April 30, 2008 
< | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | >
April 30, 2008 [00:00:05] <magnet_> DMINATOR: oh tv_interval so. let's go.[00:01:35] <DMINATOR> $t0 = [gettimeofday];[00:01:35] <DMINATOR> # do bunch of stuff here[00:01:35] <DMINATOR> $t1 = [gettimeofday];[00:01:49] <DMINATOR> maybe you can find a difference from this somehow[00:02:08] <DMINATOR> # do bunch of stuff here - here you draw[00:02:13] *** fmrfirebird has quit IRC[00:02:30] <DMINATOR> oh yes thats the thing interval i see[00:03:04] <DMINATOR> i hope it can get accurate difference in milliseconds alteast[00:07:29] *** bbeausej has joined ##OpenGL[00:07:32] *** charlie55 has joined ##OpenGL[00:10:17] *** aalex has quit IRC[00:10:26] <DMINATOR> tv_interval - Returns the floating seconds between the two times.[00:10:52] <DMINATOR> So i guess that would be 1 = 1 second , 1.5 = 1 and a half second[00:12:16] *** Burga has quit IRC[00:13:43] <DMINATOR> so you can calculate FPS = 1.0 / ( interval value in seconds )[00:14:15] <magnet_> I still have a float like 0.000125 for elapsed time[00:14:32] <DMINATOR> that gives 800 fps :)[00:14:42] <magnet_> hehe[00:14:44] <DMINATOR> can you put Delay or Sleep for 1 second there[00:14:56] <DMINATOR> so we can see if it calculates accuratelly[00:15:02] <magnet_> umm yes.[00:15:23] <DMINATOR> if it would give something like 1.0001 then it is probably accurate[00:15:36] <magnet_> 1.00043[00:15:40] <DMINATOR> great[00:16:07] <DMINATOR> so now it calculates properly[00:16:31] <DMINATOR> 1 / (1.00043) = 0.99 FPS[00:16:54] <magnet_> I commented to call to DrawScene to do this.[00:17:03] *** groton has quit IRC[00:17:14] *** rodietze has quit IRC[00:17:18] <DMINATOR> ok[00:17:41] <DMINATOR> looks like it works right now.[00:18:23] *** gpaci_ has joined ##OpenGL[00:18:47] *** gpaci has quit IRC[00:18:47] *** amz has joined ##opengl[00:19:31] <DMINATOR> this calculation however only calculates rendering time you need to change it later when you create programs that deal with input audio or other stuff and calculate total time for each part separatelly[00:19:51] <DMINATOR> but it should work fine for now[00:19:52] <magnet_> umm I feel stupid but I don't get the fps calculation part.[00:20:05] <DMINATOR> what do you mean ?[00:20:14] <DMINATOR> what exactly you don't get ?[00:20:23] <magnet_> I can't divide 1000 byt 0.000125 , and get a good result.[00:20:40] <DMINATOR> no 1000 was for milliseconds ![00:20:47] <DMINATOR> the interval returns seconds ![00:20:48] <magnet_> those are micro ![00:20:55] <magnet_> oh seconds.[00:21:02] <DMINATOR> 1 second = 1000 milliseconds[00:21:10] <magnet_> lol I'm really tired.[00:21:16] <DMINATOR> so you need 1 divide by your interval[00:21:43] <DMINATOR> ok don't worry about it you would figure it out eventually[00:22:44] <magnet_> umm it show 9500 fps now, doing 1/tv_interval[00:23:14] <DMINATOR> ok if you commented the Draw function thats probably accurate[00:23:21] <DMINATOR> uncomment it back[00:23:29] <DMINATOR> the fps should drop by a margin[00:23:48] <magnet_> yes I got 0.99 fps.[00:24:11] <DMINATOR> hmm hold on have you removed the delay ?[00:24:48] <magnet_> to have 0.99 fps I have the 1 sec delay. if I remove it I got like 10 000 fps.[00:24:55] <magnet_> sound weird.[00:25:17] <DMINATOR> well looks ok to me[00:25:26] <magnet_> really ?[00:25:29] <DMINATOR> if you aren't drawing anything thats probably accurate also[00:25:47] <magnet_> I am drawing a quad with a texture.[00:26:00] <DMINATOR> can you put a code again ?[00:26:08] <magnet_> sure.[00:26:19] *** charlie555 has quit IRC[00:26:20] <magnet_> what do you need ?[00:26:21] *** charlie555 has joined ##OpenGL[00:27:54] <DMINATOR> well put both examples ?[00:28:15] <magnet_> oki.[00:30:26] *** Yustme has quit IRC[00:30:52] <magnet_> DMINATOR: http://pastebin.ca/1002301, I just updated the pastebin.[00:31:02] <magnet_> the only function that changed is refresh_screen.[00:31:21] <magnet_> I putted the sleep(1) in comment, exept this, the 2 case are identical.[00:31:22] *** LuchoVtn3d has joined ##OpenGL[00:31:32] <DMINATOR> ok i see[00:31:59] <DMINATOR> it looks ok to me I think it works[00:32:10] <DMINATOR> put 0.2 for sleep[00:32:15] <DMINATOR> and see how much it would get[00:32:19] <magnet_> oki.[00:32:26] <DMINATOR> that should give about 5 FPS[00:33:28] *** charlie55 has quit IRC[00:33:48] <magnet_> doh[00:33:54] <magnet_> it give 90 000 .[00:34:03] <DMINATOR> :)[00:34:37] <magnet_> ah good[00:34:46] <magnet_> I corrected it.some perl problem.[00:34:54] <magnet_> I got 4.99 fps.[00:34:58] <DMINATOR> great[00:35:04] *** Lemml has quit IRC[00:35:07] <DMINATOR> thats the proof that it works perfectly[00:35:23] <DMINATOR> and 10 000 fps is pretty good for beginning :)[00:35:38] <magnet_> ok, drawglscene, now got 8771 fps with 1 image drawing on 1 quad.[00:35:47] <magnet_> that sound a lot.[00:35:54] <DMINATOR> for one quad thats probably normal[00:36:04] <magnet_> yea.[00:36:07] <DMINATOR> if you add few more it would drop pretty quickly[00:36:19] <magnet_> I think I can do that ;)[00:36:28] <DMINATOR> the usual fps would be somewhere 100-200[00:36:38] <DMINATOR> for few thousands[00:37:13] *** bbeausej has quit IRC[00:37:40] <magnet_> umpf[00:37:54] <magnet_> with 2 images I'm back to 3 fps.[00:37:55] <magnet_> lol.[00:38:13] <DMINATOR> really are you sure you removed the delay ?[00:38:43] <magnet_> yea :\[00:38:50] <DMINATOR> i can't believe 1 quad = 8771 fps , 2 quads = 3 fps[00:38:53] <DMINATOR> something is not right[00:39:09] <magnet_> yes.[00:39:19] *** rnx has left ##opengl[00:39:30] *** charlie55 has joined ##OpenGL[00:39:33] <DMINATOR> comment out the quad you have added and see if it changes[00:40:12] <magnet_> yea.[00:40:17] <magnet_> it's back to 8k[00:40:28] <DMINATOR> unbelievable :)[00:40:35] <DMINATOR> can you put the code again ?[00:40:45] <magnet_> sure[00:42:59] <magnet_> http://pastebin.ca/1002314[00:43:13] <magnet_> that the updated code. tell me if you miss something.[00:43:46] <DMINATOR> thats one quad only ?[00:44:39] <magnet_> yes, there is a comment for the 2 quad.[00:44:50] <magnet_> oh no it's 2 quad[00:44:53] <magnet_> there is no comment :)[00:45:00] <magnet_> #[00:45:00] <magnet_> image_draw($heap->{"images"}->{"2_bgtile"},0,0,640,480);[00:45:01] <magnet_> # image_draw($heap->{"images"}->{"magnet-6-2"},100,100,128,128);[00:45:17] <DMINATOR> oh i see that[00:45:34] <DMINATOR> can you put the same quad just twice[00:45:44] <magnet_> yes[00:45:45] <DMINATOR> with same parameters[00:45:51] <magnet_> same position too ?[00:45:52] <DMINATOR> yes[00:46:31] <magnet_> 3 fps still[00:46:32] <DMINATOR> and see what has changed[00:47:08] <DMINATOR> hmm[00:47:09] *** HuntsMan has joined ##opengl[00:48:34] <DMINATOR> update the code again once more.[00:50:04] <magnet_> oki.[00:50:54] <magnet_> http://pastebin.ca/1002328[00:51:37] <DMINATOR> ok can you post again the tick counts before and after ?[00:52:05] <magnet_> yep[00:54:50] *** charlie555 has quit IRC[00:56:15] <magnet_> tick1 : 1209509739,104707 tick2 1209509739, 104836 . another one : tick1 1209509739,519764 tick2 1209509739,519882 .[00:57:03] <magnet_> those are ticks get with gettimeofday .[00:59:00] <DMINATOR> hmm but that gives in seconds 0,0001 roughly 1000 fps[01:00:07] <magnet_> umm.[01:00:40] <DMINATOR> if you calculate the difference 0,000128984451 seconds[01:01:38] <DMINATOR> try to calculate manually fps from this maybe[01:01:52] <magnet_> okay[01:01:52] <DMINATOR> 1 / ( tick2 - tick1 )[01:03:41] *** charlie55 has quit IRC[01:08:06] *** trapt has joined ##OpenGL[01:09:34] <Madsy> http://www.marketwatch.com/quotes/phei[01:09:35] <Madsy> lol[01:10:11] *** ajww has joined ##OpenGL[01:11:13] <DMINATOR> how is it possible ?[01:13:58] <DMINATOR> magnet_ anyway I have to go sleep now. Just try and see what kind of values you would get. Test with delay(1.0) and delay(0.2) and see if it would give the same results as we had before.[01:14:16] <magnet_> DMINATOR: oki, thanks for your help ![01:14:41] <DMINATOR> i think that problem was in fps calculation. Continue with your program and try to animate your quads so you would see them in motion[01:14:52] <magnet_> I did[01:14:54] <DMINATOR> that would get an idea how smooth it actually is[01:15:00] <magnet_> but it was slow :\[01:15:19] <DMINATOR> maybe its because of this ? $kernel->delay( "refresh_screen", 0.05 );[01:16:12] <DMINATOR> anyway I am going to sleep now let me know if you have still some problems.[01:16:27] *** DMINATOR has quit IRC[01:20:37] *** m4ggus has quit IRC[01:20:42] *** m4ggus_ has joined ##OpenGL[01:22:09] *** Suprano has quit IRC[01:26:25] *** mm765 is now known as mm765^away[01:26:39] *** ajww has quit IRC[01:29:52] *** bbeausej has joined ##OpenGL[01:30:04] *** _linuxftw has left ##OpenGL[01:30:05] *** reviver has quit IRC[01:31:24] *** LuchoVtn3d has quit IRC[01:31:31] *** juanmabc has quit IRC[01:37:19] *** ajww has joined ##OpenGL[01:37:28] *** Walt has quit IRC[01:40:46] *** dvoid_ has joined ##OpenGL[01:45:59] *** Walt has joined ##opengl[01:47:01] *** jnd has joined ##OpenGL[01:48:17] *** yno has quit IRC[01:50:36] *** replor has joined ##OpenGL[01:52:30] *** darka has quit IRC[01:52:44] *** LtJax has quit IRC[01:54:10] *** Ademan has quit IRC[01:54:49] *** charlie5 has joined ##OpenGL[01:55:16] *** Ademan has joined ##OpenGL[01:59:32] *** elif has joined ##opengl[02:02:11] <elif> Hi, anyone could point me a public repository (or a set of them) of opengl programs that ilustrates use of diverse mecanisms of specifying ligthing, camera, shading, etc... I google up several sites with some listinings but do not found anyone good to have as reference. Thks any help.[02:02:54] <speedy1> keywords: NeHe, Lighthouse3d, open-source engines supporting opengl[02:02:59] <elif> either opengl.org has good listing because most of them are really broken links.[02:03:04] <elif> *neither[02:03:36] <elif> thks speedy1[02:04:02] <speedy1> welcome :)[02:04:17] *** slide has joined ##OpenGL[02:05:36] *** LordMetroid has quit IRC[02:10:56] *** BahamutZERO has quit IRC[02:17:51] <Rangar> elif: gamedev.net is also a vast repository of ifo, as well as gamasutra.com ( although most articles there are for experts )[02:18:37] *** gpaci_ has quit IRC[02:22:04] *** amalon has quit IRC[02:22:52] <elif> Rangar: hum, actually, I'm in a begginner level, I'm taking my first classes on computer graphics and besides my knowledge in linear algebra and some nice concepts of geometry (topology, etc..) I'm have difficult time to get on code. And unfortunatelly the classes are not oriented to code pratice... :(. So I'm trying to collect a vast set of simple programs that deal with basics of building up a specific problem of cg.[02:23:34] <Lucine> thanks, speedy1[02:23:53] <Rangar> elif: gamedev.net and nehe.gamedev.net are great places for you to start[02:24:51] <Rangar> elif: gamedev.net is broken in to sections, math, opengl, direct3d etc.. and each section subsectioned like caeras, matrices, quaternion operations etc[02:25:01] <Rangar> valuable resource inded[02:25:17] <elif> thks Rangar.[02:25:39] <Rangar> no probs[02:26:12] <Rangar> Your biggest problem is going to be what API to use for all this stuff, and on what platform[02:26:36] <Rangar> which is really a personal choice ( o course prerequisite ) issue[02:28:47] *** trapt_ has joined ##OpenGL[02:31:49] <elif> Rangar: My plattaform of choice is gnu/linux, so I'm using glut (what I think is plataform indepent) for windows issuing. And my project is to first create a simple scene where I can add up the resources slowly (lighting, shading, texture, etc..). So I have a really big problem in organizing the code to separate the concerns (define a object, manipulate a camera, set light sourc, etc...).[02:32:28] <elif> and then learn how api operates to create all that...[02:33:06] <Rangar> all that will come as you go through the tutorials :)[02:33:09] <Ragnarok> I use Qt for Opengl[02:33:16] <Ragnarok> i didn't like glut[02:33:17] <Rangar> I use wxWidgets[02:33:20] <Ragnarok> ewww[02:33:39] <Rangar> pfft][02:33:54] <Rangar> i'm not trapped with crippled licensing and it has everything I need :P[02:34:14] <Ragnarok> meh[02:34:35] <Ragnarok> yeah its buggy, and if you like the MFC crap that comes with using it[02:35:26] <Rangar> You do NOT need to use the MFC styles, not sure why people tink you do[02:36:32] <Rangar> i.e You dont need to define events... each class can just have one func called "ProcessEvent()" and every event is then sent to it and you can switch, case between them[02:36:35] <Ragnarok> i know u don't but if you do it the other way, your just using it like Qt[02:36:52] <Lucine> Except as he pointed out, the crippled licensing[02:36:57] <Rangar> i dunno, I dont program with Qt, the license put me off[02:36:59] <Lucine> is what he had problem with[02:37:04] <Lucine> not the Qt programming part[02:37:17] * Lucine nods[02:37:19] <Ragnarok> Rangar, what u like keeping ur source closed?[02:37:22] <Lucine> Kinda the same, Rangar.[02:38:23] *** ajww has quit IRC[02:38:29] <Rangar> Ragnarok: I didn't say that?.. but I may like to make a LIVING off it without paying huge ammounts of $ over to someone else I can ill afford to put out[02:38:35] <Lucine> It doesn't even necessarily have to be 'liking to keep the source closed' (which have it's argument for some things), when it could be more like "Doesn't like being forced, limited, and at the mercy of a licensing system."[02:38:38] <Ragnarok> and if u really want to make a product cross-platform u have to release ur source pretty much on linux[02:38:52] <Lucine> Not really[02:38:57] <Rangar> oh rubbish[02:39:22] * Rangar doesnt bother going any further in this religous argument[02:39:29] *** lolage0 has quit IRC[02:39:29] <Ragnarok> Rangar, thats what I used to think too anbd was the same way, but good luck building the next vidusal studio by urself[02:39:51] * Rangar blinks[02:40:25] * Rangar fails to see where building visual studio came in to this??[02:40:36] <Ragnarok> its called humour[02:40:44] <Lucine> you can release pre-compiled binaries in linux, Ragnarok.[02:40:55] <Rangar> indeed[02:41:06] <Rangar> thats why commercial companies actually do[02:41:11] <Ragnarok> Lucine, i know, but good luck finding alot of people that will use precompiled binaries[02:41:20] <Rangar> you just need to check what you are compiling against lib wise[02:41:34] <Ragnarok> im a developer i know that[02:41:41] <Rangar> hmm, odd.. I know lots of people using apps like, hmm google earth[02:41:50] <Lucine> or ATI proprietary driver :?[02:41:56] <Lucine> or some other things :/[02:42:05] <Ragnarok> Lucine, yeah for 3D[02:42:10] <Ragnarok> thats the only reason[02:42:12] *** speedy1 has quit IRC[02:42:20] * Lucine even seen peoples playing windows games on wine/cedega...[02:42:43] <Lucine> yeah, if they just want to use something, they'll use pre-compiled binaries[02:42:47] *** BahamutZERO has joined ##OpenGL[02:42:58] <Lucine> not all of them are developers and coders.[02:43:00] <Ragnarok> Lucine, yeah but wine isn't that good[02:43:07] <Rangar> Ragnarok: see, the difference here though is... "but good luck finding alot of people that will use precompiled binaries".. That depends on your target audience.. companies dont get in to this religous war like the GPL fan club do[02:43:08] <Lucine> nope, but people still use it[02:43:18] <Ragnarok> cedega i wouldn't pay when i can just restart and load vista[02:43:20] <Rangar> so if you're targeting commercial entities, it matters naught[02:44:14] <Ragnarok> Rangar, and if you do go commerical good luck against crackers :P[02:44:25] <Lucine> Vista is a few hundred dollars. :/[02:44:35] <Ragnarok> Lucine, but 90% hack it ;)[02:44:42] <Ragnarok> and don't pay[02:44:46] <Lucine> yeah, I'm not really counting those :p[02:44:51] <Ragnarok> xD[02:45:11] <Ragnarok> and then you got crackers that will discompile ur app ;)[02:45:13] <Lucine> I find it a shame that Vista 64 require licensing :/[02:45:32] <Rangar> Ragnarok: hackers are an issue no matter what platform you support[02:45:35] <Ragnarok> and plous imho wxwidgets is ugly ;)[02:45:57] <Lucine> I didn't like wxwidgets much, but it's one of the best one with very nice licensing[02:46:01] <Ragnarok> Rangar, and keygener's so have fun with that too ;)[02:46:42] *** trapt has quit IRC[02:47:03] <Lucine> Some people who have money will buy the thing rather than finding cracks for it just to avoid viruses and stuff, though.[02:47:25] <Rangar> I'm not sure where this argument is going.. first it was wxwidgets vrs qt, then it's licensing, thn it's commercial or not, hehe[02:47:28] <Rangar> alot of ground in 5 mins[02:47:30] <Lucine> :p[02:47:52] <Lucine> But yeah, I want to develop in QT, but I don't like being locked into a crippled license :/[02:47:55] <Ragnarok> imma be honest a couple of yeatrs back i used to do that shit, and i had a good paying job, but i still did it[02:47:59] * Lucine nods[02:48:26] <Rangar> Rangar: see, I have a SPECIFIC target audience, and these companies demand support etc.. everything can be chargeable[02:48:38] <Ragnarok> why not just use win32[02:48:42] <Ragnarok> or MFC ;)[02:48:54] <Rangar> I used MFC like 6 years ago[02:49:02] <Rangar> but alot of my target audience want to go to Linux[02:49:38] <Ragnarok> go use gtkmm then, wxwidgets is too ugly :( lol and slow[02:49:58] <Ragnarok> its documentation and finding tutoriakls is crappy[02:50:24] <Ragnarok> its forums are shitty, people hardly ever help you[02:50:38] <Ragnarok> i used to use wxwidgets with directx and no one would bother helping me[02:50:39] * Rangar blinks[02:51:04] <Rangar> wxWidgets people are very helpfull dude.. even the #wxwidgets channel has been great in the past[02:51:19] <Rangar> and the source comes with a multitude of sample code to look trhough for pretty much every class[02:51:30] <Rangar> and the documentation is easy to read and implement[02:51:31] <Ragnarok> and btw u can use other license besides gpl and lgpl[02:51:40] <Ragnarok> with qt[02:51:43] <Ragnarok> like bsd etc[02:51:46] <Rangar> sure.. IF you pay for it[02:51:55] <Ragnarok> nope[02:52:02] *** charlie55 has joined ##OpenGL[02:52:04] <Rangar> yes, but on a 'commercial' basis here we are talking about[02:52:32] <Ragnarok> okay are you going to pay $3000 upfront just to barely make $500[02:52:43] <Rangar> The type of software I write sells for about US $5000 per piece[02:53:09] <Ragnarok> the reason they charge is so you can help the open source community and crippe the commerical world[02:53:17] <Rangar> see, it's this free software dream world thats caused all the major US firms to outsorce the work to India and China[02:54:05] <Ragnarok> iono do whatever you want, but when you fail trying to do commerical by urself, i'd be here to say i told you so[02:54:20] * Rangar rolls his eyes[02:54:42] <Rangar> I fail to see where i ever said I wad doing anything by myself[02:54:52] *** stevosICBM has joined ##OpenGL[02:54:59] <Rangar> man you jump to huge conclusions dude[02:59:08] <Lucine> to be honest, I'm just going to write windows programs for now. SWF/WPF and ignore wx, etc... :?[02:59:31] <Lucine> but I've not made much gui stuffs, so I'm still relatively new there.[02:59:33] *** gpaci has joined ##OpenGL[02:59:45] <Ragnarok> oh btw if you do commerical better not post ur secet code in pastebin's ;) they might use it ;)[03:00:48] <Lucine> I don't think a commercial system would really be so much hurt by letting people read one or two page worth of code.[03:01:12] <Ragnarok> no but if they crack it they will ;)[03:01:37] <Ragnarok> and if the product is good you'll be cracked shortly after release[03:01:38] <Lucine> there's that possibility, true.[03:01:46] <Lucine> that'll happen no matter what[03:01:59] <elif> Rangar: I don't think it is the free software that pushs outsourcing in USA... I think outsourcing goes because of salaries paid for engeneerings working in others places (here in brazil outsourcing is growing fast, pushed by demmand of develeped countries, usa and german being the biggest. But sure india is preference, not just a high tecnical level but the languague.)[03:02:15] <Ragnarok> Lucine, you won't have to worry if its open source[03:02:49] <Rangar> Ragnarok: do you have a family?.. wife and kids, and a mortgage to pay, and trying to make a living out of what you trained for? opensource just means people get to use your hard work for free, without paying a dime... How does one live under that model?[03:03:16] <Lucine> yeah, but you don't get money anyway.[03:03:17] <Rangar> If you cannot make a living from your work, then you have to do something else[03:03:36] <Rangar> Luc; as I said, develop for companies, not 'joe bloggs users'[03:04:07] <Lucine> There's alternatives like how MySQL does it. but that's not always easily feasible, especially when you're small and hoping to earn a living. :?[03:04:25] <Rangar> i am serious though, how do you keep a family alive and a mortgage payment if what you do you have to give away for free?[03:04:49] <Lucine> have a chicken farm in the backyard, and open source the chicken :p[03:04:53] <Rangar> Lucine: MySQL will not be open source much longer - that is being ditched[03:04:54] <Ragnarok> Rangar, people pay open source people for a product thats good and to make them keep developing it[03:05:01] <Lucine> Interesting, Ragnar[03:05:20] <Rangar> Ragnarok: sure.. in a LARGE market place, as I said, my products are specific to an industry[03:05:33] <Lucine> I guess I'm moving to PostgreSQL[03:05:39] <Lucine> nah, nm[03:05:44] <Lucine> it's fine, actually as long as it stays free[03:05:51] <Ragnarok> Rangar, then get a programming job and not just a free lancer[03:06:04] <Ragnarok> well independant self programming[03:06:07] <Lucine> Ah, nm, I see what you mean now about mysql[03:06:09] <Ragnarok> not free lancer[03:06:12] <Lucine> that's a lil worrisome.[03:06:20] <Ragnarok> and if u want money go for a free lancing job[03:07:52] *** aalex has joined ##OpenGL[03:08:33] *** charlie5 has quit IRC[03:08:41] <Rangar> Ragnarok: serious... do you have a wife and kids?[03:09:33] <Ragnarok> get a fucking computer programming job[03:09:45] <Ragnarok> if u wanna make comerical products[03:09:56] * Rangar rolls his eyes[03:09:59] <Ragnarok> if u can't do that then u wouldn't be able to do it urself[03:10:20] <Ragnarok> if u do it urself there is no guarentee someone will buy it[03:10:24] <Rangar> I do not think you have listened to one thing I have said, n/m, i'm not continuing this debate[03:11:19] <Ragnarok> no ur not listening to what im saying because ur so freaking ignorant and too stubborn to hear the truth, because ur so full of shit that u tnhink ur going to make billions of dollars that u can live off from being a self programmer[03:11:47] <Ragnarok> then theres no fucking guarentee some one will buy ur crap[03:13:09] <stevosICBM> chillax Ragnarok[03:13:33] *** ajww has joined ##OpenGL[03:15:49] <Ragnarok> well im just being honest, its like those little 13 yrs old thinking they are going to build the next WoW[03:16:54] <elif> Ragnarok: I think you really didn't listen Rangar, he was based his disscussion in difficult and solid arguments to deal with... open source is adorable from my point of view but even that the arguments post there are difficult to handle, and are not the way you said it is....[03:16:57] <stevosICBM> just leave the ad-homs out of it[03:18:04] <Rangar> I have customers, I have product... and Ragnarok just jumped to huge conclusions and started insulting me with no details of the real situation, so I just drop it[03:28:18] *** t4bz has quit IRC[03:28:38] *** t4bz has joined ##OpenGL[03:29:19] *** echelon has joined ##OpenGL[03:30:42] *** hibread has joined ##opengl[03:31:02] <elif> Here in Brazil we post the discussion to the open source as a strategic movement of adoption to it, to be free of the dependence we have with big software companies from outside, what makes our development even more complicated, thinks that big organizations in brazil has to paid billions annually due to software license. So if its possible to move this software (and in some case it is) to open ones, we have to do it. But, anyway, that don'[03:32:08] <elif> So It's no guarantee to be a good solution to bust our small sized companies...[03:33:14] <elif> *burst[03:35:51] *** NeoServer has joined ##OpenGL[03:36:29] *** NeoThermic has quit IRC[03:40:05] *** blight_ has quit IRC[03:43:11] *** charlie555 has joined ##OpenGL[03:43:26] *** dvoid_ has quit IRC[03:46:28] *** trapt__ has joined ##OpenGL[03:46:29] *** trapt__ is now known as trapt[03:47:40] *** Diagmato has quit IRC[03:57:00] *** _boto2 has joined ##opengl[03:58:51] *** echelon has quit IRC[03:59:51] *** charlie55 has quit IRC[04:05:33] *** NeoServer has quit IRC[04:05:47] *** NeoServer has joined ##OpenGL[04:07:05] *** aalex has quit IRC[04:10:09] *** trapt_ has quit IRC[04:12:20] *** _boto has quit IRC[04:23:36] *** ajww has quit IRC[04:27:20] *** aalex has joined ##OpenGL[04:27:55] *** stevosICBM has quit IRC[04:45:43] *** ajww has joined ##OpenGL[04:46:15] *** miguelvmonroy has joined ##OpenGL[04:46:47] *** miguelvmonroy has quit IRC[04:53:54] *** mattn_ has joined ##OpenGL[04:55:54] *** mattn2|home has quit IRC[05:00:04] *** charlie555 is now known as charlie5[05:04:11] *** aalex has quit IRC[05:18:13] *** echelon has joined ##OpenGL[05:20:16] *** gpaci has quit IRC[05:21:31] *** hibread has quit IRC[05:34:52] *** Lucine is now known as Lucine_AFK[05:36:08] *** Lucine_AFK is now known as Lucine[05:36:21] *** Lucine is now known as Lucine_AFK[05:36:33] *** Walt has quit IRC[05:39:05] *** Lucine_AFK is now known as Lucine_Bed[05:46:56] *** Ademan has quit IRC[05:47:58] *** Ademan has joined ##OpenGL[05:48:30] *** replor has quit IRC[05:53:54] *** m4ggus_ has quit IRC[05:57:26] <NightBird> woo! my orange book finally made it[06:04:28] <Ragnarok> ?[06:04:54] <Ragnarok> the orange book really isn't nothing,its just a documentation book iirc doesn't show examples etc[06:05:27] <HuntsMan> have you read it?[06:07:03] <Ragnarok> parts[06:08:42] <HuntsMan> you probably missed the examples part :P[06:09:39] <HuntsMan> it has procedural texturing examples, lighting, fog, and mandelbrot shaders :P[06:10:02] <Ragnarok> ah[06:10:07] <Ragnarok> i like the blue book the best[06:10:08] <Ragnarok> :D[06:13:46] *** Walt has joined ##opengl[06:20:17] *** trapt has quit IRC[06:20:33] *** trapt__ has joined ##OpenGL[06:20:37] *** LiQuiDninja has quit IRC[06:20:38] *** trapt__ is now known as trapt[06:23:00] <NightBird> yeah, I looked through it a bit, and saw like... 2/3rds looked to be examples[06:24:43] <HuntsMan> yeah[06:27:17] <NightBird> that may be a gross over estimate, but... meh[06:30:14] *** pragma_ has joined ##opengl[06:30:35] *** hydoskee has joined ##OpenGL[06:30:54] <hydoskee> I've got some shapes that aren't appearing when they're called to "draw" - how can I debug that?[06:31:11] <hydoskee> they appear before they move, and during a move, but post-move, they don't appear[06:34:24] *** echelon has quit IRC[06:34:39] <Ragnarok> post code[06:34:44] <Ragnarok> or we can't help[06:34:58] <NightBird> to pastebin or something similar[06:35:03] <Ragnarok> rafb.net/paste[06:38:24] *** Lemml has joined ##OpenGL[06:39:29] *** trapt__ has joined ##OpenGL[06:42:49] *** LiQuiDninja has joined ##OpenGL[06:43:05] *** trapt__ has quit IRC[06:44:18] *** trapt__ has joined ##OpenGL[06:47:12] *** BahamutZERO has quit IRC[06:50:34] <hydoskee> Ragnarok - but it's multiple levels deep[06:50:44] <hydoskee> should I just post the openGL calls?[06:51:07] *** Madsy has quit IRC[06:51:15] <Ragnarok> just post the main code[06:52:37] *** trapt has quit IRC[06:52:54] <hydoskee> it's not going to make any sense - it's an animation routine that is multiple objects deep - a flock has birds, birds have eyes, legs, a body, a beak, each of the bird parts has shapes, shapes have vertices[06:53:06] <hydoskee> and at the top level I'm just calling flock.draw();[06:53:27] <hydoskee> so if I post "flock.draw()" it's not going to make a lot of sense[06:53:52] *** BahamutZERO has joined ##OpenGL[06:54:03] <hydoskee> I'll post my shape code, and my vertex code[06:55:08] *** BahamutZERO has quit IRC[06:55:10] <hydoskee> http://rafb.net/p/5s31VS68.html[06:56:40] <hydoskee> the parts you'll probably be interested in are vertex::draw() and shape::draw()[06:58:36] *** dolphin has joined ##OpenGL[06:59:44] *** amz has quit IRC[07:00:15] *** Shel2476 has joined ##OpenGL[07:00:16] *** Shel2476 has left ##OpenGL[07:00:35] *** mm^away has joined ##opengl[07:03:24] <hydoskee> any ideas about it?[07:03:34] *** Shel2476 has joined ##OpenGL[07:03:36] *** Shel2476 has left ##OpenGL[07:14:11] *** trapt__ has quit IRC[07:17:27] *** mm765^away has quit IRC[07:19:17] *** nokmar has joined ##OpenGL[07:19:27] *** nokmar has left ##OpenGL[07:19:45] *** bbeausej has quit IRC[07:37:31] <Jupp3> hydoskee: So basically you are drawing each single triangle or quad between its own glBegin()... and glEnd() or?[07:38:20] <Jupp3> argh...[07:38:34] <Jupp3> hydoskee: And wtf are you doing with temp colors?[07:39:50] <Jupp3> hydoskee: For some reason, you multiply the color by 100, divide by 255, and then divide by 100[07:40:13] <Jupp3> hydoskee: Why don't you just divide it by 255.0f? I can't see any sensible use for 100 in that formula...[07:44:39] *** Lemml has quit IRC[07:46:12] <Jupp3> Now I get it, that's a workaround for integer division[07:46:24] *** Plagman has quit IRC[07:47:22] <Jupp3> hydoskee: BUT as you're converting to float eventually anyway, couldn't you just do temp[0]=(float)color[0]/255.0f;?[07:47:46] <Jupp3> Although you can omit (float) as it will be done automatically when other side is in that format[07:47:57] <Jupp3> But it will be done anyway, no faster code without it or anything[07:49:06] *** thesquib has joined ##OpenGL[07:52:11] *** Walt has quit IRC[08:00:40] *** scy has joined ##opengl[08:08:04] *** hydoskee has quit IRC[08:09:42] *** XypherOrion has joined ##OpenGL[08:28:54] *** neoneye has joined ##OpenGL[08:36:19] <HuntsMan> btw, there's a pure GPU implementation of Perlin Noise?[08:36:46] <Jupp3> there is?[08:37:06] <HuntsMan> that's why i'm asking :)[08:37:23] <Jupp3> Well you just noted that "there is", but for some reason added a question mark[08:37:35] <HuntsMan> sorry my mistake[08:37:38] <Jupp3> :)[08:37:53] <HuntsMan> but you get the idea :P[08:38:07] <Jupp3> Anyway, I need to go to work now[08:38:21] <HuntsMan> and me to sleep[08:39:00] *** Roderic has joined ##OpenGL[08:39:02] *** Roderic is now known as Ingenu[08:41:36] *** Burga has joined ##OpenGL[08:48:27] *** scy has left ##opengl[08:50:48] *** XypherOrio1 has joined ##OpenGL[08:58:23] *** fargiolas has joined ##OpenGL[08:58:53] *** Yustme has joined ##opengl[08:59:11] *** fargiolas is now known as fargiolas|afk[08:59:28] *** XypherOrion has quit IRC[09:01:04] *** Walt has joined ##opengl[09:15:52] *** XypherOrio1 has quit IRC[09:19:34] *** LiQuiDninja has quit IRC[09:23:38] *** [AD]Turbo has joined ##OpenGL[09:28:13] *** Jernej has joined ##OpenGL[09:29:58] *** druggy has joined ##opengl[09:31:40] *** KU0N has joined ##OpenGL[09:32:14] <KU0N> hello[09:34:02] *** druggy_ has quit IRC[09:34:22] *** mattn_ has quit IRC[09:38:30] *** dolphin has quit IRC[09:40:28] *** groton has joined ##OpenGL[09:43:22] <[AD]Turbo> ciaa groton[09:43:29] <groton> cia'[09:43:45] *** JernejL_ has quit IRC[09:56:08] *** mattn2|home has joined ##OpenGL[09:57:38] *** mattn2|home has quit IRC[10:03:43] <quicksilver> HuntsMan: I've certainly heard them discussed. I've not atually seen one though.[10:17:16] *** Walt has quit IRC[10:26:15] *** dvoid_ has joined ##OpenGL[10:32:18] *** niraj1234 has joined ##OpenGL[10:33:08] *** niraj1234 has left ##OpenGL[10:40:43] *** belou_ has joined ##OpenGL[10:42:38] *** belou has quit IRC[10:45:13] *** Walt has joined ##opengl[10:51:48] *** scy has joined ##opengl[10:57:31] *** servus_ has joined ##opengl[11:00:18] *** BahamutZERO has joined ##OpenGL[11:11:16] *** belou has joined ##OpenGL[11:12:59] *** MatthiasM has quit IRC[11:13:06] *** MatthiasM has joined ##opengl[11:13:46] *** servus has quit IRC[11:27:20] *** belou_ has quit IRC[11:28:26] *** [AD]Turbo has quit IRC[11:33:09] *** [AD]Turbo has joined ##OpenGL[11:33:59] *** [AD]Turbo has quit IRC[11:34:13] *** [AD]Turbo has joined ##OpenGL[11:39:44] *** dolphin has joined ##OpenGL[11:49:31] *** freespac1 has joined ##opengl[11:50:41] *** echelon has joined ##OpenGL[11:51:00] *** darka has joined ##opengl[11:51:30] *** servus_ is now known as servus[11:54:27] *** freespace has quit IRC[11:54:28] *** freespac1 is now known as freespace[11:57:58] *** DarkNight has joined ##OpenGL[11:59:11] *** kaotrix has joined ##OpenGL[12:00:09] <DarkNight> Hi[12:00:55] <DarkNight> How would I setup the aspect ratio using glOrtho, so that when the window resizes, polygons etc resize in proportion?[12:02:59] <DarkNight> Coz just using glOrtho(0, w, 0, h, -1, 1) makes it clip whats in the viewport[12:04:16] <Burga> sourceforge dead?[12:05:23] <phrosty> DarkNight: just use a constant width/height[12:06:11] <DarkNight> well I tried that. but shapes dont stay in proportion[12:06:46] <DarkNight> if you change the width then it'd make a triangle squish instead of resize[12:07:25] <groton> Burga, yes[12:07:32] <Burga> grah![12:07:35] <Burga> thanks for checking for me[12:08:33] <quicksilver> DarkNight: use the real width.[12:08:42] <quicksilver> DarkNight: but calculate the height based on teh aspect ration you want to use.[12:08:54] <quicksilver> 0, w, 0, (w/aspectratio)[12:08:56] *** servus has quit IRC[12:09:41] <DarkNight> why only the height?[12:10:00] <DarkNight> shouldnt width and height use the aspec ratio?[12:11:36] *** kaotrix has quit IRC[12:11:42] *** scy has left ##opengl[12:12:14] <quicksilver> DarkNight: it's entirely up to you.[12:12:22] <quicksilver> DarkNight: you can put whatever rubbihs you want in Ortho.[12:12:33] <quicksilver> but it depends what coordinates you personally want to use in the program![12:12:39] <quicksilver> I quite often use -1,1,-1,1[12:12:48] <quicksilver> and then I have a nice square with 0,0 at the center[12:12:59] <DarkNight> well I want 0,0 to be bottom left[12:13:06] <quicksilver> if you're doing pixel accurate work it can be nice to have a 1-1 pixel mapping[12:13:40] <quicksilver> and if you don't like continually adjusting aspect ratios it's nice to ahve the aspect ratio proportion right.[12:14:47] <quicksilver> you could use (0,1,0,1/aspect) then, if you like.[12:15:52] *** LordMetroid has joined ##OpenGL[12:16:15] *** servus has joined ##opengl[12:18:45] <DarkNight> that doesnt seem to work very well[12:19:41] <DarkNight> increasing the width increases the height of an object[12:20:17] <quicksilver> well your code needs to be consistent.[12:20:24] <quicksilver> you either keep a square coordinate system.[12:20:32] <quicksilver> in which case your code doesn't need to know aspects[12:20:40] <quicksilver> or, you modify the coordinate system[12:20:43] <quicksilver> and make sure the code adjusts.[12:20:54] <quicksilver> In general it's easier to keep a square coordinate system I think.[12:21:47] <quicksilver> you can also decide, obviously, whether resizing the window makes things bigger[12:21:49] <quicksilver> or makes more visible[12:25:35] *** mm^away is now known as mm765[12:27:59] *** korff has joined ##OpenGL[12:29:32] *** speedy1 has joined ##OpenGL[12:41:49] *** Rangar has quit IRC[12:44:46] *** fargiolas|afk is now known as fargiolas[12:54:56] *** Lucine_Bed is now known as Lucine[12:56:00] *** JernejL_ has joined ##OpenGL[13:02:06] *** Rangar has joined ##OpenGL[13:03:18] *** dvoid_ has quit IRC[13:14:25] *** Jernej has quit IRC[13:18:26] *** Madsy has joined ##OpenGL[13:24:31] *** echelon has quit IRC[13:35:10] *** blight_ has joined ##opengl[13:43:44] *** rsp has joined ##opengl[13:47:44] *** Tibor__ has joined ##OpenGL[14:00:21] *** LordMetroid has quit IRC[14:06:17] *** dvoid_ has joined ##OpenGL[14:08:49] *** sanjin has joined ##OpenGL[14:15:53] *** hibread has joined ##opengl[14:18:11] *** sanjin has quit IRC[14:29:31] *** elite01 has joined ##opengl[14:29:52] <hibread> Evening boogermen![14:39:01] *** rhythm has joined ##Opengl[14:44:20] *** TheLorax has joined ##opengl[14:45:00] *** fargiolas has quit IRC[14:56:37] *** DrGonzo is now known as Gonzo[14:58:34] *** mm765 is now known as mm765^away[15:00:58] *** stuckie has joined ##opengl[15:01:51] *** dv_ has joined ##opengl[15:12:29] *** gpaci has joined ##OpenGL[15:12:31] *** [this] has joined ##opengl[15:13:12] *** [this] has quit IRC[15:21:43] *** amz has joined ##opengl[15:30:26] *** thesquib has quit IRC[15:42:39] *** elif has left ##opengl[15:45:03] *** Suprano has joined ##OpenGL[15:47:14] *** rnx has joined ##opengl[15:54:48] *** dvoid_ has quit IRC[15:58:10] *** m4ggus has joined ##opengl[16:02:13] *** pfo is now known as pfo_away_till_fr[16:02:51] *** groton_ has joined ##OpenGL[16:03:11] *** groton has quit IRC[16:03:25] *** groton_ is now known as groton[16:04:46] <DarkNight> whats the trick to rotating an object around a point?[16:05:07] <DarkNight> ive tried translating it to origin, rotating, then translating it back[16:05:24] <DarkNight> but it seems to just wanna keep revolving around the origin 0,0[16:05:44] <hibread> DarkNight: what is your approach? Ie, are you using your own matrices etc.. or using say GLs glRotate() calls etc?[16:06:05] <Seeb_> DarkNight: translate to the center point of the rotation, rotate it, translate the radius[16:06:08] <speedy1> "translating it to origin, rotating, then translating it back" -> that should work[16:06:41] *** tmccrary has joined ##OpenGL[16:07:47] <Seeb_> DarkNight: so to rotate around (x,y) with a radius r, glPushMatrix(); glTranslatef(x, y, 0.0f); glRotatef(angle, 0.0f, 0.0f, 1.0f); glTranslatef(r, 0.0f, 0.0f); /* render */ glPopMatrix();[16:08:57] *** korff has quit IRC[16:09:39] <DarkNight> http://pastebin.com/d2c0aaded[16:09:47] <DarkNight> thats basically what im doing[16:10:10] <Seeb_> DarkNight: that's not how I wrote[16:10:29] <Seeb_> you need to apply glRotatef in the center of the rotation[16:10:51] *** HuntsMan has quit IRC[16:11:10] <DarkNight> I want the object to rotate around a point on the object which is 90,50 tho[16:11:21] <Seeb_> so[16:11:24] <Seeb_> (x,y) = (90,50)[16:11:30] <Seeb_> what's your radius of the rotation?[16:12:37] <DarkNight> radius? there shouldnt be if its rotating on the spot[16:13:09] <hibread> just rotate then translate...[16:13:26] <hibread> err[16:13:36] <DarkNight> http://illusionaldesign.net/GFX01.zip[16:13:50] <DarkNight> theres what its doing at the moment[16:14:50] <Seeb_> DarkNight: do you want it to just rotate, or to rotate around something? If you want it to rotate around something you need to have a radius defined[16:14:59] <Seeb_> otherwise you can't build a circle on which to rotate[16:15:02] *** elite01 has quit IRC[16:15:33] <DarkNight> did you run the program? basically the top point of the house is what i want the rest of the house to rotate around[16:15:58] *** Sisu has quit IRC[16:17:06] <DarkNight> so im translating the top point back to origin, rotating, then moving it back[16:17:32] <DarkNight> but it still wants to just rotating the whole house around the origin[16:18:12] *** JernejL_ has quit IRC[16:18:54] <hibread> DarkNight: any chance of supplying all code? Unless you've got windows dependant gear in there[16:19:07] <DarkNight> yep just a sec[16:19:15] <Seeb_> DarkNight: don't have time to look at it, but I'm sure what I wrote will work[16:21:59] *** aalex has joined ##OpenGL[16:22:44] <DarkNight> http://illusionaldesign.net/Main.cpp[16:25:44] <speedy1> wee ATI fixed the skinning bug my game had ^^[16:26:24] *** pfo_away_till_fr has quit IRC[16:26:42] <speedy1> I could now actually un-blacklist their HD cards to perform normal GPU skinning :)[16:28:17] *** JernejL has joined ##OpenGL[16:29:08] *** niraj1234 has joined ##OpenGL[16:32:00] <DarkNight> :/ damn rotations[16:33:49] *** magnet_ has quit IRC[16:35:57] *** elite01 has joined ##opengl[16:37:00] <hibread> DarkNight: you just want the "house" to spin on itself? say around its own center?[16:39:11] <DarkNight> yes, but its own center isnt actually the center, i want the top point of the roof of the house to be the point it rotates around[16:39:14] *** sohail has quit IRC[16:39:20] <hibread> ok[16:39:52] <DarkNight> so i translated the coords of that point to the origin, then rotate, then back to original point[16:39:57] <DarkNight> so in the code[16:40:46] <speedy1> should work[16:40:47] <hibread> Firstly, create the vertices in local "model" space[16:41:14] <hibread> if you want the top of the roof to be rotated about, make that the location (0,0) in local model space[16:43:07] *** bbeausej has joined ##OpenGL[16:43:09] <speedy1> // translate so next house isnt on top of first[16:43:09] <speedy1> glTranslatef(40,60,0);[16:43:25] <speedy1> this part might be causing you troubles[16:44:16] <DarkNight> i took that out. still same issue of it rotating aroud 0,0[16:44:46] *** Bollinger has joined ##opengl[16:45:18] <hibread> http://pastebin.com/m510641e2[16:45:22] <DarkNight> hibread: im already in GL_MODELVIEW, and ive made the location 0,0 by doing the translation of the point i wanna rotate around back to the origin[16:45:23] <Ragnarok> the only thing that bothers me about opengl is trignometry :P[16:45:32] <speedy1> try switching the translatef rows[16:45:33] <hibread> DarkNight: that rotates around the point of the roof[16:45:42] <quicksilver> trigonometry is kind of essential to 3D graphics.[16:45:47] <quicksilver> it's not really opengl's fault :P[16:45:54] <Ragnarok> i never learned it in school :P[16:45:57] <speedy1> first row 70,70, third row -70,-70[16:45:59] <Ragnarok> i dropped out before[16:46:41] <quicksilver> then you can learn it now :P[16:46:46] <quicksilver> it's easier when you ave a motivation.[16:46:59] <Ragnarok> quicksilver, link?[16:47:10] <DarkNight> hibread: i took them out and the whole house still moves as if its going around 0,0[16:47:20] <quicksilver> you could try http://en.wikibooks.org/wiki/Trigonometry[16:47:25] <quicksilver> I've never looked at it though[16:47:28] <quicksilver> it coul dbe lousy :P[16:47:34] <hibread> DarkNight: took what out?[16:47:45] <DarkNight> the two translates[16:48:07] <hibread> DarkNight: did you try my code to see if thats the type of result you're looking for?[16:48:29] <Ragnarok> quicksilver, well i understand cosine and sine, which returns the location of a point from an origin :P[16:49:44] <DarkNight> hibread: ok that works. but wtf. why does changing the vertices fix it?[16:50:21] <hibread> DarkNight: modelling in local coordinates will always help[16:50:32] <hibread> err.. "good" local coordinates :)[16:51:00] <DarkNight> well i was given a set of coords so wanted to use them[16:51:16] <DarkNight> surely its doable not using local?[16:51:43] <hibread> sure it is... you really want the solution for your previous set of house vertices?[16:51:50] <hibread> ill see what i can do[16:52:17] <DarkNight> yeh i'd be very interested to know[16:52:36] <DarkNight> because i thought the translate, rotate, translate method should work for any case[16:52:50] <DarkNight> thats what we've been learning, and this is meant to show that[16:55:58] <hibread> try translating by 70,70 then rotate then -70,-70[16:56:42] <DarkNight> LOL wtf[16:56:47] <DarkNight> works[16:57:58] <DarkNight> why does that go against the theory? lol[16:59:33] <hibread> hmmmm... i admit being a little baffled myself[17:00:30] <DarkNight> very strange[17:02:04] <DarkNight> maybe the way the window is setup[17:02:17] <DarkNight> because normally rotations go clockwise yes?[17:02:32] <DarkNight> when i do 45 degrees[17:02:44] <DarkNight> the object visually rotates -45[17:03:33] <DarkNight> glOrtho screwing shit up maybe?[17:04:17] <DarkNight> or[17:04:31] <DarkNight> hm no[17:08:13] <hibread> got me stuffed[17:10:09] <DarkNight> i'll have to ask about it. but thanks for ya help mate[17:11:03] <hibread> makes no sense really. I feel dumb just thinking about it[17:11:20] <hibread> There's obviously something missing in my (and our) understanding here... quite annoying[17:12:20] <DarkNight> i know normally matrices have specific order they have to be in[17:12:52] <DarkNight> like T'.R.T.p[17:13:11] <DarkNight> where T' is translation to original[17:13:13] <DarkNight> R is rotation[17:13:22] <DarkNight> T is translation to origin[17:13:25] <DarkNight> p is the point[17:13:42] <DarkNight> in this case you have to do T.p first[17:13:59] <DarkNight> T'.(R.(T.p))[17:14:25] *** amz has quit IRC[17:14:29] <DarkNight> maybe thats how the matrix function calls in opengl work too?[17:14:52] <DarkNight> glTranslatef(70,70,0); glRotatef(theta, 0, 0, 1); glTranslatef(-70,-70,0);[17:15:08] <DarkNight> T' R T^[17:15:25] <DarkNight> just a thought, i could be totally wrong[17:16:27] <hibread> you're probably on the right track... although opengl just recieves the gl calls as they're commited in code hence the multiplication or calculations are made in that order... but err.. ill have to refresh my matrices knowledge which is quite lame[17:17:50] <DarkNight> T'.(R.(T.p)) <- that is how it should be done if you were to write it out[17:18:32] <hibread> matrix multiplication is associative though... A(BC) == (AB)C[17:18:43] <DarkNight> maybe opengl saves each transformation matrix when called, and then when a glBegin is called, it multiplies them in reverse order?[17:19:10] <DarkNight> not with transformations tho[17:19:19] <DarkNight> rotate -> translate[17:19:21] <hibread> remember its p, that you want to transform... so that would be multiplied by the concatination of the previous matrix applications[17:19:26] <DarkNight> isnt the same as doing translate -> rotate[17:19:50] <tmccrary> its because the origin (in the local coordinate system) is your pivot[17:20:20] <tmccrary> when you translate away, your pivot point is the same[17:20:33] <hibread> tmccrary: the wanted pivot is off to the left and up... so wouldn't you translate down and left? :)[17:21:10] <tmccrary> If you wanted to rotate around the center of the object... put the origin at the center and rotate[17:21:22] <tmccrary> which would usually be the inverse of the translation[17:21:54] <hibread> tmccrary: yeah thats what ive been saying, but we're trying to solve for this particular case where the object is described in such a way that the origin (0,0,0) is no where near the location in which we want to rotate about[17:23:31] <tmccrary> first translate by the inverse of the current position, then translate to the point you want rotate about, then rotate, then translate back to the final position[17:24:07] <hibread> current position with relation to the desired pivot?[17:24:44] <tmccrary> no, without[17:24:50] <tmccrary> If I am understanding you correctly[17:25:03] <hibread> so even though the vertex for use with pivot was defined as (70, 70), you translate by 70,70 first? or -70,-70? :)[17:25:04] <DarkNight> "first translate by the inverse of the current position, then translate to the point you want rotate about"[17:25:10] <DarkNight> that cancels itself out[17:25:13] <tmccrary> thats the idea[17:25:33] <DarkNight> thats pointless lol[17:25:38] <tmccrary> no, its not pointless[17:25:46] <DarkNight> it goes from 70,70 to 0,0 back to 70,70[17:25:50] <DarkNight> nothing has changed[17:26:07] <tmccrary> it goes from 70 to 0, apply local translation, rotate, then back to 70 again[17:26:24] <hibread> to go from 70 to 0, you translate by 70 or -70? :)[17:26:39] <DarkNight> "first translate by the inverse of the current position, then translate to the point you want rotate about"[17:26:42] <tmccrary> if you're at 70, you'd first translate by -70 to get to 0[17:26:49] <DarkNight> that suggests 70 -> 0 -> 70 lol[17:26:54] <hibread> tmccrary: define "you're at"[17:27:07] <tmccrary> bah, I can't help retards. Good luck![17:27:28] <DarkNight> "first translate by the inverse of the current position, then translate to the point you want rotate about, then rotate, then translate back to the final position"[17:27:50] <DarkNight> that sentence is 70 -> 0 -> 70 -> rotate -> 70[17:28:04] <DarkNight> when it should be 70 -> 0 -> rotate -> 70[17:28:24] <tmccrary> I see where you're missing the point, but I'm not helping you. /ignore[17:28:35] <Ragnarok> is there a shader ide on linux like rendermonkey etc[17:29:00] <DarkNight> pretty lame attitude tbh[17:30:12] <tmccrary> Ragnarok: lumina[17:30:52] <Ragnarok> thank you[17:31:35] <Ragnarok> iu can't seem to find it on google[17:31:44] <tmccrary> http://lumina.sourceforge.net/[17:32:26] <Ragnarok> thank you[17:33:09] <hibread> DarkNight: i think ive made some ground...[17:34:02] *** belou has quit IRC[17:34:06] <DarkNight> yeh?[17:34:31] <hibread> try this[17:34:51] <hibread> after your pushMatrix()[17:37:33] <hibread> translate by any arbitrary positive number, rotate theta about z, translate by the negated xy values of the location in the original coordinate system in which you described your model (ie, (-70,-70,0))[17:38:48] <hibread> the initial translation is not important for relocating the rotation point in the local coordinate system of the model[17:38:57] *** dvoid_ has joined ##OpenGL[17:39:01] <hibread> so you rotate first, then translate[17:39:18] <DarkNight> works but is in diff location[17:39:25] <DarkNight> rotate then trans[17:39:38] <DarkNight> ...that totally goes against the theory[17:40:08] <DarkNight> unless opengl is doing what i said earlier[17:40:37] <DarkNight> and multiplying the matrices in a diff order to what you call them[17:40:57] <DarkNight> it does keep a stack of matrices.. so maybe it does[17:41:26] <hibread> DarkNight: im pretty sure opengl doesn't store matrices at infinitem then once you want to issue a draw command that it multiplies all of them in the reverse order[17:41:37] <hibread> that stack is a different stack[17:41:49] <speedy1> it's not even a stack[17:41:50] <hibread> it definitely applies the transformations as it shows in code[17:42:05] <speedy1> it's simple matrix / matrix multiplication[17:42:53] <speedy1> the question is from which side it multiplies ;)[17:43:19] <DarkNight> not when it comes to these kind of transforms.. the order makes a difference[17:43:28] <hibread> speedy1: ahuh! nice owrk[17:43:43] <hibread> yeah... AB != BA[17:43:53] <DarkNight> visually T'.(R.(T.p)) != ((T'.R).T).p[17:44:08] <hibread> DarkNight: nah that should be the same[17:44:14] <hibread> matrices are associative[17:44:41] <DarkNight> not from what ive been taught when it comes to transformations like this[17:44:47] *** DMINATOR has joined ##OpenGL[17:45:09] <DarkNight> you always multiply the cloest matrix to the point[17:45:16] <DarkNight> eg. T'.(R.(T.p))[17:45:59] <speedy1> try inversing the order of translations, and it should work around the point[17:46:02] <DarkNight> both my linear algebra and computer gfx classes have said this[17:47:02] <DarkNight> speedy1, i originally had it as -70,-70, rotate, 70,70 and it didnt work[17:47:13] <speedy1> try 70, rot, -70[17:47:21] <DarkNight> yeh that works[17:47:28] <hibread> DarkNight: you're sure you're not mixing (AB)C == A(BC) with ABC != CBA ?[17:47:30] <speedy1> voila[17:47:43] <DarkNight> we've known this works, but we're wondering why[17:48:06] <DarkNight> because all theory on the net says to translate to origin, rotate, then translate back to original point[17:48:16] <hibread> DarkNight: yeah but the point here is that 99, rot, -70 still causes the house to rotate about the correct point[17:48:28] <hibread> the initial transformation does nothing about the point on the object which you rotate[17:48:41] *** rhythm has quit IRC[17:48:50] <DarkNight> yeh. it just means it stays where i originally wanted it ;)[17:48:53] <hibread> it does, however, affect the resultant location in "world coordinates" that it rotates[17:49:16] <hibread> DarkNight: ok no probs.. but this approach is not the ideal one :)[17:49:36] <speedy1> "99, rot, -70 still causes the house to rotate about the correct point" <- not possible[17:49:47] <DarkNight> the dilemma still stands. the rule is you translate then rotate, not otherway around.. but yet that seems to work in OGL[17:50:17] <speedy1> the question is how the transforms are accumulated and from which side are they applied to the point ;)[17:50:17] <DarkNight> which makes me wonder if ogl multiplies them internally in a diff order[17:50:25] <DarkNight> like backwards to the way you called them[17:50:33] <speedy1> if you have (A * B * C) * p[17:50:37] <speedy1> then it goes backwards[17:50:38] <DarkNight> ..makes sense but no proof that ogl actually does this[17:50:51] <speedy1> A * (B * (C*p))[17:51:04] <DarkNight> ^that is how it should be[17:51:25] <speedy1> no that's the inverse case[17:51:31] <speedy1> you specify A, B, C order[17:51:44] <speedy1> but the point gets first transformed by C then B then A[17:51:54] <DarkNight> im thinking ogl might accumulate the matrices and call them in that order[17:52:10] <Jupp3> hmm[17:52:15] <speedy1> nope it just multiplies - check the ogl spec for that[17:52:22] <Jupp3> Doesn't OpenGL always multiply by exactly one matrix?[17:52:24] <DarkNight> A * (B * (C*p)) <- is the way we've been shown in both subjects[17:52:39] <DarkNight> they insist any other way wont produce the same results[17:52:44] <DarkNight> so i dunno[17:52:51] <Jupp3> I mean, when you do glTranslatef and then glRotatef, doesn't that result in exactly one matrix, that "does both in that order"?[17:53:06] <DarkNight> lemme find an example i saw[17:53:17] *** replor has joined ##OpenGL[17:53:48] <quicksilver> opengl build up the transformationin the order you issue the commands[17:53:52] <Jupp3> and glPushMatrix() just "saves" the current matrix as backup, to which you can return with glPopMatrix()[17:53:53] <quicksilver> but it certainly "feels backwards"[17:54:05] <quicksilver> don't forget you can left multiply and right-multiply with matrices[17:54:07] <hibread> quicksilver: thats because there is AB and BA :)[17:54:12] <quicksilver> and you get different results[17:54:18] <quicksilver> hibread: right, I was just getting to that :)[17:54:22] <hibread> :)[17:54:32] <hibread> A(BC) is the same (AB)C[17:54:37] <quicksilver> right.[17:54:40] <hibread> matrices are associative[17:54:41] <quicksilver> but AB is not the same as BA[17:54:45] <hibread> but not communtative[17:54:49] <DarkNight> http://staff.psc.edu/smp/teaching/cs1566-fall2004/opengl-notes-3.pdf[17:54:49] <quicksilver> matrices rock![17:54:50] <quicksilver> ;)[17:54:52] <hibread> *commutative[17:54:55] <DarkNight> page 79 onwards[17:55:09] <DarkNight> translate/rotate != rotate/translate[17:55:16] <hibread> yeah thats right[17:55:54] <DarkNight> yeh and in all theory it says to translate to origin, rotate, trans back to original point[17:56:05] <DarkNight> however we have to do the opposite in opengl[17:56:17] <speedy1> actualy you're telling ogl to do just that[17:56:21] <speedy1> and it does[17:56:35] <speedy1> but you have to specify it in inverse order[17:56:37] *** Ingenu has quit IRC[17:56:49] <quicksilver> it's very nice for calling subroutines though[17:56:54] <quicksilver> (they way openGL does it)[17:57:02] <quicksilver> glTranslate(..to over here..)[17:57:03] <speedy1> yeah, more practical then intuitive[17:57:05] <quicksilver> drawOneOfThose()[17:57:12] <quicksilver> glTranslate(..to over there..)[17:57:15] <quicksilver> drawOneOfThose()[17:57:28] <quicksilver> with the option to push/pop periodically too.[17:57:58] <DarkNight> glTranslatef(-70,-70,0); glRotatef(theta, 0, 0, 1); glTranslatef(70,70,0); <- how they describe in all the theory[17:58:00] <Jupp3> You whould use drawTwoOfThese() instead :)[17:58:17] <DarkNight> yet you have to swap the translates for it to actually do the desired effect[17:58:40] <DarkNight> i seriously havent read anywhere that with opengl you have to call them in reverse order[17:58:51] <speedy1> dark: yeah, accept the fact that your instructions affect the vertex in inverse order ;)[17:59:00] <Jupp3> DarkNight: reverse to what?[17:59:10] <speedy1> jupp3: to intuitiveness[17:59:19] <Jupp3> I mean, in some cases, it might be the "wrong way around" (non-opengl way)[17:59:28] <DarkNight> glTranslatef(-70,-70,0); glRotatef(theta, 0, 0, 1); glTranslatef(70,70,0);[17:59:38] <ville> glTranslate(); glRotate() is probably the more intuitive way.[17:59:42] <DarkNight> that is the standard way to do it in all gfx theory[18:00:11] <DarkNight> Jupp3: OpenGl actually seems to require you to have that in inverse order[18:00:12] <Jupp3> So I'd assume at least Direct3D to be different[18:00:33] <quicksilver> you're moving the origin, not the object[18:00:34] <Jupp3> Well it felt "backwards" to me aswell[18:00:38] <quicksilver> (if it helps to think about it that way)[18:00:41] <Jupp3> But I've just learned to use it how it works[18:00:44] <ville> OpenGL and Direct3D will differ sort of. If you think about row vector vs column vector sittuation.[18:01:01] <hibread> DarkNight: i think what is confusing matters is the fact that the objects local coordinates start offset to 0,0,0...[18:01:13] <Jupp3> And you can just "switch order untill it works as you want"[18:01:19] <quicksilver> yeah, I do that sometimes :)[18:01:21] <DarkNight> that isnt the confusing part heh[18:01:28] <quicksilver> even believing I understand it, I get the order backwards sometimes.[18:01:30] <DarkNight> cause it shouldnt matter if they are offset[18:01:42] <quicksilver> DarkNight: it doesn't matter for some things.[18:01:45] <DarkNight> the trans/rotate/trans should work no regardless[18:01:46] <quicksilver> but it does for rotation.[18:01:52] <quicksilver> because rotation if about the origin.[18:01:58] <quicksilver> so if they're offset, you rotate about the wrong point[18:02:04] <quicksilver> which 'swings' the object in space.[18:02:11] <quicksilver> of course, being offest doesn't matter for translation.[18:02:19] <DarkNight> thats the whole point of translating back to origin[18:02:48] <DarkNight> you fix the offset, rotate, then put it back to where it was[18:03:26] <DarkNight> the problem was that I didnt know in opengl you had to specify the order in reverse[18:05:28] <hibread> DarkNight: Maybe your probably is thinking about transforming the coordinate space, as opposed to the "object" ?[18:06:02] <DarkNight> nah[18:06:14] <DarkNight> we went through it today in class lol[18:06:31] <DarkNight> im dead certain im talking about the object[18:07:13] <hibread> err sorry thats wha ti mean... you're really altering the coordinate system... (im confused too though.. trying to work through it on paper atm)[18:07:48] <DarkNight> every point of the "object" is translated back to the origin relative to a point you choose in the object, in my case it was 70,70, you then rotate, then you translate back by however much to get back original point[18:08:29] <DarkNight> if I could show you the diagram from the textbook it'd make more sense i think[18:08:38] <hibread> hold up a sec[18:08:49] <hibread> "you then rotate"[18:08:57] <hibread> what are you rotating about?[18:09:04] <speedy1> dark: you're missing the point that you don't translate the object, you're multiplying the model matrix with translation matrix[18:09:05] <hibread> you think the house is now in the centre of the coordinate system?[18:10:08] <DarkNight> the point 70,70 is now the centre of the system 0,0 and the other points are relative to that[18:11:14] <DarkNight> yeh im thinking there is a mix up between the standard theory and how opengl operates[18:11:31] <speedy1> nope, you just specify operations in inverse order[18:11:42] <DarkNight> yes but why lol[18:11:53] <DarkNight> that isnt the standard from what ive seen[18:11:53] <speedy1> now that's a good question :D[18:12:16] <DarkNight> this is what i meant by a mix up[18:12:30] <DarkNight> opengl operates differently to the standard theory[18:12:48] <DarkNight> eg. operations in inverse order[18:13:17] <DarkNight> maybe theres documentation on it somewhere lol[18:13:31] <speedy1> yeah, it's neatly explained in opengl spec[18:13:46] <DarkNight> ?[18:14:20] <DarkNight> link?[18:14:32] <speedy1> find it on opengl.org[18:15:45] <hibread> DarkNight: actually... have you tryed this with pen and paper?[18:16:22] <hibread> I think i can show you how it works with pictures[18:17:37] *** [AD]Turbo is now known as TurboAWAY[18:19:27] <DarkNight> where in the spec? i dont see anything specific[18:20:16] <hibread> DarkNight: priv msg[18:20:38] *** lolage0 has joined ##OpenGL[18:22:15] <DarkNight> bingo[18:22:32] <hibread> whats bingo?[18:22:33] <DarkNight> The root cause of the problem is that OpenGL matrix operations postmultiply onto the matrix stack, thus causing transformations to occur in object space. To affect screen space transformations, you need to premultiply. OpenGL doesn't provide a mode switch for the order of matrix multiplication, so you need to premultiply by hand.[18:22:37] <DarkNight> ^^[18:22:41] *** groton has quit IRC[18:23:17] <hibread> so yes.. matrix multiplication is associative but not commutative :)[18:23:25] <quicksilver> yes.[18:23:28] <quicksilver> except it isn't a "problem"[18:23:40] <quicksilver> any more than driving your car on the left side of the ride in the UK is a "problem"[18:23:49] <quicksilver> it's merely something you need to know if you don't want a head on collison.[18:24:03] <quicksilver> neither is it different from any meaningful "standard"[18:24:14] <quicksilver> mathematicians both pre and post multiply depending on their mood[18:24:17] <quicksilver> and the phase of the moon.[18:24:21] <quicksilver> one is not more common than the other.[18:24:44] <DarkNight> yeah. well i just didnt know thats how OGL dealt with it. Im used to seeing it the other way around.[18:25:15] <speedy1> yeah, more intuitive would be to "premultiply"[18:26:11] <DarkNight> Yeh well it'd be less confusing for those who learn the actual standard used in basically every gfx and linear algebra book and subject lol[18:26:29] <speedy1> for sure :)[18:26:32] <quicksilver> most gfx books at least mention opengl[18:26:36] <quicksilver> if not being based on opengl :)[18:26:43] <quicksilver> so they're likely to discuss the problem[18:27:04] <DarkNight> funny enough. the examples i saw on the net showed the premultiply way[18:27:09] <quicksilver> the linear algebra books which I have read contain plenty of multiplcation both ways around.[18:27:38] <DarkNight> yeh but specifically for translation and rotation matrices?[18:27:44] <DarkNight> or just matrices in general?[18:28:43] <hibread> I think this also has to do with column vs row matrices? the order of 'sidedness' of the multiplication?[18:28:49] <quicksilver> yes, that too[18:28:56] <quicksilver> although I think column is more common than row[18:29:06] <quicksilver> (that is, it's more common to show coordinates as columns than rows)[18:29:12] <quicksilver> at least in UK/US books[18:29:19] <quicksilver> I vaguely remember the convention is opposite in russia?[18:29:20] <DarkNight> yeh columns generally[18:29:31] <hibread> ie (AxB)transposed == (Atransposed x Btransposed) where x is multiplication[18:29:47] <quicksilver> of course some operations use column and row vectors both[18:29:48] <DarkNight> not quite[18:29:52] <hibread> errr[18:29:53] <hibread> sorry[18:29:55] <hibread> B A[18:30:00] <hibread> forgot to switch[18:30:01] <DarkNight> (AxB)transposed == (Btransposed x Atransposed)[18:30:03] <quicksilver> funky stuff like multiplying two fields together to get a scalr field[18:30:06] <DarkNight> yep :)[18:30:09] <hibread> DarkNight: yeah that :)[18:30:38] <DarkNight> qs: what?[18:31:19] <DarkNight> nvm[18:31:31] <quicksilver> you can multiply together two vector fields to get a scalar field[18:31:34] <quicksilver> dot product[18:31:42] <quicksilver> sometimes that's show by putting one on columns and one in rows[18:31:44] <DarkNight> yeh thought thats what ya meant[18:31:50] <quicksilver> (and you can put transformations in between)[18:31:56] <DarkNight> just confused by your "fields" terminology heh[18:31:58] <quicksilver> ;)[18:32:07] <hibread> Im thinking that there is no difference between these two modes of multiplications... its merely row vs column and transposed vs not transposed....[18:32:08] <quicksilver> a field is a value parameterised by space.[18:32:21] <quicksilver> something which has a value at every x,y,z coordinate[18:32:30] <hibread> you can't multply a 2x2 matrix by a 2x1 row vector[18:32:34] <quicksilver> a vector field is a vector value at every coordinate[18:32:42] <DarkNight> hibread: i'll bring it up in discussion tomorrow with my lecturer[18:32:45] <hibread> but you can multiply a 2x1 row vec by a 2x2 matrix[18:33:08] <quicksilver> hibread: a "2x1 row vec" would normally be called "1x2"[18:33:13] <quicksilver> but yes, exactly[18:34:00] <hibread> so if you're dealing with Row vectors and Row matrices you have to do pABC.. but if you're using column vectors and matrices you have to do CBAp... which results in the exact same thing, just transposed[18:34:02] <hibread> correct?[18:34:10] <quicksilver> well[18:34:17] <quicksilver> p(At)(Bt)(Ct)[18:34:18] <quicksilver> but yes.[18:34:25] *** |t4bz| has joined ##OpenGL[18:34:34] *** t4bz has quit IRC[18:34:44] <quicksilver> t for transpose[18:34:59] <hibread> so p(At)(Bt)(Ct) == (CBAp)t[18:35:01] <hibread> ?[18:35:36] <hibread> or without the t on the CBAp ?[18:36:42] <DarkNight> ok i really gotta sleep. thanks for all the help guys and the wikid big discussion. very "learningful" :)[18:36:45] <DarkNight> night night[18:37:11] <hibread> DarkNight: catchya and good luck![18:37:20] <DarkNight> cheers matey[18:37:21] <hibread> looks like we're not retards after all....[18:37:26] <DarkNight> heh[18:37:38] *** DarkNight has quit IRC[18:38:38] <quicksilver> hibread: well technically you want that t[18:38:50] <quicksilver> hibread: otherwise you have a row on the left of the = and a column on the right of it[18:38:50] *** m4ggus has quit IRC[18:38:58] <quicksilver> hibread: but, of course, what you meant would be understood.[18:39:43] <hibread> oh yep sorry, we had a resultant vector not matrix...[18:41:54] <hibread> quicksilver: in what scenario would say (assuming column vectors) performing ABCp be desired over CBAp... hmmm im not sure what im asking...[18:42:00] <hibread> :)[18:43:00] <quicksilver> never.[18:43:04] <quicksilver> it's the same thing :)[18:43:11] <quicksilver> you can't really tell them apart, typicaly.[18:43:25] <quicksilver> but sometimes it's convenient to use rows for some stuff and columsn for other stuff[18:43:44] <quicksilver> like, I dunno, rows for liquid velocity and columns for acceleration[18:44:19] <hibread> how is it the same thing if A, B and C describe different transformation?[18:45:31] <quicksilver> oh, sorry, misread[18:45:53] <quicksilver> well, it ddepends whether C happened 'before' A or after[18:45:54] <quicksilver> in some sense :)[18:46:35] *** _boto2 is now known as _boto[18:46:39] <hibread> I think its a void statement. I gues what im saying is there is only one way to do the transformation you want...[18:46:58] <hibread> And that depends on whether the api you use pre multiplies or post[18:48:58] <hibread> quicksilver: but err... hmmm if they're column matrics they should always be post multiplied shouldn't they? It doesn't make sense in the scheme of things to have your api pre multiply then finally post multply the vector with the resultant matrix[18:53:56] *** prophile has joined ##opengl[18:55:11] *** speedy1_ has joined ##OpenGL[19:07:56] *** speedy1 has quit IRC[19:22:29] *** aalex has quit IRC[19:22:47] *** amalon has joined ##opengl[19:30:29] *** TheLorax has quit IRC[19:31:30] *** KU0N has quit IRC[19:33:22] *** HuntsMan has joined ##opengl[19:35:44] *** s1d has joined ##OpenGL[19:37:01] *** Lemml has joined ##OpenGL[19:38:53] *** oxmoz has joined ##opengl[19:43:27] *** Walt has quit IRC[19:44:56] *** Tibor__ has quit IRC[19:47:26] *** Walt has joined ##opengl[19:48:34] *** fargiolas has joined ##OpenGL[19:52:15] *** pi-meson has joined ##OpenGL[19:55:00] *** Jorachim has joined ##openGL[19:55:05] <pi-meson> Is there any obvious reason / screwup that would prevent me from seeing GL_POINTs, but not anything else?[19:56:29] <hibread> are you using glBegin( GL_POINTS) ?[19:56:46] <hibread> or how ever you're using points.. GL_POINTS as opposed to GL_POINT[19:56:59] <rsp> Hi hibread[19:57:07] <pi-meson> yup, that was it[19:57:14] <pi-meson> wow, thanks, that was driving me crazy[19:57:25] <hibread> no probs! :)[19:57:30] <hibread> gday rsp![19:57:56] <dv_> GL_POINT vs. GL_POINTS is hideous[19:58:33] <rsp> :D[19:59:08] *** Suprano has quit IRC[20:00:38] *** Suprano has joined ##OpenGL[20:03:51] <Jupp3> There's GL_POINT?[20:04:02] <rsp> Yeah[20:04:09] <Jupp3> Is there also GL_TRIANGLE?[20:04:18] *** elite01 has quit IRC[20:04:27] <Jupp3> Wouldn't be surprised if some people actually prefered those[20:05:21] <Jupp3> You know, those who figure out a really neat object oriented way of drawing objects... In other words, looping each face, and drawing them separately between their own set of glBegin() ... glEnd() with a primitive suitable for amount of vertices[20:05:27] *** Ademan has quit IRC[20:09:37] *** neoneye has quit IRC[20:16:45] *** BahamutZERO has quit IRC[20:18:56] *** Ademan has joined ##OpenGL[20:21:30] *** DobosCake has joined ##OpenGL[20:29:50] *** speedy1_ is now known as speedy1[20:33:58] *** Jorachim has quit IRC[20:38:38] <rsp> What does one mean with shader quality settings a game[20:38:43] <rsp> quality?[20:39:48] <hibread> rsp: its not something like AA or AF if thats what you're thinking[20:39:54] <hibread> that you can turn up and down[20:41:11] <rsp> I know what shaders do but I don't get what quality settings mean[20:41:17] <hibread> a developer may decide to produce a few levels of shaders that require varying levels of horsepower to run well, obviously sacrificing impressiveness[20:41:19] <rsp> ugglier calculations?[20:42:41] <hibread> as a bad example, you might have per vertex lighting for a low quality shader setting or per-pixel for higher..[20:42:57] <rsp> oh[20:43:44] <hibread> I could have that all wrong, though :)[20:43:47] <hibread> im outta ere[20:43:48] <hibread> catchya![20:43:56] <rsp> C ya hibread[20:44:04] *** neoneurone has joined ##OpenGL[20:44:14] *** hibread has quit IRC[20:46:26] *** groton has joined ##OpenGL[20:47:16] *** TurboAWAY has quit IRC[20:48:32] *** Gonzo has quit IRC[20:49:15] *** Gonzo has joined ##OpenGL[21:04:09] *** Xmas| has joined ##OpenGL[21:11:21] *** DobosCake has quit IRC[21:11:57] *** quicksil1er has joined ##opengl[21:11:58] *** quicksilver has quit IRC[21:13:55] *** dv_ has quit IRC[21:14:14] *** Xmas__ has joined ##OpenGL[21:14:51] *** groton has quit IRC[21:17:47] *** mm765^away is now known as mm765[21:19:38] *** Amorphous has joined ##opengl[21:24:19] *** Xmas| has quit IRC[21:25:07] *** BahamutZERO has joined ##OpenGL[21:31:16] *** DobosCake has joined ##OpenGL[21:35:51] *** Amorphous has quit IRC[21:36:39] *** Amorphous has joined ##opengl[21:38:36] *** korff has joined ##OpenGL[21:41:40] *** Sisu has joined ##opengl[21:45:00] *** mattn2|home has joined ##OpenGL[21:45:30] *** DobosCake_ has joined ##OpenGL[21:47:58] *** prophile has quit IRC[21:50:40] *** ahelon has quit IRC[21:53:02] *** DobosCake has quit IRC[21:58:21] *** groton has joined ##OpenGL[21:58:24] *** hydoskee has joined ##OpenGL[21:58:29] <hydoskee> where can I find documentation on anti-aliasing in opengl?[21:59:32] *** groton_ has joined ##OpenGL[22:01:55] *** roche has joined ##OpenGL[22:03:50] <roche> Hello, could anyone help me with an opengl shape creation problem? :)[22:06:19] <tmccrary> No, sorry[22:09:57] *** hydoskee has quit IRC[22:10:27] <groton_> roche, your cube looks like a sphere?[22:12:58] <roche> lol not quite :P[22:14:15] <groton_> or the other way around, dunno[22:15:13] <roche> I have a class for a triangle and im trying to call the triangle but its not drawing anything when i run the program :([22:16:06] <groton_> roche, you are not an english native speaker, aint it? (me neither, btw)[22:17:24] <roche> Lol i am from england, you saying my english is crap? :P My brain is frazzled from all this programming (im a newb at it you see)[22:17:59] <groton_> really? i cant believe that. "I have a class for a triangle" and " call the triangle" make no sense to me[22:18:14] *** groton has quit IRC[22:18:25] <groton_> well, the nitpicker corner could be closed now :)[22:20:11] <roche> I have a triangle class which when called upon further down the program doesnt work. By 'doesnt work' i mean nothing is drawn to the screen when the program is running.[22:20:48] *** rutski has quit IRC[22:20:53] *** DobosCake_ is now known as DobosCake[22:22:16] <groton_> roche, well, you should have prepared a nice exposition of you problem. Then, just pop in here and paste a link to the explanation you wrote. THen wait patiently for an answer.[22:22:42] <groton_> roche, nobody could know why your unknown code does not work :)[22:23:19] <roche> True, but i was didn't want to prepare a big explanation if nobody was online to answer me:P. Just wanted to see if anyone was available to help first[22:23:30] <tmccrary> As I said earlier, no one is available to help[22:23:42] <roche> *don't know whyt i wrote 'was didnt' then :)*[22:24:08] <roche> I figured you were just saying you couldn't help[22:24:15] <groton_> roche, usualy, when one writes down the question, usually gets the answer himself coz he realizes what the problem is. I do it always like that[22:24:18] <tmccrary> No, your problem is impossible to fix[22:24:26] <tmccrary> OpenGL can't draw triangles, never could[22:24:32] *** holtmp6 has joined ##OpenGL[22:24:43] <groton_> roche, ignore tmccrary, he is a bot[22:25:29] <holtmp6> I am using glRotatef to rotate an object. However it rotates counterclockwise... how do i do a clockwise rotation?[22:25:39] <groton_> roche, if you do not write down a clear explanation of the problem cause you fear nobody could help you.... that's bad[22:25:51] <groton_> holtmp6, change the sign of the angle[22:26:00] <holtmp6> i tried doing that...[22:26:27] <tmccrary> What are you supplying for the rotation vector?[22:26:28] <holtmp6> instad of glRotatef(30.0, 0.0, 1.0, 0.0) I used glRotatef(-30.0, 0.0, 1.0, 0.0) and it had the same results[22:28:42] <holtmp6> groton_, do you have any idea what I'm doing wrong?[22:29:04] <groton_> holtmp6, you have not recompiled your program after the change :P[22:29:13] <holtmp6> haha[22:29:14] <holtmp6> i wish[22:29:55] <tmccrary> That should work I believe, although what if you tried inverting the vector?[22:30:07] <tmccrary> I haven't dealt with glRotatef in ages ;)[22:30:32] <holtmp6> ahh, how would I invert the vector?[22:31:04] <groton_> holtmp6, you should learn a bit of math and geometry before trying to do anything with ogl[22:31:26] <holtmp6> I'm sorry brain fart[22:31:37] <holtmp6> for some reason i was thinking the matrix[22:31:43] <groton_> it is just a suggestion[22:32:26] *** Walt has quit IRC[22:34:30] *** fargiolas has quit IRC[22:34:33] <roche> Where exactly can i post/host my problem to link to it? :o[22:34:52] <tmccrary> pastebin[22:34:57] <tmccrary> there are a billion[22:35:11] <roche> ok thanks[22:39:18] <groton_> reading the topic no, eh?[22:46:46] *** amalon has quit IRC[22:47:45] *** LtJax has joined ##opengl[22:48:20] <roche> ok, i think i'v done this correctly. here is the link: http://pastebin.com/m688981d2 Hopefully i've described the problem clearly. Thanks[22:54:52] <groton_> roche, sincerely, it is far from clear :)[22:58:01] <groton_> roche, please, read this short tutorial about GLUT http://lighthouse3d.com/opengl/glut/index.php?2. After studying it, i am sure you will resolve your problem yourself.[22:58:09] *** roche has quit IRC[22:58:31] *** roche has joined ##OpenGL[22:58:57] <roche> Thanks i'll have a read now[23:00:35] <roche> mm doesnt seem to have anything about calling a class.[23:01:44] <groton_> roche, yes, because you should first learn C++ :) You do not call a class, but you call a method of a class[23:02:30] <roche> Yeh i need to learn C++ more but the teacher in my college thinks we are more advanced then we are and has set us opengl instantly[23:03:27] <ville> groton_: and in C++ they are called member functions not methods.[23:04:00] <groton_> ville, the nitpicker corner has been closed already[23:04:37] <ville> It's not a nitpick, it's a fact.[23:05:10] <tmccrary> its definitely stupid though[23:05:13] <tmccrary> same difference[23:06:09] <roche> wellll basically i want to change the gl_triangles at the top of the link i sent you into a triangle class that can have its member functions called upon over and over to create multiple triangles of different sizes. If i'm still not being clear ill see if anyone in my class can do it, thanks for your help :)[23:08:15] *** Walt has joined ##opengl[23:11:08] <holtmp6> so groton_... update on the problem.. The object I was rotating was a white teapot. Turns out it was rotating correctly in the opposite direction. However, I didn't have any point of reference so it was pretty much any optical illusion...[23:11:14] <holtmp6> and yes, I am willing to accept the ragging for that one : )[23:11:49] <groton_> holtmp6, no problem, when you will have done all the possible errors, you could be called an expert :)[23:11:58] *** joakim__ has joined ##OpenGL[23:14:14] <holtmp6> Thanks groton_.. I'm on my way! : )[23:19:11] *** Jorachim has joined ##openGL[23:19:32] *** amalon has joined ##opengl[23:21:01] *** magnet_ has joined ##OpenGL[23:26:28] *** Rangar has quit IRC[23:27:27] *** m4ggus has joined ##opengl[23:29:11] *** joakim_ has quit IRC[23:31:38] *** _boto has quit IRC[23:34:34] *** johndoevodka has joined ##opengl[23:38:24] *** darka has quit IRC[23:38:48] *** darka has joined ##opengl[23:40:45] *** Jorachim has left ##openGL[23:40:51] *** Jorachim has joined ##OpenGL[23:42:47] *** sohail has joined ##OpenGL[23:43:42] *** Lucine has quit IRC[23:45:38] *** rsp has left ##opengl[23:45:39] *** roche has quit IRC[23:47:32] *** Lucine has joined ##OpenGL[23:48:03] *** tmccrary has quit IRC[23:56:35] *** bbeausej has quit IRC