00:47:45 Monster database of master branch on crawl.develz.org updated to: 0.34-a0-1914-g7991ae8747 04:31:29 Experimental (bcrawl) branch on underhound.eu updated to: 0.23-a0-5261-gd9800d219b 13:24:43 cforsyth (L18 DjFE) Crash caused by signal #6: Aborted (Swamp:1) 13:26:01 cforsyth (L18 DjFE) Crash caused by signal #6: Aborted (Swamp:2) 16:14:42 <07w​izardike> It looks like killing a spriggan rider with inner flame on it is leaving an invalid mgrid. This isn't very surprising given the questionable way we handle the death of mounted monsters 16:14:43 <07w​izardike> https://cdn.discordapp.com/attachments/747522859361894521/1454251102265675971/Screenshot_2025-12-27_115622.png?ex=69506861&is=694f16e1&hm=bfe92cb3e7a14a9a436f9d538944a49cff1192b8cd00cdbcd4210528f2df48a3& 16:14:43 <07w​izardike> https://cdn.discordapp.com/attachments/747522859361894521/1454251102991024232/Screenshot_2025-12-27_115733.png?ex=69506861&is=694f16e1&hm=715331def10faf890310274e4695a2ac12cc406e2339a3342ac02d5dc9ce54b6& 16:14:44 <07w​izardike> https://cdn.discordapp.com/attachments/747522859361894521/1454251103368777849/Screenshot_2025-12-27_115845.png?ex=69506861&is=694f16e1&hm=fcd0ddd0169720b1acfa6327453534e85d329994db587b6bd68c0a985c511a61& 16:17:31 <04d​racoomega> That actually does surprise me, tbh. When killing just 'one half' of the mounted combo, I thought it basically just created a new monster representing the half that was dying, didn't even put it on the mgrid, and then killed it immediately. And just changed the type of the monster that was on the grid. 16:32:43 <04d​racoomega> I hadn't realized that mounted monsters blew up twice from inner flame, either, though >.> 16:32:52 <04d​racoomega> Feels like that is also not how it should work 16:33:20 <04d​racoomega> (They only blow up once if they don't dismount, but a dismount can cause 2 explosions immediately) 16:43:14 <07w​izardike> The problem is that the new monster has the same position (so that it can leave a corpse), so if we ever remove it from the mgrid it will actually remove the old monster leaving the old monster floating. 16:43:51 <04d​racoomega> The offending line is this, incidentally (in explode_monster()): C++ // Detach monster from the grid first, so it doesn't get hit by // its own explosion. (GDL) env.mgrid(mons->pos()) = NON_MONSTER; 16:44:13 <04d​racoomega> I am not sure that even matter anywhere, with explosions being fineffs? 16:45:38 <04d​racoomega> It seems a reasonable safety measure to re-add the original to the grid after the mounted death, just in case. (Though I do also think the mount should not explode) 17:53:55 <02M​onkooky> it seems to me that mounted monsters should be 'shapeshifting' rather than dying and spawning a new monster 17:57:00 <07w​izardike> They do 'shapeshift' to the surviving part of the monster, but the other part of the monster still has to die and we do this by creating a new monster and immediately killing it 21:09:07 03WizardIke02 07* 0.34-a0-1915-gc2962e8470: Fix exploding mounted monsters sometimes crashing 10(3 hours ago, 1 file, 1+ 5-) 13https://github.com/crawl/crawl/commit/c2962e84704c 21:09:08 03WizardIke02 07* 0.34-a0-1916-g4e02dda68d: Fix killed monsters raised as thralls but getting xp tracking 10(14 minutes ago, 1 file, 6+ 3-) 13https://github.com/crawl/crawl/commit/4e02dda68dbc 23:35:41 Unstable branch on crawl.develz.org updated to: 0.34-a0-1916-g4e02dda68d (34) 23:59:05 Windows builds of master branch on crawl.develz.org updated to: 0.34-a0-1916-g4e02dda68d