00:00:11 Unstable branch on CRAWL.XTAHUA.COM updated to: 0.21-a0-237-g87df72b (34) 01:12:34 please god make it so that ring evokable flight can renew with recasts 01:12:45 you can do it with the spell 01:13:12 you can literally die from low duration flight 01:19:04 you can't anymore 01:19:04 also what spell 01:34:58 never mind 01:35:16 i was in .17 01:58:14 any devs awake? 02:32:03 cojito, I am not a dev, but I've tinkered. probably a 5-10% chance I can incrementally help your situation 02:32:17 unless it's dgamelaunch related, then that shoots up to maybe 50% 02:37:08 03johnstein02 07https://github.com/crawl/crawl/pull/594 * 0.21-a0-196-ga75aa36: Refactor more checks into a function 10(2 days ago, 1 file, 49+ 90-) 13https://github.com/crawl/crawl/commit/a75aa361a0ce 02:37:08 03johnstein02 07https://github.com/crawl/crawl/pull/594 * 0.21-a0-197-gcf1160a: Clean up comments 10(3 minutes ago, 1 file, 4+ 10-) 13https://github.com/crawl/crawl/commit/cf1160a06c22 02:52:49 Monster database of master branch on crawl.develz.org updated to: 0.21-a0-237-g87df72b 03:12:10 Unstable branch on crawl.beRotato.org updated to: 0.21-a0-237-g87df72b (34) 03:44:55 03johnstein02 07https://github.com/crawl/crawl/pull/604 * 0.21-a0-223-g6d06de8: Adjust some denial messages 10(66 seconds ago, 1 file, 7+ 4-) 13https://github.com/crawl/crawl/commit/6d06de8667f4 04:12:59 Hello there, is there any procedure to apply for helping to develop the game? Or should I just pick any bug on mantis and start working on it? It would be for C++ development (I'm developing games as day job, but mostly in Java, i've been wanting to get back into C++ for a while) 04:13:57 Also, are there any resources to read to dig more easily into the code? Like documents about the major classes, interactions and so on? 04:19:43 nikopol (L27 MiBe) ERROR in 'mon-util.cc' at line 1277: bogus mc (no monster data): random monster (2000) (Crypt:3) 06:34:20 Yermak (L18 HuGl) ERROR in 'mon-util.cc' at line 1277: bogus mc (no monster data): random monster (2000) (Swamp:4) 06:34:33 great 06:48:53 Morg0th: there's very little in the way of documentation unfortunately, but if you want to help then the best thing to do is definitely just to pick something that you want to fix and go for it 06:49:24 if you ask in here people will probably be able to give pointers on where to look 06:50:08 there's a mantis label of "simple bugs" (i think that's what it's called) that might be a good starting place, although they don't necessarily always turn out to be simple 06:53:53 Yermak (L18 HuGl) ERROR in 'mon-util.cc' at line 1277: bogus mc (no monster data): random monster (2000) (Swamp:1) 06:53:58 Two crashes in one game due to ghost crabs. 07:13:20 MarvinPA: ok thanks, so does that mean that there isn't an "official list" of crawl devs? as anyone can just make changes/fixes on his side and push them? 07:13:38 MarvinPA: also, how do you ensure the code quality and code standard? 07:14:01 conventions i mean 07:16:13 johnnyzero (L22 HuCK) ERROR in 'mon-util.cc' at line 1277: bogus mc (no monster data): random monster (2000) (Crypt:1) 07:16:43 there's an official list of devs with commit rights (people with +v in this channel, and also listed in CREDITS.txt and probably elsewhere) 07:17:06 but contributions from others are frequent and very welcome, usually in the form of pull requests on github or patches on mantis 07:17:59 then one of the active devs will review and merge those 07:19:35 there's a bunch of docs in docs/develop that have some general info as well as specific guides to getting stuff set up 07:20:52 contribution-process.txt and coding_conventions.txt are probably the main ones to glance at, the other stuff is mostly more specific 07:39:23 bogus mc crash with spectral mist 13https://crawl.develz.org/mantis/view.php?id=11210 by johnnyzero 07:41:48 MarvinPA: ok thanks! i'll have a look at those files 07:52:00 motkurips (L17 DsNe) ERROR in 'mon-util.cc' at line 1277: bogus mc (no monster data): random monster (2000) (Swamp:1) 07:54:19 motkurips (L17 DsNe) ERROR in 'mon-util.cc' at line 1277: bogus mc (no monster data): random monster (2000) (Swamp:1) 07:57:07 motkurips (L17 DsNe) ERROR in 'mon-util.cc' at line 1277: bogus mc (no monster data): random monster (2000) (Swamp:1) 07:58:17 removed (L17 DgFE) ERROR in 'mon-util.cc' at line 1277: bogus mc (no monster data): random monster (2000) (Swamp:3) 07:59:48 removed (L17 DgFE) ERROR in 'mon-util.cc' at line 1277: bogus mc (no monster data): random monster (2000) (Swamp:3) 08:02:18 motkurips (L17 DsNe) ERROR in 'mon-util.cc' at line 1277: bogus mc (no monster data): random monster (2000) (Swamp:3) 08:04:27 motkurips (L17 DsNe) ERROR in 'mon-util.cc' at line 1277: bogus mc (no monster data): random monster (2000) (Swamp:4) 08:05:08 Sempai (L21 TrMo) ERROR in 'mon-util.cc' at line 1277: bogus mc (no monster data): random monster (2000) (Vaults:3) 08:05:25 motkurips (L17 DsNe) ERROR in 'mon-util.cc' at line 1277: bogus mc (no monster data): random monster (2000) (Swamp:4) 08:06:40 Sempai (L21 TrMo) ERROR in 'mon-util.cc' at line 1277: bogus mc (no monster data): random monster (2000) (Vaults:3) 08:08:17 motkurips (L17 DsNe) ERROR in 'mon-util.cc' at line 1277: bogus mc (no monster data): random monster (2000) (Swamp:4) 08:22:32 !crashlog 08:22:44 17706. motkurips, XL17 DsNe, T:30028 (milestone): http://crawl.xtahua.com/crawl/morgue/motkurips/crash-motkurips-20170902-120758.txt 09:00:35 !blame amalloy 09:00:35 I pronounce amalloy... Guilty! 09:00:52 mons_genus has some asserts that don't work before random monsters are resolved 09:02:02 I think I can just move this block later in the function 09:02:08 blobert (L26 GrFi) ERROR in 'mon-util.cc' at line 1277: bogus mc (no monster data): random monster (2000) (Crypt:2) 09:05:11 hm, though I would have thought monster type should be resolved then 09:09:04 blobert (L27 GrFi) ERROR in 'mon-util.cc' at line 1277: bogus mc (no monster data): random monster (2000) (Crypt:3) 09:09:33 ah, maybe because it's a spectral thing 09:12:10 blobert (L26 GrFi) ERROR in 'mon-util.cc' at line 1277: bogus mc (no monster data): random monster (2000) (Crypt:3) 09:14:37 oh duh it's because base_type is left intact on resolving the monster type 09:17:13 03advil02 07* 0.21-a0-238-geb0acef: Don't crash on random monsters 10(2 minutes ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/eb0acef0604d 09:17:21 ??rebuild 09:17:21 rebuild[1/2]: http://crawl.akrasiac.org/rebuild/ http://underhound.eu:81/rebuild/ http://crawl.berotato.org/crawl/rebuild/ http://crawl.xtahua.com/rebuild/ https://crawl.jorgrun.rocks/rebuild/ Bug |amethyst or Nap.Kin for CDO. Use your powers wisely. 09:18:49 amalloy_ : you may want to check that that still does what you intended, I didn't try to figure out if there was a specific reason for using base_type 09:28:50 Shiggs (L27 NaFi) ERROR in 'mon-util.cc' at line 1277: bogus mc (no monster data): random monster (2000) (Pan) 09:30:44 Unstable branch on crawl.beRotato.org updated to: 0.21-a0-238-geb0acef (34) 09:30:50 I've triggered rebuilds on CAO and CBRO, can't do any of the others myself 09:32:05 Unstable branch on crawl.akrasiac.org updated to: 0.21-a0-238-geb0acef (34) 09:32:09 Kite (L18 MiBe) ERROR in 'mon-util.cc' at line 1277: bogus mc (no monster data): random monster (2000) (Swamp:4) 09:33:11 I'm rebuilding CXC 09:33:12 Kite (L18 MiBe) ERROR in 'mon-util.cc' at line 1277: bogus mc (no monster data): random monster (2000) (Swamp:4) 09:33:51 thanks 09:33:57 CUE let me log in but isn't actually responding to the rebuild button 09:34:28 Kite (L18 MiBe) ERROR in 'mon-util.cc' at line 1277: bogus mc (no monster data): random monster (2000) (Swamp:4) 09:34:44 also that CJR rebuild link doesn't seem very useful? 09:34:54 yeah I think it's been broken for a while 09:35:05 Unstable branch on underhound.eu updated to: 0.21-a0-238-geb0acef (34) 09:35:34 okay CUE rebuild did work ^ 09:35:35 Kite (L18 MiBe) ERROR in 'mon-util.cc' at line 1277: bogus mc (no monster data): random monster (2000) (Swamp:4) 09:36:26 Kite (L17 MiBe) ERROR in 'mon-util.cc' at line 1277: bogus mc (no monster data): random monster (2000) (Swamp:4) 09:37:26 Unstable branch on CRAWL.XTAHUA.COM updated to: 0.21-a0-238-geb0acef (34) 09:37:47 now we see if I actually fixed it :D 09:37:53 though I tested it and it looked right 09:40:02 cjr does rebuild every hour, but I don't think the rebuild page works 09:40:17 Kite (L18 MiBe) ERROR in 'mon-util.cc' at line 1277: bogus mc (no monster data): random monster (2000) (Swamp:4) 09:40:25 !lm kite crash x=vlong 09:40:25 6. [2017-09-02 13:39:56] [vlong=0.21-a0-237-g87df72b] Kite the Executioner (L18 MiBe of Trog) ERROR in 'mon-util.cc' at line 1277: bogus mc (no monster data): random monster (2000) (Swamp:4) 09:46:53 03advil02 07* 0.21-a0-239-g7b8ad69: Add a test that exercises ghost crabs and therefore random summoning 10(73 seconds ago, 1 file, 9+ 5-) 13https://github.com/crawl/crawl/commit/7b8ad69b8d83 10:00:23 Gotta keep those ghost crabs in good shape 10:09:09 Unstable branch on crawl.jorgrun.rocks updated to: 0.21-a0-239-g7b8ad69 (34) 11:54:31 can you just use somthing like !var for string var as a test for whether it's empty string? 11:54:37 maybe there's a .empty() 11:55:08 looks like that's what I want 12:40:34 -!- amalloy_ is now known as amalloy 12:41:07 lol rip. thanks advil 12:42:23 base_type is checked because it's supposed to apply for derived monsters like skeletons too 13:04:37 Unstable branch on crawl.akrasiac.org updated to: 0.21-a0-239-g7b8ad69 (34) 13:05:47 is there anything that makes replaying a web-tiles session impossible from a ttyrec 13:06:10 i know that functionality doesn't exist, but is there any fundamental issue with trying to replay a web-tiles replay using a ttyrec 13:07:10 if anyone knows or has thought about this, please let me know, i will be trying it in my free time 13:10:53 cojito: one barrier is that the ttyrec doesn't have the tile info. I don't know if it has version info either. so you have to find a way to bridge that and deal with any glyphs that could represent multiple tiles 13:11:21 okay, so it would have to do some kind of mapping from the text output to the webtiles 13:11:47 i misunderstood what a ttyrec was, it's just the text so things like player equipment wouldn't be possible to show 13:16:34 a ttyrec is the output part of a console session played into a file instead of over ssh. 13:16:43 so it's abunchof console screen updates 13:17:07 -!- cojito_ is now known as cojito 13:18:13 so if the player pulled up their inventory screen you'd see their equipment 13:18:50 in console format 13:26:24 cojito: there is no unique mapping from console glyphs to tiles either, so you'd need to semantically understand the game context. it'd be real hard 13:26:36 okay 13:26:56 seems like a waste of time 13:27:28 if I could make a compact web-tiles compatible replay format in a fork, would that ever be considered for merging? 13:27:49 or is this just something that nobody really cares about, which I would understand 13:28:51 there's even more fun because it does "optimal" updates --- basically onlys ends the chages, so ou'd need to reconstruct each actual console "frame" if you wanted to pull information from it. ttyrec players don't care abot this 13:29:22 (they do have to start from the nerest redraw-full-screen which is why time offsets for replays are always approximate) 13:29:48 (also why console occasionally forces full screen redraws) 15:03:31 -!- amalloy is now known as amalloy_ 15:46:07 03gammafunk02 07* 0.21-a0-240-gc4b6860: Add objstat fields for tracking whether monster are from vaults 10(37 minutes ago, 1 file, 21+ 3-) 13https://github.com/crawl/crawl/commit/c4b6860f3807 15:49:35 03gammafunk02 07[nospawn] * 0.21-a0-247-g791f81e: Merge branch 'master' into nospawn 10(39 seconds ago, 0 files, 0+ 0-) 13https://github.com/crawl/crawl/commit/791f81e82b4a 15:54:52 03gammafunk02 07https://github.com/crawl/crawl/pull/606 * 0.21-a0-247-g791f81e: Merge branch 'master' into nospawn 10(6 minutes ago, 0 files, 0+ 0-) 13https://github.com/crawl/crawl/commit/791f81e82b4a 16:09:15 Unstable branch on crawl.jorgrun.rocks updated to: 0.21-a0-240-gc4b6860 (34) 16:24:04 I’ve got the “Another game is already in progress using this save!” error. How can I reset it? 16:25:56 jameydeorio: the websocket to the process is probably disconnected, and eventually the server will see the lack of input and kill the process (after saving your game) 16:26:09 it can take 8 minutes or so 16:27:44 Hmm, my computer ran out of juice last night gammafunk, and I just opened it back up. Shouldn’t it be reset by now"? 16:28:34 <|amethyst> which server? 16:28:43 cao 16:28:51 "ghostmoth" 16:29:35 <|amethyst> I see a connection that started about 8 minutes ago 16:29:43 <|amethyst> or, rather, a game 16:29:51 Huh, maybe it was somehow connected all night...? 16:30:01 <|amethyst> or reconnected when you powered back on? 16:30:18 Yeah it’s possible 16:30:34 zircondrogin (L24 DsNe) ASSERT(!invalid_monster(&mons)) in 'mon-death.cc' at line 1910 failed. (Depths:4) 16:31:00 <|amethyst> jameydeorio: I killed it, so you should be able to reconnect again 16:31:05 Thank you! 16:31:23 there we are 16:33:23 guess that zircondrogin game just didn't transfer? 16:33:25 !crash 16:33:36 17720. ghostmoth, XL2 CeHu, T:311 (milestone): http://crawl.akrasiac.org/rawdata/ghostmoth/crash-ghostmoth-20170902-203042.txt 16:34:15 <|amethyst> !crash -2 16:34:16 No milestones for -2 (crash). 16:34:20 <|amethyst> !crash * -2 16:34:21 18043/18044. zircondrogin, XL24 DsNe, T:66105 (milestone): http://crawl.akrasiac.org/rawdata/zircondrogin/crash-zircondrogin-20170902-203034.txt 16:39:17 <|amethyst> gammafunk: it looks like the version from just before your objstat commit 16:39:34 oh, I assumed it was related to those earlier crashes 16:39:42 -!- jameydeorio is now known as ghostmoth 16:40:22 <|amethyst> pain bond again 16:40:24 looks like a grand finale that killed some nearby monsters 16:42:00 <|amethyst> a nearby monster took damage from a chunk of the exploding monster, which I think killed the original monster through pain bond before uskayaw_grand_finale specifically killed that monster 16:42:36 <|amethyst> the crash is from u_g_f trying to kill the target monster when it's already dead 16:43:21 <|amethyst> huh 16:44:06 <|amethyst> not related to this crash, but I wonder what situations would result in this "You spring back to your original position." message 16:44:36 <|amethyst> (it happens when the target monster is still alive after the call to monster_die) 16:47:46 <|amethyst> I guess the quick fix here would be to check whether the monster is already dead after throwing monster bits, and if so don't call monster_die 16:48:03 <|amethyst> but that doesn't seem right 16:50:09 <|amethyst> probably pain bond should be a final effect 17:15:18 The build was fixed. (nospawn - 791f81e #8699 : gammafunk): https://travis-ci.org/crawl/crawl/builds/271239083 17:45:19 it was never broken, travis! 17:46:42 wow 17:47:12 anyone want to hazard a guess of what percentage of monsters come from vaults vs. from level for 3-rune games? 17:47:37 er, crap 17:49:01 doh 17:50:15 well, I ended up doubling the normal monster count in objstat 17:50:22 but thankfully that doesn't ruin the data I just generated 17:58:23 03gammafunk02 07* 0.21-a0-241-g2cbab09: Fix the overall monster counts in objstat being doubled 10(89 seconds ago, 1 file, 0+ 1-) 13https://github.com/crawl/crawl/commit/2cbab0914ea5 18:02:27 !calc 1230.02 / 2611.2 18:02:28 0.47 18:02:33 percent of monsters that come from vaults 18:02:36 Unstable branch on underhound.eu updated to: 0.21-a0-241-g2cbab09 (34) 18:02:48 !calc 1052869.52 / 1842186.36 18:02:49 0.57 18:02:52 percent of XP from vaults 18:03:04 of course so much of that will come from certain rune/endgame levels 18:04:08 @?frilled lizard 18:04:08 frilled lizard (03l) | Spd: 10 | HD: 1 | HP: 1-3 | AC/EV: 0/15 | Dam: 3 | amphibious, cold-blooded | Res: 12drown | XP: 1 | Sz: tiny | Int: animal. 18:04:35 hrm, mons xp values look weird 18:05:01 will need to look into that field 18:05:19 looks like it's being halved somehow 18:09:17 Unstable branch on crawl.jorgrun.rocks updated to: 0.21-a0-241-g2cbab09 (34) 18:30:26 !calc 836675.58/790406.69 18:30:27 1.06 18:31:42 gammafunk: still distressing that nospawn only increased TotalNonVaultXP by 6% instead of the expected 18% or so 18:32:31 are monsters placed in any other way other than by vaults and by the 3d10 / 3d12 thing? 18:37:11 elliptic: I've only just gotten the data uploaded 18:37:15 so I'm taking a look now 18:37:26 https://drive.google.com/drive/folders/0B7VXhHzhWWb7eTcyOUNTQjVHTmM 18:37:36 is trunk with the objstat change 18:37:51 and then nospawn is updated: https://drive.google.com/drive/folders/0B7VXhHzhWWb7WXZoZV9GeFFwVzg 18:38:06 yeah I found those 18:40:12 for nonvault 324 trunk vs 341 nospawn 18:40:24 for all of Dungeon branch 18:40:35 so only 17 more 18:40:55 I'm more interested in the proportion more 18:41:45 we shouldn't be surprised that it is less than 3 extra monsters per level because of MONS_NO_MONS and failing to place monsters sometimes 18:42:46 yeah, but I also don't know what this chance to fail, and I don't think those two together would be giving us only 17 more vs 45 more 18:42:49 but the proportion of extra monsters should be higher for sure unless either (a) monsters become more likely to fail to place after a bunch of them have already been placed or (b) there is some way in which monsters are placed other than the 3d10 thing 18:44:27 I do see that the monster numbers increase with depth 18:44:35 level monster placement doesn't vary by depth though 18:44:58 hrm, so do number of non-vault monsters 18:44:59 bands could account for that if bands are more common on deeper levels 18:45:01 it's not just more vaults 18:45:06 oh, yeah that could be 18:45:25 seems to be a surprisingly strong effect 18:46:03 how many generations are averaged in each of these? 18:46:05 100? 18:46:42 yes, 100 in both 18:47:09 sort of an interesting drop in number of nonvault monsters on D:3 vs D:4 18:48:17 numsd is giving nan a lot 18:48:20 I don't know why 18:48:47 oh, MONS_NO_MONS for D agrees with this variance I guess 18:48:57 { 4, 7, 1000, FLAT, MONS_NO_MONSTER}, 18:48:57 { 4, 10, 4000, FLAT, MONS_NO_MONSTER}, 18:49:02 heh 18:49:14 { 6, 16, 335, SEMI, MONS_NO_MONSTER }, 18:51:22 still a bit weird that D:1-3 has fewer monsters than late D but that is at least somewhat consistent 18:53:42 gammafunk: maybe it is worth tweaking the numbers and running objstat again to get a third data point 18:54:06 increasing the level monster dice you mean? 18:54:09 yeah 18:54:38 like you increased the number of sides by 2 before, maybe increase them by 5 instead (so 3 more) 18:54:53 I can try that 18:55:49 I assume for consistency we'd want to also increase crypt/vestibul by 3 18:56:03 yeah 18:59:33 03gammafunk02 07[nospawn] * 0.21-a0-248-ge18f60a: Increase number of level monsters again 10(3 minutes ago, 1 file, 3+ 3-) 13https://github.com/crawl/crawl/commit/e18f60a32cea 18:59:38 03gammafunk02 07https://github.com/crawl/crawl/pull/606 * 0.21-a0-248-ge18f60a: Increase number of level monsters again 10(3 minutes ago, 1 file, 3+ 3-) 13https://github.com/crawl/crawl/commit/e18f60a32cea 19:03:01 I'll start that objstat job up and we can see in another 1-2 hours; I'll be back later around that time 20:15:08 03gammafunk02 07[nospawn] * 0.21-a0-250-g8a56263: Merge remote-tracking branch 'origin/master' into nospawn 10(33 seconds ago, 0 files, 0+ 0-) 13https://github.com/crawl/crawl/commit/8a56263607a1 20:15:13 03gammafunk02 07https://github.com/crawl/crawl/pull/606 * 0.21-a0-250-g8a56263: Merge remote-tracking branch 'origin/master' into nospawn 10(38 seconds ago, 0 files, 0+ 0-) 13https://github.com/crawl/crawl/commit/8a56263607a1 20:27:03 amalloy_ re derived undead hydras: ah ok. I didn't test whether my change will still do the check in that case, so it may not. 20:29:33 elliptic: those workbooks are up at https://drive.google.com/drive/folders/0B7VXhHzhWWb7cDBBY1laWVFNOWs 20:44:51 !calc 1007409.97/790406.69 20:44:52 1.27 20:45:01 !calc 16.0/11 20:45:02 1.45 20:45:25 hm 20:46:15 I guess maybe the 6% increase was partially just low due to variance 20:49:12 !calc 418.2/324.9 20:49:13 1.29 20:56:06 !calc 77601.5 / 66931.66 20:56:08 1.16 20:56:29 it's a bit more than 10% xp overall compared to trunk in current nospawn 20:56:44 but it's more than that in Dungeon, although again I guess that could be variance 20:56:57 -!- Rast-- is now known as rast 20:58:08 !calc 79326.81 / 67010.04 20:58:09 1.18 20:58:43 !calc 2047533.66 / 1842186.36 20:58:44 1.11 20:59:09 similar relationship in the swamp-snake workbook, so probably not just variance; I guess that's just because of Lair having so many pack monsters 20:59:52 er, no 20:59:59 that's not Lair, that's Dungeon 21:00:23 I'm pretty puzzled by this data 21:02:42 like, the effect of moving from 3d10 to 3d15 was much more than 2.5 times as large as the effect of moving from 3d10 to 3d12 in almost every location 21:03:28 ...oh crap 21:03:36 I just realized something 21:03:52 did nospawn change something else about generation? 21:04:18 since that is what this looks like to me, if we want to check we could move it back to 3d10 and run objstat there 21:04:32 no, it was about that objstat bug I had made 21:04:46 where I was doubling the "Num" field 21:05:06 but that wouldn't affect the fields we were looking at 21:05:20 I had to manually correct the num filed in these workbooks since it was exactly doubled 21:05:28 this is why MonsXP is half what it should be 21:05:37 since it divides by Num 21:05:41 yeah I've mainly just been looking at TotalNonVaultXP 21:05:43 but this wouldn't affect what we're looking at 21:06:12 elliptic: ok, so you think an objstat run without the level spawn increase 21:06:21 and we see if somehow generation levels are still changing 21:06:42 back to 3d10 for all levels save for Crypt/Extra vesitbul, which go to 3d8 21:06:52 as far as what could be increasing it further, I've no idea 21:06:52 that might be a good sanity check to make sure that the other things you changed in the branch aren't confounding things, yeah 21:07:06 or doing this non-linear increase thing, at least 21:07:20 let me quickly look over the commits in the branch 21:07:35 which are here: https://github.com/crawl/crawl/pull/606 21:10:07 is there any chance that the 1/8 of monsters that are placing awake aren't actually placing at all? 21:10:19 since that would be sort of close to explaining it 21:10:28 oh, that would be funny, yeah 21:10:33 or if they are placing only a small proportion of the time 21:10:34 hrm 21:10:43 and easy to test in wiz mode, to some extent 21:11:45 This was something very weird about default placement that I removed: 21:11:50 https://github.com/crawl/crawl/pull/606/commits/d2cf27e79c9d1d50879fb1c268bc9848dc8449d2#diff-337e59ae3f4d2d932ad3cef1ea8d0d4dL973 21:12:32 it would reattempt to place the monster sometimes based on the result of a random roll if the monster was close to the player 21:12:41 a linear interpolation on the total nonvault xp suggests that it should be down by roughly 8-9% if the 3d10 numbers are set back to 3d10 21:12:51 compared with the trunk numbers 21:13:31 if the 3d15 numbers are set back to 3d10, you mean? 21:13:33 yeah 21:19:26 I wonder if PROX_AWAY_FROM_STAIRS is just causing it to reject too many locations 21:20:04 I see that _place_monster_aux tries 1000 times, but only with radius 3 of a single location 21:20:13 which could easily all be forbidden by the stair condition 21:21:12 I guess that is for bands though? 21:22:57 probably for bands, yeah 21:23:01 and yeah I just made three D:2 levels 21:23:06 couldn't find a single wandering monster 21:23:12 in any of those 21:23:24 so not really sure what's happening there 21:23:38 I don't see anything clearly wrong in the code but I assume that this is what is happening, yeah 21:23:47 I just highly doubt it's actually rejecting them due to a legitimate placement issue 21:23:54 45 tries is a lot of tries 21:24:10 maybe a lot of those roll in walls though 21:24:15 I mean, a lot of the tries will be in walls yes 21:24:29 still surprising to me 21:24:37 I use grid distance and not los for this stairs check 21:24:42 I wouldn't expect also ruling out places near stairs would make that much of a difference though 21:24:44 even given that 21:24:50 yeah 21:25:04 I will have to fool with this more and figure out what, if anything is going on 21:25:17 will verify that it's actually placing these generate awake monsters before running more tests 21:25:21 with objstat etc 21:34:03 oh 21:34:06 I findally found some 21:34:19 oh, crap 21:34:30 the listing excludes status when there are multiple, nevermind 21:35:04 there must be a better way to find these 21:35:50 I mean, you could try changing the 1/8 chance to 1/1 and see if there are any monsters at all 21:36:00 indeed 21:36:38 I suspect that they do exist but are just failing to generate much more than other monsters 21:36:54 though I don't fully understand why this would be the case 21:39:30 _Detected 0 monsters. 21:39:31 sigh 21:39:52 so yeah they're not placing at all 21:47:01 I'm confused, why is LOS_RADIUS defined to be 8? I assume this has something to do with frogs 21:47:24 but that seems pretty bad to just set it for 8 and use 8 everywhere when 95% of games can use 7 instead 21:48:17 elliptic: there is a player-specific los, of course 21:48:33 right I know that effective LOS is still 7 normally 21:48:33 but generating monsters based on the player's los size could be bad, right? 21:48:42 if they cast darkness while going down or something 21:49:15 I guess it could use racial los, not sure if that's conveniently available 21:49:33 but obviously not hard to figure out since it's just a species check 21:49:35 sure but I still maintain that changing 7 to 8 every single place that uses LOS_RADIUS in code for the sake of a bad gimmick for one species was a bad idea 21:49:57 I have no idea what else might have been affected by this in minor ways 21:50:53 well, that aside, I seemed to have screwed up my iterator somehow 21:51:31 wait 21:51:37 for (distance_iterator di(mg_pos, false, LOS_RADIUS); di; ++di) 21:52:21 gah 21:52:29 foiled by poor understanding of default arguments 21:52:36 looks like I'm using radius of GDM :( 21:52:51 rip 21:53:03 which is even worse than something like 80, since it's 105 21:53:20 stairs would have to place outside the largest possible map size 21:53:25 IMO when that is fixed try rerunning objstat with the 3d13 21:53:39 3d13? 21:53:44 er, 3d12 21:53:49 we've done 3d12 and 3d15 yeah, ok 21:58:02 amalloy_: you didn't catch this bug in your review! 21:58:57 seriously though, I'm worried about this LOS_RADIUS abuse 21:59:46 if I had known that barachi were coded in such a way as to mess with minor details all over crawl in non-barachi games then I would have said something much earlier 22:00:14 various things are placed within LOS_RADIUS of player and such 22:00:33 e.g. lugonu altar when you enter abyss (sometimes) 22:02:18 the meaning of LOS_RADIUS before included "the maximum possible LOS that a species would ever have", right? 22:02:48 so I guess in that sense PF had to update it if he was going to add this mutation 22:04:05 I guess he did add LOS_DEFAULT_RANGE and changed some stuff to it 22:04:37 yeah, it wasn't crudely done from what I could, but I certainly didn't look over the code in detail either 22:04:37 *I could tell 22:05:07 (but not enough) 22:07:02 I sort of expected that it wouldn't change in a way to impact performance for other species 22:12:32 on a cursory look through I didn't find that much that LOS_RADIUS is breaking, the main things are the abyss altar placing out of sight sometimes (unless I am missing something) and barachi mysteriously getting range 8 on spells that are supposed to be range 7 22:16:11 I wonder if he intentionally did that 22:16:16 the spell range part 22:16:21 but really this seems like a lot of work for a pretty mediocre gimmick 22:16:27 possibly, though it feels weird to me 22:17:43 yeah, I initially suggested the gimmick off-hand mostly because it would amplify the slow movement and also interact with their hop ability, but PF liked the idea and implemented it 22:18:13 (since there's a gap between range 6 and range 8 then, and it's weird that extra LOS is mostly a drawback but is actually an advantage for a small subset of spells) 22:18:36 right, that I think is less defensible 22:18:52 and it's not like there aren't still things that do end up being range 8 for Ba 22:18:56 like ranged attacks 22:19:59 making them get that for spells for free is not necessary; I suppose he might see it as part of their racial thing and they are "elf-like" aka Good At Magic 22:20:13 03Floodkiller02 07https://github.com/crawl/crawl/pull/600 * 0.21-a0-214-g7b4e9e5: Fix typos 10(25 hours ago, 2 files, 7+ 6-) 13https://github.com/crawl/crawl/commit/7b4e9e5f07e3 22:20:13 03Floodkiller02 07https://github.com/crawl/crawl/pull/600 * 0.21-a0-215-gf2a98f3: Turn stat choice prompt into a function 10(46 minutes ago, 2 files, 117+ 138-) 13https://github.com/crawl/crawl/commit/f2a98f351c74 22:20:13 03Floodkiller02 07https://github.com/crawl/crawl/pull/600 * 0.21-a0-216-g4c5ead4: Unbrace. 10(40 minutes ago, 1 file, 0+ 3-) 13https://github.com/crawl/crawl/commit/4c5ead464490 22:20:13 03Floodkiller02 07https://github.com/crawl/crawl/pull/600 * 0.21-a0-217-gaab98c2: Remove 'zero drain' restriction from Shift Attributes 10(10 minutes ago, 3 files, 2+ 10-) 13https://github.com/crawl/crawl/commit/aab98c25d312 22:20:13 03Floodkiller02 07https://github.com/crawl/crawl/pull/600 * 0.21-a0-233-gc66ac31: Merge branch 'gnoll_buffs' of https://github.com/Floodkiller/crawl into gnoll_buffs 10(3 minutes ago, 0 files, 0+ 0-) 13https://github.com/crawl/crawl/commit/c66ac3117efd 22:20:17 !tell elliptic look at LOS_RADIUS = 8 sometimes 22:20:18 elliptic: OK, I'll let elliptic know. 22:22:34 <|amethyst> elliptic: the comment on LOS_RADIUS suggests that it should not be used 22:22:54 <|amethyst> and everything should say explicitly LOS_MAX_RANGE or LOS_DEFAULT_RANGE instead 22:23:38 it's used a ton in the codebase, that's for sure 22:24:53 <|amethyst> yeah 22:25:38 <|amethyst> basically every remaining use of LOS_RADIUS is something where no one has stopped to consider which it should be 22:25:48 <|amethyst> so should be viewed with suspicion 22:26:13 03gammafunk02 07[nospawn] * 0.21-a0-251-gef00001: Correctly place PROX_AWAY_FROM_STAIRS monsters 10(18 minutes ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/ef0000140df5 22:26:13 03gammafunk02 07[nospawn] * 0.21-a0-252-g8e5ebb2: Restore numbers of generated level monsters to previous values 10(14 minutes ago, 1 file, 3+ 3-) 13https://github.com/crawl/crawl/commit/8e5ebb256bcc 22:26:17 03gammafunk02 07https://github.com/crawl/crawl/pull/606 * 0.21-a0-251-gef00001: Correctly place PROX_AWAY_FROM_STAIRS monsters 10(18 minutes ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/ef0000140df5 22:26:17 03gammafunk02 07https://github.com/crawl/crawl/pull/606 * 0.21-a0-252-g8e5ebb2: Restore numbers of generated level monsters to previous values 10(14 minutes ago, 1 file, 3+ 3-) 13https://github.com/crawl/crawl/commit/8e5ebb256bcc 22:26:20 |amethyst: yeah, maybe I'll go and do that sometime 22:26:20 elliptic: You have 1 message. Use !messages to read it. 22:28:42 |amethyst: player reporting this as a swastika, although I didn't really agree: https://i.imgur.com/i6gxdPz.png 22:29:10 |amethyst: as well as this, which they may have more of a case for https://vgy.me/hHTzgN.png 22:29:52 <|amethyst> do you happen to know which vaults those are? 22:30:05 I do no, and the relevant game hasn't finished yet 22:30:17 oh 22:30:20 well, probably one of them has 22:32:04 <|amethyst> oh, I guess they're both arrival vaults, which makes them easier to find 22:32:13 yeah 22:32:56 |amethyst: ah, nzn_arrival_water_spiral 22:33:00 I guess I merged that 22:33:07 I guess I'm the real nazi 22:33:40 add a vault that says kill all whites 22:33:53 then it will be even and nobody can complain about censorship :^) 22:34:13 gonna add a vault that places all these stones 22:34:17 that spell out 22:34:29 WHATABURGER PATTY MELTS ARE HOT GARBAGE 22:34:46 even better, make an encompass vault in the image of Stalin 22:34:47 would that fit? 22:35:00 I'll make it fit 22:35:06 increase max map size if I have to 22:35:12 people need to learn the truth 22:37:03 <|amethyst> saegor_arrival_simple_f perhaps? 22:37:48 !vault saegor_arrival_simple_f 22:37:49 1/1. https://github.com/crawl/crawl/blob/master/crawl-ref/source/dat/des/arrival/simple.des#L1609 22:38:04 who do I have to trick to get a loss.jpg vault added? 22:38:12 yeah I think so 22:38:20 would be surprised if there wasnt one already 22:38:26 doesn't look like more than a spiral to me though 22:38:38 I'll add one to hellcrawl next update just for you 22:39:31 cool 22:51:37 is there an easy way to specify an open door in a vault 22:51:43 extremely important question 22:52:20 03Floodkiller02 07https://github.com/crawl/crawl/pull/600 * 0.21-a0-234-g7013bae: Fix if formatting 10(3 minutes ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/7013baea6b9d 22:53:30 <|amethyst> hellmonk: what is your definition of 'easy'? 22:53:46 will not take more than 5 minutes 22:54:05 <|amethyst> KFEAT: ' = open_door 22:54:16 cool, thanks 22:54:21 didn't know that worked 22:55:18 if that doesn't work, maybe DNGN_OPEN_DOOR? 22:57:06 well you can't specify enum values in des files 22:57:16 those look for the feature name 22:57:36 which for DNGN_OPEN_DOOR is "open_door" 22:57:37 ah 22:57:52 it's not actually the "feature name" field but called something else 22:59:06 <|amethyst> "vaultname" in fact :) 22:59:41 that's the one 23:00:25 oh, maybe I should post this morgue of an objstat crash...oh crap 23:00:32 I deleted the dir 23:01:06 ERROR in 'dbg-scan.cc' at line 717: Portal enter_abyss[54] didn't get generated. 23:01:23