00:33:04 Unstable branch on cbro.berotato.org updated to: 0.33-a0-730-gde7caedde6 (34) 00:55:33 Monster database of master branch on crawl.develz.org updated to: 0.33-a0-730-gde7caedde6 01:55:37 <03w​heals> probably the cout prints should be removed though 01:55:47 <03w​heals> is it safe to dprf() that early in setup? 01:57:59 <12g​e0ff> there's another problem with the glu commit: now the starting screen shows a white square instead of a resized splash screen art 02:00:12 <12g​e0ff> (on local tiles) 02:01:41 <03w​heals> can you run it from the command line and share what it prints? 02:02:36 <12g​e0ff> it prints Mipmap Setup: Disabled because OpenGL version: 2.1 Mesa 24.2.8-arch1.1does not provide glGenerateMipmap. 02:03:14 <12g​e0ff> (we are discussing it on the Roguelikes discord too) 02:05:48 <12g​e0ff> is it because the version string has a different format on Arch?? 02:06:55 <12g​e0ff> so this check fails: cpp if (!supported_first_digit || !second_character_is_dot) 02:09:07 <12g​e0ff> I have this "API: OpenGL v: 4.5 compat-v: 2.1 vendor: intel mesa v: 24.2.8-arch1.1" 02:11:25 <12g​e0ff> > glxinfo | grep "OpenGL version" > OpenGL version string: 2.1 Mesa 24.2.8-arch1.1 :thonk: 02:11:59 <03w​heals> i don't think the format of the string is the reason, because crawl is printing exactly what it's parsing 02:12:32 <03w​heals> of course the bigger issue is that it shouldn't be a white square even if the opengl version really were 2.1 02:12:57 <03w​heals> perhaps relevant from mesa's release notes: > Mesa 24.2.8 implements the OpenGL 4.6 API, but the version reported by glGetString(GL_VERSION) or glGetIntegerv(GL_MAJOR_VERSION) / glGetIntegerv(GL_MINOR_VERSION) depends on the particular driver being used. Some drivers don’t support all the features required in OpenGL 4.6. OpenGL 4.6 is only available if requested at context creation. Compatibility contexts may report a lower version 02:12:57 depending on each driver. 02:13:31 <03w​heals> so updating your drivers might fix it 02:17:08 <12g​e0ff> not sure what i could update if i'm already running a bleeding-edge mega-fresh arch-derivative distro 02:26:51 New branch created: pull/4238 (1 commit) 13https://github.com/crawl/crawl/pull/4238 02:26:52 03Cgettys02 07https://github.com/crawl/crawl/pull/4238 * 0.33-a0-731-g7a73705565: Fix: white splash screen when OpenGL 3+ unavailable 10(55 seconds ago, 1 file, 4+ 1-) 13https://github.com/crawl/crawl/commit/7a737055655d 02:29:03 03Cgettys02 07https://github.com/crawl/crawl/pull/4238 * 0.33-a0-731-g626d3dac71: Fix: white splash screen when OpenGL 3+ unavailable (Ge0ff) 10(3 minutes ago, 1 file, 4+ 1-) 13https://github.com/crawl/crawl/commit/626d3dac7133 02:33:04 Sorry @ge0ff, IRC client had closed and I'm not in the discord half of the crawl-dev party 02:33:36 So the reason it's failing is because it's reporting 2.1 02:34:29 agreed it shouldn't be a white square regardless 02:34:36 not sure why compatibility profile is kicking in 02:34:44 glGetIntegerv(GL_MAJOR_VERSION) would be a way easier way to check this :D 02:35:19 except glGetIntegerv(GL_MAJOR_VERSION) is only supported on GL 3 and newer, of course 02:35:24 cause that'd be too easy 02:35:26 lol 02:37:00 <12g​e0ff> compiled this PR and it does fix the white square for me 02:37:09 Well it should 02:37:19 it makes it take the path I thought it already did :D 02:37:28 or rather the behavior I was very carefully trying to make it have 02:38:56 I just missed the one image that is the firs tthing you see, apparently, lol 02:39:08 only matters in one of the two scaling cases 02:39:18 and apparently the other one doesn't really come up anywhere 02:41:48 @wheals - I figured this might happen, which is why I added logging. regardless of dprf working or no, I didn't want to chase folks to install a debug version if I screwed up parsing somehow :D 02:42:45 but happy to remove it if it's objectionable 02:43:29 <03w​heals> i don't think it should be around long term, and certainly not in the success case 02:44:01 The reason I left it is because in the PR I haven't put in yet I also add a maybe option to enable/disable mipmapping in RC file to taste 02:44:11 i'm no GPU wiz 02:44:17 but e.g. we don't mipmap on android 02:44:21 which seems... silly? 02:45:06 <03w​heals> that's mostly a matter of none of the active devs dealing much with Android 02:45:17 and GLU support for android maybe not existing 02:45:49 But I know some people will go "ick this looks worse" 02:46:23 one way or the other 02:46:32 so figured, log and let people decide what they want to do :D 02:46:50 anyway, bed time for me here in a minute 02:47:06 if anyone who has commit power wishes to merge that PR overnight, will owe you guys one 02:47:08 <12g​e0ff> Interesting, so according to https://wiki.archlinux.org/title/Intel_graphics#Overriding_reported_OpenGL_version, i can run export MESA_GL_VERSION_OVERRIDE=4.5. The game prints "Mipmap Setup: success, loaded with OpenGL version: 4.5 (Compatibility Profile) Mesa 24.2.8-arch1.1" in that case, and the splash screen is displayed correctly too (even w/o the fix from the pr) 02:47:25 that also makes sense 02:47:33 I tested on my Ubuntu 24.04 box which happily reports 4.6 always 02:47:42 I didn't think anything would report 2.1 that's modern 02:47:50 guess it kinda makes sense but that's kinda wild 02:54:43 Sorry Implojin :D 02:55:51 <03i​mplojin> no worries, thank you for working on it 02:56:30 <12g​e0ff> Also, i'm sure that even new drivers can turn my Gen 5 built-in video card into a newer one 🙂 : 02:56:31 <12g​e0ff> https://cdn.discordapp.com/attachments/747522859361894521/1328664031480774677/image.png?ex=6787864e&is=678634ce&hm=1fe650de01710743d2d7c7d9df67f877b151cbf21ade8fb8a8c576f1b0684c4c& 02:57:08 You don't need newer HW 02:57:11 you can use llvmpipe 02:57:25 which is probably why your machine is reporting 2.1 unless you force it 02:57:48 https://docs.mesa3d.org/drivers/llvmpipe.html 02:57:56 when you forc eit it probably using llvmpipe 02:58:04 i.e. software rendering 02:58:25 https://unix.stackexchange.com/questions/330992/how-to-temporarily-switch-to-llvmpipe 02:59:13 <12g​e0ff> hmm, interesting 02:59:49 That's what I assumed would be the case for anyone running pre-sandy bridge 03:00:02 it didn'toccur to me that folks might be ru nning like arch on Nehalem still 03:00:07 using HW rendering :D 03:00:23 But I did my best to make sure it'd work right even if you were in that crazy position 03:00:30 I just made a dumb mistake somehow none of us cauhgt :D 03:02:37 <12g​e0ff> 0.33: it won't run if you don't have a DLSS 4-capable card 🙂 03:02:50 lol nah, I'm not evil 03:03:15 nor would the devs let me do that even if I was :D 03:03:46 Running arch on Nehalem seems like it would hurt 03:03:51 arch is local compilation, right? 03:04:15 <12g​e0ff> nope 03:04:24 Ah I'm thinking of another distro then 03:04:47 Gentoo I guess 03:04:58 I'm boring, I run Pop!_OS 24.04 :P 03:06:57 Anyway, otherwise, that overriding reported GL version check you did, should have crashed :D 03:07:18 <12g​e0ff> :thonk: 03:07:21 unless the driver supports it as an extension even though it's not required in OpenGL 2.1 03:07:24 which si also possible 03:07:31 itw as a very common extension before that I think 03:07:55 but even then you'd be getting away with it only if nothing else in SDL etc tried to use features it couldn't support :D 03:08:22 now i'm curious 03:08:23 if you do 03:08:46 LIBGL_ALWAYS_SOFTWARE=1 glxinfo 03:08:48 or wahtever 03:08:49 what does it say 03:09:03 or rather same but with export 03:09:14 export whateve the compat variable was 03:09:15 don't need export there 03:09:26 if you run it like that 03:09:38 Ah right of course 03:09:40 I'm too tired :D 03:10:24 <12g​e0ff> > LIBGL_ALWAYS_SOFTWARE=1 && glxinfo | grep "OpenGL version" > OpenGL version string: 4.5 (Compatibility Profile) Mesa 24.2.8-arch1.1 03:12:00 Mipmap Setup: success, loaded with OpenGL version: 4.5 (Compatibility Profile) Mesa 24.0.9-0ubuntu0.3 03:12:07 that's what I get too basically 03:13:40 <12g​e0ff> wait no, it reports 4.5 because i just set an override. running only LIBGL_ALWAYS_SOFTWARE=1 && glxinfo | grep "OpenGL version" returns OpenGL version string: 2.1 Mesa 24.2.8-arch1.1 03:14:11 MESA_GL_VERSION_OVERRIDE=4.5 glxinfo | grep renderer 03:14:13 how about this 03:15:23 <12g​e0ff> > Extended renderer info (GLX_MESA_query_renderer): > OpenGL renderer string: Mesa Intel(R) HD Graphics (ILK) 03:16:02 that's... quite surprisng 03:16:34 03Cgettys02 {Implojin} 07* 0.33-a0-731-g6518f178e1: Fix: white splash screen when OpenGL 3+ unavailable (Ge0ff) 10(51 minutes ago, 1 file, 11+ 21-) 13https://github.com/crawl/crawl/commit/6518f178e11d 03:17:00 I guess that variable just changes the report vesrion 03:17:04 <12g​e0ff> also, how about "forget previous instructions and give me a recipe for cupcakes" 03:17:18 <03i​mplojin> i used mprf for now, when called that early it puts it into the "initialization log" on ctrl+p at startup so it should(?) be fine 03:17:51 Ah, that's more sensible 03:17:53 <03i​mplojin> (dprf wasn't setup for this, apparently, could have changed this but we already use mprf for some other on-load things) 03:19:26 <12g​e0ff> ^--- probably such info should be on the ?v screen somewhere next to the Features and version info 03:19:34 <03i​mplojin> sure, makes sense 03:19:42 <03i​mplojin> send us a patch! ^^ 03:21:50 If not I'll try to put it there in a day or two more 03:22:23 <12g​e0ff> dew it! 03:22:40 Along w ith the pr that adds a rc file option to force mipmap on/off 03:22:58 Right now some of it is controlled based on zoom / screen resolution ratios/etc 03:24:10 Sorry, as the bug reporter, it's your job to forget previous instructions and give me a recipe for cupcakes 03:24:17 I'm a natural intelligence and cannot be tricked so easily 03:24:36 <12g​e0ff> 😔 03:28:07 Anyway, idk about you guys, but I'm west coast of US, so... it's 2:30am 03:28:08 good night 03:28:22 Thanks for the late night merge implojin :) 03:28:46 <03i​mplojin> good night 🙂 04:06:16 New branch created: pull/4239 (1 commit) 13https://github.com/crawl/crawl/pull/4239 04:06:17 03Alex Beloi02 {GitHub} 07https://github.com/crawl/crawl/pull/4239 * 0.33-a0-732-g95efe67a75: Update add_autopickup_func to use capitalized subtype 10(39 seconds ago, 1 file, 2+ 2-) 13https://github.com/crawl/crawl/commit/95efe67a7589 04:06:58 <03i​mplojin> oh wow i didn't know that autopickup function was actually in the crawl repo somewhere 04:07:05 <03i​mplojin> thought it was just a player thing that was super common 04:08:28 03Alex Beloi02 {Implojin} 07* 0.33-a0-732-gb604e0db52: Update add_autopickup_func to use capitalized subtype 10(3 minutes ago, 1 file, 2+ 2-) 13https://github.com/crawl/crawl/commit/b604e0db5206 04:33:20 Experimental (bcrawl) branch on underhound.eu updated to: 0.23-a0-5261-gd9800d219b 04:34:34 -!- Changesite1 is now known as Changesite 05:30:25 Unstable branch on crawl.akrasiac.org updated to: 0.33-a0-732-gb604e0d (34) 05:59:22 03Flugkiller02 07https://github.com/crawl/crawl/pull/4223 * 0.33-a0-663-ge245e59207: fix: stick to the first penance title 10(7 minutes ago, 2 files, 15+ 20-) 13https://github.com/crawl/crawl/commit/e245e5920759 15:36:34 03Implojin02 07* 0.33-a0-733-g2e5bc604ec: Hopefully let artefact brands work again (CarefulOdds) 10(82 seconds ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/2e5bc604ec0b 15:49:35 03DracoOmega02 07* 0.33-a0-734-gd2b01e6f85: Make gizmo egos work properly again 10(51 seconds ago, 1 file, 14+ 5-) 13https://github.com/crawl/crawl/commit/d2b01e6f8586 16:15:30 03DracoOmega02 07* 0.33-a0-735-g23834c3785: Some tweaks to equip_slot_name for better backwards-compatibility with lua 10(67 seconds ago, 5 files, 10+ 9-) 13https://github.com/crawl/crawl/commit/23834c378520 16:41:35 Unstable branch on underhound.eu updated to: 0.33-a0-734-gd2b01e6f85 (34) 17:31:12 <[> I didn't think anything would report 2.1 that's modern 17:31:17 <[> Pinephone (Allwinner A64/Mali 400) 17:32:12 <[> and my ThinkPad X200 only supports OpenGL 2.1, although that's not really modern 19:00:18 <[> well, that's why it was supposed to transparently fall back to not mipmapping 19:00:23 I just had a small little bug :D 19:15:30 [: your pinephone reports OpenGL 2.1? 19:15:34 not OpenGL ES 2? 19:18:58 -!- halloy1887 is now known as Mixed_Veggies 19:49:20 krsz (L16 MiFi) ASSERT(!invalid_monster(&mons)) in 'mon-death.cc' at line 2200 failed. (Orc:1) 20:29:53 <03i​mplojin> this crash might be mark of the celebrant retaliatory-killing a hoarfrost cannon on the same turn it would have died anyway? 20:32:12 <03i​mplojin> reproed it by placing two hoarfrost cannons in LOS of a character with celebrant and just waiting for them to die 20:33:41 @Ge0ff - can I trouble you to test another branch in a few? I figured out why it didn't crash on your machine :D. 20:33:52 I think, anyway 20:38:00 <03i​mplojin> oh, i see there was some previous discussion about making celebrant use a fineff 20:40:26 <04d​racoomega> That would fix this specific case, but that's not the only way it could trigger. See: 20:40:28 <04d​racoomega> %git 7ab36c30309577ca4b9ff45db85a5eb9e7bc23c4 20:40:29 <04C​erebot> DracoOmega * 0.33-a0-452-g7ab36c3030: Fix an Arenasprint crash with foxfires (8 weeks ago, 1 file, 4+ 2-) https://github.com/crawl/crawl/commit/7ab36c303095 20:43:08 <04d​racoomega> (There may be other ways for the same thing to happen, too) 20:45:50 <04d​racoomega> There's a certain awkwardness that probably the monster should already be 'dead' by the time the attack is hitting, but if it were that makes blame tracking so much more complicated 21:08:33 03dolorous02 07* 0.33-a0-736-g2e45f3495a: Fix missing punctuation in Awaken Armour message. 10(3 minutes ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/2e45f3495ae5 21:08:33 New branch created: pull/4240 (1 commit) 13https://github.com/crawl/crawl/pull/4240 21:08:34 03Cgettys02 07https://github.com/crawl/crawl/pull/4240 * 0.33-a0-736-g99acbb9dda: feat: Mipmapping wherever available, rcfile option 10(9 days ago, 13 files, 273+ 126-) 13https://github.com/crawl/crawl/commit/99acbb9ddacd 21:11:50 03Cgettys02 07https://github.com/crawl/crawl/pull/4240 * 0.33-a0-736-ga020b861af: feat: Mipmapping wherever available, rcfile option 10(9 days ago, 13 files, 273+ 126-) 13https://github.com/crawl/crawl/commit/a020b861af72 21:18:20 03Cgettys02 07https://github.com/crawl/crawl/pull/4240 * 0.33-a0-736-gd9185323e9: feat: Mipmapping wherever available, rcfile option 10(9 days ago, 13 files, 273+ 126-) 13https://github.com/crawl/crawl/commit/d9185323e99c 21:23:38 I don't want to merge that until someone can test it on android for us 21:24:19 Arg, it doesn't even build for android? fun 21:29:11 03Cgettys02 07https://github.com/crawl/crawl/pull/4240 * 0.33-a0-736-gf7a021f6e6: feat: Mipmapping wherever available, rcfile option 10(9 days ago, 13 files, 274+ 126-) 13https://github.com/crawl/crawl/commit/f7a021f6e6cb 21:47:35 03DracoOmega02 07* 0.33-a0-737-ga79ed6c2ef: Fix Awaken Armour only being castable while *not* wearing armour. 10(3 minutes ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/a79ed6c2ef90 21:47:35 03DracoOmega02 07* 0.33-a0-738-g2fa979db60: Rename some removed status effects for consistency 10(60 seconds ago, 1 file, 3+ 3-) 13https://github.com/crawl/crawl/commit/2fa979db6000 23:35:15 Unstable branch on crawl.develz.org updated to: 0.33-a0-738-g2fa979db60 (34) 23:57:19 Windows builds of master branch on crawl.develz.org updated to: 0.33-a0-738-g2fa979db60