03:33:22 Experimental (bcrawl) branch on underhound.eu updated to: 0.23-a0-5261-gd9800d219b 03:37:09 03CrawlOdds02 07* 0.35-a0-693-gcd4325d217: Defer reset of monster slots 10(4 days ago, 14 files, 93+ 37-) 13https://github.com/crawl/crawl/commit/cd4325d2179a 03:37:09 03CrawlOdds02 07* 0.35-a0-694-g3667fe44f7: Keep monsters in mid cache until reset 10(4 days ago, 12 files, 35+ 68-) 13https://github.com/crawl/crawl/commit/3667fe44f751 03:37:09 03CrawlOdds02 07* 0.35-a0-695-g47c9543c1c: Don't apply enchantments posthumously 10(3 days ago, 1 file, 4+ 0-) 13https://github.com/crawl/crawl/commit/47c9543c1c17 07:58:26 03CrawlOdds02 07* 0.35-a0-696-g06d6a11a8d: Fix a wizmode forgetting crash 10(18 minutes ago, 1 file, 19+ 5-) 13https://github.com/crawl/crawl/commit/06d6a11a8df9 09:50:45 <02Monkooky> why is the inventory widget offline only? 09:53:44 <11Odds> Guessing because no one has made it work in webtiles rather than any design decision? 09:56:42 <04dracoomega> Utterly separate codepaths, yes. The fact that one exists in offline isn't even helpful for coding a webtiles version, tbh 09:57:05 remember that webtiles is a godawful hack 09:57:22 <04dracoomega> Quite a bit of webtiles UI that looks like offline tiles is still essentially reimplemented from scratch to merely look similar 09:58:03 and is still being reimplemented in fits and starts, as people contribute webtiles versions of offline tiles things 09:58:40 <04dracoomega> There's so many things about webtiles code which deeply sucks, imo >.> 09:58:48 <11Odds> Before I started knowing anything about the code I'd totally have guessed webtiles and offline tiles were essentially the same thing 09:58:54 meanwhile local tiles still doesn't have a monster list :< 09:59:00 <04dracoomega> Except it's also where the majority of the observable playerbase is 09:59:42 <04dracoomega> (Like, the fact that webtiles UI code can't even query anything about the game state that isn't bundled and passed as json messages is sometimes deeply awkward) 10:00:14 Odds: yeah it's extremely impressive how good the UI parity is if you don't know what to look for 10:02:26 <04dracoomega> Funnily enough, there's probably more UI code shared between console/local tiles than there is local tiles and webtiles. (There's a bunch of menu classes that straightforwardly work for the first two without extra steps, and require custom implementations in the latter) 10:14:55 <09gammafunk> working with the webtiles UI code does suck, but it's definitely a miracle that webtiles actually exists. You don't really see an online implementation that can run the game server-side with this level of quality in any other open source roguelike that I'm aware of 10:16:51 <09gammafunk> probably the webtiles 2.0 design that makes the most sense is the webassembly approach that mumra was toying with. Using a common SDL rending front end library for local tiles and webtiles and webtiles uses message passing to the server running the binary. Then the codebases for tiles and webtiles would be mostly shared 10:18:57 <09gammafunk> with the current webtiles javascript approach, I don't know if there's a realistic way to reduce the headaches via javascript rewrites alone 10:19:20 <04dracoomega> Yeah, I'm certainly not going to claim that this working as well as it does (and especially as many years ago as it was originally made) isn't an accomplishment. But I sure do hate interacting with it ^^; 10:27:28 <09gammafunk> There's certainly a lot to be said about just having your game always run locally with one app 10:33:00 ooc has anyone simply tried emscripten on it? 10:33:29 (would still need to bodge tiles instead of characters into it, but that seems like it'd be simpler in some sense) 10:34:37 also, I wonder if a wasm attempt now would work better, since it's evolved a bit since the last time 10:37:01 (I actually thought webtiles was just a wasm SDL implementation before I looked at the code) 10:37:13 way too old for that 10:37:51 I didn't know that! :P 10:54:41 <09gammafunk> geekosaur: mumra made a full working demo of the SDL app running in browser via wasm. But there is the issue that we actually want the game logic to run server-side 10:54:54 <09gammafunk> at least we do if we want to keep our online game tracking/scoring model 10:55:25 I was thinking invert it, a small stub fed by json from the in-browser game 10:55:32 milestones and such 10:55:54 <09gammafunk> well if you do that, you no longer have fidelity to the scoring 10:56:09 would also reduce server load quite a bit (hello, cao!) 10:56:10 <08o____0> tournaments 😔 10:56:49 or go the full way and do proper client/server separation with prediction for predictable stuff (e.g., rng dependent stuff has to be exclusively from the server). Prohibitively difficult to implement (well) though 10:56:51 <09gammafunk> in that people can run modified clients and have their scores counted. I'm not sure if there's tech to realistically detect modifications or whatnot 10:58:13 (You'd also have to account things like inputting two movements down a hallway before your next server update but there's a monster in LOS after the first move, so the proper thing to do would be to rollback the second move!) 11:02:59 <02Monkooky> I think this feels significantly worse than having to pause between moves 11:05:15 Yes it sounded much better in my head 15:42:27 Unstable branch on underhound.eu updated to: 0.35-a0-696-g06d6a11a8d (34) 22:36:08 Unstable branch on crawl.develz.org updated to: 0.35-a0-696-g06d6a11a8d (34) 23:00:16 Windows builds of master branch on crawl.develz.org updated to: 0.35-a0-696-g06d6a11a8d 23:33:30 Unstable branch on cbro.berotato.org updated to: 0.35-a0-696-g06d6a11a8d (34) 23:56:20 Monster database of master branch on crawl.develz.org updated to: 0.35-a0-696-g06d6a11a8d