January 2, 2009 
< | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | 31 | >
January 2, 2009 [00:00:05] <yates> Yes, my son?[00:00:07] * MatthiasM slaps Math` with a large sledge hammer[00:00:20] <sparky_> $1300 for 6 months is a lot[00:00:29] <Math`> Jesus?[00:00:37] <RTFM_FTW> sparky indeed[00:00:46] <tmccrary> well, the car is a monstrously fast sports sedan[00:00:49] <MatthiasM> sparky_: not even half here :)[00:00:51] <RTFM_FTW> although in MA its pretty cheap all things considered :P[00:00:59] <tmccrary> The truck has to be classified as a commercial vehicle[00:01:11] <RTFM_FTW> then again the drivers over here are (for lack of a better term) "bat-shit insane"[00:01:23] <sparky_> we were actually very lucky to get the $2172 deal[00:01:27] <sparky_> per year[00:01:36] <RTFM_FTW> and the insurance providers charge appropriately[00:02:12] <RTFM_FTW> heh I actually know someone paying close to $500 US per month for coverage :P[00:02:19] <tmccrary> Drunk driver?[00:02:23] <RTFM_FTW> nope[00:02:33] <cplusplus2> 2880$ for 1 year is not that cheap;)[00:02:35] <RTFM_FTW> new driver in MA[00:02:38] <cplusplus2> at allstar[00:02:43] <cplusplus2> allstate[00:02:56] <sparky_> cplusplus2: of course it's not cheap[00:03:06] <sparky_> but my neighbors are paying $1000 per year for two cars[00:03:08] <cplusplus2> i configured with a new audi s5 ;)[00:03:11] <sparky_> that covers both of em[00:03:16] *** grosours has joined ##OpenGL[00:03:21] *** barbar__conan has joined ##OpenGL[00:03:33] <sparky_> and I was going to be charged more than 4 times that amount :)[00:03:38] <sparky_> now it's only x 2[00:03:41] <sparky_> so I'm happy[00:03:42] *** Renderwahn has quit IRC[00:03:43] *** a_nab2 has quit IRC[00:04:41] <yates> this texture mapping is a pain in the ass[00:04:55] <yates> a bazillion options and configurations...[00:04:59] <cplusplus2> yates do u have the Platinum Package?[00:05:02] <tmccrary> My first car was one of these: http://xtreme-powersports.net/images/28003%20-%2098-02%20Pontiac%20Firebird_Trans%20Am%20Raptor.jpg[00:05:10] <tmccrary> it was around $300/mo[00:05:10] <sparky_> yates: damn the bastard who invented it yeah? :)[00:05:13] <tmccrary> for insurance[00:05:19] <RTFM_FTW> tmccrary haha awesome[00:05:24] <yates> sparky_: damn right![00:05:27] <cplusplus2> yates do you have platinum package with Enhanced Plus!?[00:05:41] <yates> cplusplus2: i have a pretty nice package, yes...[00:06:05] <MatthiasM> yates: look at shaders - even more posibilities[00:06:15] <tmccrary> audis, shh, fear the V: http://www.cadillac.com/09ctsv/[00:06:28] <tmccrary> Nürburgring® Lap Time of 7:59.32. The Fastest V8 Production Sedan in the World.[00:06:33] <yates> MatthiasM: didn't you say walk, then run the other day?[00:06:52] <MatthiasM> yep - but don't complain about too many possibilities yet :DD[00:06:58] <RTFM_FTW> hell you can simulate all of the standard GL wrap modes in the programmable pipeline[00:07:14] <yates> oh well, yeah.. ? .. ![00:07:16] <yates> hell[00:07:20] <RTFM_FTW> yep[00:07:22] <RTFM_FTW> heh[00:07:23] <cplusplus2> yates does allstate insurance have a fully comprehensive insurance?[00:07:32] <cplusplus2> that covers all[00:07:47] <yates> cplusplus2: they might, but you'd have to fart loudly to get it[00:08:21] <cplusplus2> if u have a brad new car it makes sense to take it[00:09:48] <tmccrary> You mean full coverage + broad form?[00:09:57] <yates> brad's new car has nothign to do with it[00:10:01] <cplusplus2> croad form?[00:10:05] <tmccrary> Broad form adds a lot and deductible isn't that much[00:10:22] *** sparky_ is now known as sparky[00:10:53] <tmccrary> Because if you mean full coverage, unless you're paying cash for your new vehicle, your bank will make you maintain full coverage[00:11:02] <yates> my wife wonders if she needs broad form[00:11:41] <cplusplus2> yates what happens if someone is scratches your car and escape somewhere?[00:11:44] <sparky> bloody hell, $500 per month for car insurance[00:11:47] <yates> why are we discussing car insurance?[00:11:54] <sparky> that money could get you health insurance[00:11:57] <cplusplus2> does your insurance pay for that?[00:12:04] <MatthiasM> yates: you are now in ##car-insurance :)[00:12:15] <yates> oh! stupid me.[00:12:24] <sparky> yates: because the rest of us have already learned everythnig there is to know about opengl[00:12:30] <sparky> we rock[00:12:35] <RTFM_FTW> sparky yeah no shit[00:12:38] *** k3t_ has joined ##OpenGL[00:12:41] *** k3t has quit IRC[00:12:44] <cplusplus2> yates what happens if someone is scratches your car and escape somewhere? does your insurance pay for?[00:12:44] <sparky> RFTM: shhhh! :)[00:13:10] <tmccrary> cplusplus2: make a claim, they'll fix and you own the deductible[00:13:12] <ville> Feel free to off-topic somewhere else.[00:13:27] <tmccrary> here comes ville's douchebaggery again[00:13:32] <yates> cplusplus2: it depends on how many beans you ate the night before.[00:13:40] <cplusplus2> ;)[00:13:41] <sparky> feel free to discuss on-topic opengl related stuff somewhere else![00:13:43] <RTFM_FTW> don't worry sparky I was actually referring to the "that money could get you health insurance" comment :P[00:13:57] <sparky> oh yeah, crazy isn't it?[00:14:01] <RTFM_FTW> indeed[00:14:08] <sparky> I think that's actually what sony's paying for me and my wife[00:14:17] <sparky> and sony has really good health insurance[00:14:28] <RTFM_FTW> yeah mine is pretty close to that[00:14:36] <tmccrary> You don't have to pay into your health? I have to pay like 30 bucks or something[00:14:46] <tmccrary> its blue cross blue shield[00:15:01] <tmccrary> I forget what california is, Kaiser Permenante or something like that[00:15:01] <sparky> tmc: yeah I do[00:15:08] <sparky> I think I pay about $83[00:15:11] <sparky> sony pays the rest[00:15:51] <sparky> "BlueCross PPO prudent buyer"[00:15:54] <sparky> is the one I have[00:16:11] * yates wants to mediate dog fights.[00:16:14] <sparky> covers 90% in network and 80% out of network[00:16:16] <tmccrary> Did they ask you to do a physical examination this year?[00:16:21] <sparky> nope[00:16:31] <tmccrary> They asked us to do it, I told them to fuck off[00:16:58] <RTFM_FTW> right on[00:17:28] *** barbar__conan has quit IRC[00:18:02] <sparky> is it intended as a way to maintain your health or is it intended as a way to blame you for something and not pay up when you need the coverage?[00:18:24] <RTFM_FTW> likely the latter[00:18:28] <tmccrary> The insurance people said it wouldn't change our rates, but I was like yeah right[00:18:41] <yates> my last health insurance application asked if I was an OpenGL programmer - a type of stress test, i guess...[00:18:42] <RTFM_FTW> yep[00:18:53] <MatthiasM> heh[00:19:00] <sparky> I think we all have the same rate[00:19:09] <sparky> at least they appear to be fixed[00:19:13] *** LordHavoc has joined ##OpenGL[00:19:28] <yates> come, reap, LordHavoc.[00:20:34] <yates> i was fixed a few years ago...[00:20:35] <LordHavoc> hi[00:20:45] * sparky slaps LordHavoc around a bit with a large trout[00:20:49] <sparky> welcome[00:21:18] * yates observes the sign on the OpenGL urinal wall: "We aim to please, you aim too, please."[00:21:35] <sparky> ha ha[00:21:40] <yates> he he[00:21:50] <kurt_> hehe[00:22:14] <sparky> well, I'm no Clint Eastwood but I'll give it a shot[00:22:42] <yates> ha[00:23:37] * yates wonders if you heard the one about the plastic surgeon that hung himself...[00:23:38] <sparky> one thing I really don't get is the obsession with in network vs. out of network[00:23:47] <sparky> I understand the principle[00:24:02] <sparky> those out of network don't accept the limit on the prices[00:24:07] <sparky> they're not under contract[00:24:20] <sparky> however, in practice the insurance company is going to cap it anyway[00:24:30] <yates> sparky: fascinating.[00:24:35] <sparky> and not pay about the in network price and leave the rest to you[00:24:47] <sparky> so why the fuck do they care if they are not going to pay anyway[00:25:03] <sparky> but nevertheless they will cover 90% when it's in network and only 80% when it's out of network[00:25:14] <sparky> which tells me they are paying even less when it's out of network[00:25:16] *** Yustme has quit IRC[00:25:53] <sparky> and when I asked they said we don't want you to use "out of network". But they couldn't tell me why the fuck it matters when they cap it anyway at the price they are willing to accept[00:26:10] <sparky> it's a fucked up system here in the US[00:26:23] <RTFM_FTW> yes it is[00:26:42] <sparky> plus what pisses me off even more is that even the doctor has no clue when he's guiding you and telling you want to do what the financial consequences will be for you[00:26:49] <sparky> and obviously you don't know either[00:27:04] <sparky> you just get bills a month later[00:27:07] <cplusplus2> sparky> it's a fucked up system here in the US ?[00:27:58] <sparky> the doc goes, I'd like to do this and that test on your blood[00:27:59] <tmccrary> I few months ago, i was in the hospital for an hour[00:28:03] <tmccrary> $5,000 bill[00:28:08] <sparky> yikes[00:28:18] <tmccrary> My deductible was $50[00:28:22] <RTFM_FTW> yeah just walking into an ER is ~1500.00 or more[00:28:23] <LordHavoc> tmccrary: sounds typical...[00:28:32] <RTFM_FTW> US[00:28:33] <tmccrary> but I was like good god, if I didn't have insurance I would have been so screwed[00:28:39] <RTFM_FTW> yep[00:28:40] <cplusplus2> tmccrary oh[00:28:40] <sparky> tmc: why didn't your insurance cover?[00:28:48] <tmccrary> it did, I only paid the deductible[00:28:52] <LordHavoc> insurance is the only reason the health care system in the US hasn't bankrupted every citizen[00:28:54] <yates> Happy New Year, folks![00:28:56] *** yates has quit IRC[00:29:02] <cplusplus2> $5,000 bill ... because alle the doctores got a lot of money?[00:29:04] <sparky> oh, so the bill was $5000[00:29:08] <sparky> but of that you only paid $50?[00:29:12] <tmccrary> right[00:29:40] <sparky> but it's frikken crazy, the doc is like, I'd like to give you this test and send this and that to the lab and try your bla bla bla on this and that machine[00:29:42] <sparky> and voila[00:29:53] <sparky> a month later you know if he managed to royally screw you[00:29:59] <sparky> without him even knowing it most likely[00:30:21] <sparky> it's very unusual to have a guy offer you services without telling you the price[00:30:23] *** HuntsMan has quit IRC[00:30:39] <sparky> I think the healthcare industry here in the US is the only example I know of this weird concept[00:31:02] <sparky> imagine if it were like: would you like this nice car? oh sweet sure[00:31:07] <tmccrary> yeah, its a big scam, someone is making serious cash[00:31:07] <cplusplus2> how much do you pay for healthcar per year?[00:31:11] <sparky> one month later, that'll be $80k please[00:31:35] *** twist has quit IRC[00:31:37] <LordHavoc> sparky: the dentist I go to DOES know the prices[00:31:57] <sparky> Lord: well it can be tricky because they don't know by heart all the diff insurances[00:32:01] <LordHavoc> sparky: the numbers are frightening but at least I know them in advance[00:32:09] <LordHavoc> sparky: I have no insurance[00:32:17] <sparky> well that does simplify it for you[00:32:23] <tmccrary> A few years ago, I had to get a root canal because a filling I had was messed up[00:32:27] <cplusplus2> sparky how much do you pay for healthcar per year?[00:32:30] <LordHavoc> sparky: I meant that the dentist knows the prices the labs charge, etc[00:32:40] <tmccrary> The dentist basically said, I can pull it for cheap or you can get a root canal for $1k[00:33:21] <tmccrary> The root canal was like 15 mins[00:33:30] <tmccrary> That dentist has to be doing pretty well[00:33:52] *** yates has joined ##OpenGL[00:33:57] <tmccrary> Luckily, my dental coverage paid for like $850 or something[00:34:10] <LordHavoc> the prices are just insane, in an absolute sense[00:34:16] <LordHavoc> the insurance just hides it[00:34:22] *** sparky_ has joined ##OpenGL[00:34:56] <cplusplus2> sparky how much do you pay for healthcar per year?[00:35:38] <hibread> The US has far more real issues than the ones youse put forward with the dentist example[00:35:40] <yates> when you do glTexImageiD() and subsequent glEnable(GL_TEXTURE_iD), is that for the currently active texture, a-la glActiveTexture(GL_TEXTURE0)?[00:36:14] <LordHavoc> yates: yes[00:36:39] <LordHavoc> yates: glClientActiveTexture affects only the gl*Pointer calls and glEnableClientState's[00:36:55] <SinnerG> w00t engine getting better :)[00:37:11] <SinnerG> now it supports changing to fullscreen (and it'll auto reload all textures that were in use)[00:37:25] <LordHavoc> yates: note that glEnable(GL_TEXTURE_ affects only the fixed function rendering path[00:37:30] <LordHavoc> yates: shaders ignore it entirely[00:37:48] <LordHavoc> yates: just like they ignore glTexEnv[00:38:13] *** sparky has quit IRC[00:38:17] <LordHavoc> there are less enables and texenv states than there are image and texcoord units on shader-capable hardware[00:38:30] *** itewsh has quit IRC[00:38:45] <LordHavoc> (fixed function is on life support at this point, it just does the basic features and leaves it at that)[00:39:18] <LordHavoc> I can only imagine how much pain it is to write an OpenGL driver for shader-capable hardware, because of all the fixed function translation[00:39:49] <yates> LordHavoc: i got about 80 percent of that, which is probably enough. thanks. (fixed-function is fine for my simple usage)[00:40:22] *** _sparky has joined ##OpenGL[00:40:31] <_sparky> so like I said, getting offered some product or service without getting told what it will cost you is fucked up and I don't think I've seen that crap in any other industry[00:40:33] *** iion has quit IRC[00:40:35] *** _sparky is now known as sparky[00:40:57] *** yates has quit IRC[00:41:00] <sparky> oops[00:41:05] <sparky> wth happened[00:41:08] *** Math` has quit IRC[00:41:12] <sparky> so anyway, it bites![00:41:13] <sparky> :)[00:41:14] <tmccrary> yeah, especially when its more like (Would you like to do X? You might die if you don't... but I don't know how much it costs)[00:41:49] <RTFM_FTW> LordHavoc its not fun[00:42:09] <sparky> what is it with LordHavoc?[00:42:25] <RTFM_FTW> "I can only imagine how much pain it is to write an OpenGL driver for shader-capable hardware, because of all the fixed function translation"[00:42:33] <RTFM_FTW> it isn't fun at all :P[00:42:40] <RTFM_FTW> its a gigantic PITA[00:43:57] *** MiniCow has quit IRC[00:44:27] *** glcrazy has quit IRC[00:45:35] <tmccrary> So is the fixed functionality emulation typically some kind of added code path in the driver or just a set of shaders used internally?[00:46:03] <RTFM_FTW> FF on anything near current generation is done through shaders[00:46:52] <RTFM_FTW> so it would make sense to have some sort of "table" based upon the current state which will (re)generate and cache shaders used for that translation[00:47:01] <RTFM_FTW> which is usually how its done[00:47:41] *** MouseCrazy has joined ##OpenGL[00:52:03] *** sparky_ has quit IRC[00:52:11] <LordHavoc> I find it kind of novel how Quake3 managed to encode an entire rendering state of enables and blendfunc parameters and such into one 32bit int as bit values[00:52:31] <LordHavoc> obviously for a very narrow subset, but interesting[00:52:57] <RTFM_FTW> bit-fields are hella useful[00:53:02] <RTFM_FTW> heh[00:53:06] <LordHavoc> RTFM_FTW: here's one question I have about a feature that I use which has been dropped in OpenGL 3.0 - glPolygonStipple[00:53:19] <LordHavoc> I know how to emulate it in shaders, I just find it interesting that it was dropped[00:53:20] <RTFM_FTW> thank fucking god[00:53:27] <MatthiasM> LordHavoc: use a texture and glTexGen[00:53:35] <RTFM_FTW> stipple sucks[00:53:36] <MatthiasM> or emulate it in a shader[00:53:37] <RTFM_FTW> period[00:53:39] <LordHavoc> MatthiasM: or use the window coordinates in the shader[00:53:59] <RTFM_FTW> and doing it via shaders isn't pleasant either[00:54:18] * sparky slaps Spark around a bit with a large trout[00:54:29] <LordHavoc> it's only used for one quirky feature in darkplaces engine - fake interlacing, I laughed at the guy who contributed it but accepted it[00:54:40] <RTFM_FTW> heh[00:55:24] <MatthiasM> LordHavoc: maybe render just a pattern over the result[00:55:32] <MatthiasM> to get a similar look[00:55:57] <MatthiasM> using a 2 texel texture[00:56:31] <tmccrary> I'm dealing with bitfields as we speak, but for network purposes[00:56:48] <tmccrary> when you're dealing with really large bitfields, they can be bitch when something is wrong[00:57:12] <RTFM_FTW> yep[00:59:21] <LordHavoc> MatthiasM: yeah, and they would likely perform better on ATI[00:59:42] <LordHavoc> MatthiasM: but it was meant as a joke feature[00:59:54] <LordHavoc> MatthiasM: much like v_psycho makes the gamma ramps go psychedelic[01:00:28] <LordHavoc> MatthiasM: (which stopped working on Windows 2000 and up, but works on everything else - good thing there's GLSL gamma ramp emulation in the engine now)[01:04:07] *** jcazevedo has joined ##OpenGL[01:06:23] *** OVitamiNO has quit IRC[01:06:47] *** pietia has quit IRC[01:07:19] *** HuntsMan has joined ##opengl[01:16:28] *** Candy has quit IRC[01:18:08] *** HuntsMan has quit IRC[01:27:32] *** sparky_ has joined ##OpenGL[01:28:15] <SinnerG> argh ran into an issue.. I got 100% cpu usage atm, and I see most times to prevent that, is to use a sleep to even out the 'renders' (so you dont draw 40 times first 100ms and then nthing)[01:28:50] <SinnerG> atm it just 'skips' the rendering since I got other stuff (for example processing events, calculating stuff) seperae from rending[01:29:12] <SinnerG> dont know how to solve this one.. I could add 1 thread.sleep to ease things a bit I guess[01:30:14] <MatthiasM> ??[01:32:40] <SinnerG> best way I can explain it sry ;p[01:41:03] <kurt_> kind of a race condition?[01:42:11] *** wisey has joined ##OpenGL[01:44:52] *** sparky has quit IRC[01:46:26] *** cplusplus2 has quit IRC[01:47:50] <SinnerG> more like a design flaw I fear.. I have 2 seperate counters[01:47:55] <SinnerG> FPS & RPS (runs per second)[01:48:10] <SinnerG> each 'run' the states (different parts of the game) gets 'ran'[01:48:22] <SinnerG> like 100RPS == each 10 'max' it will call the 'think' function[01:48:42] <SinnerG> problem is that I need to squeeze a sleep in there somewhere[01:48:51] <SinnerG> so I dont get 100% cpu usage[01:49:08] <SinnerG> if I do it @ the rendering part, it breaks up the RPS (aka 60FPS == 60RPS not more)[01:49:26] <SinnerG> if I do it @ rps I break the FPS (it'll be less FPS)[01:49:36] <SinnerG> pff[01:49:41] <tmccrary> hey[01:53:49] *** HuntsMan has joined ##opengl[01:56:24] *** tokemonstah^ has quit IRC[02:02:04] <MatthiasM> SinnerG: why do you need an 1:1 mapping between rps and fps ?[02:02:25] <MatthiasM> rps should be constant - and fps <= rps[02:03:08] <MatthiasM> and if the system can't handle your required rps - then you get a slow down[02:04:02] <MatthiasM> it helps if you organise your app like a multipler app - rps on server and fps on client - using some pip/socket to connect[02:04:02] *** calav3ra has quit IRC[02:04:09] <RTFM_FTW> that is where you log the following message to the console: Alright fucker... Buy a faster box... Seriously!!![02:05:56] <MatthiasM> heh[02:07:26] <SinnerG> hehe[02:07:42] <SinnerG> matthias thats what I want & got[02:07:47] <SinnerG> BUT with 100%cpu sage[02:07:51] <SinnerG> <MatthiasM> rps should be constant - and fps <= rps[02:07:52] <SinnerG> this part[02:08:11] <SinnerG> the thing is, that I need to put a sleep in there somewhere[02:08:16] <MatthiasM> well - when server and client runs on the same system - give both the same priority[02:08:22] <SinnerG> without breaking my fps/rps system[02:08:36] <MatthiasM> you don't need sleep - use a OS with preemptive multitasking[02:08:53] <MatthiasM> the fps renders when it got data from the rps[02:09:06] <MatthiasM> if it does not get data it will sleep on the socket[02:10:02] *** rutski_ has joined ##OpenGL[02:10:05] <SinnerG> well, cant fully understand where a server coems in the picture, but an idea is growing lol :P[02:10:17] *** TiLex has joined ##opengl[02:11:35] <MatthiasM> the server runs the game logic[02:12:18] <SinnerG> I know but I'm not working on that (yet) - I'm just working on a simple engine to draw 2D (and animations and such)[02:12:48] <SinnerG> bit older version of my demo app's 'main'[02:12:50] <SinnerG> http://sinnerg.nopaste.dk/p522[02:12:53] <MatthiasM> if it is multithreaded - then use client/server approch[02:13:08] <MatthiasM> if not then call the game logic update until enough tick have runs per frame[02:13:47] <SinnerG> doing something like that atm (check the code)[02:14:06] <SinnerG> if (VideoDriver.BeginRender(Color.Black)) // returns false if not rendering yet[02:14:18] <SinnerG> with this I get a stable FPS BUT I still need to add a sleep somewhere[02:14:39] <SinnerG> normally I would lock RPS == FPS because it is most simple (since I can just sleep the 'skip' time)[02:15:05] <SinnerG> but for some reason, I wanted to make it hard to myself ;p[02:15:46] <MatthiasM> you need to keep rps constant or you get issues with physics etc[02:16:11] <MatthiasM> that does not mean that you can't run less the the desired updates per second realtime - it will just slow down yopur game time then[02:16:46] <SinnerG> rps is constant with my current code[02:16:53] <SinnerG> RPS / FPS gets 'smoothed'[02:17:10] <SinnerG> so its split fairly over the full second[02:17:20] *** RTFM_FTW has quit IRC[02:17:31] *** RTFM_FTW has joined ##OpenGL[02:19:54] *** dv_ has quit IRC[02:20:12] <SinnerG> if i use a sleep @ rendering I get a stable 60 FPS/RPS with only 8% cpu usage[02:20:26] <SinnerG> remove the sleep, and switch to the skip method its still stable, but 100% cpu usage[02:21:34] <MatthiasM> you use one or two threads ?[02:21:51] <SinnerG> atm only one[02:22:13] <MatthiasM> so what's the issue with 100% cpu ?[02:22:18] <MatthiasM> enable vsync[02:23:23] <SinnerG> it is enabled, but that doesnt solve the issue[02:23:30] <SinnerG> (if I disable it I can get 400+ fps :P)[02:24:27] <SinnerG> vsync only helps if configured FPS > vsync rate[02:24:45] <SinnerG> (just tried it, if i increase FPS from 60 to 80 I actually lower cpu usage ;p)[02:24:45] <MatthiasM> if you get 100% cpu with vsync then get a better driver[02:24:59] <SinnerG> <SinnerG> vsync only helps if configured FPS > vsync rate[02:25:09] <SinnerG> since if its lower it isnt hitting the vsync limit[02:29:21] <SinnerG> maybe seperating things in 2 threads is only way for this[02:29:47] <SinnerG> I could just run them both (the RPS thread & rendering one) and sleep to even out their 'framerate'[02:29:54] <SinnerG> but ofc easier said than done[02:30:45] <MatthiasM> you could test the vsync frequency - and if it's very close to your target rate - just use the vsync rate[02:33:47] *** joe-coder has joined ##OpenGL[02:33:55] *** xp_prg has joined ##OpenGL[02:34:20] <xp_prg> hi all, I need to do fluid visualization of protein gradient diffusion, can anyone help me to know how to do this?[02:35:02] <joe-coder> xp_prg: you'll do better asking more specific questions about opengl rendering[02:35:19] <joe-coder> I'm not sure about the others in this channel, but I don't have much of a background in microbiology/chemistry[02:35:50] <joe-coder> perhaps you'd like links to some opengl tutorials?[02:36:06] <joe-coder> if so, see the channel topic.[02:36:08] <xp_prg> well imagine dropping a droplet of red dye in water, it would begin to spread out and dissipate etc...[02:36:12] <xp_prg> does that make sense?[02:36:16] <joe-coder> yes[02:36:31] <joe-coder> you could probably render it with a bunch of sprites?[02:36:32] <xp_prg> that is what I want to visualize[02:36:40] <joe-coder> (textured, translucent quads that always face the camera)[02:37:07] <joe-coder> create a few thousand of them, small enough that they look like fluid in water[02:37:20] *** TiLex has quit IRC[02:37:29] <SinnerG> ok think I got an idea - lucky for svn - I can always revert ^^[02:38:04] <xp_prg> joe-coder I was hoping there were some pre-written libraries for assisting me with such a visualization, do you know of any?[02:39:10] <joe-coder> nope, but i've never looked[02:39:44] <joe-coder> I do have a question of my own, if anone knows: I'm combining a position, rotation, and scale matrix into a single transform matrix and then passing that to opengl. however, it always applies the scale after the rotation (instead of before).[02:39:50] <RTFM_FTW> heh I've fooled around with Navier-Strokes on the GPU once or twice :D[02:39:51] <tmccrary> There were some awesome fluid simulation demos from nvidia earlier in the year[02:40:02] <tmccrary> Probably no source though[02:40:14] <joe-coder> and no matter what order i multiply my position/rotation/scale matrices, this always happens.[02:40:23] <RTFM_FTW> NS is pretty old hat (at least for the GPGPU guys)[02:41:00] <tmccrary> here we go[02:41:01] <tmccrary> http://www.cs.caltech.edu/~keenan/project_fluid.html[02:42:34] <MatthiasM> joe-coder: then your matrix math is wrong[02:42:47] <joe-coder> i'll do some prints and see what's going on[02:45:42] <tmccrary> its too bad we don't have any use for fluid sim on the current project, it'd be mess around with[02:45:46] <tmccrary> *cool to mess around with[02:48:44] <RTFM_FTW> yeah it is[02:56:06] *** Morphous has quit IRC[02:59:15] *** Morphous has joined ##opengl[03:11:33] <SinnerG> I think I got it :)[03:11:53] *** Walt has quit IRC[03:12:44] *** wisey has quit IRC[03:13:29] *** wisey has joined ##OpenGL[03:13:34] *** walt has joined ##OpenGL[03:15:12] *** Ashen has joined ##OpenGL[03:15:30] <Ashen> hello[03:15:48] <Ashen> is there any way to disable direct context rendering on a window i spawn?[03:16:24] <HuntsMan> direct context rendering?[03:16:59] <Ashen> i want to use the cpu to render the window[03:17:19] <HuntsMan> you mean direct rendering in GLX[03:17:28] <Ashen> yes[03:17:51] <HuntsMan> try the mesa software rendering proyect[03:18:16] <HuntsMan> or disable direct rendering at context creation[03:18:39] <Ashen> i tried that with the -indirect flag as an argument to glutinit[03:20:21] *** qeed has quit IRC[03:21:46] *** wisey has quit IRC[03:22:27] <Ashen> so maybe direct/indirect rendering isn't my problem --[03:23:09] <Ashen> the window of my opengl/glut application is all flashy and distorted with the original pixels from the area of the screen that it covers[03:23:51] <HuntsMan> did you clear the buffer?[03:24:14] <Ashen> i'm not sure i understand[03:24:16] <Ashen> how do i do that?[03:24:37] *** Xmas| has quit IRC[03:25:07] <HuntsMan> glClear(GL_BUFFER_BIT);[03:25:11] <Ashen> k[03:25:13] <Ashen> i'll try that[03:25:45] <Ashen> so, in my display handler, i have this:[03:25:52] <Ashen> glClear(GL_COLOR_BUFFER_BIT | GL_DEPTH_BUFFER_BIT); // erase the window[03:25:59] <Ashen> should i also add GL_BUFFER_BIT to that?[03:26:22] <Ashen> actually, that constant isn't found[03:26:28] <HuntsMan> no, it's there, my mistake :)[03:26:40] <Ashen> oh okay[03:27:48] <Ashen> an interesting thing happens when my window is animating[03:28:02] <Ashen> it seems to flash between the distorted buffer contents and the correct screen contents[03:28:08] <Ashen> but when i stop animating it's back to distorted[03:28:13] <MatthiasM> Ashen: use a double buffered context[03:28:16] <Ashen> i do[03:28:30] <MatthiasM> do you call glutSwapBuffers ?[03:28:31] <Ashen> and this code has worked on another computer[03:28:32] <Ashen> yes[03:28:42] <SinnerG> now FPS is stable, RPS is bit less than it it is configured (100 = 96 RPS, 50 = 49 RPS)[03:28:44] <MatthiasM> did you call glDrawBuffer ?[03:28:46] <SinnerG> I call that decent enough[03:29:03] <SinnerG> ow yeah, cpu usage is OK :)[03:29:27] <SinnerG> using <10% + it is a bit multithreaded :p[03:30:20] <Ashen> MatthiasM: i don't, what's the proper place for that call?[03:30:34] <MatthiasM> that's ok - you don't need to call it[03:30:38] <Ashen> i have glFlush[03:30:43] <MatthiasM> Ashen: can you paste your code ?[03:30:49] <MatthiasM> don't use glFlush[03:31:33] <Ashen> i'll paste the code, hold[03:31:45] *** AlastairLynn has quit IRC[03:32:24] <Ashen> http://pastebin.com/m7cdae94a[03:33:17] <MatthiasM> Ashen: you are calling swap buffers twice[03:33:35] <MatthiasM> after a swap the content of the back buffer is undefined[03:33:46] <Ashen> hm[03:33:48] <Ashen> aha[03:33:53] <MatthiasM> and you clear also twice[03:33:54] <Ashen> so i'll comment out the swapbuffers in displayhandler[03:34:16] <MatthiasM> you do a lot of redundant stuff[03:34:21] <SinnerG> nice nice :D[03:34:25] <SinnerG> i'm happy :)[03:34:31] <SinnerG> rps isnt rock stable but its very close :p[03:34:54] <Ashen> MatthiasM: i'll try to prune it out[03:35:04] <Ashen> the display problems stopped when i got rid of the second swapbuffers[03:35:05] <Ashen> thank you :)[03:35:49] *** MatthiasM has quit IRC[03:35:55] *** MatthiasM has joined ##opengl[03:37:03] <SinnerG> sadly enough my solution makes my engine bit more messier, but still.. I solved it quite simple :p[03:37:17] <SinnerG> StateRunner runner = new StateRunner(States);[03:37:17] <SinnerG> runner.Start();[03:37:29] <SinnerG> it spawns a new thread, and yes it has the required locking[03:39:30] *** mads- has quit IRC[03:40:32] <SinnerG> lol it can do 800k runs per second now :P[03:40:51] <Ashen> MatthiasM: the interesting part is that my error didn't show up when direct rendering was off[03:41:25] <MatthiasM> Ashen: because "undefined" may also mean does not change - or it may mean becomes totaly garbage[03:41:33] <MatthiasM> or anything between[03:41:40] <Ashen> ah[03:41:43] <MatthiasM> it may also be the previous frame[03:41:47] <Ashen> wonderful[03:41:49] <MatthiasM> it's simple "undefined" :DD[03:41:51] <Ashen> well either way[03:41:55] <Ashen> thanks :)[03:45:33] *** bijoo_osdev has joined ##opengl[03:45:33] *** bijoo_appdev has joined ##OpenGL[03:52:57] *** k3t_ is now known as k3t[03:56:28] *** eXtronuS has quit IRC[03:59:57] *** yates has joined ##OpenGL[04:00:35] *** k3t_ has joined ##OpenGL[04:00:56] <yates> ok all i'm getting is a white image when i tried implementing the 1d texture. code is here http://galois.digitalsignallabs.com/wxwaterfall/[04:01:06] *** Ashen has left ##OpenGL[04:01:17] <SinnerG> there made some changes, now it can closely match 1000RPS :) (1000 RPS / 1026)[04:01:20] <SinnerG> using nanoseconds now :P[04:01:26] <yates> see InitTextureMap[04:02:07] <yates> do i have to initialize the glPixelStore[04:03:15] <yates> and do i have to set all the glTeXParameter values? or can they default[04:03:38] <MatthiasM> yates: white basicly means the texture is invalid - check for glGetError[04:03:50] *** hang3r has joined ##OpenGL[04:04:00] <MatthiasM> you need to set texture filtering[04:04:12] <MatthiasM> it's a gltexParameter[04:04:34] <MatthiasM> and glPixelStore should also be set - but this is a global setting - not per texture[04:04:34] *** k3t_ has quit IRC[04:05:04] <MatthiasM> esp the alignment parameter - mostly you want to set it to 1[04:05:16] <RTFM_FTW> make sure that COLOR_MAP_SIZE is a power of two[04:05:21] <yates> it is[04:05:28] <HuntsMan> yates: looks you never created a valid texture ID using glGenTextures[04:05:31] <RTFM_FTW> unless you have support for ARB_texture_non_power_of_two[04:05:58] *** urnotE` has quit IRC[04:06:31] <MatthiasM> yates: you should use glGenTexture and glBindTexture[04:06:44] *** k3t_ has joined ##OpenGL[04:06:45] <yates> HuntsMan: that's not in the pyramid example in the bible[04:06:56] *** dvoid_ has quit IRC[04:06:59] <MatthiasM> but in this case I expect that wrong GL_MIN_FILTER settings[04:07:11] <yates> let me get the error and see what's going on[04:07:20] <RTFM_FTW> using hard coded texture IDs is never a good idea :P[04:07:44] <HuntsMan> RTFM_FTW: i don't see any glBindTexture also :)[04:08:11] <RTFM_FTW> furthermore use gluErrorString to make the error(s) retrieved via glGetError sensible[04:08:25] <yates> RTFM_FTW: where have i hard coded ?[04:08:31] *** k3t_ has quit IRC[04:08:54] *** k3t_ has joined ##OpenGL[04:09:27] <RTFM_FTW> I'd also recommend using shaders since it will simplify your code significantly[04:10:05] *** bijoo_osdev has quit IRC[04:10:05] *** bijoo_appdev has quit IRC[04:10:47] <RTFM_FTW> in any case for textures the usual process is: glGenTextures -> glBindTexture -> [setup texture state etc.] -> glTex[Sub]Image -> [render][04:11:15] <RTFM_FTW> with the GenTextures step and (possibly) the texture state setup being done at initialization time[04:11:18] *** amz has quit IRC[04:11:26] <RTFM_FTW> or as needed[04:12:12] *** joe-coder has quit IRC[04:12:16] <yates> aren't shaders an advanced topic?[04:12:20] <yates> have mercy - i've been writing opengl code for 1 1/2 weeks now...[04:12:32] <MatthiasM> yates: think of shaders as an extension of your program[04:12:38] <RTFM_FTW> actually shaders are a good place to start[04:12:56] <RTFM_FTW> because learning the fixed function GL pipeline is a dead end[04:13:09] <MatthiasM> you should get basicl yOpenGL running with a single texture - but don't try to get multitexturing with the fixed function pipe going - it only creates headaches[04:13:30] <RTFM_FTW> and everything that path shows you can be done *more trivially* via the programmable path[04:13:51] <MatthiasM> so basic gl -> textured triangle -> textured something bigger -> use shaders for simple texturing -> use more complex shaders[04:14:01] <RTFM_FTW> hell start with rendering a single texture mapped primitive[04:14:10] <RTFM_FTW> be it a quad, triangle, ...[04:21:35] <yates> is the "texture handle" returned by glGenTextures supposed to be used in glTeXImage1D()?[04:21:54] <MatthiasM> yates: no - only in glBindTexture and glDeleteTextures[04:21:55] <yates> i mean the second parameter in glGenTextures..[04:22:05] <MatthiasM> yates: http://www.opengl.org/sdk/[04:22:25] <MatthiasM> see under reference pages[04:22:55] <yates> yeah, but i just don't see the why this is needed.[04:22:57] <yates> what does it do?[04:23:12] <MatthiasM> it allows you to have more then one texture[04:23:18] <MatthiasM> read about glBindTexture[04:24:36] <yates> why do i need anything other than the default texture (handle 0)?[04:24:52] <HuntsMan> 0 handle means no texture[04:25:09] <MatthiasM> using the default one is not good - esp when you want to use more then one texture[04:25:17] <yates> "Teh value zero is reserved to represent the default texture for each texture target" - the bible[04:25:26] <yates> i don't want to use more than one texture[04:25:35] <MatthiasM> yep - that GL 1.0 - do it the 1.1 way[04:25:41] <MatthiasM> doesn't matter[04:25:51] <MatthiasM> learn it right from the beginning[04:25:57] *** urnotE has joined ##OpenGL[04:26:24] <MatthiasM> if you create a texture class which manages it's texture ID - you can easily create more then one texture[04:26:34] <MatthiasM> you just bind the texture ID and use it[04:26:50] *** k3t has quit IRC[04:26:57] *** k3t_ is now known as k3t[04:26:59] <MatthiasM> eg rendering one wooding cube and a marble sphere :)[04:27:12] <yates> try to remember that i am just learning this. i may have doubts/holes in my knowledge that you don't understand[04:27:28] <tmccrary> I don't think anyone here doubts that[04:27:42] <MatthiasM> that's why I say - go with the 1.1 way - even if you later use shaders - it's still using texture IDs[04:27:58] <yates> i don't know what the 1.1 way is[04:28:11] <MatthiasM> glGenTextures -> glBindTexture ....[04:28:28] <MatthiasM> and then later when rendering only glBindTexture[04:28:44] <yates> when i get to the point of actually specifrying a glTexCoor, then is it the currently bound texture that gets used?[04:28:53] <MatthiasM> yes[04:29:02] <yates> is that how i can dynamically change textures, by binding to different texture ids?[04:29:06] <MatthiasM> yes[04:29:22] *** mm765 is now known as mm765^sleep[04:29:46] <yates> ok[04:30:38] <yates> but how does the actual texture bitmap get associated with a texture id? the texture id isn't used in glTexImage1D![04:31:03] <RTFM_FTW> and when you need to use *multiple* texture units (and through that multiple textures) the process will be: ... glActiveTexture -> glBindTexture ...[04:31:06] <MatthiasM> because you also call glBindTexture before callinf glTexImage[04:31:25] <RTFM_FTW> quoting myself: in any case for textures the usual process is: glGenTextures -> glBindTexture -> [setup texture state etc.] -> glTex[Sub]Image -> [render][04:31:30] <yates> ok[04:31:33] <yates> i see[04:31:36] <MatthiasM> yates: allglTexXYZ use the current bound texture id on the currently active texture unit[04:31:39] <RTFM_FTW> carefully note the order there[04:31:46] <RTFM_FTW> because order *is* important :P[04:31:56] <MatthiasM> :)[04:32:05] <yates> you mean the index? :)[04:32:19] <yates> (it's a JOKE!)[04:33:00] <MatthiasM> yates: somehow I can't see when you joke or not - don't know why - maybe your jokes are not funny enough ?[04:33:32] <tmccrary> Maybe you could add a little (J) to the end of lines that are jokes[04:34:05] <yates> http://en.wikipedia.org/wiki/Emoticon[04:34:21] <RTFM_FTW> LOL[04:35:43] <yates> ok binding wasn't the problem - still have a nice, bland white object. ;([04:36:02] <MatthiasM> yates: set the GL_MIN_FILTER to GL_LINEAR[04:36:24] <MatthiasM> as you did not upload any mipmap - you can't use mipmaped filtering[04:36:28] <yates> ok lemme try that[04:37:09] *** amz has joined ##opengl[04:37:22] <tmccrary> you can have opengl automatically generate mipmaps on new versions of opengl. Maybe 1.5 and newer, I forget[04:37:45] <tmccrary> Not that its a great idea, but if you need to test something and you don't have mipmaps pregenerated[04:38:17] <RTFM_FTW> yep[04:38:40] <yates> MatthiasM: did you mean GL_TEXTURE_MIN_FILTER?[04:38:47] <MatthiasM> yes[04:38:59] <yates> success![04:39:03] <MatthiasM> :)[04:39:11] <yates> much grass (and cocaine too)...[04:39:52] <yates> (that's an allusion to "muchos gracias", for those of you who don't live in an illegal-immigrant-infested country)...[04:40:11] <MatthiasM> :)[04:40:49] <yates> cool, thanks everyone.[04:40:57] <yates> i don't think i'm going to learn shaders tonight.[04:41:30] <yates> call me a wussie, but my goal is to get a minimal working, useful application first. then optimize/improve later[04:41:43] <MatthiasM> good idea[04:42:06] <tmccrary> wow, microsoft might be having layoffs[04:42:07] <yates> RTFM_FTW: you're ok, i don't care what MatthiasM says about you...[04:42:14] <RTFM_FTW> when you get to shaders just start with the simplest possible (passthrough) case[04:42:30] <MatthiasM> heh[04:42:35] <tmccrary> (J)[04:42:41] <yates> i'll remember that in the several years it'll take me to get past the old testament...[04:43:00] *** GinoMan has joined ##OpenGL[04:43:02] *** yates has quit IRC[04:44:04] *** walt has quit IRC[04:44:05] *** walt_ has joined ##OpenGL[04:46:52] *** HuntsMan has quit IRC[04:54:37] *** xp_prg has quit IRC[04:55:44] *** davidc__ has joined ##opengl[04:57:53] *** sparky_ is now known as sparky[05:10:51] *** walt_ is now known as walt[05:25:53] *** reprore_ has quit IRC[05:26:15] *** reprore_ has joined ##OpenGL[05:30:14] * sparky slaps furrywolf around a bit with a large trout[05:30:47] *** MouseCrazy has quit IRC[05:31:34] *** Shel2476 has joined ##OpenGL[05:31:43] *** Shel2476 has left ##OpenGL[05:39:02] *** huperniketes1 has joined ##OpenGL[05:39:06] *** huperniketes has quit IRC[05:39:08] *** huperniketes1 is now known as huperniketes[05:55:51] *** m4ggus has joined ##opengl[06:00:39] *** m4ggus has quit IRC[06:05:20] *** pierrelux has joined ##OpenGL[06:13:24] *** vade has quit IRC[06:20:16] *** pierrelux has quit IRC[06:20:17] *** amz has quit IRC[06:20:17] *** sparky has quit IRC[06:20:17] *** phrosty has quit IRC[06:20:18] *** ixtli has quit IRC[06:20:18] *** Hypnosekroete has quit IRC[06:20:18] *** Ademan has quit IRC[06:20:18] *** Weiss has quit IRC[06:20:18] *** autonomy has quit IRC[06:20:18] *** avacore has quit IRC[06:20:18] *** LordHavoc has quit IRC[06:20:18] *** _3b_ has quit IRC[06:20:18] *** GinoMan has quit IRC[06:20:18] *** MatthiasM has quit IRC[06:20:18] *** RTFM_FTW has quit IRC[06:20:18] *** rutski_ has quit IRC[06:20:19] *** tmccrary has quit IRC[06:20:19] *** hibread has quit IRC[06:20:19] *** Deformalite has quit IRC[06:20:19] *** flazz has quit IRC[06:20:19] *** GuShH has quit IRC[06:20:19] *** Plagman has quit IRC[06:20:19] *** exDM69 has quit IRC[06:20:19] *** Icchan^ has quit IRC[06:20:19] *** ccfreak2k has quit IRC[06:20:19] *** quicksilver has quit IRC[06:20:19] *** TenOfTen has quit IRC[06:20:20] *** memfr0b has quit IRC[06:20:20] *** Stevethepirate has quit IRC[06:20:20] *** Mazon has quit IRC[06:20:20] *** dennda has quit IRC[06:20:20] *** Rangar has quit IRC[06:20:20] *** nargon has quit IRC[06:20:20] *** t4bz has quit IRC[06:20:21] *** JernejL has quit IRC[06:20:21] *** tyranis has quit IRC[06:20:21] *** jnglst has quit IRC[06:20:21] *** Gudy has quit IRC[06:20:21] *** elektr1k has quit IRC[06:20:21] *** synchris has quit IRC[06:20:22] *** servus has quit IRC[06:20:22] *** Defty has quit IRC[06:20:22] *** furrywolf has quit IRC[06:20:22] *** Kraln has quit IRC[06:20:23] *** walt has quit IRC[06:20:23] *** ol1veira_ has quit IRC[06:20:23] *** Burga has quit IRC[06:20:24] *** Tempoe has quit IRC[06:20:24] *** GNU\caust1c has quit IRC[06:20:24] *** ville has quit IRC[06:20:25] *** Seeb has quit IRC[06:20:25] *** stringfellow has quit IRC[06:20:25] *** TheFlash has quit IRC[06:20:26] *** freespace has quit IRC[06:20:26] *** DenizzzZ has quit IRC[06:20:26] *** fw0rd has quit IRC[06:20:26] *** fargiolas has quit IRC[06:20:26] *** theGreg101 has quit IRC[06:20:26] *** Tenac has quit IRC[06:20:26] *** PluribusIMP has quit IRC[06:20:26] *** tac has quit IRC[06:20:27] *** turbo24prg has quit IRC[06:20:27] *** Jupp3 has quit IRC[06:20:27] *** |chiz| has quit IRC[06:20:27] *** braden has quit IRC[06:20:27] *** pragma_ has quit IRC[06:20:27] *** sohail has quit IRC[06:20:27] *** NotWil has quit IRC[06:20:27] *** Xantoz has quit IRC[06:21:04] *** theGreg101 has joined ##OpenGL[06:24:03] *** pierrelux has joined ##OpenGL[06:24:03] *** walt has joined ##OpenGL[06:24:03] *** amz has joined ##OpenGL[06:24:03] *** sparky has joined ##OpenGL[06:24:03] *** LordHavoc has joined ##OpenGL[06:24:03] *** Rangar has joined ##OpenGL[06:24:03] *** fargiolas has joined ##OpenGL[06:24:03] *** phrosty has joined ##OpenGL[06:24:03] *** Tenac has joined ##OpenGL[06:24:03] *** ol1veira_ has joined ##OpenGL[06:24:03] *** nargon has joined ##OpenGL[06:24:03] *** t4bz has joined ##OpenGL[06:24:03] *** Burga has joined ##OpenGL[06:24:03] *** pragma_ has joined ##OpenGL[06:24:03] *** sohail has joined ##OpenGL[06:24:03] *** JernejL has joined ##OpenGL[06:24:03] *** tyranis has joined ##OpenGL[06:24:03] *** Weiss has joined ##OpenGL[06:24:03] *** avacore has joined ##OpenGL[06:24:03] *** Hypnosekroete has joined ##OpenGL[06:24:03] *** Ademan has joined ##OpenGL[06:24:03] *** ixtli has joined ##OpenGL[06:24:03] *** autonomy has joined ##OpenGL[06:24:03] *** Kraln has joined ##OpenGL[06:24:03] *** tac has joined ##OpenGL[06:24:03] *** ville has joined ##OpenGL[06:24:04] *** _3b_ has joined ##OpenGL[06:24:04] *** furrywolf has joined ##OpenGL[06:24:04] *** Xantoz has joined ##OpenGL[06:24:04] *** synchris has joined ##OpenGL[06:24:04] *** braden has joined ##OpenGL[06:24:04] *** NotWil has joined ##OpenGL[06:24:04] *** Gudy has joined ##OpenGL[06:24:04] *** elektr1k has joined ##OpenGL[06:24:04] *** Defty has joined ##OpenGL[06:24:04] *** jnglst has joined ##OpenGL[06:24:04] *** servus has joined ##OpenGL[06:24:04] *** GNU\caust1c has joined ##OpenGL[06:24:04] *** |chiz| has joined ##OpenGL[06:24:04] *** Tempoe has joined ##OpenGL[06:24:04] *** Jupp3 has joined ##OpenGL[06:24:04] *** turbo24prg has joined ##OpenGL[06:27:48] <Jonanin> what hapepned?[06:28:05] *** ixtli has quit IRC[06:28:05] *** amz has quit IRC[06:28:05] *** avacore has quit IRC[06:28:05] *** Weiss has quit IRC[06:28:06] *** phrosty has quit IRC[06:28:06] *** sparky has quit IRC[06:28:06] *** pierrelux has quit IRC[06:28:06] *** autonomy has quit IRC[06:28:06] *** Ademan has quit IRC[06:28:06] *** Hypnosekroete has quit IRC[06:28:06] *** LordHavoc has quit IRC[06:28:06] *** _3b_ has quit IRC[06:28:06] *** jnglst has quit IRC[06:28:06] *** Gudy has quit IRC[06:28:07] *** Rangar has quit IRC[06:28:07] *** elektr1k has quit IRC[06:28:08] *** JernejL has quit IRC[06:28:08] *** synchris has quit IRC[06:28:08] *** servus has quit IRC[06:28:08] *** Defty has quit IRC[06:28:08] *** t4bz has quit IRC[06:28:08] *** furrywolf has quit IRC[06:28:08] *** Kraln has quit IRC[06:28:08] *** tyranis has quit IRC[06:28:08] *** nargon has quit IRC[06:28:08] *** Tempoe has quit IRC[06:28:08] *** Burga has quit IRC[06:28:08] *** ol1veira_ has quit IRC[06:28:08] *** GNU\caust1c has quit IRC[06:28:09] *** walt has quit IRC[06:28:09] *** ville has quit IRC[06:28:09] *** Tenac has quit IRC[06:28:09] *** tac has quit IRC[06:28:09] *** fargiolas has quit IRC[06:28:09] *** turbo24prg has quit IRC[06:28:09] *** Jupp3 has quit IRC[06:28:09] *** |chiz| has quit IRC[06:28:09] *** braden has quit IRC[06:28:10] *** NotWil has quit IRC[06:28:10] *** sohail has quit IRC[06:28:10] *** Xantoz has quit IRC[06:28:10] *** pragma_ has quit IRC[06:29:04] *** reprore_ has quit IRC[06:30:30] <Jonanin> ...[06:30:31] *** pragma__ has joined ##OpenGL[06:30:31] *** Xantoz has joined ##OpenGL[06:30:31] *** pierrelux has joined ##OpenGL[06:30:31] *** walt has joined ##OpenGL[06:30:31] *** amz has joined ##OpenGL[06:30:31] *** sparky has joined ##OpenGL[06:30:31] *** LordHavoc has joined ##OpenGL[06:30:31] *** Rangar has joined ##OpenGL[06:30:31] *** fargiolas has joined ##OpenGL[06:30:31] *** phrosty has joined ##OpenGL[06:30:31] *** Tenac has joined ##OpenGL[06:30:31] *** ol1veira_ has joined ##OpenGL[06:30:31] *** nargon has joined ##OpenGL[06:30:31] *** t4bz has joined ##OpenGL[06:30:31] *** Burga has joined ##OpenGL[06:30:31] *** pragma_ has joined ##OpenGL[06:30:31] *** sohail has joined ##OpenGL[06:30:31] *** JernejL has joined ##OpenGL[06:30:31] *** tyranis has joined ##OpenGL[06:30:31] *** Weiss has joined ##OpenGL[06:30:31] *** avacore has joined ##OpenGL[06:30:31] *** Hypnosekroete has joined ##OpenGL[06:30:31] *** Ademan has joined ##OpenGL[06:30:31] *** ixtli has joined ##OpenGL[06:30:31] *** autonomy has joined ##OpenGL[06:30:31] *** Kraln has joined ##OpenGL[06:30:31] *** tac has joined ##OpenGL[06:30:31] *** ville has joined ##OpenGL[06:30:31] *** _3b_ has joined ##OpenGL[06:30:31] *** furrywolf has joined ##OpenGL[06:30:31] *** synchris has joined ##OpenGL[06:30:31] *** braden has joined ##OpenGL[06:30:31] *** NotWil has joined ##OpenGL[06:30:31] *** Gudy has joined ##OpenGL[06:30:31] *** elektr1k has joined ##OpenGL[06:30:31] *** Defty has joined ##OpenGL[06:30:31] *** jnglst has joined ##OpenGL[06:30:31] *** servus has joined ##OpenGL[06:30:31] *** GNU\caust1c has joined ##OpenGL[06:30:31] *** |chiz| has joined ##OpenGL[06:30:31] *** Tempoe has joined ##OpenGL[06:30:31] *** Jupp3 has joined ##OpenGL[06:30:31] *** turbo24prg has joined ##OpenGL[06:30:39] *** Xantoz has quit IRC[06:30:40] *** Xantoz has joined ##OpenGL[06:30:49] *** Seeb has joined ##OpenGL[06:30:49] *** stringfellow has joined ##OpenGL[06:30:49] *** TheFlash has joined ##OpenGL[06:30:49] *** fw0rd has joined ##OpenGL[06:30:49] *** freespace has joined ##OpenGL[06:30:49] *** DenizzzZ has joined ##OpenGL[06:31:10] *** bijoo_appdev has joined ##OpenGL[06:31:10] *** bijoo_osdev has joined ##OpenGL[06:31:10] *** GinoMan has joined ##OpenGL[06:31:10] *** MatthiasM has joined ##OpenGL[06:31:10] *** RTFM_FTW has joined ##OpenGL[06:31:10] *** rutski_ has joined ##OpenGL[06:31:10] *** tmccrary has joined ##OpenGL[06:31:10] *** hibread has joined ##OpenGL[06:31:10] *** Deformalite has joined ##OpenGL[06:31:10] *** flazz has joined ##OpenGL[06:31:10] *** GuShH has joined ##OpenGL[06:31:10] *** Plagman has joined ##OpenGL[06:31:10] *** Icchan^ has joined ##OpenGL[06:31:10] *** dennda has joined ##OpenGL[06:31:10] *** Stevethepirate has joined ##OpenGL[06:31:10] *** TenOfTen has joined ##OpenGL[06:31:10] *** ccfreak2k has joined ##OpenGL[06:31:10] *** exDM69 has joined ##OpenGL[06:31:10] *** Mazon has joined ##OpenGL[06:31:10] *** quicksilver has joined ##OpenGL[06:31:10] *** memfr0b has joined ##OpenGL[06:32:55] *** reprore_ has joined ##OpenGL[06:36:47] *** amz has quit IRC[06:36:48] *** pragma_ has quit IRC[06:40:29] *** xp_prg has joined ##OpenGL[06:42:43] *** bijoo_appdev has quit IRC[06:42:51] *** bijoo_osdev has quit IRC[06:46:36] *** vade has joined ##OpenGL[06:46:52] *** bijoo_appdev has joined ##OpenGL[06:46:52] *** bijoo_osdev has joined ##opengl[06:51:35] *** walt has quit IRC[06:54:34] *** walt has joined ##OpenGL[07:00:38] <Jonanin> hello, does anyone have some good resources on multisampling? fsaa maybe?[07:03:21] *** e-roder has joined ##OpenGL[07:03:25] <e-roder> hey[07:03:42] <e-roder> i'm trying to use opengl lighting and my white is coming out a brown[07:03:46] <e-roder> any ideas why?[07:06:38] *** rnx has left ##opengl[07:20:05] *** pragma__ is now known as pragma_[07:24:47] *** bijoo_appdev_ has joined ##OpenGL[07:24:47] *** bijoo_osdev_ has joined ##opengl[07:26:52] *** Candy has joined ##OpenGL[07:30:14] *** xp_prg has quit IRC[07:40:39] *** bijoo_osdev has quit IRC[07:40:40] *** Gorgoroth has joined ##OpenGL[07:40:50] *** bijoo_appdev has quit IRC[07:41:56] *** hang3r has quit IRC[07:55:40] *** Candy has quit IRC[07:56:22] *** pierrelux has quit IRC[07:59:52] *** reprore_ has quit IRC[08:03:43] *** davidc__ has quit IRC[08:05:07] *** korff_home has joined ##OpenGL[08:07:09] *** elite01 has joined ##opengl[08:07:50] *** rabbit- has joined ##OpenGL[08:10:41] <e-roder> do you guys know of any common stenciling snags?[08:10:46] <e-roder> i can't seem to get it to work[08:13:58] *** e-roder has quit IRC[08:14:22] *** reprore_ has joined ##OpenGL[08:21:30] *** reprore_ has quit IRC[08:34:26] *** Renderwahn has joined ##OpenGL[08:49:47] *** xp_prg has joined ##OpenGL[09:05:01] *** abebae has joined ##OpenGL[09:07:46] *** UUncia has joined ##OpenGL[09:09:36] *** fargiolas has quit IRC[09:09:53] *** bijoo_appdev_ has quit IRC[09:09:53] *** bijoo_osdev_ has quit IRC[09:10:13] *** fargiolas has joined ##OpenGL[09:10:17] *** pietia has joined ##OpenGL[09:10:24] *** geocalc has quit IRC[09:15:44] *** sohail has quit IRC[09:21:24] *** fargiola` has joined ##OpenGL[09:22:04] *** Spkka has joined ##OpenGL[09:23:59] *** davidc__ has joined ##opengl[09:27:23] *** fargiola` has joined ##OpenGL[09:36:55] *** fargiolas has quit IRC[09:40:23] *** neoneye has joined ##OpenGL[09:41:59] *** abebae has quit IRC[09:42:20] *** groton has joined ##OpenGL[09:43:51] *** phrosty has quit IRC[09:52:10] *** xp_prg has quit IRC[09:54:22] *** pa has joined ##OpenGL[10:00:41] *** fargiola` has quit IRC[10:00:56] *** fargiolas has joined ##OpenGL[10:01:45] *** fargiolas has quit IRC[10:01:52] *** fargiolas has joined ##OpenGL[10:02:46] *** fargiolas has quit IRC[10:12:56] *** fargiolas has joined ##OpenGL[10:13:43] *** fargiolas has quit IRC[10:22:21] *** fargiolas has joined ##OpenGL[10:22:56] *** fargiolas has quit IRC[10:23:32] *** fargiolas has joined ##OpenGL[10:25:52] *** braden has quit IRC[10:31:33] *** sparky has quit IRC[10:32:30] *** sparky has joined ##OpenGL[10:34:15] *** pist0l-fish has joined ##OpenGL[10:34:50] *** spx2 has joined ##OpenGL[10:35:48] <pist0l-fish> hello all. Really simple question, I get a segfault while executing gluBuild2DMipmaps (http://codepad.org/EYvuzfVt) can anyone with a keen eye (and more OpenGL knowledge than myself) tell me what I've done wrong? Thanks.[10:35:55] *** spx2 has left ##OpenGL[10:37:31] *** batbrat has joined ##opengl[10:40:02] <quicksilver> pist0l-fish: is that suposed to load a BMP file?[10:40:12] <quicksilver> I'm assuming so, form the name of the file :)[10:40:39] <pist0l-fish> quicksilver: yes, that's right- I was following along a tutorial that used windows bmp files[10:40:42] *** predaeus has joined ##opengl[10:41:02] <quicksilver> well, the BMP file format is more complex that just being pixel data[10:41:05] <quicksilver> you can't just load it like that[10:41:11] <quicksilver> you've got to skip the header at least.[10:41:19] <pist0l-fish> I see. I'm pretty unfamiliar with BMPs. Is the header a fixed size?[10:41:38] <quicksilver> I don't know.[10:41:42] <quicksilver> I just use a library to load images :)[10:42:08] <pist0l-fish> okay, I will look it up on google right now, but even so, that should not be causing a segfault. If there was extra information, it would simply read it and display an incorrect texture[10:42:22] <quicksilver> agreed[10:42:26] <quicksilver> I wouldn't really expect a segfault.[10:42:48] <pist0l-fish> I think the problem is from something else[10:43:12] <pist0l-fish> especially because the segfault occurs while in gluBuild2DMipmaps. Am I using that function correctly?[10:43:55] <quicksilver> looks OK to me.[10:44:01] <quicksilver> Do you have an openGL context?[10:45:26] <pist0l-fish> hmm. That may just be the problem[10:46:13] <pist0l-fish> nope, that was not the problem[10:46:21] *** Tenac_ has joined ##opengl[10:47:53] *** sparky_ has joined ##OpenGL[10:49:38] <quicksilver> I can't explain the segfault if your openGL context is OK.[10:49:44] <quicksilver> Unless width/height are absurd numbers[10:50:14] <batbrat> I have drawn an icosahedron using the technique given in the redbook edition.Instead of using surface normals and applying, I applied a different color to each face.When I try to rotate the figure, some of its faces get drawn in white and it appears somewhat see through.I have written this in Python + PyQt + PyOpenGL and pasted the code at http://pastebin.com/m4ef798bb. I have an identity projection matrix.I will be glad to supply any further details.I ha[10:50:29] <batbrat> *redbook second edititon[10:51:09] <pist0l-fish> quicksilver: no, width = 320, height =240. I tried using glTexImage2D but that also produced a segfault.[10:53:27] <quicksilver> batbrat: screenshot might help understand the problem[10:54:40] <quicksilver> pist0l-fish: paste some more code so we can see your context setup[10:54:57] <batbrat> Yes quicksilver, I'll do it and let you know.Just a minute.[10:56:51] *** hang3r has joined ##OpenGL[10:57:01] *** Tenac has quit IRC[10:57:27] *** scai has joined ##opengl[10:57:42] <pist0l-fish> quicksilver: http://codepad.org/EaYMPs6i[10:58:33] *** dv_ has joined ##opengl[11:00:30] *** dv_ has quit IRC[11:01:14] <quicksilver> pist0l-fish: do you see the window appear before it crashes?[11:01:22] <batbrat> Where should I post the screenshots? I am unable to access the place wher e I usually post them.Can you please suggest a site?[11:01:30] <batbrat> *where[11:02:20] <quicksilver> batbrat: perhaps http://imagebin.org/[11:02:55] *** tokemonstah has joined ##OpenGL[11:03:11] <pist0l-fish> quicksilver: yes, briefly[11:05:13] *** sparky has quit IRC[11:05:13] <batbrat> Thank you quicksilver. I am posting there now.[11:06:54] *** Eforen has joined ##opengl[11:08:03] <pist0l-fish> quicksilver: I don't mean to trouble you, but would you be able to compile and see if that code runs (without a segfault) on your machine?[11:09:14] <quicksilver> pist0l-fish: funnily enough, I'm trying to do that[11:09:21] <pist0l-fish> :) thank you[11:09:32] <quicksilver> pist0l-fish: just trying to remember how you do this on OSX ;) stupid non-standard apple stuff[11:10:23] *** Yustme has joined ##OpenGL[11:10:46] <pist0l-fish> quicksilver: __MACOSX directories are lurking everywhere; watch out[11:12:47] <batbrat> Ive posted a screenshot at http://imagebin.org/34807.I'll post more and tell their urls as well.[11:17:13] <batbrat> http://imagebin.org/34808 and http://imagebin.org/34810 contain the other two screenshots.[11:18:35] *** pietia has quit IRC[11:20:27] <pist0l-fish> batbrat: you have a lot of mounted devices[11:21:06] <batbrat> Yes thats because I mounted all my windows and debian partitions as well.[11:21:21] *** urnotE has quit IRC[11:21:24] <quicksilver> batbrat: doesn't look too bad to me[11:21:38] <quicksilver> batbrat: I think the only thing wrong is you're not making any effort to do hidden surface removal[11:21:45] *** sparky has joined ##OpenGL[11:21:51] <quicksilver> so you're getting backfaces in front, sometimes[11:22:30] *** fargiolas has quit IRC[11:22:38] *** fargiolas has joined ##OpenGL[11:23:15] <batbrat> quicksilver: Thanks, that explains why I thought it was see through. Thanks. So will turning on culling help?[11:24:14] <quicksilver> culling is potentially another way to do backface removal[11:24:20] <quicksilver> btu I recommend the depth buffer[11:24:35] *** xp_prg has joined ##OpenGL[11:25:47] <batbrat> Thanks for the suggestion.I haven't learn't about how to turn on the depth-buffer yet.So I'll go look it up and try to add the necessary code.[11:25:57] <pist0l-fish> quicksilver: I don't think culling would fix it- the depth buffer would. Doesn't culling in OpenGL only remove whole polygons when they are fully obscured?[11:26:25] <batbrat> I'll get back to you on the results of that in a couple of minutes.[11:27:09] *** urnotE has joined ##OpenGL[11:27:39] <quicksilver> pist0l-fish: ah well your code segfaults for me too[11:27:42] <quicksilver> that's good, in a way[11:27:45] * quicksilver looks more carefully[11:29:12] <quicksilver> pist0l-fish: for me it's definitely the 'fread' that is segfaulting[11:29:15] <quicksilver> not the glTexImage2D[11:30:01] *** XT95_ has joined ##OpenGL[11:31:22] <pist0l-fish> quicksilver: do you have a bitmap where the path points to?[11:31:35] <quicksilver> I didn't, no.[11:31:44] <quicksilver> I changed the path to point to a real file[11:31:47] <quicksilver> and now it works[11:31:58] <quicksilver> I never knew that fread on a non-existent file pointer would segfault[11:32:18] <pist0l-fish> quicksilver: yeah, you have to check if fopen worked. So it works, no segfaults?[11:32:24] *** sparky_ has quit IRC[11:32:57] <quicksilver> nope.[11:33:03] <quicksilver> I had reduced width + height to 32 and 24[11:33:05] <quicksilver> it worked then[11:33:13] <quicksilver> when I reset them to 320 and 240 it segfaults again[11:33:22] <pist0l-fish> you can't have big textures??[11:33:25] <pist0l-fish> :([11:33:30] <quicksilver> Apprently 320x240x3 is just too large to allocate on the stack[11:33:32] <quicksilver> no, silly :P[11:33:38] <quicksilver> it's the allocation of the array 'data' which is the program[11:33:49] <quicksilver> 2048x2048 and large textures are fine.[11:34:09] *** sparky_ has joined ##OpenGL[11:34:12] <quicksilver> char *data = new char[bytes];[11:34:22] <quicksilver> allocate it dynamically instead of ion the stack and it's fine.[11:34:22] <pist0l-fish> quicksilver: oh yeah, 320x240x3 is a big number >_<.[11:34:33] <quicksilver> although that's pretty weird that it segfaults instead of giving a compiler error.[11:34:36] <quicksilver> compilers suck.[11:35:13] <pist0l-fish> g++ isn't the best[11:35:36] <quicksilver> it's quite good in some ways :)[11:36:32] <pist0l-fish> well, it compiles nice code[11:36:36] <pist0l-fish> err, nice binaries[11:39:34] <pist0l-fish> thanks so much for finding that segfault[11:39:48] <quicksilver> I learnt something.[11:40:00] <quicksilver> I never knew that large stack structures caused a segfault without even a compiler warning.[11:40:33] *** boghog has quit IRC[11:41:42] <predaeus> stack size is OS specific as far as I know, you can tweak it in Linux for example.[11:42:03] * quicksilver nods[11:42:09] <quicksilver> yes, that's true.[11:42:20] <quicksilver> I just kind of expected (at least) a warning rather than silenty segfaulting coe.[11:42:27] <quicksilver> evidently I've never tried it before.[11:44:12] *** sparky has quit IRC[11:48:00] *** geocalc has joined ##opengl[11:53:47] <batbrat> Culling worked to an extent. I am still getting that white triangle (which shouldn't be possible considering my coklor calculation).If I enable depth buffering, the icosahedron looks like somebody hacked it up. I've posted the code at : http://pastebin.com/m7c6e5434 and screenshotas at: http://imagebin.org/34812 and http://imagebin.org/34813 and http://imagebin.org/34814[11:53:55] <batbrat> *color calculation[11:54:40] <batbrat> Should I post the screenshots with culling enabled? Can you guys take a look and help me again please?[11:55:10] *** LordHavoc has quit IRC[11:55:48] <batbrat> Just in case its relevant, QGLContext class automatically creates a context with depth buffering support.I read that in the documentation.However, I am unsure what exactly a 'context' is.[11:57:46] *** Yustme_ has joined ##OpenGL[11:58:38] *** Xmas| has joined ##OpenGL[12:00:30] *** rabbit- has quit IRC[12:01:30] <batbrat> http://imagebin.org/34816 and http://imagebin.org/34818 and contain the screenshots when I use front face culling... Back face messes up for some reason...[12:05:59] <quicksilver> as we said, don't use culling[12:06:02] <quicksilver> use the depth buffer[12:07:10] *** Yustme has quit IRC[12:07:29] <pist0l-fish> batbrat: culling isn't the same as depth testing- culling is a rough way to remove surfaces so that you don't ever have to draw them. It's used to save time (hidden surface removal)[12:09:13] <pist0l-fish> aka, if you have multiple objects in the scene that could be intersecting on the projection plane, culling isn't enough[12:09:47] <batbrat> pist0l-fish: Thank you.[12:10:41] <pist0l-fish> batbrat: I hope that made sense. It's explained in detail in a lot of literature (e.g. http://en.wikipedia.org/wiki/Back-face_culling)[12:11:20] <batbrat> pist0l-fish: I didn't know that. However I am not using it any more, I am using depth buffering instead. I still have some problems, so I'll try cleaning up the code and post back.[12:11:58] <batbrat> quicksilver: Yes quicksilver, I will use depth buffering and no more culling.[12:12:10] <pist0l-fish> batbrat: okay. I will probably be of little help as my OpenGL knowledge is lacking[12:13:41] *** pietia has joined ##OpenGL[12:14:49] *** walt has quit IRC[12:15:46] *** walt has joined ##OpenGL[12:19:26] *** neoneye has quit IRC[12:22:43] *** fargiola` has joined ##OpenGL[12:24:25] <batbrat> I am still having trouble after enabling depth testing. I learnt hoe to enable it from the redbook and tried it.I hope I have done it right.Can you please see whether it is correct.I am still having an error, as evidenced by the screenshot - is this the same error as before? The links are -New Code: http://pastebin.com/m16d284f4 Screenshot:http://imagebin.org/34820[12:24:31] <batbrat> *how[12:25:02] *** neoneye has joined ##OpenGL[12:25:12] <batbrat> Thanks for the link pist0l-fish that had a nice explanation.[12:25:56] <pist0l-fish> batbrat: why is this commented? #glEnable(GL_DEPTH_TEST)[12:26:21] <batbrat> One moment, that shouldn't be commented... I'll re-check[12:27:06] *** calav3ra has joined ##opengl[12:27:11] <batbrat> Sorry, I posted the wrong code... It isn't commented... I'll re post.[12:27:11] *** bijoo_osdev has joined ##opengl[12:27:11] *** bijoo_appdev has joined ##OpenGL[12:27:12] <TheFlash> It depends the language used.[12:27:37] <TheFlash> Oh, threse pastebin url...[12:27:39] <batbrat> quicksilver: I don't mean to impose, but can you please take a look at it too?[12:28:29] *** fargiolas has quit IRC[12:29:09] <batbrat> The corrected code is at: http://pastebin.com/m29a60854[12:29:21] <batbrat> that was so embarassing ...[12:30:20] <batbrat> pist0l-fish: Thanks for pinpointing that error...[12:31:51] <pist0l-fish> batbrat: no problem. Sometimes it helps to have a second set of eyes looking over your code.[12:32:00] <pist0l-fish> so does it work now?[12:32:36] <batbrat> The code still doesn't work. Same problem as I posted in the screenshot.Should I post another shot?[12:32:51] *** dvoid_ has joined ##OpenGL[12:33:30] <batbrat> TheFlash: I don't understand... I just saw your post. Sorry I didn't reply if that was targeted to me..[12:34:36] *** walt has quit IRC[12:34:41] <quicksilver> I don't really understand what's happening in that screenshot I must admit[12:34:46] <quicksilver> what does it look like as it rotates?[12:35:11] <TheFlash> I misunderstood what pist0l-fish sayed. Then I saw your pastebin link and understood the context you was talking about.[12:35:52] <TheFlash> quicksilver: There's no depthtesting nor backface culling.[12:36:06] <quicksilver> glEnable(GL_DEPTH_TEST)[12:36:07] <quicksilver> is there[12:36:24] <TheFlash> It was commented during that screenshot.[12:36:47] <quicksilver> ah well that's bit I wasn't sure about[12:36:54] <quicksilver> what does it look like now it's uncommented?[12:37:06] <batbrat> quicksilver: It looks as if a spotlight is shining on it and the same white triangle problem is thre[12:37:08] <TheFlash> batbrat: Or was it?[12:37:48] <batbrat> TheFlash, quicksilver : I am unsure now... I'll post new screenshots.[12:38:07] <batbrat> TheFlash: Thanks for explaining...[12:39:47] *** Walt has joined ##opengl[12:40:51] <batbrat> http://imagebin.org/34822 and http://imagebin.org/34823 and http://imagebin.org/34824 are the new screenshots.[12:42:47] <TheFlash> batbrat: Looks like an lighting issue. Try glEnable(GL_POLYGON_SMOOTH)[12:42:55] <quicksilver> lighting is supposed to be off[12:43:03] <quicksilver> maybe some library you import is turning it on though[12:43:09] <quicksilver> glDisable(GL_LIGHTING)[12:43:28] <batbrat> I'll try disabling lighting and enabling GL_POLYGON_SMOOTH... 1 mo[12:46:17] <batbrat> I tried them one by one and together... The error still remains... Theres yet another screenshot : http://imagebin.org/34826[12:47:55] <batbrat> Note - Both glDisable(GL_LIGHTING) and glEnable(GL_POLYGON_SMOOTH) are done in the initGL method for the screenshot...[12:49:43] <batbrat> TheFlash: What exactly does GL_POLYGON_SMOOTH do? - Polygon Antialaising?[12:51:07] <TheFlash> Yes... Well that was not what i meant, tho =)[12:51:29] <TheFlash> glLightModel was what i was thingking.[12:51:55] <TheFlash> Or was it...[12:51:56] <TheFlash> No.[12:52:15] <batbrat> I don't understand, you've lost me...[12:52:20] <pist0l-fish> I'm going to call it a night. Happy coding everyone, thanks for your help[12:52:27] *** pist0l-fish has quit IRC[12:52:28] <batbrat> Sorry, I am really new to GL[12:52:36] <TheFlash> batbrat: Sorry, just woke-up =)[12:52:46] <batbrat> Ok :)[12:53:32] <batbrat> Is there a name for this kind of error?[12:53:41] <batbrat> Rendering artifacr?[12:53:49] <batbrat> *artifacts[12:53:53] *** Walt has quit IRC[12:53:55] <TheFlash> ShadeModel[12:54:16] <batbrat> Ok so I should do glDisable(GL_SHADE_MODEL) ?[12:54:23] <TheFlash> No.[12:54:39] <TheFlash> glShadeModel(GL_SMOOTH)[12:54:47] *** gerard_ has joined ##OpenGL[12:55:04] <batbrat> Ok I'll do that ... 1 mo..[12:55:07] <TheFlash> I don't know what you're trying to archieve...[12:55:38] <gerard_> hi all, is there any open source tool to profile an opengl application on linux?[12:55:41] <TheFlash> Smooth-looking polyhedron?[12:56:17] <gerard_> I know of gDEBugger, but I'd rather not use it[12:56:43] <gerard_> I've got an ATI card by the way[12:56:46] <batbrat> I am still getting the same look. :: I am trying to get a rotating icosahedron that looks reasonable without having to do lighting ...[12:56:56] *** UUncia has quit IRC[12:57:39] *** walt has joined ##OpenGL[12:58:40] <batbrat> TheFlash: I tried glShadeModel(GL_SMOOTH) just now... I got it wrong in the previous post. However, I still have the same error.[12:58:50] <TheFlash> batbrat: Maybe try to set your projection matrix?[13:00:00] <batbrat> I'll try doing that[13:02:46] <TheFlash> Line 37: gluPerspective(...) or something like that.[13:05:01] <batbrat> TheFlash: I added it to line 38 :). gluPerspective(100,4/3.0,0.0001,1000.0). As expected, I got a blown up image of the polygon. I still have the white triangle though :([13:06:12] *** ol1veira1 has joined ##OpenGL[13:06:14] <batbrat> Should I post code/screenshots?[13:07:03] <TheFlash> Don't bother, i can't run it. (i have no PyQt).[13:07:15] <batbrat> Ok...[13:07:26] *** mattn2|home has joined ##OpenGL[13:07:32] <batbrat> What could be the reason for the white polygon?[13:08:05] <batbrat> Is it due to lack of lighting? Should I learn lighting and re-do the code?[13:08:41] <batbrat> I really doubt that having no lighting would do that though... Still I asked...[13:09:15] <quicksilver> having no lighting should make it use exactly the colours you give it.[13:09:23] <quicksilver> I think there must be a strange bug in your colour choosing code.[13:09:23] <TheFlash> Every triangle shoud have it's own color with that code you pasted before.[13:10:35] <batbrat> So I really shouldn't be having this error.[13:11:02] <batbrat> I'll go to windows try running it there and get back.Will that be useful?[13:11:39] *** boghog has joined ##opengl[13:12:09] *** eXtronuS has joined ##OpenGL[13:13:32] *** NightVisio has joined ##OpenGL[13:13:44] <NightVisio> hi all[13:14:16] <batbrat> One mo before I try going to windows, I have an idea ... I'll try using an assertion to see if I get a white colour anywher...[13:14:23] <batbrat> Hi NightVisio[13:17:10] <batbrat> TheFlash, quicksilver : I tried the assertion and the[13:17:29] <batbrat> *as well as changing the colour choosing code, to no effect.[13:17:46] *** ol1veira_ has quit IRC[13:18:07] <TheFlash> Disable lighting, if you're "manualy" coloring.[13:18:33] *** bijoo_osdev has quit IRC[13:18:33] *** bijoo_appdev has quit IRC[13:19:09] <batbrat> Ok. I'll disable lighting .. 1 second.[I did not enable it, still I'll add it because the libraries I use may have turned it on][13:20:24] *** Xmas| has quit IRC[13:21:00] <batbrat> TheFlash: I disabled lighting and am still getting the single white face.[13:22:04] <TheFlash> What kind of projection matrix did you add?[13:22:34] *** jcazevedo has quit IRC[13:23:36] <batbrat> I added Perspective projection :gluPerspective(100,4/3,0.0001,1000.0)[13:23:52] <batbrat> I think I should re-do the 4/3 aspect ratio thing...[13:24:08] *** NightVisio has quit IRC[13:24:10] *** NightVisio_ has joined ##OpenGL[13:24:53] *** bijoo_osdev has joined ##opengl[13:24:54] *** bijoo_appdev has joined ##OpenGL[13:24:55] <batbrat> I did it again with width/height (of the window)[13:25:10] <TheFlash> 4/3[13:25:11] <TheFlash> 1[13:25:21] <batbrat> Ok doing it now.[13:25:23] <TheFlash> 4/3.0[13:25:23] <TheFlash> 1.3333333333333333[13:25:51] *** bijoo_osdev has quit IRC[13:25:51] *** bijoo_appdev has quit IRC[13:26:23] <batbrat> Oh! got it.... Sorry. I've changed it to width/height now. Is that ok?[13:26:32] <TheFlash> Try smaller fov and move your farplane closer. I don't think it's your problem tho.[13:27:03] <TheFlash> batbrat: Depends of the types of width and height...[13:27:34] <TheFlash> float(width)/float(height) should do.[13:27:42] *** lewymati has joined ##OpenGL[13:29:05] <batbrat> TheFlash: I think width and height are floats already. I'll make sure by adding (width * 1.0) / height[13:29:17] <batbrat> Since Python has got duck typing...[13:29:39] *** glcrazy has joined ##OpenGL[13:32:01] <batbrat> I've changed quite a few things since the last time. I've re-posted the code at:http://pastebin.com/mdbe15f1[13:32:18] <batbrat> I had to remove the call to gluLookAt() for this to work correctly...[13:32:28] <batbrat> I still have the white triangle...[13:33:26] <batbrat> I have yet to reduce fov but I've changed the near and far clip planes to more reasonable values...[13:36:56] <batbrat> TheFlash: Ok now I've got a fov of 30 and put back the gluLookAt with correct values. I still have the same white triangle though. I wonder whether it is because of round off error?[13:39:16] *** hang3r has quit IRC[13:39:25] *** NightVisio_ has quit IRC[13:39:37] *** NightVisio has joined ##OpenGL[13:44:12] <batbrat> I've located the error!! The color choosing code gets clamped to (1,1,1) when the first face is drawn.I added some conditional color choosing code to correct it and now have a reasonable Icosahedron.[13:46:40] <batbrat> TheFlash, quicksilver : Thank you very much for your patient help. Without your tips, I would still have been floundering. Also thanks for teaching me depth buffering and the various lighting modes etc. I cant thank you guys nearly enough.[13:47:16] <batbrat> TheFlash, quicksilver : You guys rock.[13:47:39] *** twist has joined ##OpenGL[13:48:02] <batbrat> *guys/girls -- Don't know and don't want to know. So I used guys. No offence...[13:48:20] <quicksilver> heh[13:48:21] <quicksilver> ;)[13:48:44] <batbrat> :) I am calling it a day now. Bye.[13:50:06] *** batbrat has left ##opengl[13:51:36] *** NightVisio has quit IRC[13:54:05] <gerard_> is there any open source tool to profile an opengl application on linux?[13:56:18] *** Kasu has joined ##OpenGL[13:56:32] *** UUncia has joined ##OpenGL[13:57:17] *** NightVisio has joined ##OpenGL[14:10:16] *** NightVisio has quit IRC[14:11:43] <SinnerG> anyone know a good freetype irc channel?[14:16:43] <SinnerG> problem I have is while rendering text, some characters like ')' are drawn to high, others like '-' are drawn 2 low :/[14:16:56] <SinnerG> abc.. 123.. are drawn at correct (same)height[14:17:04] <MatthiasM> SinnerG: each glyph has an x and y offset[14:18:38] <MatthiasM> take a look here: http://gltt.sourceforge.net/20011027_000000.html[14:19:43] <MatthiasM> but I suggest that you use a font texture - much easier to handle - and faster - eg Angelcode's BMFont - or Hiero on slick.cokeandcode.com[14:23:05] *** tokemonstah^ has joined ##OpenGL[14:24:23] <SinnerG> omg lol found it[14:24:26] <SinnerG> was messing :P[14:24:35] <SinnerG> Gl.glTranslatef(0, glyphrec.bitmap_top * -1 + glyphrec.bitmap.rows, 0); //[14:24:43] <SinnerG> instead of Gl.glTranslatef(0, glyphrec.bitmap_top - glyphrec.bitmap.rows, 0);[14:24:53] <MatthiasM> you use translate for every glyph ?!?[14:25:14] <SinnerG> no the code I use generates a list for each char when needed[14:25:22] <SinnerG> then just draws it[14:25:26] <MatthiasM> a DL for each glyph ?!?[14:25:37] <MatthiasM> you like to waste resources[14:26:05] <SinnerG> http://sinnerg.nopaste.dk/p527[14:26:18] <SinnerG> not my code.. I dont understand FT good enough to do that yet[14:26:27] *** dmh has joined ##opengl[14:27:22] <MatthiasM> SinnerG: what are lines 856 doing ?[14:27:25] <MatthiasM> and 857[14:27:44] <SinnerG> it caches the calls[14:28:02] <MatthiasM> huh?[14:28:06] <SinnerG> and yeah I added disable/enable ot make sure its enabled..[14:28:11] <SinnerG> lets just say its not perfect yet ;)[14:28:22] <MatthiasM> why disable before enabling then ?[14:28:44] *** eXtronuS has quit IRC[14:28:55] <SinnerG> since the cache gets corrupted sometimes.. I actually add that as precaution[14:28:56] <MatthiasM> well - I don't think you this code will provide acceptable performance ;)[14:29:00] <SinnerG> might work without the disable :P[14:29:28] <SinnerG> its more like a placeholder anyhow - I'll look into it myself later[14:29:30] <MatthiasM> you reset state that is already set each loop[14:29:45] <MatthiasM> you put push/pop matric into a DL for a single quad[14:33:07] *** GuShH has quit IRC[14:33:22] *** GuShH has joined ##OpenGL[14:33:36] *** tokemonstah has quit IRC[14:34:32] <SinnerG> yeah fixed that blending thing now :P[14:34:35] <SinnerG> was indeed overkill[14:34:56] <SinnerG> now it draws correctly, but not at the position where I want it.. looking into it[14:39:01] *** rabbit- has joined ##OpenGL[14:50:16] *** reprore_ has joined ##OpenGL[14:58:36] *** calav3ra has quit IRC[15:01:48] <SinnerG> temp fix I found was using some capital letter's position, and base the rest on that[15:01:50] *** dmh has quit IRC[15:02:38] <SinnerG> still not perfect :/[15:04:07] <SinnerG> good enough for now .. I'll create ticket to fix it later[15:08:53] *** walt_ has joined ##OpenGL[15:09:07] *** johndoe has joined ##opengl[15:10:04] *** burak575 has joined ##OpenGL[15:12:27] *** reprore_ has quit IRC[15:14:06] *** PainBank has joined ##OpenGL[15:15:07] *** walt_ has quit IRC[15:15:10] *** dmh has joined ##opengl[15:15:46] *** walt_ has joined ##OpenGL[15:16:42] *** walt has quit IRC[15:16:43] *** glcrazy has quit IRC[15:16:56] *** glcrazy has joined ##OpenGL[15:17:20] <SinnerG> http://www.youtube.com/watch?v=ocFpe_nngA4[15:19:53] *** walt_ has quit IRC[15:20:32] *** walt_ has joined ##OpenGL[15:22:11] *** Gorgoroth has quit IRC[15:23:36] *** Sudi has joined ##OpenGL[15:24:42] *** fargiola` is now known as fargiolas[15:27:02] *** Bollinger has joined ##OpenGL[15:29:35] *** grosours has quit IRC[15:33:24] *** gerard_ has left ##OpenGL[15:41:31] *** sohail has joined ##OpenGL[15:48:22] *** reprore_ has joined ##OpenGL[15:48:31] *** nitrotrigger has joined ##OpenGL[15:52:24] *** rabbit- has quit IRC[15:55:54] <burak575> do "kare" has a meaning?[15:59:02] <Adrinael> Karet has a meaning.[15:59:26] <Adrinael> http://en.wikipedia.org/wiki/KARE[16:00:11] <burak575> I asked because SinnerG was posted a youtube video with named Kare2D :)[16:05:32] *** rnx has joined ##opengl[16:06:44] <TheFlash> Using copyrighted background pic and music is... Sad.[16:07:08] *** djork has joined ##opengl[16:15:15] *** cplusplus has joined ##OpenGL[16:15:54] *** Wagoo has joined ##opengl[16:20:20] *** calav3ra has joined ##opengl[16:32:20] *** m4ggus has joined ##opengl[16:40:28] *** UUncia has quit IRC[16:46:21] *** groton has quit IRC[16:48:42] *** philou has joined ##OpenGL[16:49:32] *** rnx has quit IRC[16:49:56] *** rnx has joined ##opengl[16:50:54] *** philou has quit IRC[16:52:54] *** Xmas| has joined ##OpenGL[16:55:08] *** qeed has joined ##opengl[16:57:42] *** philou has joined ##OpenGL[16:58:52] *** philou has quit IRC[17:04:19] *** xslr has joined ##OpenGL[17:04:46] *** philou has joined ##OpenGL[17:22:42] *** glcrazy has quit IRC[17:22:51] *** glcrazy has joined ##OpenGL[17:23:25] *** ol1veira1 has quit IRC[17:26:09] *** ol1veira_ has joined ##OpenGL[17:29:11] *** walt_ is now known as walt[17:29:27] *** tokemonstah^ has quit IRC[17:30:17] *** pierrelux has joined ##OpenGL[17:33:28] *** urnotE has quit IRC[17:33:56] *** urnotE has joined ##OpenGL[17:34:08] *** urnotE has quit IRC[17:34:34] *** urnotE has joined ##OpenGL[17:34:47] *** reprore_ has quit IRC[17:35:55] *** johndoe has quit IRC[17:40:24] *** lewymati has quit IRC[17:41:47] *** urnotE has quit IRC[17:42:15] *** urnotE has joined ##OpenGL[17:46:17] *** dusted has joined ##OpenGL[17:47:28] *** goruka has joined ##OpenGL[17:51:19] *** sparky_ is now known as sparky[17:51:54] *** dmh has quit IRC[17:55:41] *** Tenac has joined ##opengl[18:02:40] *** Tenac_ has quit IRC[18:03:17] *** AlastairLynn has joined ##opengl[18:03:46] *** walt has quit IRC[18:04:05] *** xslr has quit IRC[18:04:33] *** midwaltz has joined ##OpenGL[18:04:45] *** walt has joined ##OpenGL[18:04:49] *** midwaltz has quit IRC[18:05:09] *** zwiep` has quit IRC[18:06:23] <sparky> good morning[18:11:20] *** glcrazy has quit IRC[18:11:26] *** glcrazy has joined ##OpenGL[18:14:53] *** XT95__ has joined ##OpenGL[18:17:10] *** XT95_ has quit IRC[18:20:38] *** philou has left ##OpenGL[18:23:04] *** elite01 has quit IRC[18:34:55] *** itewsh has joined ##OpenGL[18:49:07] <goruka> Here's a question![18:49:23] <goruka> I am doing hardware skinning[18:49:46] <goruka> so, if my skeleton has something like.. 100 bones[18:50:04] <goruka> is it feasible to send 100 matrices as uniforms to the shader? (one for each)[18:50:28] <quicksilver> I believe the answer is no.[18:50:35] <quicksilver> I think you want to send them as a texture[18:50:40] <quicksilver> which probably requires various extensions.[18:51:01] *** hackkitten has quit IRC[18:51:03] <goruka> that.. isn't nice[18:51:08] <TheFlash> 400 uniform vectors...[18:51:11] <goruka> not even in SM4.0 ?[18:51:12] *** dusted has quit IRC[18:51:28] <TheFlash> But where do you need 100 bones?[18:51:32] <quicksilver> doesn't seem un-nice to me.[18:51:41] <goruka> I mean not even on gf8xxx level hardware is possible?[18:51:41] <quicksilver> much faster to upload 100 matrices in a big block of data[18:51:49] <quicksilver> than as a 100 individual mini-data[18:51:58] <goruka> well, i need a matrix per bone[18:52:03] <quicksilver> so, "texture" (which really just means "big block of data") is the way to go.[18:52:10] <RTFM_FTW> use bindable uniform[18:52:43] *** wisey has joined ##OpenGL[18:52:49] <quicksilver> RTFM_FTW: oh, that's nice[18:52:58] <quicksilver> I never heard of that extension before[18:53:01] <goruka> RTFM_FTW, bindable uniform? how do i get info on that?[18:53:20] <quicksilver> google it :P I iddid.[18:53:24] <quicksilver> http://opengl.org/registry/specs/EXT/bindable_uniform.txt[18:53:35] <RTFM_FTW> http://opengl.org/registry/specs/EXT/bindable_uniform.txt[18:53:38] <RTFM_FTW> right there[18:53:46] <goruka> great, thanks![18:53:46] <RTFM_FTW> EXT_bindable_uniform[18:53:53] <RTFM_FTW> use it.[18:54:16] <RTFM_FTW> all DX10 level hardware supports this extension[18:54:55] <quicksilver> RTFM_FTW: without that extension, is using a sampler as an array a good way? or does all hardware which supports that support this?[18:55:34] <RTFM_FTW> umm using bindable uniform *is* the good way[18:55:36] <RTFM_FTW> heh[18:55:36] <goruka> awesome[18:56:12] <goruka> RTFM_FTW, any idea if opengl 3.1 will make this standard (as well as FBOs, multisampled FBOs, geometry shaders, etc?)[18:56:25] <RTFM_FTW> and as I said earlier anything DX10 level (i.e. ATI R6xx+, Nvidia G8x+) has the necessary (HW) support for this extension[18:57:41] <TheFlash> Quite high requirements...[18:57:45] <RTFM_FTW> although you already have FBO (+ blit and MSAA)[18:57:49] <RTFM_FTW> in GL 3.0[18:58:07] <RTFM_FTW> ARB_framebuffer_object solved that problem for you[19:02:26] *** phrosty has joined ##opengl[19:10:41] *** walt has quit IRC[19:19:28] *** wisey has quit IRC[19:20:31] *** mm765^sleep is now known as mm765[19:20:54] *** nettikkcah has joined ##opengl[19:21:05] *** x0rx0r has joined ##OpenGL[19:22:36] *** nettikkcah is now known as hackkitten[19:22:53] *** Candy has joined ##OpenGL[19:25:20] *** Candy has quit IRC[19:26:56] *** gaffa has joined ##opengl[19:30:06] *** phrosty has quit IRC[19:32:55] *** Sudi has quit IRC[19:39:52] *** glcrazy has quit IRC[19:46:52] *** phrosty has joined ##opengl[19:49:10] *** calav3ra has quit IRC[19:50:18] *** zwiep` has joined ##opengl[19:58:41] *** Gorgoroth has joined ##OpenGL[20:00:10] *** glcrazy has joined ##OpenGL[20:11:48] *** HideousNashimoto has joined ##OpenGL[20:12:22] <HideousNashimoto> Hey, I wonder can I use array and then that array draw on screen like some kind of image?[20:13:24] *** ol1veira1 has joined ##OpenGL[20:13:32] <MatthiasM> HideousNashimoto: glDrawPixels[20:17:00] <RTFM_FTW> you don't need DrawPixels for that :P[20:17:02] <RTFM_FTW> use a texture[20:17:49] <RTFM_FTW> learn to re-structure your problem around the use of texture mapped primitives (a single quad for example)[20:18:06] *** pietia has quit IRC[20:18:24] *** PainBank has left ##OpenGL[20:21:02] *** ol1veira_ has quit IRC[20:23:37] *** lewymati has joined ##OpenGL[20:24:12] *** lewymati has quit IRC[20:30:54] *** huperniketes has quit IRC[20:32:05] *** HuntsMan has joined ##opengl[20:34:49] *** groton has joined ##OpenGL[20:35:48] *** itewsh has quit IRC[20:38:05] *** jfroy has joined ##OpenGL[20:40:42] *** stringfellow has quit IRC[20:47:00] *** eXtronuS has joined ##OpenGL[20:50:03] *** neoneye has quit IRC[20:51:23] *** nargon has quit IRC[20:52:34] *** nargon has joined ##OpenGL[20:53:31] <HideousNashimoto> Has anyone used oglconsole?[20:53:41] <HideousNashimoto> the quake style console?[20:54:00] *** jcazevedo has joined ##OpenGL[20:54:11] <HideousNashimoto> I wonder how to set it up so that it pop ups automaticly, without pressint tilde(~)[20:59:44] *** GuShH has quit IRC[21:05:20] *** itewsh has joined ##OpenGL[21:12:02] *** Jupp3- has joined ##OpenGL[21:27:58] <HideousNashimoto> Is there any easy way to visualize glVertex3f(...); because I don't have idea how to create triangle, square, or whatever, I know that first argument is X, second is Y, and third is Z[21:30:06] *** eXtronuS has quit IRC[21:31:01] <HideousNashimoto> I can enter some values, then compile app, see results, and adjust, but that takes forever[21:37:47] *** Xmas| has quit IRC[21:39:38] *** amz has joined ##opengl[21:42:25] <MatthiasM> HideousNashimoto: get yourself a piece of graph paper and paint what you want on it - then read the X/Y coordinates[21:45:23] <rnx> or just use a 3d modeller and load/convert the model[21:45:49] <HideousNashimoto> rnx, what kind of 3d modeller do you reccomend?[21:46:51] <rnx> wings, blender, ... depends on goal, skill, platform, ...[21:47:43] <glcrazy> 3DS Max 2009 :D[21:48:34] <tmccrary> max is so late 90's[21:49:18] *** genom has joined ##OpenGL[21:49:29] <genom> Hello =)[21:49:30] <HideousNashimoto> rnx, well I am skilled in programming, but I don't know anything about modellinx[21:49:33] <HideousNashimoto> *moddeling[21:49:36] <HideousNashimoto> modeling[21:49:41] *** dv_ has joined ##opengl[21:50:33] <glcrazy> 3ds max 2009 price: $3495 :o:o:o[21:50:35] <HideousNashimoto> rnx, I want some modeler that will print me out all object vertices, so that I can just C/P it in my OGL application[21:50:59] <HideousNashimoto> glcrazy, I guess you haven't heard about "certain" websites[21:51:08] <HideousNashimoto> ;)[21:51:11] <glcrazy> :))[21:51:28] <glcrazy> i never heard of them[21:51:30] <glcrazy> :D[21:51:33] <HideousNashimoto> hehe[21:51:48] <genom> :)[21:52:26] <rnx> pretty much every modeller will export to several plain text formats like .obj[21:52:26] *** geocalc has quit IRC[21:52:27] <MatthiasM> HideousNashimoto: learn to read data files[21:53:22] *** geocalc has joined ##opengl[21:55:35] *** ccfreak2k has quit IRC[21:56:57] <HideousNashimoto> MatthiasM, show me how and I will![21:57:11] <HideousNashimoto> there are all kinds of data files[21:57:53] <tmccrary> the obj is a great place to start, as previously mentioned[21:57:58] <tmccrary> it's easy to read and understand[21:58:04] <tmccrary> it doesn't get any simpler than that[21:58:12] <genom> +1[22:04:33] <genom> i have a question[22:05:20] *** predaeus has quit IRC[22:05:57] <genom> is it possible to create custom interface with opengl ? Like this -> http://frenzic.com/desktop-video/[22:06:38] <RTFM_FTW> yes[22:06:44] <RTFM_FTW> that was done in OpenGL :)[22:06:56] <genom> awww =)[22:07:08] <genom> but i need sdl right ?[22:07:15] <genom> or glut[22:08:01] *** qeed has quit IRC[22:08:47] <genom> ya[22:09:10] <genom> but instead of the rectangular window (with the widgets inside)[22:09:22] <genom> i would like to create a custom window[22:09:39] <genom> but grat job ;-)[22:09:42] <genom> great*[22:10:51] <MatthiasM> well - I have not yet seen a way to create a none rectangular OpenGL window :)[22:11:00] <genom> damn ><[22:11:50] <genom> so there is no way to create custom window (non rectangular) with SDL or opengl or another multi plateform lybraries ?[22:12:14] <MatthiasM> opengl core does not handle window creation at all[22:12:33] <MatthiasM> and I know that on windows you can create non rectangular non-gl windows[22:12:33] *** mm765 is now known as mm765^away[22:12:34] <genom> so i need to take a look at SDL or glut ?[22:12:41] <genom> yes[22:12:43] <MatthiasM> glut can't for sure[22:12:52] <genom> ok =)[22:12:53] *** lewymati has joined ##OpenGL[22:13:06] <HideousNashimoto> I am using SDL and OpenGL, and when I try SDL_SaveBMP, I get segfault :([22:13:13] <HideousNashimoto> It doesn't happen when I am using SDL only[22:13:38] <MatthiasM> HideousNashimoto: this is most likly caused by a bad pointer from you[22:13:45] <RTFM_FTW> clever use of transparency could be used to give the appearance of a non rectangular view[22:14:04] <RTFM_FTW> transparency and masking actually[22:14:15] <genom> @RTFM_FTW i have thought about this thing =)[22:14:18] <HideousNashimoto> Damn genom, I opened that link while I was listening to my music, I thought I am going crazy[22:14:35] <genom> HideousNashimoto: =D[22:14:55] <genom> Thanks for all tips ;-)[22:18:18] <genom> MatthiasM: Does 2d (opengl) programming is hard ?[22:18:38] <HideousNashimoto> MatthiasM, it seems like you can't use SDL_SaveBMP, when using OpenGL[22:18:39] <MatthiasM> no[22:18:57] <MatthiasM> HideousNashimoto: does it say so in the documentation ?[22:18:58] <genom> ok ;-)[22:19:05] <HideousNashimoto> http://www.gamedev.net/community/forums/topic.asp?topic_id=343376[22:19:10] <HideousNashimoto> Checkout last post[22:19:53] <MatthiasM> well - bad luck then :)[22:20:13] <MatthiasM> I have code on my website to write a PNG screenhsot from OpenGL[22:20:36] <genom> MatthiasM: what is the URL ?[22:20:47] <MatthiasM> www.matthiasmann.de[22:20:54] <genom> Thank you =)[22:26:12] *** Zenja_Solaja has joined ##OpenGL[22:26:45] <Zenja_Solaja> Morning / afternoon / evening[22:30:56] *** djork has quit IRC[22:31:39] *** lewymati has quit IRC[22:31:55] <Zenja_Solaja> I have a simple question - the project I'm working on does not use glIndexPointer data at all, hence I do not use glDrawElements. I do however use glDrawArrays which just ploughs through the raw data, even though there duplicate vertices. I am aware that my solution consumes more memory. The question I have is "for future projects, should I switch to using indices and glDrawElements instead?" Anyone have any perfo[22:32:40] <Zenja_Solaja> For my current project, it's too late to switch, but I'd like to get into the habit of doing things the right way.[22:33:03] <Zenja_Solaja> Assuming that using glDrawElements performs better than glDrawArrays[22:33:13] <Zenja_Solaja> I haven't benchmarked that yet[22:33:17] <Zenja_Solaja> Looking for advice.[22:33:27] <MatthiasM> use glDrawRangeElements[22:35:00] *** wisey has joined ##OpenGL[22:35:06] <Zenja_Solaja> Why do you think glDrawRangeElements is faster than glDrawElements, if the arrays are the same size?[22:36:16] <RTFM_FTW> heh if you were curious :P[22:37:26] <burak575> lol it look like real random :D[22:37:33] <Zenja_Solaja> MatthiasM: OK, i'm reading the red book and it says that the ICD can prefetch data with range elements. Couldn't it do the same with plain old draw elements.[22:38:10] *** HuntsMan has quit IRC[22:38:21] <MatthiasM> Zenja_Solaja: do you think they would have added this method if it would not give any gain ?[22:38:58] <Zenja_Solaja> MatthiasM: what I'm interested in is will it be faster (small data sets) to plough through more data (glDrawArrays) than to hop around with less data (glDrawElements)?[22:39:43] <MatthiasM> well - how knows :)[22:39:48] <MatthiasM> *who[22:40:39] <RTFM_FTW> only the vendor(s) know :D[22:41:30] <vade> yea RTFM_FTW . im not so interested in noise, but structure[22:41:40] *** genom has quit IRC[22:41:40] <vade> so some of the things in the video I showed you interest me more[22:41:50] <vade> but i am also curious about dynamic movement[22:41:51] <Zenja_Solaja> MatthiasM: I'm googling around and discovered that a few people also claim that their benchmarks show that glDrawRangeElements is faster than glDrawElements - interesting.[22:42:02] <vade> parking on address + offset and watching memory change[22:42:32] <MatthiasM> Zenja_Solaja: you tell OpenGL which vertex range your indices will use - so there is a potential saving on processing[22:43:46] <MatthiasM> this gets interresting when you render a subset of the vertices with each call[22:44:27] <Zenja_Solaja> MatthiasM: but you also supply a range with glDrawElements - start and count. glDrawRangeElements supplies start and end. End can be calculated as start+count. So whats the difference?[22:44:29] <RTFM_FTW> well you could always add a bit of that randomness into what you are already generating[22:44:37] <MatthiasM> eg you have a character with vertex skinning - but because of uniform limits you can't render it in one batch - but this don't prevent you from putting all vertices into one VBO[22:44:46] <RTFM_FTW> that might be interesting[22:45:10] <MatthiasM> Zenja_Solaja: the start+end is for vertices - while offste and count is for indices[22:47:23] <Zenja_Solaja> I see (just read read book). You still have pointer and count, but you can restrict the indices range. ie, if you have 20 indices, you can restrict drawing to only indices 5-10. Interesting[22:47:31] <Zenja_Solaja> 2nd read = red[22:47:44] <MatthiasM> Zenja_Solaja: NO![22:48:03] <Zenja_Solaja> start and end are GLuint[22:48:14] <Zenja_Solaja> not void *[22:48:32] <MatthiasM> Zenja_Solaja: start and end are for the referenced vertices[22:48:42] <Zenja_Solaja> agree[22:49:33] <MatthiasM> it's just a hint for OpenGL to save computing or moving data[22:50:13] <Zenja_Solaja> Getting back to my original question, I can now see the performance benefits of glDrawElements vs glDrawArrays. The pipeline can cache transformetd/lit vertices, allowing them to be reused.[22:50:45] <MatthiasM> yes[22:50:55] <MatthiasM> Zenja_Solaja: http://home.comcast.net/~tom_forsyth/papers/fast_vert_cache_opt.html[22:51:32] <Zenja_Solaja> MatthiasM: thanks for the link - will read right away.[22:51:39] *** burak575 has quit IRC[22:54:26] *** pierrelux has quit IRC[22:55:17] <RTFM_FTW> in any case I wouldn't expect to see huge differences in the vertex processing rate between the index / no index data cases[22:55:53] <MatthiasM> except if you use a very long and slow vertex shader :)[22:55:57] <RTFM_FTW> OTOH the addition of index data does help the driver and allows for further (index data specific) optimizations on the server side[22:56:00] <RTFM_FTW> which is good[22:58:01] <Zenja_Solaja> MatthiasM: I have one final question for glIndexPointer. What happens to normals for shared vertices? A simple example is a cube - a index can reference 3 faces. Any way to have seperate normals, or do I have to 'combine' the normals?[22:58:43] <MatthiasM> either you combine the normal and get a "soft" corner - or you duplicate the vertex[23:00:51] <Zenja_Solaja> If you duplicate the vertex, you end up in a situation where you can stick with glDrawArrays and ignore indices (for the cube scenario). Or soft corners, as you've already pointed out.[23:00:53] <vade> http://aras-p.info/blog/2008/12/30/arb-draw-buffers/[23:02:44] *** jfroy has quit IRC[23:07:05] <RTFM_FTW> LOL[23:07:17] <RTFM_FTW> the ARB_draw_buffers spec is pretty tame :D[23:07:30] *** m4ggus has quit IRC[23:07:30] *** m4ggus_ has joined ##opengl[23:07:43] *** m4ggus_ is now known as m4ggus[23:09:56] *** Eforen has quit IRC[23:13:36] *** HuntsMan has joined ##opengl[23:16:55] *** XT95___ has joined ##OpenGL[23:20:08] <Zenja_Solaja> MatthiasM: I've finished reading the link you've sent me. Interestingly, he claims that indexed lists (optimised for HW caches) can be faster that indexed strips.[23:21:30] <Zenja_Solaja> First time I've read an article which says "dont bother with strips" - with the caveat of preprocessing the triangles to be cache friendly. He used XBOX1 and above generation hardware.[23:22:20] *** XT95__ has quit IRC[23:28:11] <RTFM_FTW> yeah it really depends[23:28:28] <RTFM_FTW> and it will vary between vendors and between ASICs[23:29:11] <RTFM_FTW> hell many "budget level" ASICs out there don't even have dedicated vertex caches for example :D[23:29:36] <RTFM_FTW> in the end this is worth profiling[23:29:41] <HuntsMan> RTFM_FTW: such as?[23:29:44] *** xp_prg has quit IRC[23:33:37] *** wisey has quit IRC[23:34:32] *** Spkka has quit IRC[23:36:36] <cplusplus> RTFM_FTW do you study?[23:38:21] *** twist has quit IRC[23:39:26] *** Zenja_Solaja has left ##OpenGL[23:39:34] *** iion has joined ##OpenGL[23:49:37] *** Renderwahn has quit IRC[23:51:30] *** dv_ has quit IRC[23:53:59] *** nargon has quit IRC[23:56:39] *** nargon has joined ##OpenGL