00:47:37 the talk was 100% about why pizza tornado could never be released 02:58:44 Monster database of master branch on crawl.develz.org updated to: 0.26-a0-660-gc07710193c 03:07:17 -!- johnnyzero is now known as johnny0 03:10:41 Monster database of stone_soup-0.25 branch on crawl.develz.org updated to: 0.25.1-2-g60ff931df9 03:29:00 -!- jilles_ is now known as jilles 03:30:58 Fork (bcrawl) on crawl.kelbi.org updated to: 0.23-a0-3570-gb561bb485a 03:38:40 Fork (bcadrencrawl) on crawl.kelbi.org updated to: 0.22.1-3026-gd860270ee5 04:32:30 Monster database of master branch on crawl.develz.org updated to: 0.26-a0-660-gc07710193c 05:42:55 -!- amcnicky is now known as Guest75484 05:43:29 !messages 05:43:29 No messages for aMcNicky. 05:50:37 -!- amalloy is now known as amalloy_ 08:47:00 steves (L11 KoAr) Crash caused by signal #6: Aborted (Lair:1) 09:01:03 <03wheals> &rc steves 09:01:05 <04Cerebot> https://crawl.kelbi.org/crawl/rcfiles/crawl-git/steves.rc 09:01:34 !crashlog steves 09:01:35 4. steves, XL11 KoAr, T:19527 (milestone): https://crawl.kelbi.org/crawl/morgue/steves/crash-steves-20201006-124647.txt 09:04:53 the same crash when closing the skill menu after hitting a skill target, with the same person, but apparently not related to anything special in their rc 10:57:59 -!- ProzacElf_ is now known as ProzacElf 11:32:55 -!- cebolla is now known as spicycat 11:34:53 wheals there's a crash originating in 0.25-a (from the patch that does exact skill targeting) where if you close the skill target menu after hitting a skill target during world_reacts, the game crashes; the fix for this is rather complicated because basically no UI should be happening during world_reacts (as it is impossible to save in the middle of, and interactive UI should only happen at points when you can save) 11:34:53 I think xp gain needs to be refactored so that it is saved up during world_reacts and applied afterwords, sort of like a fineff 11:37:05 this is that crash, the signature is that after the disconnect skill menu code repeats "You need to enable at least one skill for training." in a tight loop 11:38:17 well, this wasn't world_reacts I guess, but it's a similar issue 11:38:59 basically gain_exp / the skill menu needs to not be a side effect of monster_die 12:53:15 alexjurkiewicz: if you clock on a ttyrec link at e.g. https://crawl.project357.org/ttyrec/Senshi/ it shows it in-browser; is it possible to improve the mime type for those urls so that it won't think it's text? 12:53:15 s/clock/click/ 14:11:44 is console rendering laggy somehow, recently? 14:12:05 I'm seeing delays in many cases that feel like several hundred milliseconds, where profiling is not detecting anything 14:13:19 and I guess also go away if I hit keys very rapidly 14:13:52 iirc there were some changes to how console was buffered 14:17:18 I was able to record this so I guess it's not just me hallucinating, but has anyone else been noticing this? 14:18:59 %git 0b5ff556b1e2c4 14:18:59 07Aidan Holm02 * 0.26-a0-394-g0b5ff55: Fix excessive redraw when autofiring ranged weapons 10(3 months ago, 1 file, 23+ 5-) 13https://github.com/crawl/crawl/commit/0b5ff556b1e2 14:20:11 I guess this is the most likely suspect if it is not just something system-specific 14:54:34 haven't noticed a delay but also haven't been testing too much in console 14:56:29 what kind of delay? I notice several on this slow ancient laptop 15:35:42 it manifests as a redraw delay on actions like moving 15:37:36 I see that with most such actions, only at the end for multiple-AUT stuff 15:37:49 no idea what it's doing in between 15:39:03 here's a video of me pressing move left at an even pace, about once per second: https://www.dropbox.com/s/crv58jjzbo8rafx/laggy_delay.mov?dl=0 15:39:17 (ignore the last bit where it took me a while to figure out how to stop the screen recording) 15:39:53 (and the window title which is meaningless, leftover from being logged into cbr2 in that window) 15:44:18 hm, that's fairly smooth here unless something new comes into view 15:44:57 did you notice it appears to move two tiles at a time often :-) 15:45:03 yes, I saw that 15:45:12 I'm pretty sure that's new 15:45:32 oh or do you mean you aren't seeing that lag and it's smooth on your end 15:45:46 hm, come to think of it I have not rebuilt since Sunday (and I no longer have time to do so today, it takes close to an hour) 15:46:08 this isn't very bleeding edge (it's a branch off a few weeks ago), so that shouldn't matter 15:46:10 right, it's smooth over here aside from delays when new things come into view or my HP reach max or etc. 15:46:54 are you still on mac? 15:46:54 and that's only like a quarter second delay 15:46:54 no, this is an old 1st generation netbook running linux 15:46:54 ok 15:46:54 I wonder if there is some weird interaction with mac ncurses 15:47:13 this is bad enough that I feel like there should have been some complaints if it's not just me 15:47:33 see if the same happens with h instead of the arrow? 15:47:40 oh I meant h 15:47:54 I tried with both here, both worked okay 15:48:12 it's not specific to moving, i noticed it because it was affecting other stuff I'm working on to the point I loaded up my profiler to figure out what I'd done wrong 15:48:28 I guess bisecting comes next 15:49:23 perhaps I should do things like restart terminal first 18:06:19 -!- amrqx is now known as yesno 20:14:51 %git 2f00e6de5b3949 20:14:51 07bh02 * 0.13-a0-2960-g2f00e6d: Take MP before casting spells. 10(7 years ago, 3 files, 36+ 20-) 13https://github.com/crawl/crawl/commit/2f00e6de5b39 20:15:00 does anyone know why we do this? 20:19:00 <10PleasingFungus> not i 20:19:04 <10PleasingFungus> ask bh? 20:19:13 <10PleasingFungus> oh, 7 years ago 20:20:02 the only thing I can figure out from that commit is that there may have been a djinni-related reason 20:21:25 <10PleasingFungus> it’s immediately followed by a bug report that veh didn’t give mp if you were full on mp before casting 20:21:25 <10PleasingFungus> but causality seems iffy there 20:21:25 <10PleasingFungus> (looking at crawl-dev logs) 20:21:25 hm 20:21:25 <10PleasingFungus> oh 20:21:28 <10PleasingFungus> actually that was why 20:22:07 <10PleasingFungus> “i wrote the bug so people would know why the heck i wrote the patch... on second thought i could have written a better commit message “ 20:22:11 <10PleasingFungus> now yoy know 20:22:21 lol 20:22:33 I started showing uselessness reasons directly in the targeter, which is called after the decrement, so there's an annoying problem the other way at low mp 20:23:09 I guess this does seem like the kind of thing that I wouldn't want to mess with changing 20:24:28 am I going to need to unwind_var 20:25:28 anyways, thanks pleasingfungus 20:29:18 I could abuse fake_spell 20:30:11 lol someone already thought of unwind_var 20:32:41 this is possibly crawlcode worthy: https://github.com/crawl/crawl/blob/master/crawl-ref/source/spl-cast.cc#L1097 20:41:35 <10PleasingFungus> hurts my brain to look at it 20:41:40 <10PleasingFungus> might need too much context to be a good tweet tho 21:05:06 currently when you're in targeter, the amount of mp the spell will cost is shown as deducted in the HUD -- that started happening recently iirc so it's not directly due to the commit, but it is nice 22:01:10 svendre (L6 GhGl) Crash caused by signal #6: Aborted (Bailey) 22:02:26 I like the new plant-dying mechanic. 22:02:32 Also Palentongas are fun. 23:18:24 so much qol on them pesky plants now 23:36:28 any pure ownage shit happen today?