IRC logs for #openttd on OFTC at 2024-08-15
⏴ go to previous day
00:39:07 <_glx_> should work, but may need improvements
02:19:14 *** D-HUND has quit IRC (Ping timeout: 480 seconds)
02:53:08 *** gnu_jj_ has joined #openttd
02:56:14 *** gnu_jj has quit IRC (Ping timeout: 480 seconds)
04:44:01 <DorpsGek> - Update: Translations from eints (by translators)
05:12:48 *** SigHunter has quit IRC (Read error: Connection reset by peer)
05:13:50 *** SigHunter has joined #openttd
05:21:54 *** SigHunter has quit IRC (Ping timeout: 480 seconds)
05:22:03 *** SigHunter has joined #openttd
05:36:05 <andythenorth> slow Horse compile is slow
05:36:16 <andythenorth> I suspect the per-variant name string handling
05:37:03 <andythenorth> 19326 instances of `STR_PARENTHESES` in iron-horse.nml
05:40:33 <andythenorth> hmm not a very significant difference
06:13:42 <andythenorth> ok approx 35% of the compile time is due to intense use of vehicle variants
06:14:24 <andythenorth> ~7000 vehicles in Horse, of which ~5000 are variants
06:15:08 <andythenorth> knocking those out of the grf reduces compile time from ~40s to ~25s
06:17:43 <andythenorth> "all working as expected"
06:57:01 *** HerzogDeXtEr has joined #openttd
07:38:36 *** verzide68 has joined #openttd
07:38:36 <verzide68> dumb question - where can I find the road "tileset" for the game?
07:43:14 <_pruple> between sprites 1313 and 1405 in trg1r.grf
08:25:12 *** mindlesstux has joined #openttd
08:31:57 *** SigHunter_ has joined #openttd
08:33:54 *** SigHunter has quit IRC (Ping timeout: 480 seconds)
08:41:34 *** SigHunter_ has quit IRC (Ping timeout: 480 seconds)
08:41:59 *** SigHunter has joined #openttd
09:00:54 *** SigHunter has quit IRC (Ping timeout: 480 seconds)
09:04:36 *** SigHunter has joined #openttd
09:28:50 *** SigHunter_ has joined #openttd
09:31:39 *** SigHunter has quit IRC (Ping timeout: 480 seconds)
09:37:39 *** SigHunter_ has quit IRC (Ping timeout: 480 seconds)
09:42:21 *** SigHunter has joined #openttd
09:56:50 <peter1138> Step 1: make reversing trains work properly at all...
10:04:05 <andythenorth> I think we should add lots of flags for magic
10:04:22 <andythenorth> preferably ones that aren't orthogonal and have surprise combinatory effects
10:04:33 <andythenorth> or we could just reverse the whole consist
10:04:45 <andythenorth> it will all be done by christmas?
10:08:28 <reldred> reversing trains reversing trains reversing trains
10:08:40 <reldred> finally ill be able to build terminus stations again
10:11:30 <johnfranklin> Sometimes we feel computers do whatever they want…
10:13:21 *** akimoto has joined #openttd
10:15:13 <peter1138> The reversing issue is partly "what tile is an engine actually on?"
10:25:23 *** mindlesstux has quit IRC (Ping timeout: 480 seconds)
10:26:56 <peter1138> To fix this, train movement code needs to track front and rear position (maybe only of the consist) explicitly instead of only the vehicle centres.
10:29:15 *** mindlesstux has joined #openttd
10:30:19 <peter1138> Which track is the train on?
10:30:19 <andythenorth> can vehicles do that relative to consist?
10:30:59 <peter1138> Only the middle vehicles can be moved.
10:31:47 <peter1138> Moving the first and last vehicles could place the vehicles on different tiles.
10:36:46 <peter1138> The train is... not on the water tile.
10:39:52 <peter1138> Also in terms of v->tile, all 3 of those vehicles are on the same tile.
10:40:09 <peter1138> But the train is 1.25 tiles long
10:42:20 <andythenorth> so it's not just a one line change to do this? 😮
10:43:32 <reldred> andythenorth: if it was then I would have done it 😛
10:43:41 <FLHerne> in C everything can be a one-line change, just remove all the newlines
10:43:58 <reldred> disgusting, I love it
10:44:00 <FLHerne> this is why Python is a bad language
10:45:17 <verzide68> i'm curious, are separate parts of the rail considered different, because if not how would the trains pathfind across? at least where can i find the code that deals with that?
10:48:43 <gwyd4016> FLHerne: Strong agree
10:49:09 <gwyd4016> White space should mean nothing outside of a string
10:50:24 <_jgr_> There are plenty of things to not like about Python, but the indentation/whitespace is just a non-issue
10:50:48 <FLHerne> I mostly agree /;#]**************
10:51:22 <FLHerne> I mostly agree, although whitespace-only blocks make it painfully easy to
10:51:34 <andythenorth> lol those who still get bothered by python
10:51:47 <andythenorth> and then everyone else, getting things done
10:51:53 <FLHerne> ... get code in the wrong place
10:52:10 <FLHerne> I only complain about it for the purposes of bad humour though
10:52:44 <gwyd4016> I complain about it because it is the singular issue that I have built my life around
10:53:20 <FLHerne> verzide68: question isn't very clear, but yes, the different segments of rail on each tile are represented by `trackbits`, see e.g. src/pathfinder/follow_track.hpp
10:54:03 <peter1138> IIRC python would get upset by mixed tabs and spaces.
10:54:17 <peter1138> But then again, it is perfectly normal to get upset by that.
10:55:07 <FLHerne> the actual pathfinding is in src/pathfinder/yapf/ which is notoriously obtuse
10:55:38 <andythenorth> enjoy a big diff
10:56:32 <FLHerne> again, if people would simply sed s/\n//, all diffs would be one-line changes as well
10:56:39 <FLHerne> peter1138: yes please :D
10:57:07 <peter1138> Lets get rid of the map array.
10:57:21 <peter1138> Keep it in `std::list`
10:57:57 <peter1138> So a 4096x4096 map could require 2^24 pointer transversals 😄
11:36:48 <peter1138> Hmm, maybe I will have to just build a PC.
11:39:38 *** SigHunter has joined #openttd
13:46:35 <peter1138> > Chats are temporarily unavailable
14:08:13 <Rubidium> FLHerne: the C++ compiler might not require newlines, but the preprocessor requires them so compiling OpenTTD without newlines is not going to work
15:10:06 <andythenorth> should I bug report that?
15:10:21 <andythenorth> it's cb results, rather than action 0, not sure those are searched
15:11:26 <peter1138> The callback result is not persisted.
15:11:47 <peter1138> So sorting or filtering would need to execute the callback result again for every item.
15:11:53 <peter1138> Probably not that much of a problem.
15:12:21 <peter1138> Oh, Minecraft is trying to do bundles again, eh?
15:33:46 *** alfagamma0007 has joined #openttd
15:33:46 <alfagamma0007> peter1138: damn, never knew you can do the *stuff* by literally flooding it around
17:03:01 *** Flygon has quit IRC (Read error: Connection reset by peer)
18:01:24 *** XYZ has quit IRC (Ping timeout: 480 seconds)
18:30:57 *** scramjet has quit IRC (Quit: User went offline on Discord a while ago)
20:36:43 *** nielsm has quit IRC (Ping timeout: 480 seconds)
21:03:58 *** HerzogDeXtEr has quit IRC (Read error: Connection reset by peer)
21:28:38 *** keikoz has quit IRC (Ping timeout: 480 seconds)
21:41:49 *** akimoto has quit IRC (Remote host closed the connection)
21:45:23 *** ChanServ sets mode: +v tokai
21:52:13 *** tokai|noir has quit IRC (Ping timeout: 480 seconds)
continue to next day ⏵
