00:41:53 Fork (bcrawl) on crawl.kelbi.org updated to: 0.23-a0-4907-g988a294422 03:22:07 Experimental (bcrawl) branch on underhound.eu updated to: 0.23-a0-4907-g988a294422 05:27:37 Unstable branch on crawl.akrasiac.org updated to: 0.30-a0-1123-gabc949c (34) 05:33:21 <06advil> hm I should be able to connect to the last tournament db on cdo at the command line using the settings in loaddb.py, right? 05:39:33 <06advil> lol also please tell me this supports py3 06:59:43 <06advil> dup from roguelikes, I ran some basic counts on banners from t0.19: * least popular: trog (brute force, 36+6+9=51), vehumet (ruthless efficiency, 35+32+1=68), lugonu (spiteful, 89+3+7=99), qazlal (prophet, 104+63+22=189) * most popular: ely (pious, 883+152+16=1051), chei (slow and steady, 996+65+23=1084), gozag (avarice, 894+475+4=1373), tso (vow of courage, 1371+94+19=1484) * most popular t3 are hepl, zin, makhleb * player banner counts 06:59:43 by tier: 10236, 2626, 817 * for all that heretic has been getting some hate, it's solidly middle of the pack in terms of popularity: 263 + 157 + 14 = 434 07:00:33 <06advil> er 07:00:36 <06advil> t0.29 07:48:08 <06advil> for various banners, but especially wjc (sifu), kiku (lord of darkness), ru (ascetic), sif (lorekeeper), there is a strong skew towards the tier 1, which could indicate some need for rebalancing. For the other tiers nothing pops out like that except that it appears that for makhleb (speed demon) essentially no one who does it stops at t1 07:48:39 <06advil> maybe most of the players who get any makhleb banners are speedrunning whole games 08:52:53 <06advil> Alright, here's my collected notes on changes for this tournament: https://github.com/crawl/dcss_tourney/issues/13#issuecomment-1518690661 09:00:03 <06advil> I should say also that I kinda think we overall just have too many banners 09:00:11 <10PleasingFungus> plausible 09:00:14 <06advil> though that proposal doesn't reduce them by much 09:02:01 <06advil> I suppose all three of prophet/heretic/lord of darkness mentioned at the end of those notes could simply be removed 09:51:52 <08Implojin> anecdotally I think it would be a mistake to get rid of lord of darkness 3, even if it doesn't see many people actually completing it; it's something that people seem to get excited about finishing 09:52:18 <08Implojin> https://old.reddit.com/r/dcss/comments/4kgmfq/lord_of_darkness3/ 09:52:22 <08Implojin> https://old.reddit.com/r/dcss/comments/p69nnh/027_tournament_lord_of_darkness_3_minotaur/ 09:52:28 <08Implojin> https://old.reddit.com/r/dcss/comments/p0z86r/yavp_baee_lord_of_darkness_run/ 09:52:29 <08Implojin> etc 09:57:40 <12ebering> damn that was one of my finest games 09:57:49 <12ebering> LoD t3 is a good banner imo 09:58:14 <12ebering> having a spread of t3 banners from the "easy for a newplayer" to the "legendary" I think helps spice up the banners purpose in the t 09:58:21 <08Implojin> i vaguely remember doing one where i found arga early and decided to go for it but i'm not sure which t it was to pull up the morgue offhand 09:59:25 <12ebering> both LoD (kiku) and Vow of Courage (TSO) are some of the best t3 banners imo 09:59:39 <08Implojin> i'd kind of like to reduce the sheer realtime requirement for the megazig banner but I haven't come up with any good ideas to accomplish this that wouldn't overlap somehow with other tc or realtime things, or wouldn't make meteoran mandatory for it 10:00:41 <08Implojin> err, megazig category rather 10:02:11 Hello Crawl devs! Just letting you know that the game has been available on Flathub for numerous year. It might be time to update the Linux instructions on your website. https://github.com/flathub/org.develz.Crawl 10:04:18 <06advil> to be clear, "not many" in the case of t0.29 was 2 10:04:28 <08Implojin> dang 10:04:52 <06advil> although both harvest and ruthless efficiency only had 1 t3 10:05:04 <06advil> harvest I think still has some ui issues we never fixed? 10:05:32 Discord|: I started packaging it when it was at 0.21. 10:06:06 <06advil> arteymix[m]: sorry, that was responding to some earlier stuff, not to you 10:06:12 <06advil> thanks for packaging flathub! 10:07:36 <06advil> I would say that LoD stands out because it is 431 t1 (a normal number), 12 t2, and 2 t3 10:08:44 <06advil> arteymix[m]: I think actually a recent commit (https://github.com/crawl/dcss-website/commit/bdaa21fd40c802e2ba85c779f52b82ac2a86ff13) does link to the flathub package, but it hasn't gone live on the crawl website 10:09:03 <06advil> (also, Discord| is a relay bot) 10:12:00 <06advil> vow of courage seems in very reasonable shape wrt the numbers, it's also the most popular t1 10:13:46 <08Implojin> out of curiosity, what do the numbers look like for nature's ally 3? 10:14:36 <08Implojin> i think it's maybe not surprising that banners that force players down uncomfortable/nonstandard paths are less popularly completed, but they're also probably more memorable? 10:14:42 <06advil> I'll just post a version of this table as a gist 10:15:18 <06advil> https://gist.github.com/rawlins/e8c8d5e1cbda09924741d9139f5d2b20 10:15:51 <06advil> people definitely do fed, it's sort of middle of the pack as a t3 (but as you'll see, a lot of people don't do t3s at all!) 10:16:09 <06advil> the first two lines there are some kind of metadata, ignore them 10:21:35 <06advil> yeah, to be clear I don't have a problem with pushing people towards non-standard paths, but I'm just looking for the most extreme cases, and also cases where the tiers seem out of whack with each other 10:24:00 <06advil> for LoD maybe the biggest problem is the t2 numbers, not the t3; though synergy with the hypothetical new vaults rune banner might push this up a bit in 0.30 10:25:08 <06advil> I'd complete forgotten about the harvest ui issues (iirc still there's no way to see what uniques you're missing? this was an oldt feature that we keep failing to port) 10:27:49 <08Implojin> harvest also has the issue that it's pretty random, you can play the entire t and still not find murray 10:30:48 <08Implojin> (looks like this happened to sergey last t) 10:30:49 <08Implojin> https://crawl.develz.org/tournament/0.29/unique-harvesting.html 10:33:30 <06advil> yeah, though way more people did it in the old ui. In 0.24 it looks like 66 people got harvest 3 10:33:43 <06advil> annoyingly the 0.25 t pages don't have rankings for banners I guess 10:35:06 <06advil> in t0.26 it's easier to see, and 4 people got all 81 checks for that category 10:37:40 <06advil> maybe we tightened our criteria? 10:38:37 <12ebering> yeah old harvest was just over n, not all iirc 10:38:48 <12ebering> might be good to go back to over 72 uniques for harvest t3 10:38:53 <12ebering> to smooth out that randomness 10:39:22 <08Implojin> yeah i was thinking maybe make it 80 or something 10:41:40 <08Implojin> maybe a higher level question to ask here is are these banners supposed to be extremely rare? on the one hand you have things like lord of darkness or whatever, on the other hand, most of the banners clearly look like they're targeted as easier picks for players who maybe have a hard time winning, to let them still participate 10:41:54 <06advil> huh it's hard to check specific possibilities because the json kill list doesn't dedup 10:42:33 <06advil> I think there's supposed to be a good range, with many t1s possible for weaker players (including ones that don't win) 10:42:34 <08Implojin> maybe some of these harder banners could be shifted into a new challenges category 10:43:10 <06advil> essentially all t3s are currently very hard (in some sense) and require reasonably consistent winning 10:45:35 <06advil> no t1s require winning (except maybe the score one?), and most should be doable by enyone who puts some time into it 10:45:40 <08Implojin> like, i'm wondering if the reason some of these t3s are rarely completed isn't that players don't find them fun, but that they're not incentivized strongly enough to attempt them if they're trying to play for tournament points 10:47:17 <06advil> that seems possible, though I would say on the whole the banners and even t3s aren't primarily aimed at points players? 10:49:17 <06advil> only one player in t0.29 got max banner points 10:49:46 <06advil> it did work out for them I guess, that is narz in 2nd place 11:20:40 <09gammafunk> makes me wonder if a good feature would be to have Botessa respond to Dicord| pings from irc users with instructions on using @ to ping discord users 11:21:15 <09gammafunk> er Discord| 11:47:52 <06advil> maybe, that was also from matrix so I'm not sure how it looks 11:48:11 Discord|: test 11:48:27 <06advil> yeah, it's even easier to make that mistake from matrix, you just click 12:12:45 <09gammafunk> ah, so it was, although I've seen non-matrix irc users do it as well 13:11:50 03PleasingFungus02 07* 0.30-a0-1124-g6e8f0f562f: Reduce shield penalties with Str instead of size (Lici) 10(5 weeks ago, 2 files, 11+ 11-) 13https://github.com/crawl/crawl/commit/6e8f0f562f66 13:17:37 -!- MakMorn_ is now known as MakMorn 13:19:16 Unstable branch on crawl.kelbi.org updated to: 0.30-a0-1124-g6e8f0f562f (34) 14:36:48 oh, the bridge will make @foobar into a ping? 14:45:15 <08raxraxraxraxrax> CAO disk replaced and RAID rebuilding. Assuming this all works fine, there are other things I could look at doing post-tournament (upgrade the version to... debianoldoldstable or something else significantly newer..., increase the size of the partition, try to upgrade the xen host to see if that provides any improvements), but the chance of non-trivial or even sizable downtime from those is higher, so I'm going to wait until the 14:45:16 tournament is over unless it looks like we need to do something else. 14:52:30 <09gammafunk> perryprog: yes, it does that already 14:54:24 <09gammafunk> fantastic! Plan for post-tournament upgrade sound nice as well. An upgrade to a more recent debian would be wonderful indeed, much needed even if it does require some downtime, however I'll leave coordination of that to you and advil, since he's far more aware of the issues/benefits than I am. 14:55:15 <08raxraxraxraxrax> I know it's possible to get a VM on that machine at least to stretch, because I've done that, but... that's... still very very old. I may experiment with my own VM first :P 15:25:44 Unstable branch on underhound.eu updated to: 0.30-a0-1124-g6e8f0f562f (34) 16:29:12 <08nicolae> i have noticed a bizarre phenomenon where the vaults layout does not seem to want to rotate vaults 🤔 17:10:02 <09gammafunk> I take it that it's a vault room? 17:10:21 <09gammafunk> and by mirror you mean it's never horizontal/vertical flipping the vault in placement? 17:11:13 <08nicolae> yes, big V Vaults, and it does not flip the vault, even after regenerating the level a few dozen times 22:28:35 <09gammafunk> re: tournament, not sure if this was discussed already, but in the list of 0.29 tournament changes: > Meteorans are newly added in 0.29, but unlike other species, they are not permitted to set class high scores. They are allowed to set their combo high scores and the meteoran species high score normally. This aspect may be re-assessed in the 0.30 tournament, especially if high score becomes based on action duration (arbitrary units of 22:28:36 time) instead of turns. 22:30:15 <09gammafunk> obviously aut scoring will forever remain a pipe dream, like SSO, but worth looking at whether to keep that 22:33:44 <10PleasingFungus> don't think anything's changed that would invalidate the justification there 22:35:01 Unstable branch on crawl.develz.org updated to: 0.30-a0-1124-g6e8f0f562f (34) 22:57:13 Windows builds of master branch on crawl.develz.org updated to: 0.30-a0-1124-g6e8f0f562f 23:15:19 Unstable branch on cbro.berotato.org updated to: 0.30-a0-1124-g6e8f0f562f (34) 23:54:31 Monster database of master branch on crawl.develz.org updated to: 0.30-a0-1124-g6e8f0f562f