <exDM69> it should be really simple[13:06:44] *** Mathematicz has joined ##OpenGL[13:07:27] <HideousNashimoto> exDM69, do you maybe know, any link to example?[13:12:23] *** Eforen has joined ##opengl[13:13:31] *** lewymati has joined ##OpenGL[13:13:42] <Spkka> HideousNashimoto http://nehe.gamedev.net/data/lessons/lesson.asp?lesson=06[13:13:59] <Spkka> http://gpwiki.org/index.php/SDL:Tutorials:Using_SDL_with_OpenGL[13:14:04] <Spkka> not that much work was it?[13:14:07] <lewymati> hi spkkie :D:D:D[13:14:11] <Spkka> hi lewy :)[13:14:40] <lewymati> what the HideousNashimoto's question was?[13:14:54] <Spkka> OMG, is there any tutorial how to Texture cube on linux, using SDL and Opengl, I have been searching whole morning[13:15:44] <lewymati> ^^[13:17:48] *** Mathematicz has quit IRC[13:17:55] *** cplusplus has joined ##OpenGL[13:18:06] *** cplusplus has left ##OpenGL[13:19:50] *** positron2007 has joined ##OpenGL[13:21:31] <MatthiasM> it's somehow strange that a lot people are not able to adapt one tutorial/example to their needs[13:23:58] <HideousNashimoto> What kind of joke is this printf("warning: image.bmp's width is not a power of 2\n");[13:24:03] <HideousNashimoto> So What?!!?'[13:24:20] <Spkka> its the secret of google hiding things for certain ppl[13:24:22] <HideousNashimoto> If damn image is not whatever of 2, then I can't display it, well FU[13:24:27] <Spkka> nah[13:24:29] <Spkka> you can :)[13:24:36] <Spkka> some videocards dont allow it or sth[13:24:37] <HideousNashimoto> I can't I get black screen[13:24:47] <Spkka> try making it a power of 2?[13:24:57] <HideousNashimoto> How to do that?[13:25:13] <HideousNashimoto> I don't know math, is power of 2: 2^2?[13:25:45] <MatthiasM> HideousNashimoto: round it up to the next POT size and then use glTexSubImage2D after allocation storage using gltexImage2D(...., NULL);[13:25:49] <Spkka> 128,256,512,1024[13:26:10] <Spkka> 2,4,8,16,31,64,128 etc :)[13:26:22] <MatthiasM> HideousNashimoto: POT == 2^x where x=0...log2(MAX_TEXTURE_SIZE)[13:28:17] <MatthiasM> or use ARB_texture_rectangle if you have support for that[13:32:42] *** mm765^away is now known as mm765[13:33:29] *** UUncia_ has quit IRC[13:33:40] *** UUncia has joined ##OpenGL[13:36:55] *** Tr3B has joined ##OpenGL[13:39:12] *** X-Scale has joined ##OpenGL[13:43:26] *** elite01 has joined ##opengl[13:43:37] *** dv_ has quit IRC[13:53:14] *** xp_prg has quit IRC[14:03:32] *** Suprano has joined ##OpenGL[14:03:41] *** sohail has joined ##OpenGL[14:05:34] *** nitrotrigger has joined ##OpenGL[14:06:03] *** dusted has joined ##OpenGL[14:07:22] *** walt has quit IRC[14:23:57] <HideousNashimoto> All camera rotation examples use GLUT, adn I need SDL :([14:29:29] <MatthiasM> HideousNashimoto: GLUT does not care about camera[14:29:43] <MatthiasM> HideousNashimoto: you need to disable the examples and combine them[14:32:57] *** walt has joined ##OpenGL[14:36:52] <HideousNashimoto> glRotatef before drawing rotates the object, right[14:36:57] <HideousNashimoto> how to do the same with camera[14:37:04] <dusted> there is no camera[14:38:29] <HideousNashimoto> gluLookAt is tehre same function that would do like this glut function, but without glut[14:38:41] <dusted> not in sdl[14:38:58] <MatthiasM> HideousNashimoto: glu != glut ![14:39:16] <HideousNashimoto> I mean glu[14:39:36] *** UUncia_ has joined ##OpenGL[14:39:47] <MatthiasM> so - what's the issue then ?[14:39:50] <HideousNashimoto> dusted, and how to do it in opengl, I just want that my camera rotates around couple of objects, so that it can be seen it is 3D[14:39:57] <Renderwahn> HideousNashimoto: should be possible to use glu in sdl[14:40:00] *** UUncia_ has quit IRC[14:40:22] <Renderwahn> HideousNashimoto: you don't rotate a camera, you rotate everything else[14:40:57] <dusted> Renderwahn, it is[14:41:23] <dusted> just just have to decide which one you want to use to setup the rendering context and to flip the buffer[14:44:56] *** calav3ra has joined ##opengl[14:49:19] *** McLovin` has joined ##opengl[14:50:15] *** UUncia has quit IRC[14:50:54] <HideousNashimoto> I finally understood it[14:51:23] <HideousNashimoto> I never didn't know how to properly draw cube in school, like I 3D blind[14:51:29] <HideousNashimoto> *i am[14:52:29] *** Suprano has quit IRC[14:54:09] *** duzt has joined ##OpenGL[14:54:44] *** jberg has joined ##OpenGL[14:57:46] *** amz has joined ##opengl[14:58:30] *** kalopsian has joined ##OpenGL[15:00:51] *** Kasu has joined ##OpenGL[15:01:50] *** Suprano has joined ##OpenGL[15:02:29] *** groton has quit IRC[15:06:45] *** zwiep` has quit IRC[15:09:13] *** aep has joined ##OpenGL[15:09:19] <aep> greetings[15:09:36] <aep> i'm trying to use opengl with X11, maybe somone can help me out[15:09:52] <aep> basicly i dont receive expose events when creating a window with glx context[15:10:44] <aep> i guess thats becouse the glx drawable is outside the control of X11, but how do i know when to redraw then?[15:16:43] <MatthiasM> aep: take a look how glut does it[15:16:52] <MatthiasM> *freeglut[15:17:38] *** UUncia has joined ##OpenGL[15:17:53] *** m4ggus has joined ##opengl[15:19:31] *** calav3ra has quit IRC[15:21:03] *** rutski_ has joined ##OpenGL[15:21:11] *** calav3ra has joined ##opengl[15:25:15] *** jcazevedo has joined ##OpenGL[15:26:58] <aep> good idea[15:28:08] <batbrat> I am trying to build a camera that simulates a pilot's view.I got interested in the topic after reading the redbook's pilotView example. I didn't understand it correctly and wrote a bit of code to test it and play with.However, I couldn't get the axes of rotation quite right.I posted about it this afternoon.After researching more, I heard about Quaternions. Do I have to use a Quaternion camera for this case?[15:28:35] <aep> um looks like i AM receiving expose evnts[15:29:06] <aep> its just that XCopy from the opengl drawable to the X11 drawable fails[15:29:22] <aep> it copys nothing at all[15:30:06] <MatthiasM> batbrat: it can work without quaternions[15:31:52] <batbrat> MatthiasM: Thanks. In that case, will doing this sequence of rotations work ok? - Roll about z axis. then Pitch about x axis and Heading change i.e. yaw about Y axis.[15:32:22] <MatthiasM> yes[15:33:29] <MatthiasM> but remember that the last matrix operation is applied first to the objects[15:33:38] <MatthiasM> and for a camera you want to do everything inverse[15:33:51] <batbrat> Ok thanks. I was mightily confused because the redbook does Roll-z + Pitch - Y + Heading X, which seemed counterintuitive.Is this a mistake?[15:33:59] *** Ademan has quit IRC[15:34:11] <batbrat> I'll remember to reverse the sequence and reverse the rotations :)[15:34:20] <MatthiasM> it just depends on what you define as forward[15:35:01] <batbrat> Ok. So if I define the axes differently, the redbook code would be right? If so thats a relief...[15:35:12] <MatthiasM> batbrat: as you want to rotate all objects first around "-heading" so that they are in front of the camera and then pitch/roll[15:35:24] <MatthiasM> you want to write the heading as last line of code[15:36:17] <batbrat> Ok... Makes sense. I have gotten it wrong in my code.[15:36:32] <batbrat> I have a question: Is yaw different from heading?[15:36:47] *** qeed has joined ##opengl[15:36:48] <MatthiasM> no[15:37:04] <MatthiasM> I call it heading if it apllies to an moving objects's forward vector[15:37:23] <batbrat> Ok so it is like rotating about the cameras up vector.That really clarifies it.[15:37:35] <MatthiasM> yes[15:38:50] <batbrat> Thanks MatthiasM. I have another question though- What about gimbal lock?[15:39:19] <MatthiasM> gimbal lock does not matter if you define your rotation as yaw/roll/pitch[15:39:36] <MatthiasM> but if you want to move freely in 3d space then I suggest quaternions[15:40:03] <MatthiasM> eg you want to limit your pitch to +- 180°[15:40:25] <MatthiasM> because if you do a looping it acts strange[15:40:42] <batbrat> Ok... I'm listening[15:41:22] <batbrat> So I should use quaternions to limit my pitch or should I use them to allow looping?[15:42:17] <MatthiasM> looping gets easier with quaternions[15:43:41] <batbrat> Ok. I have one last query: In one site, They used roll-pitch-heading like you said but they said that it should be roll-yaw-pitch if you use an anglee relative to the vehicle longitudinal axis.[15:43:53] <batbrat> *they also said.[15:44:40] <batbrat> Why the difference? I am trying to think in terms of a local coordinate system.What you say is now clear to me.However, I do not get the second case.[15:44:41] <MatthiasM> it depends on what you want[15:45:03] <batbrat> I don't mean to impose. But can you please elaborate.[15:45:28] <MatthiasM> no idea what the 2nd case should do[15:46:17] <batbrat> Ok thanks for your patient help MatthiasM.I really appreciate it.[15:46:24] <MatthiasM> np[15:54:36] *** pragma_ has quit IRC[15:55:06] *** Tr3B has quit IRC[15:55:20] *** Tr3B has joined ##OpenGL[16:08:35] <batbrat> I have finally understood all the different cases.I'll post a summary here: 1.If you want to point the vehicle somewhere and then roll/pitch it, what MatthiasM said works issue calls for Roll+Pitch+heading.If you are talking in terms of an airplanes "yaw" i.e. deviation from its longitudinal axis, then the calls should be Roll/YAW/Pitch.This is because you lose control over the heading when you yaw when the plane is at 90 degrees pitch.A much better expla[16:09:32] <MatthiasM> you got cut off after "a much better expla"[16:11:00] <batbrat> Ok.. I'll post the remaining part[16:11:14] <batbrat> *A much better explanation is at -http://dremba.bol.ucla.edu/3dg_camera.htm. And once again a nod to MatthiasM who helped me get this after nearly 16 hours of effort.[16:11:28] <MatthiasM> :)[16:11:53] <batbrat> MatthiasM: I'm still trying the polar coordinates example. I got the order of rotations there.[16:12:41] <batbrat> However, all rotations use the negative of the angle supplied except the azimuth.If you're free, can you help me with that please?[16:19:15] *** Tenac_ has joined ##opengl[16:19:21] *** wisey has joined ##OpenGL[16:28:55] <batbrat> Sorry.I am attending a call.Forgot to se t away .Sorry[16:34:01] *** dv_ has joined ##opengl[16:35:34] *** Tenac has quit IRC[16:36:29] *** x0rx0r has joined ##OpenGL[16:36:41] *** k3t_ has joined ##OpenGL[16:36:52] *** k3t has quit IRC[16:38:07] *** k3t_ is now known as k3t[16:39:28] *** XT95_ has quit IRC[16:40:58] *** hang3r_ has joined ##OpenGL[16:42:12] *** hang3r_ has quit IRC[16:42:29] *** hang3r has joined ##OpenGL[16:44:08] <aep> so. i have no clue about opengl. just copyed some primitive examples. problem is i only do 2d drawing and only need primitive rasterisation of a couple of shapes plus radiants and tranformation. any idea were to start?[16:44:29] <aep> well shapes look easy. radiants and transformation i have yet to discover[16:44:54] *** Tr3B has left ##OpenGL[16:45:41] <hang3r> aep: Your best place to start would be nehe's tutorials, don't let anyone convince you otherwise either: http://nehe.gamedev.net[16:45:50] <MatthiasM> lol[16:45:59] <aep> yeah i tryed. its all complex 3d stuff[16:46:06] <hang3r> Start from tutorial 1 and work right through them all[16:47:05] *** a has joined ##OpenGL[16:47:06] <hang3r> "tried" being the keyword, keep working through it until you understand opengl, then read the redbook[16:47:22] <aep> i really dont want to know about 3D. isnt there a simple list of commands to use with 2D drawing. i copy and pasted some 2D matrix thing from a forum, but i'd prefer understanding it.[16:47:28] <MatthiasM> you should first read the red book[16:47:35] <MatthiasM> it explains things better[16:48:11] <aep> its not like i'm going to study opengl for a year, read 10 books and finally be a professional, so i can finally draw a stupid plane[16:48:39] <aep> unless there's a 2d drawing book[16:48:45] <dusted> aep, the basics are the same for 2D and 3D..[16:48:45] <aep> thats might be helpfull[16:48:52] <aep> they are? oh[16:49:05] <dusted> my first 2D game was in 3D mode, I just didn't notice :P[16:49:06] <aep> ok how much time does it take to get the basics?[16:49:07] <hang3r> yes they are[16:49:21] <aep> a day, a week, a year?[16:49:27] <MatthiasM> aep: when you setup an orthogonal projection matrix - then the depth does not affect the location on the screen[16:49:42] <hang3r> opengl is a state machine, I got the concept within a day, you may well do the same[16:50:00] <MatthiasM> aep: just follow the redbook examples[16:50:01] <aep> MatthiasM: yeah got that so far. although i'd like to understand how that matrix works[16:50:13] <MatthiasM> and you need to know your math[16:50:24] <aep> actually i just want to add more transformations[16:50:31] <aep> sure.[16:50:34] <batbrat> aep: I am learning OpenGL too. You'll be able to do what you need within a few days. I was.[16:50:36] <MatthiasM> aep: the OpenGL spec has all involved math[16:50:39] <aep> well whats that red book you spoke of?[16:50:49] <aep> batbrat: sounds reasonable[16:50:55] <MatthiasM> aep: care to read the channel topic[16:51:00] <batbrat> aep: I am actually learning OpenGL now ... Like you are[16:51:00] <aep> sorry[16:51:38] <kurt_> What kind of math should one know more about?[16:51:52] <aep> i guess basic geometrics?[16:52:07] <aep> like whats a plane, intersection, how does one build shapes in space etc[16:52:08] <MatthiasM> kurt_: if you want to go 3d - you need linear algebra[16:52:16] <aep> oh thats whats its called[16:52:24] <aep> we had that in school, i belive thats enough[16:52:28] <kurt_> ok :) Another thing I need to learn.. hehe[16:52:34] * aep goes reading that famous book[16:52:52] *** UUncia has quit IRC[16:52:52] <hang3r> aep, the redbook will not disappoint[16:53:06] <aep> thanks guys :)[16:53:21] <batbrat> All the best aep[16:53:34] <kurt_> I have some math knowledge, but I have a brother that is quite good.. So hopefully he'll teach me some more :)[16:53:43] <batbrat> MatthiasM: Sorry.I left. No offence.[16:54:00] <batbrat> kurt_: You're lucky then. I am doing it by myself ...[16:56:04] <kurt_> batbrat, yeah :) Luckily, he is very interested in math.[16:57:04] <batbrat> I have another viewing related question. If azimuth=angle in XY measured from +z elevation= angle in YZ from +ve z and twist=angle viewing vol is rotated about its line of sight.The redbook suggests rotate calls - Rotate -elevation,Rotate -twist, Rotate + azimuth.Shouldnt it be +elevation?[16:57:46] <batbrat> kurt_: I am making myself interested in math :)[16:58:41] <kurt_> hehe good :)[16:58:59] <kurt_> so am I :)[17:00:49] <batbrat> MatthiasM: Thanks for pointing out the split post. I've probably had a lot of other split posts without realizing. I am adding an extension to take care of it right away[17:01:03] <MatthiasM> :)[17:02:35] *** XT95_ has joined ##OpenGL[17:08:15] *** k3t has quit IRC[17:12:20] *** suppahasrv has joined ##OpenGL[17:13:14] <batbrat> Ok goodnight everybody. I am leaving for tonight.[17:13:16] *** batbrat has left ##opengl[17:17:42] *** sparky_ has joined ##OpenGL[17:18:07] *** sparky is now known as Guest11623[17:18:13] *** sparky_ is now known as sparky[17:21:21] *** mm765 is now known as mm765^away[17:23:28] *** suppahsrv has quit IRC[17:30:31] *** Guest11623 has quit IRC[17:32:41] *** jberg has quit IRC[17:33:14] *** Jophish has joined ##OpenGL[17:42:22] <HideousNashimoto> Is tehre OpenGL function to draw 3D text?[17:42:34] <sparky> no[17:44:12] <MatthiasM> try ftgl[17:45:48] *** clrtxt has joined ##OpenGL[17:46:27] <clrtxt> are gl headers essentially the way for C to interface with the graphics card?[17:46:55] <MatthiasM> no - to interface with OpenGL[17:47:16] <MatthiasM> and you allso need a few .a / .lib files[17:47:24] *** pragma_ has joined ##opengl[17:47:55] <clrtxt> MattiasM: so C interfaces with OpenGL. Does openGL just use whatever graphics driver is running X to control the graphics card?[17:47:59] *** reprore has quit IRC[17:48:29] <MatthiasM> in order to use GL in X - you need GLX[17:48:55] <MatthiasM> it might use the driver for HW GL if it can do it - or use Mesa for software GL[17:48:58] *** reprore has joined ##OpenGL[17:49:23] *** reprore has quit IRC[17:51:20] <clrtxt> MatthiasM: the gl headers I have in /usr/include/GL/ look like they are Mesa headers. I have an nvidia graphics card and I'm wondering if I need to do any extra work to make sure OpenGL uses the card[17:51:44] <MatthiasM> did you install the drivers from nVidia ?[17:51:51] <MatthiasM> execute glxinfo[17:51:59] <MatthiasM> and check what driver it uses[17:53:22] <clrtxt> it appears that is does: http://pastebin.com/m598d3ac[17:54:10] *** reprore_ has joined ##OpenGL[17:54:52] <MatthiasM> clrtxt: gz - you have full GL 2.1 support[17:55:10] <clrtxt> huh, nice[17:55:19] <clrtxt> I wish I knew how that happened[17:57:24] *** reprore_ has quit IRC[17:57:35] <clrtxt> so I assume hw gl directly interfaces with the hardware. What does a software gl do?[17:57:54] *** gaffa has joined ##opengl[17:58:08] <MatthiasM> it renders into a byte[] using software and then displays this buffer[17:59:42] <clrtxt> is that how graphical linux installers display the GUI when installing on a computer with an unsupported graphics card?[17:59:56] *** izike has joined ##OpenGL[18:00:12] <MatthiasM> well - normal 2d X graphics are a lot easier in SW then full GL[18:00:33] *** reprore_ has joined ##OpenGL[18:00:54] <izike> i have wrote 2d drawing library using opengl for glx systems, now with nvidia cards it work very well/fast while with fglrx it work soooooo slow, much slower than doing it in software, how can i know why is it so slow in fglrx? why could that happen? (i dont do anything complicated there)[18:02:21] <MatthiasM> izike: you don't have HW GL[18:02:26] <MatthiasM> izike: try glxinfo[18:02:43] <izike> MatthiasM:i do glxinfo return Direct rendering[18:02:51] <izike> MatthiasM:i even checked glXIsDirect()[18:03:19] <MatthiasM> izike: direct rendering only means that the glx server is running on the same host[18:03:38] <izike> ok, glxinfo say i do have direct rendering..[18:03:48] <MatthiasM> and that it can use shared memory and other tricks to speed it up[18:04:01] <MatthiasM> izike: what driver does it tell you for GL[18:04:14] <MatthiasM> izike: this paste is from clrtxt: http://pastebin.com/m598d3ac[18:04:17] <HuntsMan> MatthiasM: generally it means you have HW acceleration[18:04:20] <izike> sure, sec[18:04:37] <MatthiasM> HuntsMan: no - software mesa also uses direct rendering today[18:04:59] <HuntsMan> what software mesa?[18:05:03] <HuntsMan> the offscreen sw renderer?[18:05:28] <MatthiasM> when mesa is providing the glX support without having HW drivers[18:05:32] <HuntsMan> mesa has DRI drivers you know :)[18:05:40] <MatthiasM> but not for all HW[18:06:29] <HuntsMan> sure, that's when you don't have direct rendering :P[18:06:53] <HuntsMan> besides, that's what DRI reports[18:07:01] <HuntsMan> i can create a dummy driver that reports direct rendering[18:07:01] *** izike_ has joined ##OpenGL[18:08:02] <izike_> MatthiasM:sorry the system crushed, will reboot it and paste it again[18:08:24] <HuntsMan> that's no good ;)[18:08:31] *** izike has quit IRC[18:08:38] <HuntsMan> it crashed when you ran glxinfo?[18:09:17] <izike_> MatthiasM:no, when i tried to run my program as root, to see if there are premiisions problems...[18:09:46] <HuntsMan> lol[18:09:49] <HuntsMan> run glxinfo first[18:11:23] <izike_> ok, pastebin.com/m27f9681b[18:12:10] <MatthiasM> HuntsMan: http://rafb.net/p/CQtCaw68.html see[18:12:12] *** hang3r has quit IRC[18:13:09] <MatthiasM> izike_: ATI + linux => bad luck[18:13:15] <HuntsMan> MatthiasM: so? That's just misleading information from mesa[18:13:23] <HuntsMan> doesn't change what direct rendering means :)[18:13:34] <MatthiasM> no - because direct rendering does not tell you anything about HW or not HW[18:13:50] <izike_> infact[18:14:00] <MatthiasM> it tells you if you have a network connection between glx server or not[18:14:04] <izike_> it crushed again[18:14:07] <izike_> wow...[18:14:36] <izike_> i guess they have bugs with multithreaded applications using glx, beacuse i keep geting "assert" on locking from xlib[18:14:52] *** itewsh has joined ##OpenGL[18:14:54] <HuntsMan> you have multithreaded operations in your app with GLX?[18:14:58] <izike_> yea[18:15:04] <izike_> one thread is drawing[18:15:08] <izike_> one thread copy to the screen[18:15:32] <izike_> (when their is multiple screens, one thread is drawing per screen...) but it crush with just one screen as well[18:15:38] <izike_> with nvidia it work very well[18:16:05] <HuntsMan> one context per thread?[18:16:14] <izike_> yea[18:16:26] <MatthiasM> shared context?[18:16:32] <izike_> display list...[18:16:37] <izike_> not shared context[18:16:38] <HuntsMan> than is probably bugs in the ati driver ;P[18:16:51] <izike_> err i want to kill this guys[18:17:06] <izike_> i wrote this library just beacuse they didnt implement the xrender operations in their driver[18:17:11] <MatthiasM> if you want to use linux buy nVidia[18:17:17] <izike_> so i thought at least they implemanted opengl operations[18:17:31] <HuntsMan> izike_: they do[18:17:33] <izike_> MatthiasM: problem is that thinclients doesnt come with nvidia...[18:17:46] <izike_> HuntsMan: i have tried to use xrender_scale_something...[18:17:46] <HuntsMan> izike_: look forward to nvidia ion then[18:17:57] <HuntsMan> izike_: both ati and nvidia implement them[18:18:07] <izike_> well not when i looked back then[18:18:13] <izike_> it was falling for fall back[18:18:46] <HuntsMan> well it's best to use OpenGL than XRender[18:18:54] <izike_> XTransforScaleVector? something like that...[18:19:04] <izike_> again it worked perfrectly good with nvidia... :-)[18:19:23] <izike_> but with ati, it was faster to do the scalling using software :)[18:20:04] <MatthiasM> probably because you scale using a 2d scaling algo - not with a generic texture mapping one[18:20:09] <predaeus> let's hope the open sourced driver will fix all this crap, I want to by an ATI next.[18:20:18] *** jfroy has joined ##OpenGL[18:20:19] <RTFM_FTW> if you have a vendor specific problem stop bitching on IRC and *please* take the time to file a fucking bug[18:20:31] <predaeus> :-D[18:20:33] <RTFM_FTW> this goes for *all* vendors and *all* hardware[18:20:39] <RTFM_FTW> and *all* OSes[18:20:59] <tmccrary> I am having a problem with my s3 ViRGe's performance[18:21:10] <tmccrary> it does not appear to be working at maximum efficiency[18:21:20] <izike_> RTFM_FTW: i didnt come here to complain about the bugs in ati, just to trying figure how can i make my prog work fast[18:21:39] <tmccrary> it's supposed to be the next big thing in rendering technology[18:21:39] <RTFM_FTW> AFAIC if you never filed a bug then you have no right to bitch about the problems you might be encountering :P[18:21:59] <izike_> RTFM_FTW: btw i didnt report to ati about their performence problem with xrender, and they have never replaied me[18:22:04] <HuntsMan> RTFM_FTW: where do you file a bug for the fglrx driver?[18:22:48] <RTFM_FTW> "I didn't report ... " ... "and they have never replied to me" -- well fuck imagine that :P[18:22:55] <MatthiasM> heh[18:23:01] <RTFM_FTW> fucking Christ[18:23:14] <izike_> errr[18:23:16] <izike_> didnt->did[18:23:38] <RTFM_FTW> in any you can file issues though developer relations[18:23:55] <RTFM_FTW> err in any case[18:24:09] <tmccrary> you want him to have relations with a developer to get bugs fixed?[18:24:10] <tmccrary> yeesh[18:25:27] <RTFM_FTW> yeah I'd probably use goat sacrifice for Nvidia issues[18:25:42] <RTFM_FTW> I've heard that works well[18:25:55] <izike_> :)[18:26:34] <predaeus> reminds me of the voodoo times...[18:27:35] *** Nescafe has joined ##OpenGL[18:27:58] *** Orphis_ has joined ##opengl[18:28:02] *** Orphis has quit IRC[18:29:55] *** xp_prg has joined ##OpenGL[18:43:41] *** HuntsMan has quit IRC[18:46:36] *** Shel2476 has joined ##OpenGL[18:46:45] *** Shel2476 has left ##OpenGL[18:48:49] <tmccrary> begun the gas wars have[18:53:00] *** Eforen has quit IRC[18:57:05] *** pierrelux has joined ##OpenGL[18:57:10] *** dolphin has joined ##OpenGL[19:02:17] *** wisey has quit IRC[19:02:35] *** pietia has quit IRC[19:03:54] *** izike_ has quit IRC[19:13:48] *** bijoo_osdev has joined ##opengl[19:13:48] *** bijoo_appdev has joined ##OpenGL[19:15:20] *** walt has quit IRC[19:17:33] *** walt has joined ##OpenGL[19:26:21] *** walt has quit IRC[19:30:40] *** walt has joined ##OpenGL[19:37:51] *** l3dx has joined ##OpenGL[19:40:12] *** wisey has joined ##OpenGL[19:40:58] *** fargiola` is now known as fargiolas[19:46:58] *** jfroy has quit IRC[19:48:17] *** amz has quit IRC[19:57:07] *** calav3ra has quit IRC[19:57:27] *** a has quit IRC[19:59:42] *** a has joined ##OpenGL[20:13:48] *** GuShH has quit IRC[20:15:38] *** ol1veira_ has joined ##OpenGL[20:21:05] *** XT95__ has joined ##OpenGL[20:21:50] *** HuntsMan has joined ##opengl[20:24:48] *** rnx has joined ##opengl[20:26:03] *** nitrotrigger has quit IRC[20:26:08] *** nitrotrigger has joined ##OpenGL[20:26:46] *** itewsh has quit IRC[20:29:11] *** ol1veira1 has quit IRC[20:31:24] *** Mathematicz has joined ##OpenGL[20:31:49] <Mathematicz> Is knowledge of another language needed to use OpenGL? To make games? Or can you create "just as good" games with just Gl?[20:32:27] <dusted> ogl is an api[20:32:36] <buggs> Mathematicz, you need at least some language which exposes opengl[20:32:48] <Mathematicz> Kk[20:32:49] <tmccrary> lol[20:32:54] <tmccrary> get it through your skull[20:32:55] <Mathematicz> I'm going c++ -> OGL[20:32:57] <tmccrary> jesus[20:33:05] <tmccrary> rofl[20:33:18] <buggs> Mathematicz, that's rather standard then :)[20:33:22] <Mathematicz> :)[20:33:33] <Mathematicz> Still gotta learn it lol.. Need my new pc to come rgh[20:33:51] <buggs> but there are nice java, python, etc. interfaces too[20:34:11] <Mathematicz> I like python, but it's slower than c++, and java, well that is too kinda[20:35:04] *** Mathematicz is now known as Mathematics[20:35:06] <dusted> well, C++ seems pretty fast to me :) and there is all the C below it to get your hands dirty :P[20:35:18] *** Mathematicz has joined ##OpenGL[20:35:21] <buggs> but you can create at least two games in those languages before you finished the first c++ one[20:35:37] <dusted> but it's more fun :P[20:35:56] <dusted> "Woooop, did I free that ? Hmm, where does this pointer point?"[20:36:26] <dusted> "Oh wait, did I just increase that pointer without dereferencing it, hihi" :P[20:36:35] <Mathematicz> So what you saying? Learn python? People on this channel were trying to get me to learN Haskell lol, I just want a powerful quick engine in the end[20:36:55] <Renderwahn> Mathematicz: than use a prebuilt engine![20:36:59] <buggs> Mathematicz, the use an existing one ;)[20:37:25] <dusted> OpenGL is not a game engine, or any other engine[20:37:44] <tmccrary> I think he's a bot, he's been coming in asking these retarded questions over and over for days[20:38:08] <Mathematicz> I know[20:38:16] <Mathematicz> I'm no bot[20:38:17] <tmccrary> I helped him the first few times, but I have him on ignore now[20:38:18] <dusted> It's the "I need a job and want to be a cool programmer" syndrome, it's basically people who does not realize the existence of india ? :P[20:38:31] <Mathematicz> I know it's not that simple[20:38:50] <Mathematicz> I'll learn c++ then OpenGL, but only baby steps, like simple boxs at first[20:39:21] *** XT95_ has quit IRC[20:39:34] <Renderwahn> Mathematicz: stop talking abount it, do it![20:39:36] <dusted> Good idea, gogogog :) nothing's holding you back, google have all the info you need[20:39:45] <Mathematicz> Render[20:39:55] <Mathematicz> I will when got new PC lol, matter of days[20:40:06] <dusted> you can write code on this one?[20:40:11] <dusted> You don't need a new pc to write c++[20:40:16] *** suppahasrv is now known as suppahsrv[20:40:17] <tmccrary> Or better yet, use direct3d and Visual Studio[20:40:19] <Mathematicz> I know[20:40:27] <tmccrary> Direct3d is far superior to OpenGL is every possible way for games[20:40:29] <dusted> and if linux is your OS, it's very likely that you allready have everything installed to just go and start coding[20:40:30] <tmccrary> You can't even compare the too[20:40:31] <Mathematicz> VS free?[20:40:39] *** bhagabhi has joined ##OpenGL[20:40:40] <tmccrary> and Direct3d's api is C++[20:40:47] <tmccrary> So you "won't have to learn C"[20:40:58] <tmccrary> *the two rather[20:40:58] <tmccrary> ;)[20:40:58] <Mathematicz> Hmm, whats the advantages D3D compared to OpenGL?[20:41:11] *** l3dx has quit IRC[20:41:30] <dusted> that you use a non-free closed source, api that is only supported on one platform, it's the bomb![20:41:33] <Mathematicz> Oh and the reaosn im not coding/learningf now is that this is a family pc, people always talking, ill be reading accelerated c++ and coding what im reading, and itll be quiet ;)[20:41:33] <tmccrary> There are other channels and irc networks that have better support for direct3d, than we have for opengl here[20:41:34] <tmccrary> Go ask them[20:41:45] <tmccrary> Direct3d = The best games graphics API by FAR[20:42:13] *** XT95___ has joined ##OpenGL[20:42:41] <AlastairLynn> dusted: heh[20:42:48] <Mathematicz> isn't that 3D as well?[20:43:03] <RTFM_FTW> D3D is a wonderful API actually[20:43:13] <tmccrary> It's awesome[20:43:22] <AlastairLynn> dusted: all that you see of Direct3D is the API, whereas all of you see of OpenGL is... the API...[20:43:22] <TheFlash> D3D is the best, if you want "state of the art" -bullshit.[20:43:23] <AlastairLynn> i think i'm missing your point here[20:43:31] <Plagman> APIs are trivial[20:43:33] <tmccrary> Mathematicz: Use Direct3d it's for pro game developers[20:43:42] <tmccrary> OpenGL is for poor newbies, you don't want to learn it[20:43:43] <RTFM_FTW> heh[20:43:51] <tmccrary> No one in the games industry likes opengl or uses it[20:44:01] <Mathematicz> whys that[20:44:06] <Plagman> you can ultimately do the exact same thing using both D3D and OpenGL using a trivial compatibility layer[20:44:19] <TheFlash> Except every one who develops for Mac or PS3 =)[20:44:20] <RTFM_FTW> for the most part yes[20:44:24] <Plagman> so it really amounts to the same end-result[20:44:27] <tmccrary> Direct3d is designed or games and has rich support for games specific functionality[20:44:39] <tmccrary> OpenGL is for CAD and scientific apps, you don't want to use opengl[20:44:39] <dusted> hmm, I remember some gamedev writing about how he hated that he had to put up with the pain of developing in d3d at work, and how he relaxed and refound the fun in coding at home, in his freetime where he used opengl[20:44:41] <tmccrary> its garbage for games[20:44:49] <tmccrary> That's what the G stands for[20:45:05] <Mathematicz> Direct3D is C style[20:45:07] <Mathematicz> *?[20:45:11] <tmccrary> No, that's opengl[20:45:16] <tmccrary> Direct3d is a C++ API[20:45:20] <tmccrary> C++ = real game developer language[20:45:24] <Mathematicz> http://www.directxtutorial.com/Tutorial9/B-Direct3DBasics/dx9B1.aspx#still looks it[20:45:28] <Mathematicz> "Knowledge of WIN32"[20:45:31] <AlastairLynn> i know of at least one company that ported their game from OpenGL to Direct3D via a bunch of #defines and stuff because they used OpenGL as-is throughout their code rather than in a separate rendering engine[20:45:34] <TheFlash> D3D is an ugly C++-api.[20:46:44] <tmccrary> Mathematicz: windows is THE gaming platform, you should use direct3d and C++[20:46:53] <Mathematicz> Kk[20:46:55] <TheFlash> AlastairLynn: Garage games had a D3D-OpenGL -wraper.[20:46:57] <tmccrary> There are other irc networks that have much better support for games development than this channel[20:47:06] <Renderwahn> lol[20:47:06] <Mathematicz> Should I use Visual Studio even if I just use C++?[20:47:14] <tmccrary> Yeah, use Visual Studio[20:47:17] <Mathematicz> tmccrary[20:47:20] <Mathematicz> like which channels?[20:47:29] <TheFlash> #haskel[20:47:35] <tmccrary> #gamedev on efnet[20:47:36] <Renderwahn> rofl[20:47:37] <dusted> tmccrary, you are a strange person, unless you just opened a can of bullshit to get him to leave :P[20:47:41] <Mathematicz> Is it not free o.O?[20:47:53] <RTFM_FTW> several firms have D3D -> GL layers... Transgaming being one of the larger ones[20:48:00] <tmccrary> Check gamedev.net's irc[20:48:03] <kurt_> you can find vs for free..[20:48:04] <tmccrary> they can help you[20:48:05] <TheFlash> It costs 1 million bucks.[20:48:10] *** walt has quit IRC[20:48:20] <kurt_> I don't remember the link, but there is a version of vs that is free[20:48:26] <tmccrary> OpenGL is for old guys with beards who work on scientific apps, it has no role in modern games[20:48:34] <replaced> there is an express edition for free download[20:48:36] <replaced> ymmv[20:48:38] <dusted> tmccrary, how long is your beard?[20:48:53] <tmccrary> It's pretty long, some times I get mistaken for amsih[20:48:55] <tmccrary> amish[20:49:07] *** walt has joined ##OpenGL[20:49:18] <tmccrary> But yeah, Mathematicz, use direct3d/C++/VS, it's for elite hackers, not use poor slobs[20:49:26] <Mathematicz> Lol[20:49:29] <TheFlash> World of Goo is a great scientific app.[20:49:36] <tmccrary> It's the truth, now get going[20:49:45] <Mathematicz> Hopefully, In a few days when get PC i'll be coding like mad and not bugging you guys :D[20:50:08] <tmccrary> You should ask these kinds of questions in a channel for pros, like a d3d channel[20:50:09] <Plagman> don't try to react, the guy obviously is trolling :p[20:50:12] <tmccrary> We can't help you[20:50:29] <dusted> Plagman, it's like a itching, but I kind of see where he is going[20:50:45] <dusted> Thanks to tmccrary my next game will be in gimp scheme[20:50:56] <dusted> (((()))(())()())())))))(()))+++))())[20:51:11] <Plagman> he's going the way where he wants a newbie out of here :p[20:51:15] <kurt_> replaced, yeah, that was the one :)[20:51:25] *** buggs has quit IRC[20:51:43] <Mathematicz> Would creating a simple box inside a box with being able to move it left right etc accoridng the arrow keys being pressed be a "simple" task or "semi-advanced"?[20:51:48] <dusted> assembler is the preferred way to write 3d game engines however[20:52:27] <Mathematicz> assembler?[20:52:29] <Mathematicz> that a language?[20:52:31] <TheFlash> Mathematicz: If they are cartboard boxes, no.[20:52:35] <Renderwahn> opcode, noone uses assembler in games![20:52:37] <AlastairLynn> Mathematicz: yes, it is[20:52:37] <tmccrary> Direct3d would let you do this in around 15 minutes with Visual Studio[20:52:53] <TheFlash> Mathematicz: If heavy-duty steel boxes, yes.[20:52:56] <dusted> assembler is THE language, Duke3D was only realized because the important parts of it were written in assembler[20:53:01] <tmccrary> uh[20:53:10] <tmccrary> An assembler creates binaries[20:53:20] <tmccrary> "Assembly" is a low level machine language[20:53:20] <Mathematicz> visual studio download is a .iso[20:53:23] <TheFlash> Out of assembly.[20:54:08] <kurt_> Mathematicz, yeah, and you burn that .iso on a cd/dvd and install it from that..[20:54:25] <Mathematicz> I might cancel the download[20:54:36] <Mathematicz> it's only on 3% and its been on 5 mins[20:54:43] <kurt_> Mathematicz, get on with it :)[20:54:43] <Mathematicz> I might just download when got new PC[20:54:47] *** walt has quit IRC[20:54:59] <dusted> Mathematicz, write a mail to ms and ask them for a better torrent[20:55:22] <kurt_> Mathematicz, if you download it now, you can play around with easy c++ code, and even have it ready for the new pc[20:55:24] <Mathematicz> lol[20:55:41] <Mathematicz> any of you'se got ac++?[20:55:45] <Mathematicz> accelerated[20:55:48] <kurt_> yeah[20:56:02] <TheFlash> 2 of them.[20:56:09] *** Ingenu has joined ##OpenGL[20:56:20] <dusted> TheFlash, can I have one? or do you just have 2 copies of the pdf ? :P[20:56:21] *** XT95__ has quit IRC[20:56:22] <TheFlash> Oh, it was tic++, not ac++[20:56:26] <dusted> :([20:56:53] <Mathematicz> I find the second paragraph abit confusing[20:57:18] <kurt_> oh? what is it?[20:57:40] *** a has quit IRC[20:58:00] *** walt has joined ##OpenGL[20:58:14] <TheFlash> I bet, it's introduction or something similar.[20:58:29] <Mathematicz> Lol no[20:58:38] <TheFlash> Maybe part of forewords.[20:58:58] *** k3t has joined ##OpenGL[20:59:10] <Mathematicz> just why he uses std::string::size_type instead of int, even though they got the same size? ah i can't explain it[20:59:28] <TheFlash> Intro/forewords tend to be at beginning of the book and they contain more than 2 paragraphs, normaly.[21:00:08] <Mathematicz> Ah, has anyone here actually read it?[21:00:24] <TheFlash> Mathematicz: std::string::size_type is a typedef of int, or some similar datatype.[21:00:28] <kurt_> Mathematicz, that's because you wan't to now the size of a string. You can't simply say int something, when that something is not an int[21:00:40] <kurt_> now= know[21:01:23] <Mathematicz> Whats the advantages of using Visual Studio?[21:01:57] <TheFlash> Mathematicz: std::string::size_type mah_string_len = mah_string.size(); std::cout << "mah string haz " << mah_string_len << " letters"; // Got it?[21:02:45] <Mathematicz> I know how to use it, It's just.. I don't know why you'd want it a string, and not a int[21:02:55] <Mathematicz> Because they use it to store how many rows will be printed[21:03:00] <TheFlash> Mathematicz: VS hides you things and keeps you confused.[21:03:12] <Mathematicz> I take it you don't like it?[21:03:16] <Mathematicz> What would you reccomend?[21:03:56] <TheFlash> It is an integer! Not a string. Just the detail of it's type is "hidden"![21:04:11] <kurt_> Mathematicz, because you can't count the number of characters in that string without making a class or the likes that does the same thing..[21:04:25] <Mathematicz> Oh ok..[21:09:01] <kurt_> Mathematicz, it's like std::string::size_type says count the characters in the string that I provide, and give me the number as an int that I can use. Instead of you making a large class with lots of code that do count all the characters and give you an int.. :)[21:10:22] *** HuntsMan has quit IRC[21:10:47] <Mathematicz> Is there good D3D books?[21:12:30] <Ingenu> no clue[21:12:39] <Ingenu> ShaderX, GPU Gems,...[21:12:51] <Ingenu> cover D3D[21:14:31] *** iion has joined ##OpenGL[21:14:49] *** HuntsMan has joined ##opengl[21:15:20] *** Mathematicz has quit IRC[21:16:02] <kurt_> Mathematicz left?[21:20:19] *** bijoo_osdev has quit IRC[21:20:20] *** bijoo_appdev has quit IRC[21:22:55] *** _THEGOD__ has joined ##OpenGL[21:26:25] <kurt_> that's good :)[21:26:40] <tmccrary> By switching to geico?[21:26:52] <RTFM_FTW> no actually by switching to USAA[21:28:40] <_THEGOD__> where you were before ?[21:28:43] <_THEGOD__> niggereia ?[21:29:29] <tmccrary> Is MA a no fault state?[21:30:19] <RTFM_FTW> yep[21:30:20] *** pragma_ is now known as oranges`[21:30:23] <RTFM_FTW> as I recall[21:33:16] <_THEGOD__> damn[21:33:20] <_THEGOD__> crasday[21:33:27] <_THEGOD__> does not works on my pc[21:33:37] <_THEGOD__> it says something is missing about directx[21:33:44] <_THEGOD__> and nor flat out 2[21:34:05] <_THEGOD__> while it works with need for speed.[21:34:15] <_THEGOD__> which i hate a bit .[21:34:42] <_THEGOD__> do anyone knows some cool games like that ?[21:34:47] <_THEGOD__> which uses opengl .[21:34:53] <_THEGOD__> not directx[21:35:30] *** X-Scale has quit IRC[21:38:31] *** lewymati has quit IRC[21:39:51] <duzt> _THEGOD__, good games?[21:39:58] <Renderwahn> _THEGOD__: dwarf fortress[21:40:31] <duzt> quake3[21:40:44] <duzt> anything id software for the moment[21:40:55] <dusted> id<3[21:40:57] <duzt> but quake3 is opensrc which may be what your looking for[21:41:09] *** X-Scale has joined ##OpenGL[21:41:10] <dusted> actually, ioQuake3 is what you're looking for :P[21:42:25] <duzt> tru[21:43:04] <tmccrary> why is id less than 3[21:43:25] <dusted> Because 3 is the magic number[21:43:39] <tmccrary> one is the loneliest number[21:43:45] <dusted> yes :([21:44:16] *** wisey7 has joined ##OpenGL[21:44:50] <Renderwahn> ε> <3[21:46:38] *** druggy has quit IRC[21:46:49] *** druggy has joined ##opengl[21:48:24] *** oranges` is now known as pragma_[21:53:07] <Ingenu> <3, share it[21:53:12] <Ingenu> ;)[21:53:26] *** Stevethepirate has quit IRC[21:55:49] *** JoshDreamland has joined ##OpenGL[21:56:00] <JoshDreamland> hola, tacos[21:56:30] <JoshDreamland> just looking for a way to copy draw buffer into a buffer in memory[21:56:40] <JoshDreamland> in order to print it to a file.[21:56:54] *** wisey has quit IRC[21:57:20] <dusted> glReadPixels ?[21:57:32] <JoshDreamland> *smacks forehead*[21:57:34] <JoshDreamland> thanks.[21:57:52] <dusted> np :)[21:57:58] <JoshDreamland> wow, I even used that before in case framebuffer extension failed[21:58:02] <JoshDreamland> I wonder how I just forget..[21:58:02] *** bhagabhi has quit IRC[21:58:07] <JoshDreamland> forgot, too.[22:03:31] *** HuntsMan has quit IRC[22:10:53] <aep> despite i set GL_LINE_SMOOTH_HINT to GL_NICEST, lines still apear every rough. any idea what to enhance?[22:13:35] <aep> ah blending[22:13:38] <aep> looks nice now[22:15:03] *** wisey7 has quit IRC[22:15:49] <HideousNashimoto> http://imagebin.ca/view/IG8zsh.html[22:16:09] <HideousNashimoto> aep , how did you done it?[22:16:16] <HideousNashimoto> I have same problem[22:16:54] <aep> http://www.glprogramming.com/red/chapter06.html#name2[22:16:56] <aep> its mentioned there[22:17:26] <aep> basicly i used an ARGB visual and enabled blending[22:23:24] *** Tenac_ is now known as Tenac[22:25:13] *** _THEGOD__ has quit IRC[22:25:34] *** Nescafe has quit IRC[22:25:38] *** iion has quit IRC[22:38:06] *** Cultz has joined ##OpenGL[22:44:22] *** reprore_ has quit IRC[22:45:56] *** dusted has quit IRC[22:48:53] *** xp_prg has quit IRC[22:49:46] *** predaeus has quit IRC[22:50:57] *** HuntsMan has joined ##opengl[22:51:44] *** Ingenu has quit IRC[22:55:47] *** dolphin has quit IRC[23:03:05] *** nitrotrigger has quit IRC[23:09:51] *** ExcelsisD has joined ##OpenGL[23:12:30] *** walt has quit IRC[23:12:49] *** walt has joined ##OpenGL[23:15:17] *** Jonanin has quit IRC[23:19:44] *** X-Scale has quit IRC[23:19:45] *** kbotnen has joined ##OpenGL[23:23:15] *** walt has quit IRC[23:27:32] *** Renderwahn has quit IRC[23:33:16] *** eu-prleu-peupeu has joined ##OpenGL[23:33:20] <eu-prleu-peupeu> hi[23:33:44] <eu-prleu-peupeu> how do i know that im triggering a shader recompilation by changing uniform values in runtime ?[23:34:19] <eu-prleu-peupeu> if i specify those values as varying will it be ok ?[23:34:24] <RTFM_FTW> you don't[23:34:52] <RTFM_FTW> OTOH profiling should give you a reasonable idea of where any (possible) performance issues might be[23:37:24] *** itewsh has joined ##OpenGL[23:37:58] <eu-prleu-peupeu> :/ ok[23:45:14] *** ExcelsisD is now known as X-Scale