March 28, 2019 
< | 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 | >
March 28, 2019 [00:07:02] <Elronnd> is there a great performance penalty for rendering to a renderbuffer rather than directly to the screen? Or doing a few nested renderbuffer renders -- renderbuffer, to renderbuffer, to renderbuffer, to screen?[00:07:11] <Elronnd> Obviously there will be some, but is it significant?[00:25:18] <Stragus> You get extra some blit/copy operations, but the overhead of switching render targets. Don't abuse it and it's very manageable[00:40:30] *** upgrdman <upgrdman!~upgrdman@blender/artist/upgrdman> has quit IRC (Quit: Leaving)[00:46:00] *** fatalhalt <fatalhalt!~fatalhalt@c-67-163-60-93.hsd1.il.comcast.net> has joined ##OpenGL[01:03:28] *** BPL <BPL!~BPL@102.56.27.77.dynamic.reverse-mundo-r.com> has quit IRC (Quit: Leaving)[01:13:39] *** xx2 <xx2!~xx2@212.253.8.72> has quit IRC (Quit: Leaving)[01:23:29] *** KAHR-Alpha <KAHR-Alpha!~Alpha@2a01cb0d024a6600a5cdd878672cded9.ipv6.abo.wanadoo.fr> has quit IRC (Ping timeout: 250 seconds)[01:27:55] <derhass> a dozen of different passes is nothing special nowadays[01:50:03] *** learningc <learningc!~learningc@mti-37-145.tm.net.my> has joined ##OpenGL[01:51:56] *** davr0s <davr0s!~textual@host81-147-72-11.range81-147.btcentralplus.com> has quit IRC (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)[01:56:56] *** soulz <soulz!~soulz@unaffiliated/soulz> has quit IRC (Ping timeout: 252 seconds)[02:07:07] *** slidercrank <slidercrank!~slidercra@ircpuzzles/2015/april-fools/fifth/slidercrank> has quit IRC (Ping timeout: 245 seconds)[02:08:51] *** soulz <soulz!~soulz@unaffiliated/soulz> has joined ##OpenGL[02:23:48] *** ratchet_freak <ratchet_freak!~ratchetfr@ptr-82s3g7l9r42569r41f4.18120a2.ip6.access.telenet.be> has quit IRC (Ping timeout: 250 seconds)[03:05:03] *** groton <groton!~groton@unaffiliated/groton> has joined ##OpenGL[03:09:21] *** ville_ <ville_!~ville@212-149-214-47.bb.dnainternet.fi> has joined ##OpenGL[03:11:13] *** ville <ville!~ville@212-149-214-47.bb.dnainternet.fi> has quit IRC (Ping timeout: 246 seconds)[03:11:20] *** ville_ is now known as ville[03:14:51] *** tomha <tomha!~tomha@c-73-137-210-182.hsd1.ga.comcast.net> has joined ##OpenGL[03:17:15] *** esrse <esrse!~nil@175.126.171.165> has joined ##OpenGL[03:23:26] *** glYoda <glYoda!~MTLYoda@c-73-25-27-206.hsd1.or.comcast.net> has joined ##OpenGL[03:34:19] *** tomha <tomha!~tomha@c-73-137-210-182.hsd1.ga.comcast.net> has quit IRC (Read error: Connection reset by peer)[04:09:57] *** USER_ <USER_!~lucid-lul@66.115.169.237> has joined ##OpenGL[04:13:00] *** lucid-lullaby <lucid-lullaby!~lucid-lul@66.115.169.237> has quit IRC (Ping timeout: 250 seconds)[04:19:39] *** Elysion <Elysion!~Elysion@124-148-129-247.dyn.iinet.net.au> has joined ##OpenGL[04:42:30] *** Lord-Kamina <Lord-Kamina!~Lord-Kami@pc-250-87-104-200.cm.vtr.net> has quit IRC (Quit: Textual IRC Client: www.textualapp.com)[04:48:24] *** zalt <zalt!~lambda443@unaffiliated/lambda443> has quit IRC (Ping timeout: 250 seconds)[04:56:21] *** Twipply <Twipply!~Twipply@unaffiliated/twipply> has quit IRC (Quit: Leaving)[04:59:59] *** groton <groton!~groton@unaffiliated/groton> has quit IRC (Quit: groton)[05:04:25] *** lucid-lullaby <lucid-lullaby!~lucid-lul@66.115.169.237> has joined ##OpenGL[05:04:34] *** USER_ <USER_!~lucid-lul@66.115.169.237> has quit IRC (Ping timeout: 250 seconds)[05:07:01] *** groton <groton!~groton@unaffiliated/groton> has joined ##OpenGL[05:22:45] *** layeredcurkin <layeredcurkin!~reisima@2001:1530:1000:ad1e:74:95c5:654e:d143> has joined ##OpenGL[05:24:47] *** fatalhalt <fatalhalt!~fatalhalt@c-67-163-60-93.hsd1.il.comcast.net> has quit IRC (Quit: fatal halt)[05:33:55] *** immibis <immibis!~immibis@125-238-72-168-fibre.sparkbb.co.nz> has joined ##OpenGL[05:34:16] <layeredcurkin> there still might be a method to just have a path from registers to i-buffers or i-cache, but it looks like I need to work more on it. Perhaps this can have second dispatch NOP'ed on shorthened pipeline to ignore the flags. to boost the perf.[05:42:57] *** slime <slime!~slime73@24.215.81.93> has quit IRC (Quit: This computer has gone to sleep)[05:44:18] *** pie___ <pie___!~pie_@unaffiliated/pie-/x-0787662> has joined ##OpenGL[05:47:33] *** pie__ <pie__!~pie_@unaffiliated/pie-/x-0787662> has quit IRC (Ping timeout: 246 seconds)[06:00:27] *** LunarJetman2 <LunarJetman2!LunarJetma@5ec16e1a.skybroadband.com> has quit IRC (Ping timeout: 245 seconds)[06:02:15] *** cfoch <cfoch!uid153227@gateway/web/irccloud.com/x-qxnxrrhukjznretj> has quit IRC (Quit: Connection closed for inactivity)[06:05:29] <layeredcurkin> NVIDIA has a clever method, to substitute some of the L1 cache with the shared memory upon request. L1 is quite fast, but instruction parts of that are read-only, so on shortened pipeline on AMD cards it needs an extension to access or migrate it.[06:05:51] <layeredcurkin> since there is no need anymore to access it.[06:17:47] *** layeredcurkin <layeredcurkin!~reisima@2001:1530:1000:ad1e:74:95c5:654e:d143> has quit IRC (Read error: Connection reset by peer)[06:49:52] *** grim001 <grim001!~grim001@ip70-189-228-129.lv.lv.cox.net> has quit IRC (Ping timeout: 250 seconds)[07:08:30] *** qeed <qeed!~qeed@cpe-172-248-136-12.socal.res.rr.com> has quit IRC (Quit: qeed)[07:30:41] *** glYoda <glYoda!~MTLYoda@c-73-25-27-206.hsd1.or.comcast.net> has quit IRC (Quit: glYoda)[07:45:41] *** snyp <snyp!~Snyp@202.83.17.8> has joined ##OpenGL[08:02:34] *** stefkos <stefkos!~Pawel@79.184.19.214.ipv4.supernova.orange.pl> has joined ##OpenGL[08:06:37] *** stefkos <stefkos!~Pawel@79.184.19.214.ipv4.supernova.orange.pl> has quit IRC (Ping timeout: 246 seconds)[08:07:12] *** stefkos <stefkos!~Pawel@79.184.19.214.ipv4.supernova.orange.pl> has joined ##OpenGL[08:08:26] *** groton <groton!~groton@unaffiliated/groton> has quit IRC (Quit: groton)[08:26:45] *** pie__ <pie__!~pie_@unaffiliated/pie-/x-0787662> has joined ##OpenGL[08:27:34] *** neure <neure!~timo@mobile-access-b04807-94.dhcp.inet.fi> has joined ##OpenGL[08:28:46] *** pie___ <pie___!~pie_@unaffiliated/pie-/x-0787662> has quit IRC (Read error: Connection reset by peer)[08:28:47] *** Dave13h <Dave13h!~Dave@gibson.int13h.com> has quit IRC (Ping timeout: 245 seconds)[08:30:33] *** slvn_ <slvn_!~slvn_@c2s31-1-78-245-90-111.fbx.proxad.net> has joined ##OpenGL[08:31:08] *** neure <neure!~timo@mobile-access-b04807-94.dhcp.inet.fi> has quit IRC (Remote host closed the connection)[08:40:15] *** davr0s <davr0s!~textual@host81-147-72-11.range81-147.btcentralplus.com> has joined ##OpenGL[08:50:34] *** Foaly <Foaly!~Foaly@ppp-88-217-60-172.dynamic.mnet-online.de> has joined ##OpenGL[08:52:08] *** excalith <excalith!~excalith@195.175.206.218> has joined ##OpenGL[09:12:04] *** slidercrank <slidercrank!~slidercra@ircpuzzles/2015/april-fools/fifth/slidercrank> has joined ##OpenGL[09:28:02] *** davr0s <davr0s!~textual@host81-147-72-11.range81-147.btcentralplus.com> has quit IRC (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)[09:42:36] *** davr0s <davr0s!~textual@host81-147-72-11.range81-147.btcentralplus.com> has joined ##OpenGL[09:42:36] *** glYoda <glYoda!~MTLYoda@c-73-25-27-206.hsd1.or.comcast.net> has joined ##OpenGL[09:49:12] *** Matthijs <Matthijs!~quassel@unaffiliated/matthijs> has joined ##OpenGL[09:53:13] *** zalt <zalt!~lambda443@unaffiliated/lambda443> has joined ##OpenGL[09:53:22] *** begui <begui!~bj@108-200-216-48.lightspeed.dybhfl.sbcglobal.net> has quit IRC (Ping timeout: 245 seconds)[09:58:22] *** BPL <BPL!~BPL@102.56.27.77.dynamic.reverse-mundo-r.com> has joined ##OpenGL[10:07:58] *** halbeno <halbeno!~halbeno@d50-93-37-115.abhsia.telus.net> has quit IRC (Ping timeout: 245 seconds)[10:11:33] *** halbeno <halbeno!~halbeno@d50-93-37-115.abhsia.telus.net> has joined ##OpenGL[10:35:07] *** slidercrank <slidercrank!~slidercra@ircpuzzles/2015/april-fools/fifth/slidercrank> has quit IRC (Ping timeout: 255 seconds)[10:41:19] *** immibis <immibis!~immibis@125-238-72-168-fibre.sparkbb.co.nz> has quit IRC (Ping timeout: 246 seconds)[10:48:33] *** davr0s <davr0s!~textual@host81-147-72-11.range81-147.btcentralplus.com> has quit IRC (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)[10:51:21] *** mukunda <mukunda!mukunda@cpe-104-175-176-150.socal.res.rr.com> has quit IRC (Ping timeout: 246 seconds)[11:14:06] *** Serus <Serus!~Serus@unaffiliated/serus> has quit IRC (Ping timeout: 246 seconds)[11:15:33] *** User_ <User_!~learningc@mti-37-145.tm.net.my> has joined ##OpenGL[11:18:46] *** learningc <learningc!~learningc@mti-37-145.tm.net.my> has quit IRC (Ping timeout: 255 seconds)[11:19:19] *** User_ <User_!~learningc@mti-37-145.tm.net.my> has quit IRC (Read error: Connection reset by peer)[11:20:41] *** learningc <learningc!~learningc@mti-37-145.tm.net.my> has joined ##OpenGL[11:24:07] *** User_ <User_!~learningc@mti-37-145.tm.net.my> has joined ##OpenGL[11:25:31] *** Serus <Serus!~Serus@unaffiliated/serus> has joined ##OpenGL[11:25:56] *** learningc <learningc!~learningc@mti-37-145.tm.net.my> has quit IRC (Ping timeout: 244 seconds)[11:46:52] *** User__ <User__!~learningc@mti-37-145.tm.net.my> has joined ##OpenGL[11:49:42] *** User_ <User_!~learningc@mti-37-145.tm.net.my> has quit IRC (Ping timeout: 244 seconds)[11:52:57] *** User__ <User__!~learningc@mti-37-145.tm.net.my> has quit IRC (Ping timeout: 245 seconds)[11:53:28] *** learningc <learningc!~learningc@mti-37-145.tm.net.my> has joined ##OpenGL[11:55:48] *** User_ <User_!~learningc@mti-37-145.tm.net.my> has joined ##OpenGL[11:55:55] *** User_ <User_!~learningc@mti-37-145.tm.net.my> has quit IRC (Max SendQ exceeded)[11:58:16] *** learningc <learningc!~learningc@mti-37-145.tm.net.my> has quit IRC (Ping timeout: 250 seconds)[12:00:05] *** learningc <learningc!~learningc@mti-37-145.tm.net.my> has joined ##OpenGL[12:04:46] *** slime <slime!~slime73@blk-215-81-93.eastlink.ca> has joined ##OpenGL[12:05:19] *** learningc <learningc!~learningc@mti-37-145.tm.net.my> has quit IRC (Ping timeout: 246 seconds)[12:06:00] *** learningc <learningc!~learningc@mti-37-145.tm.net.my> has joined ##OpenGL[12:11:11] *** User__ <User__!~learningc@mti-37-145.tm.net.my> has joined ##OpenGL[12:11:52] *** learningc <learningc!~learningc@mti-37-145.tm.net.my> has quit IRC (Ping timeout: 255 seconds)[12:17:15] *** User_ <User_!~learningc@mti-37-145.tm.net.my> has joined ##OpenGL[12:17:31] *** User_ <User_!~learningc@mti-37-145.tm.net.my> has quit IRC (Max SendQ exceeded)[12:19:20] *** learningc <learningc!~learningc@mti-37-145.tm.net.my> has joined ##OpenGL[12:19:58] *** User__ <User__!~learningc@mti-37-145.tm.net.my> has quit IRC (Ping timeout: 255 seconds)[12:21:07] *** slime <slime!~slime73@blk-215-81-93.eastlink.ca> has quit IRC (Quit: This computer has gone to sleep)[12:24:16] *** learningc <learningc!~learningc@mti-37-145.tm.net.my> has quit IRC (Ping timeout: 250 seconds)[12:25:13] *** Foaly <Foaly!~Foaly@ppp-88-217-60-172.dynamic.mnet-online.de> has quit IRC (Quit: Now 'mid shadows deep falls blessed sleep.)[12:31:34] *** zalt <zalt!~lambda443@unaffiliated/lambda443> has quit IRC (Ping timeout: 272 seconds)[12:50:30] *** zalt <zalt!~lambda443@unaffiliated/lambda443> has joined ##OpenGL[12:52:40] *** davr0s <davr0s!~textual@host81-147-72-11.range81-147.btcentralplus.com> has joined ##OpenGL[13:01:05] *** davr0s <davr0s!~textual@host81-147-72-11.range81-147.btcentralplus.com> has quit IRC (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)[13:04:45] *** eetsi123 <eetsi123!~eetu@91-152-205-56.elisa-laajakaista.fi> has quit IRC (Quit: WeeChat 2.4)[13:06:28] *** hellozee <hellozee!~hellozee@116.73.36.92> has joined ##OpenGL[13:14:23] *** begui <begui!~bj@108-200-216-48.lightspeed.dybhfl.sbcglobal.net> has joined ##OpenGL[13:24:25] *** excalith is now known as excalith[afk][13:25:50] *** gareppa <gareppa!~gareppa@unaffiliated/gareppa> has joined ##OpenGL[13:28:10] *** gareppa <gareppa!~gareppa@unaffiliated/gareppa> has quit IRC (Remote host closed the connection)[13:39:09] *** excalith[afk] is now known as excalith[13:47:37] *** rocketmagnet <rocketmagnet!~username@unaffiliated/rocketmagnet> has joined ##OpenGL[13:48:03] <rocketmagnet> hi everyone, i got me the bluebook but i'm not able to compile the examples i'm on linux, what packages do i need to install ?[13:48:13] <rocketmagnet> i get the error undefined reference to `glXGetProcAddress'[13:48:25] <rocketmagnet> and undefined reference to `dlopen'[13:51:49] <derhass> this is not a matter of packages to install, this is a matter of actually linking the required libraries[13:56:21] *** zalt_ <zalt_!~lambda443@unaffiliated/lambda443> has joined ##OpenGL[13:59:03] *** learningc <learningc!~learningc@2001:d08:d6:3c3d:f820:4eed:acf8:f7ca> has joined ##OpenGL[14:00:14] *** zalt <zalt!~lambda443@unaffiliated/lambda443> has quit IRC (Ping timeout: 268 seconds)[14:02:35] *** learningc <learningc!~learningc@2001:d08:d6:3c3d:f820:4eed:acf8:f7ca> has quit IRC (Client Quit)[14:02:54] *** learningc <learningc!~learningc@2001:d08:d6:3c3d:f820:4eed:acf8:f7ca> has joined ##OpenGL[14:04:48] *** Twipply <Twipply!~Twipply@unaffiliated/twipply> has joined ##OpenGL[14:07:10] *** Foaly <Foaly!~Foaly@ppp-88-217-60-172.dynamic.mnet-online.de> has joined ##OpenGL[14:07:32] *** esrse <esrse!~nil@175.126.171.165> has quit IRC (Ping timeout: 250 seconds)[14:18:49] *** zalt__ <zalt__!~lambda443@unaffiliated/lambda443> has joined ##OpenGL[14:19:15] *** cfoch <cfoch!uid153227@gateway/web/irccloud.com/x-iqdaqmldgeicmogc> has joined ##OpenGL[14:22:16] *** zalt_ <zalt_!~lambda443@unaffiliated/lambda443> has quit IRC (Ping timeout: 250 seconds)[14:24:56] *** excalith is now known as excalith[afk][14:27:23] <rocketmagnet> hi everyone, i got me the opengl blue book but i get the error sb6code/include/sb6.h:283:17: error: variable or field ‘GLFWCALL’ declared void[14:28:05] <Stragus> I have no idea what sb6code or sb6.h are, do you need that stuff?[14:29:48] <Yaniel> probably some framework used by the book?[14:33:27] <derhass> https://github.com/openglsuperbible/sb6code[14:34:58] *** excalith[afk] is now known as excalith[14:44:09] *** davr0s <davr0s!~textual@host81-147-72-11.range81-147.btcentralplus.com> has joined ##OpenGL[15:00:04] *** diwr <diwr!~diwr@unaffiliated/diwr> has joined ##OpenGL[15:01:14] *** grouse <grouse!~grouse@83-233-9-2.cust.bredband2.com> has joined ##OpenGL[15:04:51] *** snyp <snyp!~Snyp@202.83.17.8> has quit IRC (Ping timeout: 272 seconds)[15:07:03] *** Lord-Kamina <Lord-Kamina!~Lord-Kami@pc-250-87-104-200.cm.vtr.net> has joined ##OpenGL[15:10:36] *** diwr <diwr!~diwr@unaffiliated/diwr> has quit IRC (Quit: Konversation terminated!)[15:14:15] *** LunarJetman2 <LunarJetman2!LunarJetma@5ec16e1a.skybroadband.com> has joined ##OpenGL[15:15:02] *** soulz <soulz!~soulz@unaffiliated/soulz> has quit IRC (Quit: leaving)[15:22:04] *** stefkos <stefkos!~Pawel@79.184.19.214.ipv4.supernova.orange.pl> has quit IRC (Quit: Going offline, see ya! (www.adiirc.com))[15:23:22] *** soulz <soulz!~soulz@unaffiliated/soulz> has joined ##OpenGL[15:31:16] *** zalt_ <zalt_!~lambda443@unaffiliated/lambda443> has joined ##OpenGL[15:34:38] *** charlie5 <charlie5!~rod@pa49-180-139-151.pa.nsw.optusnet.com.au> has quit IRC (Ping timeout: 245 seconds)[15:34:58] *** zalt__ <zalt__!~lambda443@unaffiliated/lambda443> has quit IRC (Ping timeout: 246 seconds)[15:49:12] *** charlie5 <charlie5!~rod@pa49-180-41-4.pa.nsw.optusnet.com.au> has joined ##OpenGL[15:56:31] <Serus> I'm struggling with getting 2 really simple shaders to compile on a raspberry pi (GLES2), it compiles simple.vert, but glGetShaderiv with GL_COMPILE_STATUS returns false for simple.frag[15:56:33] <Serus> https://gist.github.com/DragoonX6/cb2d88a9159f1046291bc571f50eb5f4[15:57:20] <Stragus> Did you ask for a shader compilation log?[15:57:32] <Stragus> It will tell you why it doesn't like the shader[15:57:36] <Serus> yes, and it's the most useless thing ever[15:57:43] <Serus> It literally says "Compiled"[15:59:18] <Stragus> Ahah[15:59:49] <Stragus> It has been a long time since I have touched GLSL 1.0, but nothing wrong pops up to me[16:00:13] <Serus> https://www.khronos.org/opengl/wiki/Data_Type_(GLSL)#Vector_constructors[16:00:17] <Serus> Is this supported?[16:00:31] <Stragus> Maybe the problem is in your code, the shader compiled but you are querying something else[16:00:31] <Serus> A friend asked if vec4(1.0) is supported[16:01:01] <Stragus> I believe that was supported since day one[16:01:33] <Serus> I first call glCompileShader, then glGetError, and then glGetSahderiv with GL_COMPILE_STATUS[16:08:27] <Serus> ok what[16:08:33] <Serus> now I'm really confused[16:08:54] <Serus> if I assert on the compile status[16:08:57] <Serus> I crash[16:09:58] <Serus> if I printf it, with fprintf(stderr, "Compiled %s", compiled == GL_TRUE ? "successfully" : "erroneously"); it outputs "Compiled successfully"[16:10:26] <Serus> Then the warnings I get are "Compiled"[16:10:43] <Stragus> If that's really what happens, it's a serious case of undefined behavior due to some earlier bug[16:11:04] <Serus> And the program fails to link with "ERROR:LINK-9 (link time, line -1) Missing main function for shader"[16:15:23] <Serus> Ok, it magically gets past the asserts now, I think I may have forgotten to update the shaders earlier[16:15:41] *** upgrdman <upgrdman!~upgrdman@blender/artist/upgrdman> has joined ##OpenGL[16:23:08] *** excalith is now known as excalith[afk][16:24:50] <Serus> Am I supposed to attach shaders after compiling?[16:24:56] <Serus> Or can I do this before compiling?[16:27:58] <Stragus> Hrm... Perhaps you can attach without linking before compiling, but that's a bit weird. Just compile, attach, link[16:29:53] *** excalith[afk] is now known as excalith[16:30:10] <Serus> I do attach, compile, link, right now[16:31:00] <Stragus> I guess that works, I don't remember the specs and never have done it that way[16:31:39] <derhass> the spec doesn't care[16:32:06] <derhass> you need to compile before link, and obviously, also attach before link[16:32:22] <derhass> you can compile after it is attached, and also re-compile[16:32:39] <derhass> Stragus: so show the _actual_ code[16:32:43] <derhass> sorry[16:32:52] <derhass> Serus: so show the actual code[16:33:06] <Serus> Uh, how much of it?[16:33:20] <derhass> the relevant parts (tm).[16:33:40] <Serus> I'll do my best[16:33:56] <derhass> I can assure you that simple shaders do work on the RPi[16:34:14] <derhass> (above that, it becomes tricky, but that is another story)[16:34:39] <Serus> I know that shaders do work, it works in the samples[16:35:12] <derhass> well, we suspect there is some issue with you client code[16:35:26] <Serus> obviously, yes[16:35:42] <Serus> Well, it's going to be a bit much, but I hope it provides enough context[16:37:23] *** Twipply <Twipply!~Twipply@unaffiliated/twipply> has quit IRC (Quit: Leaving)[16:38:47] *** Foaly <Foaly!~Foaly@ppp-88-217-60-172.dynamic.mnet-online.de> has quit IRC (Quit: Now 'mid shadows deep falls blessed sleep.)[16:41:43] <Serus> https://gist.github.com/DragoonX6/cb2d88a9159f1046291bc571f50eb5f4[16:41:48] <Serus> It's basically all of it[16:42:08] <Serus> But I don't want to redact anything that can be of help in understanding what is there[16:42:49] <Serus> As I wrote in a comment, UniqueHandle<T> is simply a move only wrapper for handles, that sets the moved from to 0 when moving[16:43:03] <Serus> Observer<T> == T*[16:54:31] *** foolsh <foolsh!~quassel@89.187.181.164> has joined ##OpenGL[16:54:41] <Serus> derhass, Stragus, any idea?[16:54:53] <Serus> If you're reading it at all[16:55:57] <Stragus> It's a bit too much code for me, and I'm pretty sure it's a code/C++ problem rathern than a GL problem[16:56:17] *** qeed <qeed!~qeed@cpe-172-248-136-12.socal.res.rr.com> has joined ##OpenGL[16:56:22] <Serus> Well, I made sure the shaders are loaded properly by outputting them[16:56:32] <Serus> They compile properly[16:56:53] <Serus> Not to say I don't make mistakes, but I went over this so much now I'm sure the code is fine[16:57:04] <Serus> At least memory management wise[17:04:23] *** Foaly <Foaly!~Foaly@ppp-88-217-60-172.dynamic.mnet-online.de> has joined ##OpenGL[17:06:41] *** davr0s <davr0s!~textual@host81-147-72-11.range81-147.btcentralplus.com> has quit IRC (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)[17:11:06] *** grouse <grouse!~grouse@83-233-9-2.cust.bredband2.com> has quit IRC (Quit: Leaving)[17:15:58] <derhass> Serus: I was distracted by something else, I will have a look now[17:21:15] *** Lunatrius <Lunatrius!~Lunatrius@77.38.91.100> has quit IRC (Ping timeout: 246 seconds)[17:22:06] *** Lunatrius <Lunatrius!~Lunatrius@77.38.77.180> has joined ##OpenGL[17:26:32] *** Lunatrius` <Lunatrius`!~Lunatrius@77.38.39.142> has joined ##OpenGL[17:26:54] <derhass> Serus: I find your code seriously dangerous[17:27:07] *** Lunatrius <Lunatrius!~Lunatrius@77.38.77.180> has quit IRC (Ping timeout: 245 seconds)[17:27:13] *** englosh <englosh!5bba3e9e@gateway/web/freenode/ip.91.186.62.158> has joined ##OpenGL[17:27:17] *** Lunatrius` is now known as Lunatrius[17:27:47] <englosh> how do I make a rectangle with custom width and height?[17:28:01] <Matthijs> How do you make a rectangle without custom width a height/[17:30:53] <englosh> I have this so far (the first triangle of the rectangle) : -1, 1, 1, 1, -1, -1, and when I make it this -1, 1, 1, 1, -1, -height, and height is 0.8 for example then it kinda works but when the height is 0, the triangle has the height of the half screen[17:30:56] <englosh> but it shouldnt be visible then[17:30:59] <englosh> because the height is 0[17:32:00] <Matthijs> You don't specify width/height, you specify x/y[17:32:20] <Matthijs> So it's not "0" height, it's "start at -1", "end at 0"[17:32:53] <englosh> what? but shouldn't I use the x and y values for the translate matrix?[17:33:37] <englosh> I basically want the middle (0) to be on the top of the screen instead of actually in the middle[17:33:50] <englosh> so that height 0.5 would be the half[17:33:50] <englosh> not 0[17:36:03] *** snyp <snyp!~Snyp@1.39.138.101> has joined ##OpenGL[17:37:02] <Matthijs> You can either configure the viewport to accomplish that (glViewport), or multiply/translating the values in you matrix: value * 2 - 1 is what you need (0.5 * 2 - 1 = 0)[17:37:31] <Serus> derhass: what about it?[17:39:40] <derhass> Serus: The way your abstractions work. The destructs for GL object. I see that you rely on this UniqueHandle thing and move semantics, but it is easy to blow in your face if you make some sublte mistake somewhere[17:40:23] <Serus> hmmm, I'll get to there eventually[17:40:39] <Serus> But anything that stands out in terms of the gl code?[17:41:11] <derhass> I am too far busy following all that C++ stuff to look at the actual GL[17:44:19] <Serus> :/[17:47:22] <derhass> Serus: I don't see anything obvoiusly broken with the GL calls there[17:48:31] <Serus> hmmm[17:49:53] *** excalith is now known as excalith[afk][17:50:18] <derhass> Serus: but just for fun, I would just print the id in the constructors and desctuctors of the Shader and ShaderProgram, just to be sure[17:50:32] <derhass> also, I'd print the actual strings passed to glShaderSourcem and the size[17:50:56] <Serus> I already confirmed those with the debugger, but I'll add that[17:51:41] <Serus> wait, hmm[17:51:44] <Serus> maybe the size[17:51:58] <Serus> and I guess I can fetch the source, to see if it's actually there[17:52:51] *** Matthijs <Matthijs!~quassel@unaffiliated/matthijs> has quit IRC (Quit: *waves frantically*)[17:58:39] <Serus> aha[17:58:42] <Serus> this explains a lot[17:59:19] *** excalith[afk] is now known as excalith[18:00:51] <Serus> aw shit, it works now[18:03:18] *** slvn_ <slvn_!~slvn_@c2s31-1-78-245-90-111.fbx.proxad.net> has quit IRC (Remote host closed the connection)[18:05:03] *** slvn_ <slvn_!~slvn_@c2s31-1-78-245-90-111.fbx.proxad.net> has joined ##OpenGL[18:15:41] *** davr0s <davr0s!~textual@host81-147-72-11.range81-147.btcentralplus.com> has joined ##OpenGL[18:17:02] *** excalith is now known as excalith[afk][18:19:12] <derhass> Serus: so what was it?[18:31:43] <Serus> the size of the shader source was 0[18:32:33] <Serus> In my file reading routine I had an erroneous check, where I checked if the capacity of the vector was enough to hold the data, I should've checked the size[18:32:48] <Serus> I just removed the check entirely and just resize[18:36:59] *** wildlander <wildlander!~wildlande@unaffiliated/wildlander> has joined ##OpenGL[18:38:29] *** xx2 <xx2!~xx2@212.253.8.72> has joined ##OpenGL[18:50:04] *** excalith[afk] is now known as excalith[18:59:57] *** Foaly <Foaly!~Foaly@ppp-88-217-60-172.dynamic.mnet-online.de> has quit IRC (Quit: Now 'mid shadows deep falls blessed sleep.)[19:01:36] *** ratchet_freak <ratchet_freak!~ratchetfr@ptr-82s3g7l9r42569r41f4.18120a2.ip6.access.telenet.be> has joined ##OpenGL[19:02:52] *** xx2 <xx2!~xx2@212.253.8.72> has quit IRC (Quit: Leaving)[19:04:47] *** learningc <learningc!~learningc@2001:d08:d6:3c3d:f820:4eed:acf8:f7ca> has quit IRC (Ping timeout: 240 seconds)[19:08:34] *** LunarJetman2 <LunarJetman2!LunarJetma@5ec16e1a.skybroadband.com> has quit IRC (Ping timeout: 250 seconds)[19:28:21] *** davr0s <davr0s!~textual@host81-147-72-11.range81-147.btcentralplus.com> has quit IRC (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)[19:31:12] *** Foaly <Foaly!~Foaly@ppp-88-217-93-222.dynamic.mnet-online.de> has joined ##OpenGL[19:31:45] *** daniels <daniels!sid30311@gateway/web/irccloud.com/x-njbytfawluxwmeao> has quit IRC (Excess Flood)[19:31:57] *** daniels <daniels!sid30311@gateway/web/irccloud.com/x-mktrbegoinlfkjmr> has joined ##OpenGL[19:43:24] *** LunarJetman2 <LunarJetman2!LunarJetma@5ec16e1a.skybroadband.com> has joined ##OpenGL[19:43:28] *** Albori <Albori!~Albori@216-229-75-117.fidnet.com> has quit IRC (Ping timeout: 246 seconds)[19:45:03] *** krystcich <krystcich!~krystcich@088156132005.dynamic-ww-04.vectranet.pl> has joined ##OpenGL[19:45:50] *** Albori <Albori!~Albori@216-229-75-117.fidnet.com> has joined ##OpenGL[19:50:37] *** davr0s <davr0s!~textual@host81-147-72-11.range81-147.btcentralplus.com> has joined ##OpenGL[19:51:18] *** R2robot <R2robot!~R2robot@unaffiliated/r2ro> has quit IRC (Quit: Pull the lever, Kronk. Wrong leverrrrrrr!)[19:54:41] *** R2robot <R2robot!~R2robot@unaffiliated/r2ro> has joined ##OpenGL[20:01:22] *** zalt__ <zalt__!~lambda443@unaffiliated/lambda443> has joined ##OpenGL[20:04:09] *** zalt_ <zalt_!~lambda443@unaffiliated/lambda443> has quit IRC (Ping timeout: 244 seconds)[20:22:32] *** KAHR-Alpha <KAHR-Alpha!~Alpha@2a01cb0d024a660074ad48bc729385c6.ipv6.abo.wanadoo.fr> has joined ##OpenGL[20:25:20] *** gareppa <gareppa!~gareppa@unaffiliated/gareppa> has joined ##OpenGL[20:26:22] *** gareppa <gareppa!~gareppa@unaffiliated/gareppa> has quit IRC (Client Quit)[20:27:41] *** Matthijs <Matthijs!~quassel@unaffiliated/matthijs> has joined ##OpenGL[20:30:07] *** Twipply <Twipply!~Twipply@unaffiliated/twipply> has joined ##OpenGL[20:31:42] <Matthijs> When I make my window very wide and not so high (say 2500x500), and turn the camera, my models get warped towards the side http://puu.sh/D6FJ0/52c75bc23a.jpg They also appear closer. I update the viewport and aspect ratio on a resize, but it seems something is going wrong. Or not? I'm not sure what I should even expect here...[20:32:58] <Serus> Matthijs: do you update your matrices to reflect the new dimensions?[20:37:24] *** snyp <snyp!~Snyp@1.39.138.101> has quit IRC (Ping timeout: 250 seconds)[20:40:27] <Matthijs> Serus: Yes... but maybe I just found the bug as I checked that. When computing the projection matrix I divided the fov by 2, for some reason.[20:40:31] *** slidercrank <slidercrank!~slidercra@ircpuzzles/2015/april-fools/fifth/slidercrank> has joined ##OpenGL[20:40:48] *** snyp <snyp!~Snyp@1.39.132.205> has joined ##OpenGL[20:41:42] *** excalith <excalith!~excalith@195.175.206.218> has quit IRC (Quit: excalith)[20:44:27] <Matthijs> And I suspect I should take the aspect ratio into account there.[20:44:32] *** immibis <immibis!~immibis@125-238-72-168-fibre.sparkbb.co.nz> has joined ##OpenGL[20:45:15] <zy]x[yz> you probably divided the fov by two because you want to represent the angle from edge to edge, but your arithmetic wants a center-to-edge angle[20:46:15] <Matthijs> Removing the division actually makes it worse.[20:48:17] <zy]x[yz> yeah, most frustum matrix formula i've seen assumed a center-to-edge angle value[20:48:46] <zy]x[yz> or all, rather[20:49:12] <derhass> Matthijs: show the code[20:49:14] *** Twipply <Twipply!~Twipply@unaffiliated/twipply> has quit IRC (Ping timeout: 250 seconds)[20:52:58] <Matthijs> Code: http://puu.sh/D6GsF/9b24b5030f.txt[20:53:33] *** Twipply <Twipply!~Twipply@unaffiliated/twipply> has joined ##OpenGL[20:57:03] <snyp> m[3,3] should be 1.[20:57:15] <zy]x[yz> not for a projection matrix[20:57:43] <snyp> nvm. forgot.[20:58:18] <zy]x[yz> i only had that on the top of my head because i made that mistake recently[21:01:09] <snyp> I see. I've had a long-standing confusion sitting in my head for a while which I cleared a couple days back. You can project (mentally at least) to the near plane and then lerp into the NDC [-1, 1] range, or project to the [-1, 1] plane parallel to the near plane (located at 1/tan(yfov / 2) distance along -z axis).[21:02:25] *** immibis <immibis!~immibis@125-238-72-168-fibre.sparkbb.co.nz> has quit IRC (Ping timeout: 255 seconds)[21:03:15] <derhass> snyp: I totally agree to the first part. for the part beginning with "or project ...", I don't know what you even mean[21:05:27] <snyp> derhass: I mean the plane located at z = -1/tan(yfov/2) has a y extent of -1 to 1 (where it cuts the frustum, well not the "frustum", the square pyramid in this case)[21:05:37] <derhass> snyp: I think I get it[21:05:51] <derhass> snyp: the phrasing [-1 ,1] plane" threw me off[21:05:57] <snyp> oh[21:06:36] <derhass> but I do get your point now[21:06:45] <derhass> one could certainly look at things that way[21:07:51] <derhass> any plane parallel to the near plane is as good an image plane as any other anyway (with the exception of z=0, and also the mirroring for behind the camera)[21:09:22] <snyp> I see. Yeah, approaching the projection matrix in the those two ways (for transforming the x and y components) you see you don't depend on the near plane's location on z axis (the 'near_z' variable cancels out in the indirect approach).[21:10:09] <derhass> yeah, the classic frustum definition just uses the AABB oin the near plane[21:12:19] <snyp> Lengyel's book defines the projection matrix using the near plane rectangle (top, bottom, left, right in view space) to arrive at the projection matrix, so directly using that 1/tan(...) thing was weird to me.[21:13:11] <snyp> (again.. 'mentally')[21:13:52] <Matthijs> Mentally is the easy part ;)[21:16:30] *** slime <slime!~slime73@blk-215-81-93.eastlink.ca> has joined ##OpenGL[21:18:26] *** davr0s <davr0s!~textual@host81-147-72-11.range81-147.btcentralplus.com> has quit IRC (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)[21:19:48] *** davr0s <davr0s!~textual@host81-147-72-11.range81-147.btcentralplus.com> has joined ##OpenGL[21:20:56] *** eetsi123 <eetsi123!~eetu@91-152-205-56.elisa-laajakaista.fi> has joined ##OpenGL[21:21:53] *** eetsi123 <eetsi123!~eetu@91-152-205-56.elisa-laajakaista.fi> has quit IRC (Client Quit)[21:23:51] *** mandeep <mandeep!~mandeep@unaffiliated/mandeepb> has joined ##OpenGL[21:35:29] *** Matthijs <Matthijs!~quassel@unaffiliated/matthijs> has quit IRC (Quit: Matthijs)[21:47:44] *** wildlander <wildlander!~wildlande@unaffiliated/wildlander> has quit IRC (Read error: Connection reset by peer)[21:56:59] *** Foaly <Foaly!~Foaly@ppp-88-217-93-222.dynamic.mnet-online.de> has quit IRC (Quit: Now 'mid shadows deep falls blessed sleep.)[21:59:50] *** slvn_ <slvn_!~slvn_@c2s31-1-78-245-90-111.fbx.proxad.net> has quit IRC (Quit: Leaving)[22:15:36] *** glYoda <glYoda!~MTLYoda@c-73-25-27-206.hsd1.or.comcast.net> has quit IRC (Quit: glYoda)[22:26:43] *** hellozee <hellozee!~hellozee@116.73.36.92> has quit IRC (Remote host closed the connection)[22:33:17] <Serus> hmmm, I have a triangle I want to render[22:33:27] <Serus> if I render it with glDrawArrays, it works[22:33:39] <Serus> but if I use an element buffer, it doesn't draw anything[22:34:09] *** bayoubengal <bayoubengal!~bayoubeng@rrcs-50-84-94-202.sw.biz.rr.com> has quit IRC (Ping timeout: 246 seconds)[22:34:16] <Serus> my element buffer containts 0, 1, 2, specifying the 3 vertices of the triangle[22:35:34] <Serus> I get no errors, just nothing on the screen[22:35:46] <Serus> do I have to specify 1 vertex twice?[22:36:35] <derhass> no[22:36:41] <derhass> show the code[22:37:14] <zalt__> does depth testing improve performance?[22:38:16] *** mandeep <mandeep!~mandeep@unaffiliated/mandeepb> has quit IRC (Ping timeout: 255 seconds)[22:38:17] <derhass> zalt__: by early rejecting fragments (and potentially even more than single fragments)[22:38:41] *** CoolerZ <CoolerZ!~coolerext@14.139.38.135> has joined ##OpenGL[22:38:48] <Serus> https://gist.github.com/DragoonX6/cb2d88a9159f1046291bc571f50eb5f4#file-render_system-cpp I added it to the old gist, derhass[22:39:12] <zalt__> hmm, i have a simple 2d renderer that is utilizing 100% of the gpu while being only at 20% of the cpu usage[22:39:41] <zalt__> i wonder what is the bottleneck, even though i'm rendering way more than anyone would realistically want[22:40:17] <chrisf> point your gpu profiler at it, apply palm to face[22:40:18] <zalt__> the fps reach 30 at 60,000 sprites, this is without depth testing, and with alpha blending[22:40:53] <Serus> derhass: RenderComponent just contains a pointer to the model and material[22:42:04] <zalt__> i have many possible culprits in mind, one of them is the gpu bandwidth which is unlikely, for my card it is 250gb/s[22:42:27] <derhass> Serus: your last parameter to glDrawElements is wrong[22:42:30] <zalt__> the other is alpha blending itself, is it an expensive operation?[22:42:51] <Serus> oh?[22:42:58] <derhass> Serus: it is the byte offset in the currently bound ELEMENT_ARRAY_BUFFER[22:43:06] <derhass> it is supposed to be, at least[22:43:07] <Serus> oh[22:43:08] <chrisf> zalt__: on what basis have you determined that you're saturating the GPU?[22:43:14] <Serus> lol I see[22:43:22] <zalt__> chrisf, msvs's gpu profiler which is somewhat broken[22:43:44] <Serus> yeah, there we go[22:43:45] <zalt__> it doesn't show many useful details, but it says gpu utilization was at 100%[22:43:47] <Serus> a triangle[22:44:38] <zalt__> is there a better profiler to use?[22:44:43] <chrisf> zalt__: what hardware?[22:44:50] <zalt__> i have a gtx 1070[22:44:57] <chrisf> go download nsight then[22:45:12] * zalt__ has nsight[22:45:17] <zalt__> i never used it though[22:47:20] *** mandeep <mandeep!~mandeep@unaffiliated/mandeepb> has joined ##OpenGL[22:47:20] <chrisf> https://devblogs.nvidia.com/the-peak-performance-analysis-method-for-optimizing-any-gpu-workload/[22:49:18] *** snyp <snyp!~Snyp@1.39.132.205> has quit IRC (Quit: WeeChat 2.4)[22:51:25] *** xx2 <xx2!~xx2@212.253.8.72> has joined ##OpenGL[22:58:05] *** KAHR-Alpha is now known as Guest64622[22:58:07] *** KAHR-Alpha_ <KAHR-Alpha_!~Alpha@2a01cb0d024a66005474f06642e55b50.ipv6.abo.wanadoo.fr> has joined ##OpenGL[22:58:07] *** krystcich <krystcich!~krystcich@088156132005.dynamic-ww-04.vectranet.pl> has quit IRC (Quit: Going offline, see ya! (www.adiirc.com))[22:58:21] *** snyp <snyp!~Snyp@1.39.132.205> has joined ##OpenGL[23:00:37] *** ShadowIce <ShadowIce!~pyoro@unaffiliated/shadowice-x841044> has joined ##OpenGL[23:01:11] *** Guest64622 <Guest64622!~Alpha@2a01cb0d024a660074ad48bc729385c6.ipv6.abo.wanadoo.fr> has quit IRC (Ping timeout: 258 seconds)[23:10:45] <zalt__> chrisf, thanks, i just installed a newer version of it, the one i had was broken[23:11:10] *** glYoda <glYoda!~MTLYoda@c-73-25-27-206.hsd1.or.comcast.net> has joined ##OpenGL[23:15:36] *** englosh <englosh!5bba3e9e@gateway/web/freenode/ip.91.186.62.158> has quit IRC (Quit: Page closed)[23:19:32] <zalt__> chrisf, on a frame with 40k sprites all actions took 17ms, i don't know how to interpret all of the statistics in there :O[23:26:13] <chrisf> let's see some of it[23:28:49] *** upgrdman <upgrdman!~upgrdman@blender/artist/upgrdman> has quit IRC (Quit: Leaving)[23:28:56] *** mandeep <mandeep!~mandeep@unaffiliated/mandeepb> has quit IRC (Quit: Leaving)[23:30:55] <chrisf> what i would initially suspect is (1) ineffective batching, or (2) massive overdraw[23:40:46] <zalt__> chrisf, i don't know what's relevant to screenshot but here's one: https://i.imgur.com/cSaFp5H.png[23:42:27] <zalt__> what do you mean by overdraw, because if it means what i think it means then yes, there are huge amounts of it[23:42:59] <Yaniel> overdraw means that you draw geometry in places where you already have rasterized something[23:43:15] <zalt__> i'm wondering does it matter whether there are lots of primitives overlapping on the same pixel space vs if they were scattered around the screen?[23:43:17] <Yaniel> if you have lots of semitransparent layers that would be tons of overdraw[23:43:19] <Yaniel> or particles[23:43:38] <zalt__> like for the same number of triangles, would it be faster to have them uniformly spread?[23:43:47] *** mandeep <mandeep!~mandeep@unaffiliated/mandeepb> has joined ##OpenGL[23:44:00] * zalt__ should learn how to cheat in benchmarks[23:44:28] <Yaniel> IIRC triangles are rasterized "in the order they were defined"[23:45:06] <Yaniel> which would mean yes, it is faster if they don't overlap[23:45:06] <zalt__> yep, and i imagine me having blending and no depth tasting would make this even worse[23:45:07] <derhass> or as if[23:55:15] <chrisf> well, i suppose the good news is you dont have ineffective batching :)[23:56:30] <zalt__> Yaniel, chrisf, thank you for teaching me how to cheat[23:56:49] <zalt__> changing the sprite's size to 15 pixels by 15 pixels, i can now spawn 100k entities at 60fps :D[23:57:02] <zalt__> so, it is overdrawing that is the reason[23:57:04] <derhass> that's not cheating[23:57:08] <zalt__> :P[23:57:38] <chrisf> tell us what's going on with your texturing[23:58:14] <derhass> a single number never is able to represent complex interrelations[23:58:28] <zalt__> chrisf, what do you mean?[23:58:40] <zalt__> now i get solid 28fps with 200k sprites[23:58:41] <chrisf> "you miss the texture cache so much you cry yourself to sleep"[23:58:56] *** sparr <sparr!~sparr@pdpc/supporter/active/sparr> has joined ##OpenGL[23:59:53] <sparr> having trouble compiling a random game, type conflicts between multiple declarations of PFNGLFRAGMENTLIGHTIVSGIXPROC in the game's glew.h and the system's include/GL/glext.h. Might a different version of GL be useful here? Or is there some other obvious way to reconcile this sort of problem?[23:59:58] <derhass> sprites should have a quite efficient texture acess pattern