IRC logs for #openttd on OFTC at 2024-11-25
⏴ go to previous day
00:07:30 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
00:09:01 <_glx_> digitalsoma: I'm not sure how to do it in vscode, but you set cmake variable in configuration file for openttd (I do it in CMakePresets.json, but it can be done with CMakeSettings.json too)
00:43:20 *** Flygon has quit IRC (Read error: Connection reset by peer)
00:53:06 <digitalsoma> I'll look into those for vscode, thanks!
01:46:53 <digitalsoma> Ok, with Visual Studio 2022, I was able to get it to find the libraries listed under compiling.md. However, Cmake "Could NOT find Grfcodec (missing: GRFCODEC_EXECUTABLE GRFID_EXECUTABLE NFORENUM_EXECUTABLE)".
01:46:53 <digitalsoma> Also, in the CMake output, it says it found zlib and the package is compatible with the targets, but in the CMakeSettings.json, under variables, the "name": "zlib_include_dir" has "value": "zlib_include_dir-notfound" and the CMAKE_TOOLCHAIN_FILE has a "property name not allowed by schema" warning. Are these issues I have to resolve?
02:21:16 <_glx_> grfcodec and nforenum are not required, unless you want to modify openttd.grf (you don't want 😉 )
02:32:28 *** tokai|noir has joined #openttd
02:32:28 *** ChanServ sets mode: +v tokai|noir
02:39:08 *** tokai has quit IRC (Ping timeout: 480 seconds)
04:48:34 <DorpsGek> - Update: Translations from eints (by translators)
06:17:29 *** keikoz has quit IRC (Ping timeout: 480 seconds)
07:17:38 <DorpsGek> - Add: summary for week 47 of 2024 (by OpenTTD Survey)
10:18:16 <peter1138> [2024-11-25 10:17:54] dbg: [sprite:3] ReserveSpriteCache: Reserved capacity for 557594 sprite cache entries (22303760 bytes)
10:18:27 <peter1138> [2024-11-25 10:18:01] dbg: [sprite:1] 554287 total sprites, 1248 extra sprites, 93 from baseset, 1155 from fallback
10:56:36 <xarick> the widening is not widening enough
11:01:03 <xarick> talltyler: would be disappointed
11:50:14 *** XYZ has quit IRC (Ping timeout: 480 seconds)
11:51:25 <peter1138> What about your crash?
11:55:05 <xarick> i was getting the tile height of the wrong thing
12:18:48 <exceptik> ah, reminds me of road ownerships with 500 companies 😆 when using already existing bits in tile data aint a great idea
12:23:12 <peter1138> "Just change MAX_COMPANIES"
12:24:15 <xarick> tropic generator needs to look for water to know whether to set a tile rainforest
12:24:36 <xarick> but at this stage, with my changes, there is no sea
12:25:14 <xarick> GetTileZ(tile) == 0 does the same thing, but it costs a lot of performance 😦
12:25:23 <xarick> what alternatives do i have?
12:26:44 *** SigHunter has joined #openttd
12:27:23 <peter1138> Do the tropic generator later.
12:48:42 <merni> xarick: What a dystopian world
12:58:46 *** XYZ has quit IRC (Ping timeout: 480 seconds)
13:05:53 <xarick> it is a strategy I found to counteract terraforming on water tiles
13:14:33 <xarick> can't postpone desert/rainforest generation after river generation because river generation needs to know where are the rainforest to place a spring
13:25:35 *** asymptotically2 has quit IRC (Remote host closed the connection)
13:30:23 <xarick> I have a tendency to make unneeded changes
13:30:31 <xarick> just because it would look better
13:37:47 <xarick> something akin to... "if IsTileFlat can return the height of the tile and there is another TileHeight on the tile, why not change the code so the height is only done once?".
13:38:07 <xarick> then this inflates the diff
13:38:39 <xarick> then ppl complain about "irrelevant" change
13:38:55 *** toktik has quit IRC (Remote host closed the connection)
13:39:57 *** asymptotically2 has joined #openttd
14:08:00 <peter1138> Maybe you need to set dependabot up properly?
14:09:42 <peter1138> I know nothing about it, but it should only be running against your main branch.
14:10:04 <peter1138> (In my limited understanding of what it does)
14:12:30 <_glx_> yeah only current version should be checked, and if upgrade is needed on release then you do a new release
14:33:00 <peter1138> Back to fighting Bootstrap's grid system.
14:34:05 <peter1138> I want 3 columns, "a, b, c", unless the screen is too narrow, in which case I want two columns, with c below b, "a, b/c"
14:34:50 <peter1138> But also "a" may not exist, in which case "b, c" is good unless the screen is very narrow.
14:46:42 <xarick> there is no GetTileSlopeMaxZ 😦
14:56:03 <peter1138> Hmm, I need col-md-6 or col-md-4 depending on what else is visible. This system sucks.
15:04:35 <xarick> there are stalls during town generation
15:04:48 <LordAro> peter1138: css grid clearly
15:12:02 <peter1138> Yeah, already working much better.
15:12:50 <peter1138> Although I think I still need two layouts for the optional column a.
15:14:46 <xarick> what is the official english word for river connection to the sea?
15:15:20 <xarick> inclined slope of river with tile height 0
15:20:28 <xarick> > Change: Postpone sea conversion to after river generation
15:20:28 <xarick> > To ensure that rivers get properly widened at the mouth via a terraforming command that can't destroy water, postpone conversion of the tiles at level 0 to after river generation.
15:23:14 <xarick> Copilot for the rescue
15:23:18 <xarick> > Change: Postpone sea conversion to after river generation
15:23:18 <xarick> > To ensure that rivers are properly widened at the mouth using a terraforming command that doesn't affect water, postpone the conversion of land tiles at height level 0 to sea tiles until after river generation. This change ensures that the river widening process is not disrupted by premature sea conversion.
15:24:36 <xarick> I'm surprised it kept the word "terraforming"
15:30:26 *** XYZ has quit IRC (Ping timeout: 480 seconds)
15:33:35 <xarick> generating tropic is now slower, enjoy
15:36:18 <xarick> trying to figure a way to make it not slower
15:41:19 <xarick> 1 check was no successful
15:41:55 <_glx_> it's the usual stuff, you add more "details" which implies more things to do, so it's slower
16:05:14 <xarick> 3 times slower at CreateDesertOrRainForest
16:10:32 <xarick> hmm it's not even doing a 100% match
16:12:29 <xarick> oh, only the tropic isn't
16:14:29 <peter1138> We've never really focused on making sure game seeds produce the same world.
16:17:03 <kuhnovic> Self-imposed requirements
16:19:09 <xarick> hmm yeah, there are differences
16:23:33 *** benjaminv has joined #openttd
16:24:53 <peter1138> The TGP tropical zone selection is actually really bad anyway.
16:51:36 <peter1138> The original generator's strategy of making the map half rain forest, half desert worked fine. At 256x256.
17:06:12 <mnhebi> I mean if you really wanted to model how deserts form in real life you'd take a year to generate a map :3
17:40:08 <xarick> I have a suspicion trees are getting in the way
17:43:05 <xarick> I solved the problem with brute force
17:44:44 <xarick> probably an even slower tropic gen as a result, but now it 100% matches
17:49:38 <xarick> yep... 0.4 seconds more required
17:56:29 <xarick> now I am wondering which tiles were not matching GetTileZ != 0
18:05:02 <xarick> why was it so obvious?...
18:05:11 <xarick> why didn't I think of them
18:12:50 <xarick> hmm apparently there's more
18:25:27 <xarick> it's one of those 3 corner raised slopes
18:26:29 <xarick> it may or may not be converted to sea depending on the slopes surrounding it
18:28:06 <xarick> it's at sea level, but it's not sea
18:28:13 <xarick> and is not converted to sea
18:28:41 <xarick> GetTileZ however gives me a 0
18:34:03 *** nielsm has quit IRC (Remote host closed the connection)
18:34:15 *** gelignite has joined #openttd
19:59:34 <peter1138> IMHO, EndSegmentReason and EndSegmentReasonBits are reversed.
20:05:54 <xarick> can I edit "table/genland.h"
20:06:12 <xarick> I have an "optimization" idea
20:09:03 <xarick> it the diffs become ordered, then on Map::Iterate(), i only need to iterate the diffs once. the next tile from map::iterate would only be requiring to check the last item on those diffs
20:11:28 <xarick> the last item of a X coordinate, or all X items if I switch to the next Y
20:22:43 <xarick> what's cheaper, faster? Tile(t) or t.base()?
20:28:20 *** Wormnest has joined #openttd
20:49:18 <xarick> don't know how to use that
20:50:05 <xarick> I'm in the process of removing redundant code from RiverMakeWider
20:50:43 <xarick> there was a lot of care about water tiles which I believe DC_NO_WATER solves, so some code needs clean up
21:31:55 <xarick> i found a very strange assert
21:34:16 <xarick> oohhhh... I forgot that edge tiles actually flood during map gen
21:34:50 <xarick> makedesert stuff runs a tile loop for a bit, enough to create sea
21:54:34 <peter1138> Hmm, this station thing is annoying.
22:04:55 <xarick> im gonna turn this into a
22:05:24 <xarick> there's more than meets the eye
22:06:36 *** gelignite has quit IRC (Quit: Stay safe!)
22:16:24 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
22:19:07 *** Wormnest has quit IRC (Quit: Leaving)
22:21:28 *** keikoz has quit IRC (Ping timeout: 480 seconds)
22:24:33 <peter1138> Also, where did the limping behaviour come from?
22:26:50 <peter1138> Hmm, 14.1 does it as well, I never noticed before.
22:30:32 <xarick> is that with some newgrf?
22:30:54 <xarick> maybe the rail type? engine?
22:31:33 <xarick> my asserts are awesome!
22:33:28 <xarick> try a clean openttd.cfg
22:38:55 <xarick> previously it was sacrificing other non inclined river tiles
22:39:30 <xarick> and it's not sea here, i made sure there was no sea on the map, only rivers
22:51:16 <xarick> wants to raise East Corner of 547937
22:51:24 <xarick> that would touch water
22:51:31 <xarick> terraform doesn't touch water
22:53:14 <xarick> previous code would make this look more elegant indeed
22:53:40 <xarick> the river would have 2 inclined slopes
22:53:53 <xarick> and that dead end tile would disappear
22:55:47 <xarick> 547937 would likely also be turned into a river tile once flat
22:56:08 <xarick> hmm... disappointing what my code do
22:58:14 <xarick> wish there was a way to forward forbidden tiles to terraform command
23:10:28 <_glx_> xarick: you always open PR way too early
continue to next day ⏵
