November 8, 2014 
< | 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 | >
November 8, 2014 [00:01:06] *** hexagoxel has joined ##OpenGL[00:01:51] *** jdolan has joined ##OpenGL[00:02:08] <aethersis> anyway, Yaniel, foobaz, Bloodlust, I'm very thankful for your help[00:02:44] *** Ryp has joined ##OpenGL[00:05:51] *** TechEffigy has joined ##OpenGL[00:11:23] *** TechEffigy has quit IRC[00:12:48] *** konom has joined ##OpenGL[00:15:30] *** sreich_ has joined ##OpenGL[00:16:45] *** sreich has quit IRC[00:17:49] *** Ad1 has quit IRC[00:17:55] *** Ryp has quit IRC[00:17:55] *** roboman2444 has joined ##OpenGL[00:18:17] *** Match has quit IRC[00:19:07] *** Zupo has joined ##OpenGL[00:23:34] *** Zupo has quit IRC[00:24:13] <japro> tensors are like the monads of physics :[00:24:16] *** jdolan has quit IRC[00:24:21] <aethersis> lol[00:24:27] <japro> everyone constantly goes on about them but no one actually gets what they are[00:24:57] <aethersis> I also find it hard to understand them[00:26:04] *** Alina-malina has quit IRC[00:26:43] *** Alina-malina has joined ##OpenGL[00:29:44] <Yaniel> at least monads have successfully been compared to burritos[00:30:06] *** _leb has quit IRC[00:32:19] *** nszceta has joined ##OpenGL[00:32:21] *** stefkos has quit IRC[00:35:06] <aethersis> lmao[00:35:24] <Yaniel> the great part about that is that it actually makes sense[00:35:30] <Yaniel> the comparison that is[00:36:01] <aethersis> weird xD[00:36:06] <aethersis> do they make you fart?[00:36:22] *** japro has quit IRC[00:36:29] *** meoblast001 has quit IRC[00:38:01] *** Crehl_ has joined ##OpenGL[00:38:10] *** aethersis has quit IRC[00:40:21] *** staylor has quit IRC[00:40:24] *** PasNox has quit IRC[00:40:53] *** Crehl has quit IRC[00:41:16] *** foreignFunction has quit IRC[00:44:20] *** ShadowIce has quit IRC[00:44:20] *** TheBunnyZOS has quit IRC[00:45:54] *** b4b has joined ##OpenGL[00:51:44] *** phr34k has quit IRC[00:55:16] *** seangrove has joined ##OpenGL[00:58:44] *** Waynes1 has quit IRC[01:10:08] *** kuldeepdhaka has joined ##OpenGL[01:12:06] *** bkre__ has quit IRC[01:12:28] *** Orion] has quit IRC[01:15:44] *** tcsc has joined ##OpenGL[01:16:14] *** Jack__ has quit IRC[01:16:16] *** shintah has quit IRC[01:18:46] *** hahuang65 has quit IRC[01:20:22] *** TunnelCat has quit IRC[01:23:08] *** cmr has joined ##OpenGL[01:27:18] *** RyanPridgeon has joined ##OpenGL[01:29:39] *** Biliogadafr has quit IRC[01:35:46] *** petervaro has joined ##OpenGL[01:36:59] *** seangrove has quit IRC[01:37:04] *** Orion] has joined ##OpenGL[01:43:08] *** DMJC-M has quit IRC[01:44:53] *** hexagoxel has quit IRC[01:51:37] *** SleekoNiko has joined ##OpenGL[01:52:43] *** Bigpet_ has quit IRC[02:01:01] *** Lemml has quit IRC[02:01:24] *** charlie5 has quit IRC[02:03:06] *** Jack__ has joined ##OpenGL[02:12:20] *** huntercool has quit IRC[02:13:00] *** petris has quit IRC[02:13:35] *** petris has joined ##OpenGL[02:13:52] *** hahuang65 has joined ##OpenGL[02:16:41] *** slidercrank has quit IRC[02:17:59] *** Keniyal_ has joined ##OpenGL[02:18:18] *** huntercool has joined ##OpenGL[02:18:21] *** Gamecubic has quit IRC[02:20:40] *** Gamecubic has joined ##OpenGL[02:21:07] *** sreich__ has joined ##OpenGL[02:21:30] *** Dom__ has joined ##OpenGL[02:21:33] *** pigeon_ has joined ##OpenGL[02:21:52] *** zerd_ has joined ##OpenGL[02:21:59] *** vottini has quit IRC[02:21:59] *** pigeon has quit IRC[02:22:04] *** sreich_ has quit IRC[02:22:05] *** nszceta has quit IRC[02:22:06] *** Keniyal has quit IRC[02:22:07] *** bkc_ has quit IRC[02:22:10] *** Dom_ has quit IRC[02:22:10] *** jhuntley has quit IRC[02:22:11] *** Yaniel has quit IRC[02:22:12] *** zerd has quit IRC[02:22:13] *** Atomim has quit IRC[02:22:14] *** vottini has joined ##OpenGL[02:22:18] *** jhuntley has joined ##OpenGL[02:22:37] *** Yaniel has joined ##OpenGL[02:22:49] *** bkc_ has joined ##OpenGL[02:23:12] *** Atomim has joined ##OpenGL[02:26:05] *** charlie5 has joined ##OpenGL[02:27:54] *** Demon_Fox has quit IRC[02:28:09] *** Demon_Fox has joined ##OpenGL[02:28:39] *** slime has quit IRC[02:32:42] *** _leb has joined ##OpenGL[02:39:42] *** jdolan has joined ##OpenGL[02:50:32] *** qeed has joined ##OpenGL[03:00:58] *** jdolan has quit IRC[03:01:41] *** Misu has quit IRC[03:04:43] *** jdolan has joined ##OpenGL[03:06:45] *** b4b has quit IRC[03:08:10] *** Lucretia has quit IRC[03:08:27] *** Lucretia has joined ##OpenGL[03:11:24] *** akiress has joined ##OpenGL[03:11:39] *** Ad1_RnR has joined ##OpenGL[03:13:11] *** Garner has quit IRC[03:14:32] *** Leates has quit IRC[03:15:17] *** Ad1_RN has quit IRC[03:17:10] *** seangrove has joined ##OpenGL[03:17:10] *** b4b has joined ##OpenGL[03:18:26] *** MLM has joined ##OpenGL[03:21:59] *** bjz has joined ##OpenGL[03:23:10] *** seangrove has quit IRC[03:26:28] *** Crehl_ has quit IRC[03:27:46] *** Shogun has quit IRC[03:29:02] *** roboman2444 has quit IRC[03:29:27] *** charlie5 has quit IRC[03:31:59] *** akiress has quit IRC[03:40:23] *** puerum has joined ##OpenGL[03:40:28] *** phr34k has joined ##OpenGL[03:48:31] *** HuntsMan2 has joined ##OpenGL[03:52:28] *** Hunts has quit IRC[03:55:27] *** slime has joined ##OpenGL[03:58:18] *** b4b_ has joined ##OpenGL[04:01:41] *** b4b has quit IRC[04:02:46] *** b4b_ has quit IRC[04:03:44] *** b4b has joined ##OpenGL[04:06:51] *** jdolan has quit IRC[04:08:29] *** b4b has quit IRC[04:09:22] *** phr34k has quit IRC[04:09:40] *** b4b has joined ##OpenGL[04:11:21] *** jdolan has joined ##OpenGL[04:14:17] *** b4b has quit IRC[04:15:47] *** b4b has joined ##OpenGL[04:17:46] <b4b> weird problem I can only login into sshd when I’m conencted to the router the server is connected otherwise I get permission denied when entering my password[04:17:59] <b4b> does it have to do with ssh keys?[04:18:39] *** jdolan has quit IRC[04:21:30] *** jdolan has joined ##OpenGL[04:21:49] *** b4b_ has joined ##OpenGL[04:22:31] *** b4b has quit IRC[04:22:31] *** b4b_ is now known as b4b[04:23:31] *** Shogun has joined ##OpenGL[04:24:35] *** jdolan has quit IRC[04:26:58] *** xissburg has quit IRC[04:27:07] *** b4b has quit IRC[04:31:17] *** b4b has joined ##OpenGL[04:31:47] *** agorecki has quit IRC[04:32:36] <Demon_Fox> centrinia, I'm in here to[04:32:47] <centrinia> Cool.[04:33:06] <Demon_Fox> If I got more sleep, I would have been more help[04:33:16] *** vegard1992 has quit IRC[04:33:23] <centrinia> I guess I could use scissor. :p[04:34:08] *** Burga has joined ##OpenGL[04:34:24] *** devbug has joined ##OpenGL[04:35:14] *** b4b_ has joined ##OpenGL[04:36:25] *** b4b has quit IRC[04:36:26] *** b4b_ is now known as b4b[04:39:27] <cmr> Anyone know of a mirror for etcpack? http://devtools.ericsson.com/etc seems to be perpetually down...[04:42:03] *** qeed has quit IRC[04:42:30] *** Keniyal_ has quit IRC[04:42:46] *** b4b_ has joined ##OpenGL[04:42:50] *** urraka has quit IRC[04:42:54] *** b4b has quit IRC[04:42:54] *** b4b_ is now known as b4b[04:43:33] *** YarnFive8 has quit IRC[04:44:00] *** Cabanoss- has joined ##OpenGL[04:44:10] *** kidnapped_robot has quit IRC[04:44:12] *** sreich__ is now known as sreich[04:47:14] *** RyanPridgeon has quit IRC[04:47:20] *** Cabanossi has quit IRC[04:47:21] *** Cabanoss- is now known as Cabanossi[04:47:22] *** b4b has quit IRC[04:48:02] *** b4b has joined ##OpenGL[04:50:02] *** Twinklebear has joined ##OpenGL[04:52:46] *** b4b has quit IRC[04:55:52] *** b4b has joined ##OpenGL[04:57:07] *** petervaro has quit IRC[04:59:50] *** Orion] has quit IRC[05:01:53] *** Burga has quit IRC[05:02:48] *** charlie5 has joined ##OpenGL[05:02:50] *** samrat has joined ##OpenGL[05:05:54] *** seangrove has joined ##OpenGL[05:08:39] *** jdolan has joined ##OpenGL[05:11:29] *** seangrove has quit IRC[05:17:18] *** shingshang has joined ##OpenGL[05:17:40] *** kestasjk has joined ##OpenGL[05:18:32] <kestasjk> hey guys, I write the visualization software for a maritime simulator and we really need to up our game when it comes to 3D water[05:18:52] *** shingshang has quit IRC[05:19:18] *** shingshang has joined ##OpenGL[05:19:19] <kestasjk> at the moment it's just a bunch of quads that are sent over and over to the card, some shading for basic reflections etc[05:20:23] <kestasjk> we're using openscenegraph, and Im not sure whether to really go back to basics and figure it all out, or if there's some more prescriptive way of doing great looking 3D water[05:22:25] <kestasjk> how long does it usually take someone to become a real opengl artist?[05:25:41] <Stragus> Usually, one becomes an OpenGL programmer, not an OpenGL artist :)[05:25:47] <kestasjk> Im a CS grad and when it comes to OO coding, .NET, C++, SQL, UI design, text processing, data-driven web, etc, etc I'm golden, but 3D graphics is still very hard for me to grasp. all the parellelism and how so much seems to just be tweaking and knowing the tricks e.g. solar flare and reflections. anyone else crossed that divide[05:26:09] *** jdolan has quit IRC[05:26:27] <Stragus> Good looking water isn't too hard, there are many techniques out there[05:27:00] <kestasjk> "Any sufficiently advanced technology is indistinguishable from magic[05:27:48] *** derhass_ has joined ##OpenGL[05:27:53] <Stragus> Have you played with vertex and fragment shaders?[05:28:03] <kestasjk> what complicates things is for us water isn't just eye candy; in the simulator the water behaviour has to mimic exactly the behavior experienced by the ships[05:28:14] <kestasjk> so I can't just drop in e.g. Triton[05:28:26] <Stragus> Okay, a ship going through an ocean will be harder[05:29:23] <kestasjk> Stragus, briefly, to get reflections working and draw thrusts onto the ocean, but a lot of it building off work others have done and hacking at it until it works (a technique I hate). and tesselation I'm lost at[05:31:04] *** derhass has quit IRC[05:31:33] <Stragus> Ah, so you need reflections[05:31:42] <kestasjk> the shame of it is we have so many really powerful GPUs are our disposal that I just know are being horribly underutilzed, meanwhile the GPUs are working like crazy updating low-res wave grids[05:32:24] <Stragus> I wasn't sure what kind of quality level you were aiming at. If you want to simulate wave grids with reflections, it's going to be pretty intensive[05:32:45] <kestasjk> reflections are nice, most of the time if it's calm enough that you see a clear reflection you're not doing emergency exercises, but thrusts are must have[05:32:51] <Stragus> There are other, simpler techniques which are fairly cheap and look okay : http://www.rayforce.net/water00.png[05:33:51] *** pizthewiz has quit IRC[05:36:10] <b4b> glYoda will be charged with crimes against the irc community[05:36:19] <kestasjk> they do look nice, but making them match the sim experience is tricky. you need choppiness and foam due to high wind, thrusts, etc[05:37:07] <glYoda> "glYoda will be charged with crimes against the irc community" ...WTF?[05:37:13] <kestasjk> Im just worried Im going to have to learn this stuff inside out[05:37:28] <Stragus> Right, so you are aiming at the high-end. Creating that water might take a while even for someone with plenty of shader experience[05:37:38] <Stragus> kestasjk, I think you will, for what you are aiming at[05:37:49] <kestasjk> damn[05:38:24] <glYoda> b4b why are you here? …I don't think I've seen you contribute anything of value in this channel for quite some time now[05:45:14] *** pragma- is now known as p[05:45:41] *** gaganjyot has joined ##OpenGL[05:45:41] *** p is now known as pragma-[05:46:15] <kestasjk> are there any consultants who can deal with virtual water really well ?[05:46:30] *** Jack_ has joined ##OpenGL[05:47:58] <kestasjk> okay next question then: visual studio 2010, what are he best possible tools to make shader development straightforward[05:49:24] *** Jack__ has quit IRC[05:49:25] <Stragus> I'm sure there are persons with experience of realistic water around, but answers might be a little delayed; it's midnight in the Americas and very early in Europe[05:54:52] <XMPPwocky> virtual water how?[05:55:10] <XMPPwocky> smooth flowing water?[05:55:12] <kestasjk> http://kestas.kuliukas.com/12.png http://kestas.kuliukas.com/13.png[05:55:13] <XMPPwocky> "noisy" waterfalls?[05:55:31] <XMPPwocky> kestasjk: 404[05:55:32] <Stragus> 404[05:55:40] *** konom has quit IRC[05:56:55] <kestasjk> my mistake: http://kestas.kuliukas.com/12.jpg http://kestas.kuliukas.com/13.jpg[05:57:04] *** Textmode has joined ##OpenGL[05:57:50] <kestasjk> with no waves from a height it looks okay, but when on a tug with large swell and high wind things look very odd[05:58:42] <kestasjk> the only flowing water worth worry about is the chaotic wake water developed by thrusts[05:59:28] <XMPPwocky> kestasjk: add a tristrip or something for wakes[05:59:37] * Textmode huggles ##OpenGL :3[05:59:48] <kestasjk> and wakes[06:00:05] *** samrat has quit IRC[06:00:35] <Stragus> kestasjk, do you presently do the normal perturbations with tesellation or shaders, or both?[06:02:03] <kestasjk> we have a bump map which gives the normal stuff so that it doesn't look flat, then we have the quads on top of those. wakes and thrusts have no effect on the quads and so are basically painted onto the wave texture, as with reflections[06:02:05] <XMPPwocky> that just looks like a dumpmap[06:02:06] <XMPPwocky> yeah[06:02:11] <XMPPwocky> bumpmap*[06:02:46] *** HOS_cr`nge has quit IRC[06:03:06] <kestasjk> the problem is this technique looks nice in this circumstance, but in a choppy sea with a decent swell it looks a bit silly[06:03:11] <Stragus> I assume the bump map is animated one way or the other, people usually merge multiple normal maps to give a changing, chaotic look to the surface[06:03:16] <Stragus> Indeed[06:03:39] <Stragus> You'll want shader-based normal perturbations on top of geometry-based perturbations for bigger changes[06:03:46] <kestasjk> and this isnt like a flight sim where they don't know what it actually looks like, they know better than me[06:05:10] <Stragus> The geometry-based perturbations will need some basic wave propagation and dissipation, that might already do the trick[06:05:25] <XMPPwocky> you could be clever and abuse render to texture[06:05:25] *** karswell` has joined ##OpenGL[06:05:27] <kestasjk> I think the bump map is shifted with time, not altered but meaning of colors are sort of shifted so normals slowly change[06:05:32] <XMPPwocky> and solve differential equations that way[06:05:58] <XMPPwocky> like actually solve the wave equation, it would probably look decent if you do your things correctly[06:06:08] *** karswell has quit IRC[06:06:09] <Stragus> Ah, some kind of animated cyclic normal map, that can work... It can get a little noticeable though[06:06:46] <Stragus> Bottom line, you really need geometry based perturbations on top of the shader normal perturbations for wakes and other serious effects[06:07:08] <Stragus> Particle effects on the side would probably help in some cases[06:07:19] *** TunnelCat has joined ##OpenGL[06:07:28] <kestasjk> so tesslation of some sort to increase the geometry without taxing the CPU?[06:08:02] <Stragus> Yes. You could even do the wave dynamics on the GPU if you wanted[06:08:11] <kestasjk> nice wind wash stuff would be the icing on the cake, but at the moment even when very choppy it looks so flat. this is the the #1 issue[06:08:14] <XMPPwocky> do you actually need tesselation[06:08:18] <Stragus> Tesselation can be dynamic if you want to adapt to local surface complexity... giant quads when it's smooth[06:08:25] *** b4b has quit IRC[06:08:29] <XMPPwocky> like, if all you see is water[06:08:33] <Stragus> XMPPwocky, yes he does[06:08:52] <XMPPwocky> you can't tell a cleverly-shaded quad apart from "real" geometry[06:09:03] <kestasjk> I do, but it has to be in sync with all other machines, becuase each sim has ~8 vis machines and the facility has ~7 sims[06:09:13] <kestasjk> so waves can't be different from one place to another[06:09:56] <XMPPwocky> why are good-looking waves so important? :P[06:10:07] <kestasjk> XMPPwocky, unfortunately they just are[06:10:26] *** razieliyo has quit IRC[06:10:45] <kestasjk> XMPPwocky, my manager asks me all the time, but the clients just see crappy visuals and think crappy simulator, no matter how good the models are[06:10:58] *** slidercrank has joined ##OpenGL[06:11:08] <Stragus> Typical, eye-candy is what sells :p[06:11:22] <Stragus> Even in the physics simulation business, tell me about it[06:11:51] <kestasjk> yup[06:12:48] <kestasjk> lightening is pracitcally meaninless to these guys, but we want to have lightening with the delay thunder and show-casting-flash just because it looks nice[06:13:04] <kestasjk> *shadow-casting-flash[06:13:31] <XMPPwocky> i think an animated bumpmap, on top of displacement-mapped vertices[06:13:43] <XMPPwocky> w/ dynamic tess for LODing[06:14:23] <kestasjk> wish I had more than a vague idea what you mean, but Ill look into it[06:14:55] <XMPPwocky> if you're generating the displacement map from some noise function[06:15:15] <kestasjk> would that approach allow interference to be added from ship wakes/thrusts/bow waves ?[06:15:30] <XMPPwocky> the displacement map could simply be a texture[06:15:42] <XMPPwocky> or really you could get away with a height map, possibly[06:16:55] <kestasjk> food for thought, thanks[06:18:16] <Stragus> I would rather suggest tesselation than displacement maps... seems easier and cheaper, but there are advantages to each approach[06:18:25] <Stragus> Just look into the various ways[06:18:36] <kestasjk> please elaborate[06:19:19] <XMPPwocky> Stragus: i mean tesselation to LOD "quads" which sample from the dispmap[06:20:12] <Stragus> Ah. So you mean looking up the elevation at each point from sampling a texture in the vertex shader[06:20:20] <kestasjk> when I think height map I think resending a height map to 60 or so vis machines hundreds of times a second[06:20:21] <XMPPwocky> right[06:20:24] <Stragus> None of that intersection with displacements in the fragment shader[06:20:25] <XMPPwocky> or geometry shader[06:20:49] <XMPPwocky> Stragus: like you sometimes do for facial animation[06:20:54] <Stragus> I would rather use a regular attrib array of "heights" as input to the vertex shader[06:21:10] <XMPPwocky> Stragus: yeah, but then you can't LoD it as "magically"[06:21:13] <Stragus> With variable tesselation per "area", if necessary[06:21:21] <XMPPwocky> if it's a 2d texture (across the surface of the water)[06:21:42] <XMPPwocky> emitting more verts from tesellation / geom shader just samples more for each vertex[06:21:50] <XMPPwocky> and texture filtering does the rest[06:21:52] *** seangrove has joined ##OpenGL[06:22:12] <Stragus> Hum yes, there are advantages to both. Texture sampling from the vertex shader can be a little slow on some hardware[06:23:42] <XMPPwocky> kestasjk: if you can come up with a decent noise function that looks sufficiently ocean-y you can just sync a seed for that[06:23:49] <Stragus> kestasjk, regarding the height map, each visualization machine should probably compute its own[06:24:04] <Stragus> There's no need to send that information if it can be computed cheaply[06:24:46] <Stragus> Be very careful if you use floating point and expect consistent results though[06:26:15] <kestasjk> well I dont undestand most of what you just said, but having them compute the waves independently will be tricky. we have different models and some would be straightforward. others where waves go around breakwaters are precomputed and can't be done on the fly, and it's definitely not about generating something that looks oceany if it doesn't map our specific sea-state conditions[06:27:03] *** seangrove has quit IRC[06:27:20] <kestasjk> what kind of limit is there for loading new texture maps?[06:27:57] <kestasjk> doing it every 1/50 seconds or so would be pretty ample[06:28:08] <XMPPwocky> you can stream data to a texture pretty easily[06:28:17] <Stragus> You can do it every frame, if you do it intelligently and it's a reasonable size[06:28:24] <XMPPwocky> you want to declare it with the right usage hints[06:28:39] <kestasjk> how do you mean?[06:28:41] <Stragus> More exactly, don't upload a same texture every frame; cycle between like 3 textures[06:30:26] <kestasjk> so based on this height bitmap the tesselator shader then generate some extra geometry?[06:30:48] *** Zeioth has quit IRC[06:31:01] <kestasjk> probably where I need to do some reading anyway[06:31:21] <Stragus> You could also just define all vertices, without a tesselation shader[06:31:41] <Stragus> If your water has a complex edge, I would rather pre-compute its exact geometry[06:32:04] <Stragus> (In other words, no need to render water under a mountain)[06:32:47] <kestasjk> good point[06:37:02] *** pizthewiz has joined ##OpenGL[06:40:52] *** centrinia has quit IRC[06:41:27] *** gaganjyot has quit IRC[06:42:59] *** gaganjyot has joined ##OpenGL[06:47:05] *** BreadProduct_ has joined ##OpenGL[06:50:44] *** BreadProduct has quit IRC[06:50:53] *** BreadProduct_ is now known as BreadProduct[06:53:09] *** HOS_cr`nge has joined ##OpenGL[06:53:12] *** BreadProduct_ has joined ##OpenGL[06:53:15] *** samrat has joined ##OpenGL[06:53:48] *** DMJC-M has joined ##OpenGL[06:53:55] *** samrat has quit IRC[06:54:03] *** BreadProduct__ has joined ##OpenGL[06:55:03] *** centrinia has joined ##OpenGL[06:55:43] *** BreadProduct has quit IRC[06:55:52] *** BreadProduct__ is now known as BreadProduct[06:57:44] *** BreadProduct_ has quit IRC[07:05:44] <Textmode> odd question from beyond the bounds of freenode; but is it reasonable to call glViewport(…) multiple times in a frame, e.g. to do split-screen or PiP?[07:06:11] *** karswell` has quit IRC[07:10:53] *** samrat has joined ##OpenGL[07:14:39] *** Jack_ has quit IRC[07:18:49] *** MLM has quit IRC[07:20:37] *** phr34k has joined ##OpenGL[07:20:40] <chrisf> Textmode: yes, that is fine.[07:21:03] *** DMJC-M has quit IRC[07:21:40] *** MLM has joined ##OpenGL[07:24:29] *** slidercrank has quit IRC[07:26:54] *** DMJC-M has joined ##OpenGL[07:27:02] *** roboman2444 has joined ##OpenGL[07:35:48] *** roboman2444 has quit IRC[07:40:25] *** MLM has quit IRC[07:42:09] *** MLM has joined ##OpenGL[07:43:36] *** damir__ has joined ##OpenGL[07:44:27] *** phr34k has quit IRC[07:45:41] *** zajfy has quit IRC[07:48:51] *** samrat has quit IRC[07:49:13] *** gaganjyot has quit IRC[08:00:59] *** gaganjyot has joined ##OpenGL[08:01:42] *** tcsc has quit IRC[08:02:45] *** pabb has quit IRC[08:03:35] *** Gamecubic has quit IRC[08:04:42] *** telex has quit IRC[08:06:49] *** telex has joined ##OpenGL[08:09:04] <centrinia> I finally got stereoscopic 3D working with glViewport: http://i.imgur.com/gCZDVGt.png[08:10:16] *** Lemml has joined ##OpenGL[08:18:15] <Yaniel> Textmode: quite common even[08:21:49] *** devbug has quit IRC[08:21:55] <cmr> Is there much of a disadvantage to using[08:22:22] <cmr> grr... something like Cg, compiling to ARB programs, and then submitting those instead of GLSL shaders, on modern drivers?[08:22:29] *** jhuntley has quit IRC[08:22:54] <Yaniel> as far as I know Cg is getting a bit behind the times[08:23:23] <Yaniel> and I don't think ARB shaders have been updated in a long time[08:25:09] <cmr> neither did I, but then I saw http://developer.download.nvidia.com/opengl/specs/GL_NV_gpu_program5.txt[08:25:43] *** b4b has joined ##OpenGL[08:25:46] <Yaniel> oh[08:25:51] <cmr> nvidia keeps extending ARB programs to support features on their new GPUs[08:26:03] <Yaniel> I guess then it is fine at least on nv[08:26:15] <Yaniel> if the Cg compiler is kept up to date too[08:26:44] *** jdolan has joined ##OpenGL[08:27:22] <cmr> I haven't been able to find anything for AMD's stuff, so I think compiling to glsl instead of an assembly language might be best.[08:27:29] <cmr> (or intel, for that matter)[08:31:22] *** glYoda has quit IRC[08:31:42] *** glYoda has joined ##OpenGL[08:32:01] *** jdolan has quit IRC[08:37:43] <chrisf> cmr: if all you care about is nvidia, then maybe things like GL_NV_gpu_program5 are viable. unextended arb programs are very limited.[08:38:57] <cmr> chrisf: yeah, that's the conclusion I've come to. they're the only vendor doing anything like it[08:39:19] <foobaz> what are you compiling to GLSL?[08:39:28] <chrisf> glsl and cg are pretty similar; if you can write cg you can adapt easily[08:39:42] <cmr> foobaz: a hypothetical graphics-api-"agnostic" shading language.[08:39:45] <slime> cmr: also keep in mind the asm used in those still doesn't match hardware[08:40:01] <foobaz> oh right, i tend to forget directx exists[08:40:06] <Yaniel> the only real reason to use Cg is if you want to use the same shaders with both OpenGL and D3D[08:40:39] <cmr> foobaz: not just d3d, but now also metal, maybe mantle too, maybe gl next-gen... I foresee a proliferation of shading languages[08:41:11] <cmr> there isn't really any good tooling for converting between these things, it's easier to just rewrite the shaders[08:41:17] <foobaz> i would like a functional shading language[08:41:30] <slime> cmr: https://github.com/aras-p/glsl-optimizer can compile glsl to metal[08:42:25] <foobaz> writing regular programs in a functional language feels limiting, but shader programs already have all those limitations by nature[08:43:26] <chrisf> increasingly not though -- plenty of side effects in modern glsl[08:44:07] *** Garner has joined ##OpenGL[08:45:31] <cmr> slime: nice, didn't see that.[08:45:47] <foobaz> true, but for this language to be cross-platform it could not express opengl-specific side effects, only the functionality that is common to all its targets[08:46:37] *** TyrfingMjolnir has quit IRC[08:47:55] *** phr34k has joined ##OpenGL[08:50:41] *** pizthewiz has quit IRC[08:55:15] *** Twinklebear has quit IRC[09:01:21] *** charlie5 has quit IRC[09:07:27] *** b4b has quit IRC[09:14:00] *** samrat has joined ##OpenGL[09:19:05] *** charlie5 has joined ##OpenGL[09:33:21] *** ShadowIce has joined ##OpenGL[09:33:58] *** groton has joined ##OpenGL[09:38:20] *** _leb has quit IRC[09:41:44] <dahlia> how about this thing? https://icculus.org/mojoshader/[09:43:11] <cmr> dahlia: it has one of the problems that most of these sorts of tools have: poor feature support[09:43:21] <cmr> only SM3[09:43:25] <cmr> valve's tools are similar.[09:43:34] <dahlia> ah[09:44:18] *** samrat has quit IRC[09:45:10] *** onatbas has joined ##OpenGL[09:46:07] <onatbas> (join ##c++[09:46:13] <onatbas> oops.[09:46:34] *** japro has joined ##OpenGL[09:55:36] *** HunterD has joined ##OpenGL[09:59:06] *** charlie5 has quit IRC[09:59:35] *** katlogic has joined ##OpenGL[09:59:44] *** stoopdapoop has quit IRC[10:00:54] *** stoopdapoop has joined ##OpenGL[10:00:54] *** stoopdapoop has joined ##OpenGL[10:01:30] *** katlogic_ has quit IRC[10:02:19] *** shingshang has quit IRC[10:05:09] *** katlogic has quit IRC[10:06:39] *** katlogic has joined ##OpenGL[10:09:13] *** Demon_Fox has quit IRC[10:13:04] *** MLM has quit IRC[10:15:30] *** charlie5 has joined ##OpenGL[10:17:22] *** pazul has joined ##OpenGL[10:18:12] *** shingshang has joined ##OpenGL[10:27:07] *** onatbas has quit IRC[10:28:05] *** Khlorghaal has quit IRC[10:28:21] *** Slion has joined ##OpenGL[10:33:52] *** hexagoxel has joined ##OpenGL[10:34:29] *** stefkos has joined ##OpenGL[10:35:59] *** Textmode has quit IRC[10:37:17] *** charlie5 has quit IRC[10:37:48] *** viggo2 has joined ##OpenGL[10:39:50] *** Shogun has quit IRC[10:41:03] *** Shogun has joined ##OpenGL[10:41:15] *** _Cid has quit IRC[10:41:23] *** viggo has quit IRC[10:42:32] *** Dark_Confidant has joined ##OpenGL[10:44:48] *** _Cid has joined ##OpenGL[10:44:54] *** foreignFunction has joined ##OpenGL[10:45:14] *** ChrisUK has quit IRC[10:45:33] *** HunterD has quit IRC[10:45:43] *** Dark_Confidant|m has quit IRC[10:46:43] *** ChrisUK has joined ##OpenGL[10:47:34] *** onatbas has joined ##OpenGL[10:50:23] *** onatbas has quit IRC[10:53:23] *** charlie5 has joined ##OpenGL[10:55:18] *** CainJacobi has quit IRC[11:03:48] *** Shogun_ has joined ##OpenGL[11:06:43] *** Shogun has quit IRC[11:09:12] *** groton has quit IRC[11:12:50] *** Xeethrax- has quit IRC[11:17:22] *** Match has joined ##OpenGL[11:18:09] *** phr34k has quit IRC[11:19:46] *** charlie5 has quit IRC[11:21:29] *** Kane has joined ##OpenGL[11:23:54] *** Alina-malina has quit IRC[11:24:03] *** Alina-malina has joined ##OpenGL[11:24:26] *** Alina-malina has quit IRC[11:24:41] *** Alina-malina has joined ##OpenGL[11:24:56] *** Alina-malina has quit IRC[11:24:56] *** Alina-malina has joined ##OpenGL[11:27:25] *** Khlorghaal has joined ##OpenGL[11:30:34] *** karab44 has joined ##OpenGL[11:34:29] *** charlie5 has joined ##OpenGL[11:35:02] *** Leates has joined ##OpenGL[11:35:34] *** karab44 has quit IRC[11:38:01] *** shintah has joined ##OpenGL[11:41:17] *** _Cid has quit IRC[11:41:34] *** Ad1 has joined ##OpenGL[11:41:41] *** Khlorghaal has quit IRC[11:52:03] *** BreadProduct has quit IRC[11:52:26] *** Keniyal has joined ##OpenGL[11:55:27] *** _Cid has joined ##OpenGL[12:00:21] *** aethersis has joined ##OpenGL[12:00:26] *** aethersis is now known as dreamminder[12:00:37] <dreamminder> hi :)[12:01:24] <dreamminder> I have a question: When I call some function inside of fragment shader, does that function do something just for the single fragment being processed or for all fragments? Is there some way to make it work for all fragments at once?[12:03:59] <centrinia> Do you know how fragment shaders work?[12:04:08] <Bloodust> umm yes sort of[12:04:19] <centrinia> If you have a lot of data sharing, you should use an uniform.[12:05:59] *** t4nk661 has joined ##OpenGL[12:06:12] <dreamminder> I will explain what I need[12:06:43] <dreamminder> let's say there are 2 filters. The first filter does something, and the second filter needs the image to be already processed by the first filter[12:07:23] <centrinia> You need to render the result of the first filter before applying the second filter.[12:07:39] <dreamminder> so I'd have to have separate shaders for each filter?[12:08:02] <t4nk661> thats up to how you handle it in the shader, you can set "flags"[12:08:16] <centrinia> Unless the second filter processes pixels individually.[12:08:24] <dreamminder> what are these flags[12:08:45] <t4nk661> like an int "pass"[12:08:51] <t4nk661> 0 = filter 1, 1 = filter 2[12:08:57] <dreamminder> the thing I actually have is a fluid solver. Each function requires that all the cells were processed by the previous function[12:08:58] <t4nk661> simple if loop inside the shader[12:09:31] <dreamminder> hm smart, but won't it stop once it processes all fragments?[12:09:38] <centrinia> You might want to use something more compute oriented like OpenCL or compute shaders.[12:09:40] <t4nk661> ofc[12:09:43] <t4nk661> you need to use framebuffer objects[12:09:51] <dreamminder> that's what I want to do[12:10:01] <Yaniel> you usually need to do separate passes[12:10:12] <Yaniel> and then it makes sense to have one shader per pass[12:10:14] <t4nk661> in the second filter pass you bind the fbo texture to the shader[12:10:17] <t4nk661> and process that image[12:10:44] <dreamminder> I see[12:10:50] <dreamminder> maybe it'd be easier to have a few shaders[12:11:07] <centrinia> Why do you need to use OpenGL?[12:11:11] <t4nk661> well thinking long term, a deferred pipeline makes such things very easy[12:11:13] <dreamminder> that'd be less clear what's going on[12:11:36] <dreamminder> I don't really need to but I want to learn OpenGL by doing it[12:11:46] <dreamminder> I don't need to learn OpenCL[12:11:51] <dreamminder> and I want to learn OpenGL[12:12:06] <centrinia> What about compute shaders?[12:12:15] <dreamminder> they are limited to newer hardware[12:12:31] <Yaniel> DX11 level hw[12:12:40] <dreamminder> yeah, I dont want it[12:12:46] <Yaniel> it's not *that* new though[12:12:57] <dreamminder> starting from gtx 600 series I guess[12:13:00] <dreamminder> that's pretty new[12:13:42] <dreamminder> besides once I get it running on simple fragment shaders, then I will be able to run it on WebGL[12:14:28] <Yaniel> please, don't do this to me :D[12:15:27] <dreamminder> why? :o[12:16:12] *** junaidnaseer2 has joined ##OpenGL[12:17:26] <dreamminder> could I do something like this?: shader1 -> render to Fbo_1, shader2 ->render modified fbo_1 to fbo_2 , shader3->render modified fbo_2 to fbo_1 , final shader -> display fbo_1 ?[12:18:07] <dreamminder> and I'd use/unuse all of these shaders in render loop[12:18:14] <dreamminder> one after another[12:18:18] <dreamminder> would that work efficiently?[12:20:43] *** clauswitt has joined ##OpenGL[12:22:18] *** clauswitt has quit IRC[12:22:54] *** clauswitt has joined ##OpenGL[12:23:14] <dreamminder> Yaniel?[12:23:47] *** Jeanne-Kamikaze has joined ##OpenGL[12:25:40] *** Jeanne-Kamikaze has quit IRC[12:27:56] <ClaudiusMaximus> dreamminder: looks like a fairly standard approach, using fbo to cache results of previous processing and using the bound texture as input to the next pass[12:28:06] *** Jeanne-Kamikaze has joined ##OpenGL[12:28:14] <dreamminder> thank you ClaudiusMaximus[12:29:53] *** Jeanne-Kamikaze has quit IRC[12:30:37] *** Misu has joined ##OpenGL[12:30:58] *** DMJC-M has quit IRC[12:32:18] *** gaganjyot has quit IRC[12:32:47] *** clauswitt has quit IRC[12:33:06] *** clauswitt has joined ##OpenGL[12:37:10] <Yaniel> yes, that is how multipass techniques are usually done[12:37:21] <dreamminder> could you take a look at the code I have?[12:37:34] <dreamminder> https://github.com/aethersis/fluid-solver/blob/master/solver.cpp[12:37:50] <dreamminder> the 2 functions at the very bottom - vel_step and dens_step[12:38:44] <Yaniel> maybe in an hour or so[12:39:08] <centrinia> So it only looks at the nearest neighbors?[12:39:51] *** srjil_ has joined ##OpenGL[12:40:06] <dreamminder> yes, always moore neighbourhood[12:40:16] <dreamminder> almost every function does it[12:40:52] *** Slion has quit IRC[12:42:01] <centrinia> dreamminder: http://nullprogram.com/webgl-game-of-life/[12:42:36] <centrinia> He even made a blog entry: http://nullprogram.com/blog/2014/06/10/[12:43:31] *** Bigpet_ has joined ##OpenGL[12:44:20] *** qeed has joined ##OpenGL[12:45:14] *** robot-beethoven has quit IRC[12:48:54] *** gaganjyot has joined ##OpenGL[12:50:00] <dreamminder> niiice[12:50:02] <dreamminder> that will be very useful[12:50:07] *** paperManu_ has joined ##OpenGL[12:50:37] *** lbt has joined ##OpenGL[12:53:00] *** hdon has quit IRC[12:57:47] *** gaganjyot has quit IRC[13:04:26] *** phr34k has joined ##OpenGL[13:04:42] *** hexagoxel has quit IRC[13:07:22] *** HOS_cr`nge has quit IRC[13:07:33] *** Sound has joined ##OpenGL[13:10:10] *** hexagoxel has joined ##OpenGL[13:10:36] *** _Cid has quit IRC[13:11:04] *** _Cid has joined ##OpenGL[13:12:22] *** _Cid has quit IRC[13:17:45] *** pazul has quit IRC[13:19:06] *** centrinia has quit IRC[13:21:18] *** HOS_cr`nge has joined ##OpenGL[13:26:39] *** Waynes has joined ##OpenGL[13:29:24] *** paperManu_ has quit IRC[13:29:32] *** pazul has joined ##OpenGL[13:36:41] *** kuldeepdhaka has quit IRC[13:40:38] *** Orion] has joined ##OpenGL[13:42:22] *** Beetny has quit IRC[13:43:20] *** centrinia has joined ##OpenGL[13:45:54] *** clauswitt has quit IRC[13:46:45] *** clauswitt has joined ##OpenGL[13:48:22] *** gaganjyot has joined ##OpenGL[13:50:28] *** slidercrank has joined ##OpenGL[13:51:23] *** phr34k has quit IRC[13:52:14] *** gaganjyot has quit IRC[13:54:59] *** newguise1234 has quit IRC[13:55:41] <dreamminder> hmmhm anyone ready to help? xD[14:02:28] *** t4nk661 has quit IRC[14:03:09] *** antonp has joined ##OpenGL[14:04:27] *** razieliyo has joined ##OpenGL[14:04:27] *** razieliyo has joined ##OpenGL[14:05:06] <Sound> Uhm, I have a program for displaying meshes. On Linux, I can see the back faces, so there's some Z depth problem. On Mac and Windows it works fine. I know this information is not much for helping, but any hint? I have glEnable(GL_DEPTH_TEST); glDepthFunc(GL_LEQUAL); glEnable(GL_CULL_FACE);[14:05:49] <Sound> culling works, but opacity doesn't[14:05:50] *** newguise1234 has joined ##OpenGL[14:06:03] <japro> are you requesting a depth buffer at context creation?[14:06:24] <Sound> japro: this one? glClear(GL_COLOR_BUFFER_BIT | GL_DEPTH_BUFFER_BIT);[14:07:04] <japro> no, when you create your opengl window/context[14:07:42] <Sound> japro: I don't think so - I can't see any explicit command for doing that in my code. How would that look like?[14:07:57] <japro> how do you create your window?[14:08:27] <Sound> japro: it's a wxGLCanvas, I'm using wxWidgets. It provides the context[14:08:30] <Sound> japro: this is my code: https://github.com/alexrj/Slic3r/blob/master/lib/Slic3r/GUI/PreviewCanvas.pm#L449[14:09:00] *** BitPuffin has joined ##OpenGL[14:10:21] <Sound> japro: I'm checking out the wxGLCanvas documentation about depth buffer[14:12:18] <srjil_> Sound: Do you have a screenshot of what the problem looks like?[14:12:56] <Sound> srjil_: sure: https://github.com/alexrj/Slic3r/issues/2197[14:13:04] <srjil_> Sound: A couple of ideas come to mind that could be issues. For instance, are you specifying your vertices in the same order that you have culling set up for, i.e. clockwise versus counter-clockwise?[14:15:02] <Sound> srjil_: they are specified counter-clockwise (i.e. normal pointing towards front side)[14:15:36] <Sound> srjil_: and I have glCullFace(GL_BACK);[14:16:59] <srjil_> Sound: In your screenshot, is the issue the plane appearing in front of the model? The bunny and frog models look okay for instance, unless Iḿ missing something[14:18:12] <Sound> srjil_: yeah, exactly. The plane should be hidden by the models since it's below them. It's being drawn *after* the models, actually. If I draw it before, the models will cover it. But that doesn't sound too correct[14:18:51] <Sound> srjil_: also, I have another screenshot that I'm trying to upload that shows a convex model where I can see also some internal faces (whose normal points to the front)[14:18:52] *** hackkitten has quit IRC[14:20:13] <Sound> srjil_: here it is: http://i.imgur.com/1S34fNx.png[14:21:10] <Sound> srjil_: the plane is not above anymore because I'm just drawing it before the model as a workaround. But you can see there's a transparency that makes me think culling works, but Z depth doesn't[14:21:53] *** Crehl has joined ##OpenGL[14:23:08] *** Polarina has quit IRC[14:23:12] *** centrinia has quit IRC[14:24:12] *** urraka has joined ##OpenGL[14:24:54] <srjil_> Soud: Just as a test, you could try changing to CULL_FRONT to see what it looks like. I also try turning off culling all together to track down if it´s just a depth buffer issue.[14:25:25] *** Polarina has joined ##OpenGL[14:25:37] *** Polarina has joined ##OpenGL[14:26:41] <Sound> srjil_: GL_FRONT now shows a black-ish silhouette instead of the object, not much useful[14:27:43] <Sound> srjil_: disabling cull changes things a bit but still shows the transparency[14:28:18] <Sound> I'm trying to understand whether the context is initialized with a depth buffer or not[14:34:31] <Sound> that was it[14:35:15] *** stefkos has quit IRC[14:35:19] <Sound> japro: your hint was good! It looks like wxWidgets doesn't request a depth buffer by default on Linux.[14:35:26] <newguise1234> I've just ventured into the joy that is compiling in different environments and I am now getting a Compile failure in my vertex shader.[14:35:29] <Sound> japro, srjil_: thank you for your help[14:35:52] *** pazul has quit IRC[14:35:53] <newguise1234> I don't understand why this would happen if I'm specifying the opengl version with glew and have the shader version set to the same as on my other system[14:36:55] <HuntsMan2> newguise1234: many reasons, what is the compiler error message?[14:37:31] *** hdon has joined ##OpenGL[14:37:50] *** junaidnaseer2 has quit IRC[14:38:01] <newguise1234> error C0000: syntax error, unexpected '(', expecting "::" at token "("[14:38:23] <newguise1234> the line is a layout(location = 0) in vec4 vertexCoord;[14:38:47] <Yaniel> I don't know of anything in GLSL where you'd use ::[14:38:58] <newguise1234> thats part of why I'm confused lol[14:39:02] *** roboman2444 has joined ##OpenGL[14:39:03] <Yaniel> so I'm going to assume that error is actually from visual studio?[14:39:08] *** Garner has quit IRC[14:39:20] <newguise1234> I'm in linux building my project with a makefile[14:40:27] <newguise1234> the only thing I can think of is that when I compiled glew, I had to hide my nvidia libGL because it was tossing undefined references to me[14:41:10] <Yaniel> ummm[14:41:27] <Yaniel> that sounds like something you should investigate first[14:42:44] *** Garner has joined ##OpenGL[14:49:36] *** jdolan has joined ##OpenGL[14:51:12] *** jdolan has quit IRC[14:52:15] *** newguise1234 has quit IRC[14:52:41] *** jdolan has joined ##OpenGL[14:53:39] *** samrat has joined ##OpenGL[14:55:00] *** HuntsMan2 has quit IRC[14:55:18] *** HuntsMan has joined ##OpenGL[15:00:06] *** Match_it has joined ##OpenGL[15:03:19] *** timsche has joined ##OpenGL[15:04:34] <Match_it> is there any lib to load blender object with bones/skeleton ?[15:06:04] <Yaniel> can assimp do that?[15:06:30] <Match_it> I try to see[15:07:03] *** stefkos has joined ##OpenGL[15:07:10] *** newguise1234 has joined ##OpenGL[15:07:15] *** newguise1234 has left ##OpenGL[15:07:21] *** timsche has quit IRC[15:07:29] *** newguise1234 has joined ##OpenGL[15:07:37] *** timsche has joined ##OpenGL[15:09:25] <Match_it> Yaniel, seems yes[15:09:29] <Match_it> I try with this[15:10:57] <newguise1234> so, I've fixed that error, glew is compiled using my nvidia libGL.so, but I still have the vertex shader compilation issue[15:11:10] <newguise1234> pointers as to where to start approaching this?[15:14:06] <newguise1234> for reference: "0(7) : error C0000: syntax error, unexpected '(', expecting "::" at token "(""[15:17:53] *** hexagoxel has quit IRC[15:20:48] <HuntsMan> newguise1234: you should pastebin code instead of just the error[15:21:13] *** antonp has quit IRC[15:24:02] <dreamminder> it simply says there's an error in line 7 of your shader lol[15:24:15] <dreamminder> how can they help you if tey don't see the code[15:24:20] <dreamminder> paste it here www.pastebin.org[15:24:29] <dreamminder> *pastebin.com[15:24:36] <newguise1234> working on it sorry[15:25:40] <newguise1234> pastebin.com/SRehuiEc[15:28:10] <newguise1234> it works fine on my other computer[15:28:33] <newguise1234> and I added a cout statement to make sure it was actually reading the file in and it was[15:28:35] *** timsche_ has joined ##OpenGL[15:29:02] *** timsche_ has quit IRC[15:29:28] *** timsche_ has joined ##OpenGL[15:29:45] *** Jeanne-Kamikaze has joined ##OpenGL[15:30:03] *** Captain_Crow has joined ##OpenGL[15:30:15] *** timsche_ has quit IRC[15:30:17] <krnlyng> do i need to upload a GLSL uniform every iteration (in my draw function)? or is it enough if i upload it during the initialisation (if it does not change). currently i get a black screen if i only upload my projection matrix during init and i don't know why (even if my projection matrix is the identity matrix)[15:30:23] <HuntsMan> #version 120 <- first problem[15:30:26] *** timsche_ has joined ##OpenGL[15:30:51] <HuntsMan> newguise1234: which GL version are you targeting?[15:30:53] *** timsche has quit IRC[15:31:25] <Yaniel> krnlyng: uniform values are part of the shader program[15:31:54] <Yaniel> program state, that is[15:32:02] <Captain_Crow> is it possible to get shadows on a large scale without them looking blocky close up?[15:32:05] <newguise1234> HuntsMan: my laptop only supports up to 2.1 with its intel gfx-card, I'm now on a computer that has run code requiring 3.3[15:32:08] <Yaniel> once the program is linked you can set them and they will stay the same until you change them[15:32:25] <newguise1234> HuntsMan: the code compiles/runs fine on my laptop[15:33:07] <HuntsMan> newguise1234: that doesn't really say anything, drivers have bugs that allow invalid code[15:33:10] <HuntsMan> specially NVIDIA[15:33:37] *** RyanPridgeon has joined ##OpenGL[15:34:11] <krnlyng> Yaniel: hm okay... so there must be something else wrong with my code, thanks[15:37:23] *** ozette has joined ##OpenGL[15:39:23] <newguise1234> well, if I change the version to 330, it works fine[15:39:49] *** charlie5 has quit IRC[15:40:27] <Yaniel> might be because the code is actually 3.30 GLSL?[15:40:41] *** timsche_ has quit IRC[15:40:54] <Yaniel> or anything starting from 1.30[15:41:34] *** timsche has joined ##OpenGL[15:42:03] *** konom has joined ##OpenGL[15:42:31] *** Ryp has joined ##OpenGL[15:43:23] <newguise1234> Yaniel: it was working fine on my laptop with #version 120 though, might be the case with explicit locations or something though[15:44:15] <Yaniel> yeah those were extensions at first[15:46:04] <newguise1234> well thanks yall, I'll get back to trying to do stuff now[15:47:18] <krnlyng> Yaniel: do you see something wrong here: https://bpaste.net/show/c483a567ca3c ?[15:48:17] <Captain_Crow> is it possible to have shadows in large scenes without them being blocky?[15:48:42] <newguise1234> possible to have shadows in x Bigpet_[15:48:44] <newguise1234> | large scenes[15:48:49] <newguise1234> sorry :/[15:48:57] <Bigpet_> what?[15:50:39] <dreamminder> omg[15:50:49] <dreamminder> I call this thing glTexImage2D(GL_TEXTURE_2D, 0, GL_RGB, texture_width, texture_height, 0, GL_RGB, GL_UNSIGNED_BYTE, 0); and it displays junk from memory![15:50:55] <Captain_Crow> is there any information on different types of shadows in opengl?[15:51:25] <Yaniel> Captain_Crow: blur them[15:51:34] <Yaniel> or use cascaded shadow maps[15:55:22] *** ClarusCogitatio has quit IRC[15:56:17] <ClaudiusMaximus> dreamminder: sure, there's no automatic clearing or anything (seems a potential information leak, but i guess GL drivers aren't security-paranoid)[15:56:24] *** charlie5 has joined ##OpenGL[15:56:27] <dreamminder> I see[15:56:30] <dreamminder> that's interesting[15:57:15] *** ClarusCogitatio has joined ##OpenGL[16:01:28] <krnlyng> Yaniel: if i remove "* projection_matrix" it works... but even if projection_matrix is the identity i get a black screen[16:08:05] *** Gamecubic has joined ##OpenGL[16:09:33] *** Binero has joined ##OpenGL[16:09:55] <Binero> Hello[16:10:30] <Binero> I have just read about OpenGL differences between platforms, and I'm wondering, how do people still considered OpenGL platform independent?[16:10:50] <Binero> There are so many differences, especially related to multithreading between platforms[16:11:56] <Binero> Even opengl.com says it's "unsure" how some stuff should be done[16:12:08] <Binero> *.org[16:12:15] <Binero> and that's the official documentation[16:12:15] <danhedron> some concepts remain similar[16:12:19] <Binero> Yeah the functions are[16:12:29] <Binero> but how and when then functions work is platform specific[16:12:40] <danhedron> could you give a concrete example[16:12:53] <Captain_Crow> is there some way to fake raytraced quality shadows in realtime?[16:12:58] <Binero> functions and their relation to contexts[16:13:03] <Bigpet_> OpenGL is a platform independent specification[16:13:36] <Bigpet_> the spec is platform independent, things outside of the spec, like context creation and multi-threading naturally differ between implementations[16:13:39] <Binero> on windows functions are context specific, on X they aren't[16:13:47] <Binero> not just context creation[16:13:53] <Binero> but how contexts work after they are created[16:13:59] <Binero> that's platform specifi c[16:14:08] <Bigpet_> still, not within the scope of the spec[16:14:18] <Binero> That's an easy way out though[16:14:54] <Bigpet_> so what? do you think "undefined behavior" is an easy way out for C and C++ too?[16:15:00] <Binero> it almost feels like OpenGL was built completely ignoring multithreading[16:15:11] <Bigpet_> yeah, it was[16:15:12] <Yaniel> it was in the beginning[16:15:56] <Binero> I do not see how to build a safe cross platform application with this[16:16:01] <Binero> it's bound to have unsafe behaviour[16:16:17] <danhedron> not if you read the relevant specifications[16:16:43] <Binero> there is none[16:16:52] <Binero> there is vendor specifications[16:17:23] <danhedron> which are relevant[16:17:24] <Binero> but because they are so fundamentally different from each other it's either making something unsafe, either not going with cross platform[16:17:50] *** JoshSyn has joined ##OpenGL[16:19:30] <Binero> it's like the implementations are just trying to cause segfaults on their competitors[16:19:44] <Binero> which I'm sure they are[16:20:04] *** Jeanne-Kamikaze has quit IRC[16:20:28] <krnlyng> is 0 a valid uniform location?[16:20:31] <danhedron> such as?[16:20:38] <Yaniel> krnlyng: yes[16:20:54] <Yaniel> negative ones are not[16:20:55] *** Jeanne-Kamikaze has joined ##OpenGL[16:21:01] <krnlyng> okay[16:21:33] <Yaniel> (getuniformlocation returns -1 on error)[16:22:19] <Jeanne-Kamikaze> does anyone see why would glTexImage3D(GL_TEXTURE_3D, 0, GL_R8UI, 256, 256, 256, 0, GL_RED_INTEGER, GL_UNSIGNED_BYTE, data) fail when glTexImage3D(GL_TEXTURE_3D, 0, GL_RED, 256, 256, 256, 0, GL_RED, GL_UNSIGNED_BYTE, data) doesn't?[16:22:20] <krnlyng> then i really don't understand why my projection matrix produces a black screen... (i have tried hardcoding projection_matrix=identity into my vertex shader and that worked.., but why won't it work if i define it as uniform upload it using glUniformMatrix4fv[16:22:34] <krnlyng> ?)[16:22:35] <Jeanne-Kamikaze> "fail" as in the texture is all black[16:22:41] <Jeanne-Kamikaze> no gl errors[16:25:29] *** jbud has joined ##OpenGL[16:26:06] <krnlyng> i upload other stuff via glBindBuffer&glBufferData and that works, but why could it fail for my uniform mat4?[16:28:07] *** kidnapped_robot has joined ##OpenGL[16:30:08] <krnlyng> is there a way to get some debug output from glsl and gles?[16:31:53] *** Jeanne-Kamikaze has quit IRC[16:32:36] *** Jeanne-Kamikaze has joined ##OpenGL[16:33:47] *** clauswitt has quit IRC[16:34:27] *** clauswitt has joined ##OpenGL[16:34:33] <krnlyng> ah damn... i was just missing the glUseProgram call before glGetUniformLocation...[16:36:20] *** clauswitt has quit IRC[16:36:59] *** clauswitt has joined ##OpenGL[16:38:49] *** Involuntary has joined ##OpenGL[16:39:04] *** Jeanne-Kamikaze has quit IRC[16:40:41] *** clauswitt has quit IRC[16:41:09] *** Jubb has quit IRC[16:41:38] *** clauswitt has joined ##OpenGL[16:45:46] *** phr34k has joined ##OpenGL[16:46:03] *** Involuntary has quit IRC[16:47:47] *** kuldeepdhaka has joined ##OpenGL[16:50:04] *** Twinklebear has joined ##OpenGL[16:50:12] *** Jubb has joined ##OpenGL[16:52:01] *** BitPuffin has quit IRC[16:55:15] <Captain_Crow> are there any options for large scale shadows other than cascade shadows and blurring?[16:55:33] *** BitPuffin has joined ##OpenGL[16:56:29] *** kidnapped_robot has quit IRC[16:56:37] *** Zeioth has joined ##OpenGL[16:57:05] *** Majiet has quit IRC[16:58:46] *** roboman2444 has quit IRC[17:00:30] *** kidnapped_robot has joined ##OpenGL[17:01:48] *** PasNox has joined ##OpenGL[17:03:37] *** Binero has quit IRC[17:06:19] <kestasjk> what are the standard tools of the trade when writing opengl intensive code?[17:06:28] *** ibqn has joined ##OpenGL[17:07:39] <krnlyng> hi, i have an application which i would like to make aspect ratio independant. i have calculated an orthogonal matrix like this: ortho_matrix(projection_matrix, -aspect_ratio, aspect_ratio, -1.0f, 1.0f, -1.0f, 1.0f); with ortho_matrix defined like this: https://bpaste.net/show/a83cb3c84b51 now if i draw circles they are actual circles so this seems to work. but now some parts of my drawings are outside of the screen... what am i doing wrong?[17:08:03] *** roboman2444 has joined ##OpenGL[17:08:25] <Yaniel> Captain_Crow: well you can bake shadow maps into the meshes[17:08:33] <Yaniel> or do shadows in some other way[17:08:37] <derhass_> krnlyng: well, you can't habe it both ways[17:08:55] <Yaniel> kestasjk: tons of patience and your favourite text editor[17:09:15] <krnlyng> derhass_: hm why? and then, how am i supposed to do it?[17:09:16] <Yaniel> codeXL or apple's opengl profiler are useful too, if you can get your hands on them[17:09:35] <Yaniel> valve's vogl might turn out to be great as well once it matures[17:09:53] <derhass_> krnlyng: you simply must define which part of the scene you want to see[17:10:28] <Stragus> kestasjk, do you feel a need for anything specific besides a good code editor?[17:10:36] <Stragus> There are some debuggers and profilers[17:10:53] <dawik> you could scale it so that it fits within the viewport[17:11:05] <derhass_> krnlyng: and the aspect ratio will put a further constraint on this[17:12:06] <krnlyng> derhass_: so i need to figure out how i should define the vertex positions of my objects?[17:13:00] <derhass_> no, you should simplye define a projection matrix[17:13:24] <derhass_> krnlyng: think about it. you have you sceen in some arbitrary coordinates.[17:13:51] *** karab44 has joined ##OpenGL[17:13:57] <derhass_> krnlyng: you can now just define some axis-aligned rectangle representing your window/viewport[17:14:08] <derhass_> krnlyng: and you can make that as big or small as you like[17:14:48] <krnlyng> derhass_: i have a projection matrix, it's the orthographic matrix i mentioned earlier, or is that wrong? sorry i don't understand it... yet[17:15:00] *** kestasjk has quit IRC[17:15:09] <derhass_> krnlyng: well. an ortho matrix is not wrong when you want an ortho projection :)[17:15:11] *** kestasjk has joined ##OpenGL[17:15:24] <kestasjk> havent had to ghost anyone in a while ..[17:15:27] *** kestasjk_ has joined ##OpenGL[17:15:29] <derhass_> I don't know what you try to do, you must know that :)[17:15:49] <kestasjk> anyway I just want all the help I can get[17:15:58] *** Majiet has joined ##OpenGL[17:16:21] <kestasjk> if people generally use text editors fine, but if someone asked me how to write C# Id tell them get visual studio[17:18:34] <krnlyng> derhass_: i have a sceene where i am drawing octangles (replace octangles with circles if thats more convenient) but they are in egg like shape and i want them to be round. now i thought of adding an orthographic projection to include the aspect ratio so my octangles are round, but now some of the octangles i am drawing are off screen[17:19:41] <derhass_> krnlyng: what kind of projection did you use before?[17:19:48] <krnlyng> derhass_: none[17:19:53] <derhass_> meaning identity[17:19:58] <krnlyng> yes#[17:20:09] <derhass_> that would be like using ortho(-1, 1, -1, 1, 1, -1)[17:20:28] <krnlyng> yes[17:20:55] <derhass_> so with the ortho you described, you should actually see a larger area as long as the aspect >= 1, so the typical case of a window which is wider than it is tall[17:22:16] <derhass_> if that is not the case, then there is something wrong[17:24:01] <krnlyng> derhass_: hm i have aspect ratio <=1, did i calculate the aspect ratio wrongly? const float aspect_ratio = screen_widthf / screen_heightf;[17:24:49] <derhass_> well.[17:25:08] <derhass_> don't know what "screen" is in that context[17:25:54] <krnlyng> screen_width=960, screen_height=540 and screen_{width,height}f are the same just cast to floats[17:26:07] <krnlyng> ah wait[17:26:12] <derhass_> krnlyng: well, 960/480 should be > 1 in my book[17:26:56] <krnlyng> yeah i just recognized that ;) but i have a rotation hack going on here, actually screen_width=540 and screen_height=960[17:27:18] <krnlyng> (rotation hack due to an adreno driver bug)[17:28:03] <derhass_> krnlyng: well, is your screen/window/viewport wider than it is tall, or isn't it[17:28:32] *** dreamminder has quit IRC[17:28:45] <derhass_> krnlyng: if aspect < 1 is correct, you will see less than before. but you can do ortho (-1,1, -1/aspect, 1/aspect, ...) in that case[17:29:15] <derhass_> krnlyng: same as with your tv: letterboxing versus cropping if the aspect of the program does not match that of the device[17:29:28] *** telex has quit IRC[17:30:45] *** telex has joined ##OpenGL[17:31:02] <krnlyng> derhass_: basically i render on a surface with width=540 and height=960 and rotate the screen contents so the game runs in landscape mode (all due to the adreno bug)[17:33:01] <derhass_> krnlyng: well, the method I suggesteed should work, then[17:33:12] *** gaganjyot has joined ##OpenGL[17:33:20] *** karab44 has quit IRC[17:33:21] <krnlyng> derhass_: okay this makes it so there is more black on the left and right of the screen :)[17:33:31] <derhass_> krnlyng: yes.[17:33:45] <krnlyng> just confirmed it :)[17:33:50] <derhass_> krnlyng: it of course does. I already told you: you can't have it both ways[17:34:24] <krnlyng> but now if i actually want to have some off the octangles to be at the very edge of the screen?[17:35:23] *** MLM has joined ##OpenGL[17:35:32] <krnlyng> how would i do that "correctly"?[17:35:45] <derhass_> krnlyng: depends on what you are trying to achieve[17:36:17] <derhass_> krnlyng: and especially, if you design your scene for a specific aspect ratio, or if you want to support different ratios[17:36:41] *** pabb has joined ##OpenGL[17:37:12] <derhass_> krnlyng: you simply cannot have the same scene with the same image section visibily on different displays with different aspect ratios without introducing distortion[17:38:05] <krnlyng> derhass_: okay the octangles are actually buttons (visual ques for a touch input plugin for an emulator) i want the octangles to be exactly there on the screen where the input mechanic detects the touch input[17:38:43] <Stragus> With an orthographic matrix, it should be pretty easy to figure out where your stuff will be rendered[17:38:47] <derhass_> krnlyng: well, the OS APIs will probably provie you with window coordinates for the touch position[17:38:57] <derhass_> *provide[17:39:11] <krnlyng> SDL provides me with the coordinates yes[17:39:40] *** ozette has quit IRC[17:40:19] *** jhuntley has joined ##OpenGL[17:41:00] *** Ryp has quit IRC[17:41:35] *** charlie5 has quit IRC[17:42:03] <krnlyng> but how do i translate to opengl coordinates for the vertices?[17:42:04] *** Biliogadafr has joined ##OpenGL[17:42:13] *** karab44 has joined ##OpenGL[17:42:24] <krnlyng> with the orthographic projection such that the buttons are round[17:45:10] <derhass_> krnlyng: you can simply set tup an ortho prjection to match your window coords[17:45:44] <krnlyng> oh that sounds good, how would i do that?[17:45:45] <derhass_> krnlyng: so ortho (0, width, height, 0, -1, 1) and you can directly use pixel coords for drawing[17:46:20] <derhass_> note that this also flips y, so that (0,0) is the upper left corner, as window systems typically use that convention[17:46:53] *** tcsc has joined ##OpenGL[17:56:46] *** JoshSyn has quit IRC[17:58:13] *** charlie5 has joined ##OpenGL[17:58:21] <krnlyng> derhass_: now the buttons are off screen (to the left side of the screen, i just see one because it overlaps over the edge) is my vertices calculation off? https://bpaste.net/show/31936b952b32 (the commented out is the old calculation, k is just for the color calculation)[17:59:54] <derhass_> krnlyng: well, x+radius * cos(something) looks like it will produce values < 0 for half of the circle[18:01:36] *** Dudi has joined ##OpenGL[18:02:11] *** PasNox has quit IRC[18:02:21] <krnlyng> oh well, maybe it's related to my rotation hack, i just disabled it and the octangles are drawn, but in rotated mode which they actually shouldn't...[18:03:24] <krnlyng> they should be drawn in portrait mode[18:03:48] *** RyanPridgeon has quit IRC[18:04:04] *** Maya-sama has joined ##OpenGL[18:06:19] *** Maya-sama is now known as hackkitten[18:07:03] *** JoshSyn has joined ##OpenGL[18:11:12] *** karab44 has quit IRC[18:20:30] *** jokoon has joined ##OpenGL[18:20:44] <krnlyng> ok got it correct now for the non rotated case[18:20:44] <jokoon> is it the proper place for shader questions ?[18:20:53] *** zajfy has joined ##OpenGL[18:21:09] *** b4b has joined ##OpenGL[18:21:10] <Stragus> jokoon, as long as it's OpenGL shaders. :)[18:22:05] *** kuldeepdhaka has quit IRC[18:22:17] <jokoon> I want to change pixels of a texture depending on the surrounding pixels[18:22:25] <jokoon> to have some sort of propagation effect[18:22:39] <jokoon> just like some life form developing on a surface[18:22:59] <jokoon> I mean it's totally doable right[18:23:36] <Stragus> I don't think you can read and write to the same texture at the same time[18:23:48] <Stragus> But you could use a "double buffering" texture[18:23:53] <Stragus> (i.e. two textures)[18:24:40] <Stragus> I believe there's some barrier extension to both read and write to the same texture, but I have never used it and I have no idea how supported it is[18:25:14] <jokoon> barrier extension ?[18:25:47] <jokoon> you meaning "removing that barrier"[18:25:48] <Stragus> https://www.opengl.org/registry/specs/NV/texture_barrier.txt[18:26:36] <Stragus> I would just use two textures and swap continuously[18:26:51] <jokoon> yeah seems simpler[18:27:14] <jokoon> I guess it might be slower but I don't really care as I'm a beginner with shading[18:28:19] <jokoon> so I use 2 textures, at one stage I read from A and write to B, and at second stage I do vice versa etc[18:28:26] <Stragus> Yup[18:29:13] <jokoon> any tutorial on the most basic available function of fragment shaders ?[18:29:28] <jokoon> like how to get the pixel position, etc[18:29:47] <jokoon> I could not find anything really relevant[18:30:00] *** karswell` has joined ##OpenGL[18:30:15] <Stragus> If you want to get some "position" for each fragment, interpolate from values defined in the vertex shader[18:31:05] <Stragus> I couldn't recommend tutorials, but perhaps others can (I learned OpenGL a long time ago)[18:33:43] <jokoon> how do I pass them from the vertex shader to the fragment shader ?[18:34:24] *** JoshSyn has quit IRC[18:35:23] <Stragus> Shader tutorials would clear that up, and a lot more. Just google for OpenGL shader tutorial and you should find good stuff[18:35:25] <kestasjk> the "in" storage qualifier specifies that data is going into a shader[18:35:32] <kestasjk> and "out" specifies that data goes out[18:35:32] <jokoon> maybe what I want to do is a little rough to learn[18:35:45] <jokoon> oh ok[18:35:54] <Stragus> It's not hard, but it's a lot of information to share over IRC when good guides exist[18:36:46] <jokoon> yeah I'm not going to ask someone to explain it to me over skype or soemthing like that[18:38:07] <jokoon> I might be ready to buy a book on that[18:38:29] <kestasjk> Ive got "OpenGL Programming Guide Eight Edition"[18:38:54] <jokoon> I got the red bible bought 2 years ago I don't even know if it talks about shaders[18:38:59] <kestasjk> and also the openscenegraph book, because raw opengl programming is such a nightmare[18:39:12] <Stragus> The Red Book paper edition is excellent, the online free version is very obsolete[18:39:24] <kestasjk> this red bible is basically all about shaders, its opengl 4.3[18:39:44] <Stragus> Tsk, raw OpenGL programming is no nightmare. :) It's fairly easy and very powerful[18:40:05] <Stragus> OpenSceneGraph is terribly limited for so many things[18:40:18] <kestasjk> like what?[18:41:25] <Stragus> Some project at work has trouble with OSG due to the lack of support for occlusion conditinal rendering, CUDA interfacing, and many other features[18:41:38] <Stragus> OSG still uses display lists and immediate mode by default for rendering[18:42:29] <kestasjk> you can put occluders in OSG, and you can call opengl directly if you need to[18:42:51] *** Term3 has joined ##OpenGL[18:43:02] <Stragus> There are occlusion queries (which should never be used), but no occlusion conditional rendering as far as I know[18:43:24] <kestasjk> and you dont have to keep track of a million C calls and handles[18:43:31] <Stragus> Also, I'm very wary of any library that needs 120000 malloc() calls to render a cube[18:44:09] <Stragus> (I'm not even exagerating, count of active memory allocations was in the 6 digits)[18:45:35] <kestasjk> well we use it at work and render highly detailed ships and terrain, and when we're not getting over 60fps malloc calls arent the bottleneck[18:46:04] <Stragus> It's just a symptom of poor design to me[18:47:37] <kestasjk> registering a callback on a node so that you can update it when necessary, getting culling for free, support for multiple file formats (including FLT support which is vital for simulations), etc etc[18:47:49] <kestasjk> all seems like beautiful design to me[18:48:17] <Stragus> Well, you can keep on using OSG if you want, hopefully you won't run into the troubles we have had :)[18:48:41] <Stragus> And I'm aware of that 120000 active malloc() calls because I had to track down a memory corruption bug in OSG, it was fun.[18:48:49] <kestasjk> havent in the last 3 years but we'll let you know if you do[18:49:41] <kestasjk> I mean frankly theres no alternative when you need FLT support for free, it's OSG or vegaprime[18:49:47] <kestasjk> opengl isnt an alternative[18:53:06] <Stragus> I had to look that up... Multigen OpenFlight format? Okay, that's rather specific[18:53:21] <jokoon> I have the 7th edition[18:53:53] <kestasjk> jokoon: surprised it wouldnt cover shaders decently[18:54:50] <kestasjk> Stragus: yep very common in the simulation industry. things like level of detail, geospatial positioning are build into the format[18:55:19] <kestasjk> Stragus: and navigation lights, which need to blink at certain rates, be seen at certain angles, etc, etc[18:55:55] <jokoon> yes nothing about shaders it seems[18:56:18] <kestasjk> which opengl version does it mention?[18:56:31] <jokoon> 3 and 3.1[18:56:40] *** karswell` is now known as karswell[18:57:34] <kestasjk> jokoon: that sucks. from the back "...has been almost completely rewritten and provides definitive, comprehensive information on OpenGL and the OpenGL Shading Language/"[18:57:52] <jokoon> isn't it a huge book ?[18:57:59] <kestasjk> "For the first time, this guide completely integrates shader techniques, alongside classic, function-centric techniques"[18:58:00] <jokoon> how can it cover shaders ?[18:58:57] <kestasjk> it does cover shaders, Im just surprised the previous edition didn't since it's so prominent in this edition[18:59:17] <jokoon> how much of the book is about shaders ?[18:59:20] <jokoon> like 20% ?[19:00:27] <kestasjk> pm me your email and Ill send the contents[19:01:22] *** RyanPridgeon has joined ##OpenGL[19:02:04] <kestasjk> k sent[19:02:43] <kestasjk> interested how different it is from the last edition[19:04:17] *** HOS_cr`nge has quit IRC[19:04:21] *** hexagoxel has joined ##OpenGL[19:05:36] *** shingshang has quit IRC[19:07:24] <jokoon> im so mad[19:07:41] <jokoon> that book cost freaking $60[19:07:45] *** xish has quit IRC[19:08:18] <jokoon> I'm sure it doesn't even exist in pdf version[19:08:19] <kestasjk> when did you buy it?[19:08:29] <jokoon> 3 years ago or more[19:08:32] <jokoon> can't remember exactly[19:08:36] <Bloodust> which book[19:08:44] <jokoon> red book 7th edition[19:08:54] <kestasjk> well to be fair its a fast moving industry, shaders werent so important back then[19:09:06] <jokoon> in 2010 ?[19:09:20] <kestasjk> Ive bought 3 C# books in that period[19:09:37] <jokoon> shaders have been important for a long time now[19:09:46] <Bloodust> I have red book 7 too :D[19:09:56] <Bloodust> I bought it when it was released[19:10:04] <Bloodust> which was ages ago[19:10:33] <jokoon> there are 2 or 3 chapters about shaders in the 8th[19:10:49] <jokoon> I can't pay $60 for those 3 chapters :/[19:12:14] *** mazde has joined ##OpenGL[19:12:16] <Bloodust> you can just google the 8th book if you really want to read it[19:12:24] <kestasjk> then you really need to get your finances in order :|[19:12:42] <Bloodust> jokoon enjoy http://www.ics.uci.edu/~gopi/CS211B/opengl_programming_guide_8th_edition.pdf[19:12:45] <jokoon> I'm poor[19:12:52] <Bloodust> 4th search result for me[19:13:09] <jokoon> wow thanks[19:13:11] <derhass_> i could afford that, but I would never do spend that amount on such a topic[19:13:51] <Bloodust> id just read tutorials[19:14:20] <jokoon> I like to have well put explanations[19:14:37] <Bloodust> then arcsynthesis tutorial[19:14:40] <Bloodust> its like a book[19:15:01] <jokoon> I don't like to waste time understanding something because it might be better explained somewhere else. I'm a maniac I know[19:15:08] <Captain_Crow> can opengl use stencil shadows?[19:15:38] <derhass_> Captain_Crow: "use" is the wrong word. but yes, you can implement that technique with GL[19:17:41] *** Alina-malina has quit IRC[19:19:22] <kestasjk> what are you writing out of interest jokoon ?[19:19:45] <Captain_Crow> can stencil shadows block out specular light?[19:20:01] <Yaniel> sure[19:20:29] <Yaniel> the idea behind them is to just not process shadowed fragments at all during the lighting pass[19:20:56] <jokoon> kestasjk, I want to write a image where a virus propagates on a texture[19:21:04] <jokoon> depending on some perlin noise[19:21:32] *** Alina-malina has joined ##OpenGL[19:21:32] *** Alina-malina has joined ##OpenGL[19:21:34] <jokoon> so with a simple equation I increase the color or a pixel depending on the values of nearby pixels[19:21:45] <jokoon> I tried doing that on CPU and it's very slow[19:21:54] <jokoon> even for the 8 surrounding pixels[19:22:09] <Yaniel> you need two fbos[19:22:25] <jokoon> I know[19:22:40] <jokoon> since I need to read/write somebody explained it to me[19:22:49] <Yaniel> yep[19:22:51] <jokoon> I'm using SFML too[19:23:13] <Stragus> That could be fast on CPU with SSE2 processing 8 pixels in parallel (assuming 16 bits won't overflow)[19:23:19] <Stragus> But that isn't the channel's topic :)[19:23:47] <jokoon> well yeah but it will be faster on shaders I guess[19:23:53] <Yaniel> meh fragment shaders will still be probably close to an order of magnitude faster[19:23:56] <derhass_> well, 8 pixels in parallel is quite lame when you can also do that on a GPU[19:23:58] <jokoon> it's a good opportunity to learn shaders too[19:24:19] <Stragus> Sure sure, no doubt GPUs will crush that[19:24:21] *** Ad1 has quit IRC[19:24:55] <kestasjk> is it a conways game of life thing?[19:25:03] <jokoon> kinda yeah[19:25:12] <jokoon> except with discrete values[19:25:33] <kestasjk> sounds interesting[19:25:34] <jokoon> I guess anybody could invent that[19:26:07] <jokoon> already working with C++, but it's very slow[19:26:45] <jokoon> should I patent it ?[19:26:45] *** Biliogadafr has quit IRC[19:27:21] <jokoon> I might get some money but I'm too modest... nah I guess somebody already invented it...[19:27:22] <derhass_> jokoon: if it has rounded corners, apple already patented it[19:27:23] <kestasjk> probably not if you cant afford to buy a book[19:28:03] <jokoon> well should I call a lawyer, I might still be able to make money haha[19:28:13] <jokoon> patents are so weird, I don't understand them[19:28:26] <derhass_> well, patents aren't that expensive, acutally. except when lawyers get involved...[19:28:48] <Stragus> Patents are expensive. Even if you did invent something, the reviewer can be an idiot and think it's similar to something else, and send everything back for review[19:28:54] <derhass_> but it will greatly depend on yout local laws[19:29:05] <TheFlash> => Your lawyer get your money and you only loose some money ;)[19:29:19] *** Biliogadafr has joined ##OpenGL[19:29:26] <kestasjk> so you steal the hard work someone puts into writing a book then try and think of a way to make money off some derivative academic idea .. classy[19:29:30] <derhass_> Stragus: hehe, I had exactly that kind of situation recently[19:29:30] *** YarnFive8 has joined ##OpenGL[19:30:25] <jokoon> sometimes I really can't believe I have ideas that great[19:30:37] <jokoon> or they might already be patented anyways[19:30:58] <derhass_> jokoon: you totally overestimate the requirements for a patent[19:31:26] *** kernelpanic has joined ##OpenGL[19:31:26] <derhass_> jokoon: even people like me get those[19:31:35] <jokoon> I don't know if I'm modest or just am in a narcissistic thought of being a genius[19:32:33] <kestasjk> *cringe*[19:32:43] <Captain_Crow> can stencil shadows be casted onto the same object thats casting the shadow? like a mountain & terrain on the same mesh for example[19:33:02] <jokoon> Yes I think I just think I'm smarter that I really am[19:33:31] <derhass_> jokoon: " I really can't believe I have ideas that great" sounds like the opposite[19:34:22] <kestasjk> Captain_Crow: Id assume so[19:34:35] <jokoon> some guy said my propagation thing was interesting[19:34:58] <kestasjk> derhass_: ideas like some variant of conways game of life, of which there are dozens, and which have no practical application?[19:35:14] <jokoon> I'm unemployed, so I often want to believe there was soemthing that could get me a job[19:35:43] <derhass_> kestasjk: well. I meant that in a more general way[19:35:53] <jokoon> In a small smartphone game it could be interesting[19:36:01] <jokoon> it depends if you use it right[19:36:02] <Stragus> jokoon, produce open-source software and be noticed, don't patent weird stuff[19:37:13] <jokoon> I'd prefer getting money than getting noticed[19:37:35] <Stragus> Patenting weird stuff won't give you money, it will cost you money[19:37:43] <jokoon> I know[19:37:46] <Yaniel> open source it, get noticed, get hired to work on it more[19:37:50] <derhass_> hehe, yes[19:37:55] <jokoon> but I don't know how patents work so I'm wondering anyhow[19:37:57] <Yaniel> the ideal situation[19:37:58] *** jdolan has quit IRC[19:38:00] *** charlie5 has quit IRC[19:38:07] <Yaniel> sw patents generally don't work[19:38:09] <kestasjk> or better yet look in the job listings and find a job[19:39:05] <jokoon> why don't I strap my job helmet, slip into a job cannon[19:39:10] <jokoon> and fire off into job land[19:39:20] <jokoon> where job grow on jobbies[19:39:21] <Yaniel> heh[19:39:28] *** harha_ has joined ##OpenGL[19:39:32] *** Shogun_ has quit IRC[19:39:37] <Bloodust> :P[19:39:53] <kestasjk> where do you live? want me to do a google search for you?[19:39:59] <krnlyng> derhass_: btw, thanks for your help :) i've got it now how i want it to be[19:40:05] <jokoon> I put all my code on bitbucket, it's a huge pile of mess[19:40:25] <Yaniel> the best way to get a job IMO is if you know someone who works at someplace interesting and can +1 your application so it isn't trashed immediately[19:40:33] <jokoon> france, don't bother :p the job market and laws are messy here[19:41:07] <kestasjk> excuses[19:41:11] <Yaniel> got my coolest summer job so far that way[19:41:15] <jokoon> I should create my startup here...[19:41:47] <Stragus> If you produce good code, you'll get noticed[19:41:57] *** jdolan has joined ##OpenGL[19:42:00] <jokoon> I don't produce good code so :p[19:42:07] <Yaniel> shh[19:42:12] <jokoon> I just like the result[19:42:25] *** roboman2444 has quit IRC[19:42:30] <Stragus> I even had an interview proposal just for fixing a bug in the libbfd library (some very low-level Linux executable library)[19:43:28] <jokoon> in france you can't fire people[19:43:44] <jokoon> people are very productive but you can hardly fire them[19:43:58] <jokoon> so you don't have real quality control[19:43:59] *** Shogun has joined ##OpenGL[19:44:08] *** Term3 has left ##OpenGL[19:44:39] <jokoon> I don't want to make an excuse but the US will always be better for IT jobs[19:45:00] <jokoon> or where it's more open towards free entreprise[19:45:09] *** Sound has quit IRC[19:45:13] <kestasjk> then move[19:45:22] <Stragus> Or work remotely[19:45:22] <kestasjk> youre in the EU[19:46:09] <jokoon> french people have a very bad reputation as english talkers[19:46:11] *** roboman2444 has joined ##OpenGL[19:46:15] *** sreich has quit IRC[19:46:47] <derhass_> jokoon: I recently ordered something from amazon.fr. without speaking much french, it was quite a tough challenge[19:46:55] <jokoon> it pains me to give so many excuses I know I'm a sack of shit for not trying more things[19:47:03] <Stragus> I speak both french and english (I'm from Quebec), and I couldn't even understand their announcements in english in the airport or trains of Paris :)[19:47:37] <Stragus> For some reason, they insist on speaking english as if they were reading french[19:47:38] <jokoon> france has very high level education, but its private sector is very weird[19:48:16] <Stragus> Regulations and the mighty unions choke businesses in France[19:48:17] <kestasjk> the worst thing is how they act so indignant when you go over there and dont speak fluent french, but when they come over here they speak angry french 100 words / minute and are annoyed noone understands them[19:48:29] <jokoon> here its 7:45 PM, there are construction worker doing a lot of noise at this hour just near my apartment[19:49:19] <jokoon> yeah people are quite lefty here, it has many advantages, but it also has disdvantages.[19:49:51] <jokoon> I understand how surprised you can be hearing some french people about some stuff.[19:49:55] <Stragus> The french union culture sees businesses as evil exploiters, rather than as parterns[19:49:59] <Stragus> partners*[19:50:38] <jokoon> well in the US corporations are not that all good[19:50:56] <Stragus> Was it a Goodyear tire factory that they once burned down because they wouldn't raise the wages as much as they wanted?[19:51:00] <jokoon> but the private sector in the breathing system of any society[19:51:53] <jokoon> no it's because there was a CEO that secretely moved all the factory tools to another place without any warning[19:52:10] <jokoon> knowing employees would be angry to be all fired[19:52:16] <jokoon> it's complicated[19:52:24] <Stragus> Okay. I could be mixing up various stories[19:52:56] *** charlie5 has joined ##OpenGL[19:53:13] <jokoon> it's not black or white, but for a sector like IT that lives on innovation and where you need quality control, it's not adequate[19:53:46] <jokoon> there are some good companies, like free.fr or few others[19:53:53] <jokoon> but they're exceptions[19:54:12] <Stragus> You could come to Quebec, jokoon. We speak french, we have free enterprise in the American style, a social democracy, and healthy food[19:54:21] *** Zupo has joined ##OpenGL[19:54:36] <jokoon> it's hard getting a visa, you need $3000 on your account in order to land[19:54:54] <jokoon> they say they welcome you but it's a half truth[19:55:02] <Stragus> That's the work visa?[19:55:22] <jokoon> I mean, they want to be sure you have saving before coming to canada[19:55:35] <jokoon> they don't let you in if you don't have money[19:55:49] <jokoon> it allows immigrants to spend money in canada[19:55:51] <jokoon> I guess[19:55:53] <Stragus> As a tourist, they let in anyone from the EU[19:55:59] *** BreadProduct has joined ##OpenGL[19:56:13] <Stragus> So I assume you refer to the work visa[19:56:18] <jokoon> they call it a safety measure, so it's not really a "land of opportunity"[19:56:28] <jokoon> yes the work visa[19:56:35] <jokoon> I don't want to come as a tourist[19:56:54] <Stragus> Students also have the right to work part-time[19:57:44] <jokoon> I wonder if I really can take a one way flight and just get a job[19:57:50] <jokoon> I'm not really sure[19:57:54] <jokoon> I don't know[19:58:17] <Stragus> You'll certainly need a visa[19:58:22] <jokoon> Indeed[19:59:08] <jokoon> anyhow I don't have a lot of savings, I could not stay very long[19:59:09] <Stragus> It's easy if some business invites you, but that comes back to the "get noticed" part[19:59:25] <jokoon> maybe staying 3 month could give me enough time to find something[20:00:02] <jokoon> yeah but it's not really me in the flesh[20:00:31] <jokoon> not guarantees of getting some interview[20:00:36] <jokoon> and where[20:01:22] <jokoon> and I don't have any skill in java, c# or jquery or css3[20:01:29] <Stragus> Me neither :p[20:01:50] <Stragus> Pure C, CUDA, and assembly (okay, and some C++, relunctantly)[20:01:50] <jokoon> isn't that a huge drag?[20:02:19] <Stragus> Not really, depends on the industry, you can't aim to do any and all jobs in IT[20:02:29] <kestasjk> whats a huge drag is not getting off your ass and doing what it takes to get a job[20:02:36] <kestasjk> we're not life coaches[20:02:38] <jokoon> yeah but interviewers always want me to already have experience in java or C# or jquery for those jobs[20:02:55] *** Keniyal_ has joined ##OpenGL[20:03:07] <Stragus> Either learn them or find a different kind of interview[20:03:13] <jokoon> kestasjk, yes you're right :)[20:03:31] <dawik> I want to move to Canada too[20:04:03] <dawik> somewhere around BC, where the grass is the greenest[20:04:46] *** Sound has joined ##OpenGL[20:05:07] <jokoon> for now I just code at home, I try stuff, I'm not paid and all I get is welfare. I don't really like to learn something that doesn't seem interesting to me just to get a job, especially if I don't have any guarantee of getting hired.[20:05:12] <mazde> Could somebody please explain me what are those: http://pastebin.com/NfcuUVFb[20:05:14] <mazde> ?[20:05:15] <jokoon> I'm just comfortable now[20:05:46] <Stragus> mazde, an array of 9 floats?[20:05:47] <jokoon> the state put me back to school recently for 1 year... so it's not so bad after all[20:05:59] <Bloodust> mazde normal code[20:06:13] <mazde> Stragus: It says in a tutorial it's to draw a triangle but I don't understand what each number represents[20:06:23] <Stragus> Ah, must be 3 vertices[20:06:31] <Stragus> XYZ, XYZ, XYZ[20:06:50] *** Keniyal has quit IRC[20:07:00] <mazde> Stragus: -1.0f[20:07:05] <mazde> Stragus: What is that ? only[20:07:07] <jokoon> but the "getting off your ass" advice, yeah sorry that doesn't help me at all. I might be slow because of the country I'm in, but on the other hand, I don't see any welcoming hand[20:07:22] <kestasjk> negative 1 in floating point[20:07:39] <mazde> kestasjk: I know but what part of triangle is it ?[20:08:02] <jokoon> gotta go[20:08:03] <Bloodust> xyz coordinates[20:08:10] *** jokoon has quit IRC[20:08:14] <Bloodust> in that order[20:08:24] <mazde> Bloodust: -1.0f is X ?[20:08:45] <Bloodust> ...[20:09:11] <kestasjk> there's more than one -1, but why not try modifying the code and seeing what happens[20:09:58] *** Shogun has quit IRC[20:10:03] *** Shogun_ has joined ##OpenGL[20:10:18] <mazde> kestasjk: I'll surely do that but can you tell me please what the numbers of the first line are ?[20:10:47] <Bloodust> we just did[20:10:51] <Bloodust> twice already[20:11:50] <mazde> Bloodust: Ah xyz coordinates, my window is small, sorry[20:17:34] *** kestasjk has quit IRC[20:17:41] *** harha_ has quit IRC[20:18:44] *** Dudi has quit IRC[20:21:40] *** Dudi has joined ##OpenGL[20:21:40] *** Sound has quit IRC[20:22:10] *** cden has joined ##OpenGL[20:24:15] *** Dudi has quit IRC[20:24:46] *** DMJC-M has joined ##OpenGL[20:26:53] *** hexagoxel has quit IRC[20:28:06] *** clauswitt has quit IRC[20:28:28] *** harha_ has joined ##OpenGL[20:29:27] *** clauswitt has joined ##OpenGL[20:30:22] *** Yaniel has quit IRC[20:31:01] *** Yaniel has joined ##OpenGL[20:31:06] *** phr34k has quit IRC[20:31:50] *** bkre__ has joined ##OpenGL[20:32:01] *** pabb has quit IRC[20:39:17] *** zoraj has quit IRC[20:39:58] *** katlogic has quit IRC[20:42:44] *** charlie5 has quit IRC[20:44:43] *** katlogic has joined ##OpenGL[20:46:01] *** theripper93 has joined ##OpenGL[20:46:02] <Captain_Crow> what happens if theres another light in a stencil shadow? would the light be visible?[20:47:30] <Captain_Crow> can there be multiple lights and stencil shadows without things looking glitchy?[20:49:23] <Yaniel> afaik each light is rendered separately with stencil shadows[20:50:35] *** Waynes has left ##OpenGL[20:51:07] <Captain_Crow> is that good or bad?[20:52:14] <Yaniel> both I guess[20:52:26] <Yaniel> the shadows don't interfere with each other[20:52:34] *** Slion has joined ##OpenGL[20:52:45] <Yaniel> but you have to re-generate the stencil mask for each light every time[20:53:13] *** harha_ has quit IRC[20:53:37] <Stragus> Stencil shadows have generally been abandoned these days[20:54:09] <Stragus> Computing proper silhouettes for complex meshes is expensive. You can't have shadow anti-aliasing. It's very expensive to have multiple lights. Soft shadows are way out of the question.[20:57:41] *** charlie5 has joined ##OpenGL[21:04:39] *** Slion has quit IRC[21:04:56] *** samrat has quit IRC[21:05:38] <Captain_Crow> is there a way to get raytraced quality shadows on a large scale without baking it to a mesh?[21:07:29] *** razieliyo has quit IRC[21:08:09] <Yaniel> no[21:08:24] <Yaniel> well, you can always raytrace for real[21:09:02] *** prophile has joined ##OpenGL[21:09:40] *** gaganjyot has quit IRC[21:11:19] <Captain_Crow> is there a way to make shadows get softer based on the distance of the object from the surface its casting on?[21:11:35] *** charlie5 has quit IRC[21:12:21] *** CainJacobi has joined ##OpenGL[21:12:21] <Captain_Crow> something like this http://www.cs.utexas.edu/~fussell/courses/cs384g/projects/raytracing/ray_examples/soft_shadows.jpg ?[21:12:46] <Yaniel> no reallt good way[21:14:01] <Captain_Crow> i know its possible using several lights, but i was wondering if theres a shader that could do that more efficiently[21:14:54] <Yaniel> variance shadow maps try to do that[21:17:06] *** samrat has joined ##OpenGL[21:17:28] *** konom has quit IRC[21:19:09] *** Leates is now known as Guest54225[21:24:24] *** leates has joined ##OpenGL[21:25:35] *** DrBenway has quit IRC[21:25:56] *** charlie5 has joined ##OpenGL[21:26:16] *** Vge has joined ##OpenGL[21:28:49] *** DrBenway has joined ##OpenGL[21:31:11] *** ScruffyFurn has joined ##OpenGL[21:34:04] *** linuxuz3r has joined ##OpenGL[21:37:49] *** leates is now known as Guest11112[21:40:23] *** theripper93 has quit IRC[21:42:05] *** Leates has joined ##OpenGL[21:43:17] *** Cheery has joined ##OpenGL[21:44:03] <Cheery> going to create some opengl bindings.[21:44:16] <Cheery> downloading the specification files already[21:44:39] <Cheery> I wonder what's the right way to call it all.[21:44:59] <Cheery> unique functions via getprocaddress?[21:45:05] <Cheery> and remaining normally?[21:45:31] <derhass_> what are "unique functions"?[21:46:17] <Cheery> things like ARB, EXT, NV[21:46:57] <Yaniel> those are extensions[21:47:12] <derhass_> and the distinction is quite useless[21:47:33] <derhass_> you'll need to query function pointers for core functions, too, on many platforms[21:47:58] *** linuxuz3r has quit IRC[21:48:58] <Cheery> ok.[21:51:19] *** konom has joined ##OpenGL[21:53:11] *** hexagoxel has joined ##OpenGL[21:53:49] *** bkre_ has joined ##OpenGL[21:54:29] *** Khlorghaal has joined ##OpenGL[21:54:39] *** cden has quit IRC[21:55:17] *** bkre__ has quit IRC[21:55:36] *** DrBenway has quit IRC[21:56:32] *** charlie5 has quit IRC[21:56:33] *** ibqn has quit IRC[21:57:23] *** DrBenway has joined ##OpenGL[21:57:59] *** Beetny has joined ##OpenGL[22:02:03] *** groton has joined ##OpenGL[22:07:44] *** Waynes1 has joined ##OpenGL[22:09:38] *** samrat has quit IRC[22:10:53] *** kernelpanic has quit IRC[22:11:15] *** Zeioth has quit IRC[22:11:33] *** Zeioth has joined ##OpenGL[22:12:07] *** charlie5 has joined ##OpenGL[22:13:33] *** TyrfingMjolnir has joined ##OpenGL[22:13:34] *** xissburg has joined ##OpenGL[22:17:12] *** xish has joined ##OpenGL[22:18:02] *** icedp has quit IRC[22:20:06] *** icedp has joined ##OpenGL[22:25:24] *** prophile has quit IRC[22:26:05] *** prophile has joined ##OpenGL[22:31:20] *** flan3002 has joined ##OpenGL[22:34:05] <flan3002> Hey everyone! I'm trying to... display a white (very ambitious, I know...) window with SDL2, GLES2 and Rust on ArchLinux. Assuming my context is setup correctly, what other calls than glClearColor and glClear would I need? If it's not setup correctly, how do I test that? Source: https://gist.github.com/17a13dd78e9dce19f315[22:35:05] <slime> you need to use SDL_GL_CreateContext after creating the window[22:40:41] <flan3002> slime: I've done that. Changed code: https://gist.github.com/160e3d92825e181916f3 Yet, nothing is drawn. The window seems transparent, until something changes.[22:41:16] <flan3002> slime: Note that it doesn't panic, so it should be safe to assume the context was created successfully. But... I'm not using that context. :/[22:41:48] <flan3002> I guess I can't expect something to change if I don't use it. Let me try something...[22:42:25] *** katlogic has quit IRC[22:42:33] *** Dark_Confidant has quit IRC[22:42:33] *** Dark_Confidant|m has joined ##OpenGL[22:47:04] *** katlogic has joined ##OpenGL[22:47:37] *** theripper93 has joined ##OpenGL[22:49:24] *** Textmode has joined ##OpenGL[22:50:16] *** Demon_Fox has joined ##OpenGL[22:51:44] <flan3002> :'([22:52:27] *** meoblast001 has joined ##OpenGL[22:54:44] <Stragus> flan3002, what language is that?[22:55:12] <flan3002> Stragus: That's Rust. rust-lang.org[22:55:40] <Stragus> I see, I don't think we'll able to help much with how to setup SDL using wrappers for exotic languages[22:55:42] <flan3002> Stragus: That snippet admittedly looks pretty alien. :)[22:56:08] <Stragus> But I assume whoever wrote the wrapper must have some kind of working demo/test[22:56:46] <XMPPwocky> flan3002: do you swap buffers?[22:56:59] <flan3002> Stragus: I'm trying to rule out failure-points. Right now, there are still too many. The demos work, but they use the SDL renderer, which seems too limited for what I want to do. :/[22:57:22] <flan3002> XMPPwocky: I was hoping that window.gl_swap_window(); does that.[22:57:40] <XMPPwocky> yeah[22:58:07] <flan3002> XMPPwocky: I mean... it does that. It calls the correct function, I validated that.[22:58:16] <XMPPwocky> flan3002: what does .gl_create_context actually give you[22:58:44] <flan3002> XMPPwocky: In the end, it's a pointer. I don't know what to do with that pointer, though.[22:59:01] <XMPPwocky> okay, it seems you don't need to do anything there[22:59:20] <XMPPwocky> flan3002: turn off culling and depth testing, first, just to rule that out[22:59:52] <XMPPwocky> oh, and print something out from inside your draw loop[22:59:52] <flan3002> Did that. Nope... :/[22:59:58] <XMPPwocky> so you see it's running[23:00:21] <flan3002> Prints. A lot.[23:00:39] <XMPPwocky> try removing the call to glViewport?[23:01:08] <flan3002> Done, nothing.[23:01:25] <Stragus> Are you sure it's really calling an OpenGL context somewhere?[23:01:59] <Stragus> Do you get anything from glGetString(GL_VENDOR) ?[23:02:05] *** Dark_Confidant|m has quit IRC[23:02:14] <XMPPwocky> flan3002: #rust-gamedev on irc.mozilla may be more helpful fwiw[23:02:14] <Stragus> ... really creating* an OpenGL context ...[23:02:17] *** Dark_Confidant has joined ##OpenGL[23:02:27] <flan3002> That's exactly what I was looking for. Some non-graphical way to see if /something/ works. I'll try that.[23:03:49] <flan3002> XMPPwocky: I already asked there. :/[23:04:20] <flan3002> Stragus: Well... that's a null pointer. Now we're getting somewhere...[23:04:21] <XMPPwocky> flan3002: so yeah, check glGetString, and also see what glGetError gives[23:04:31] <Stragus> Null pointer, no OpenGL context[23:05:00] <XMPPwocky> flan3002: look into SDL_GL_MakeCurrent (in C)- it should be exposed from sdl2-rs somewhere[23:05:02] <flan3002> glGetError returns 0.[23:05:26] <XMPPwocky> you can try that w/ the thing you get from .gl_create_context[23:05:31] <Stragus> Make sure you are creating a SDL window and not some... other kind of window[23:08:00] <XMPPwocky> flan3002: oh, maybe you need to tell it it's GLES 2.0[23:08:04] <flan3002> XMPPwocky: You're awesome! Yay! A white window. Thanks everyone! ;)[23:08:08] <XMPPwocky> https://wiki.libsdl.org/SDL_GL_SetAttribute[23:08:12] <XMPPwocky> oh, nvm then :P[23:08:17] <flan3002> It was the make_current thingy.[23:08:25] <Stragus> Cool.[23:09:27] <flan3002> Working code in case this channel is logged: https://gist.github.com/b2b3ff8e195ef57c91bb ;)[23:09:57] *** ivan\ has quit IRC[23:10:10] <XMPPwocky> flan3002: if you want you could be nice and submit a documentation issue on sdl2-rs for that[23:11:56] <flan3002> XMPPwocky: I might. I'll finish this example and put it somewhere on Github, then maybe put a link to it somewhere in the issue. Right now, none of the examples use OpenGL directly.[23:13:05] <Captain_Crow> is it possible to use variance shadows with cascade shadow mapping?[23:14:52] *** theripper93 has quit IRC[23:15:33] *** b4b has quit IRC[23:19:19] *** ScruffyFurn has quit IRC[23:28:01] *** groton has quit IRC[23:38:53] *** det has quit IRC[23:39:02] *** det has joined ##OpenGL[23:41:42] *** Kane has quit IRC[23:42:33] *** void256 has joined ##OpenGL[23:50:12] *** Plasmastar has quit IRC[23:52:16] *** Plasmastar has joined ##OpenGL[23:56:00] *** hdon has quit IRC[23:56:57] *** jdolan has quit IRC[23:59:35] <void256> is glBlendFunc(GL_SRC_ALPHA, GL_ONE_MINUS_SRC_ALPHA) actually one of the potter & duff composition operations? I can't find it in their table o_O and if not, what would it be called?