IRC logs for #openttd on OFTC at 2022-11-05
⏴ go to previous day
00:14:55 *** sla_ro|master has quit IRC ()
00:31:48 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
00:31:52 *** WormnestAndroid has joined #openttd
01:46:11 <reldred> just merge it, it's perfect, perfect little cinnamon bun of a feature
02:12:01 *** Wormnest has quit IRC (Quit: Leaving)
03:34:35 <reldred> there we go, working parameterized urban ground underlay, for the japanese maglev tracks at least.
03:34:52 <reldred> I still wanna do something aboiut the fences though
03:35:04 <reldred> the fences in the south korean trackset are absolutely sublime
04:01:06 *** D-HUND has quit IRC (Ping timeout: 480 seconds)
04:10:03 *** wallabra_ has joined #openttd
04:12:33 *** wallabra has quit IRC (Ping timeout: 480 seconds)
04:12:33 *** wallabra_ is now known as wallabra
04:58:03 <reldred> now will I ever play a single game of openttd long enough to use them? Probably not.
05:51:53 *** Tirili has quit IRC (Quit: Leaving)
06:38:56 *** HerzogDeXtEr has joined #openttd
06:56:12 *** wallabra_ has joined #openttd
06:59:22 *** wallabra has quit IRC (Ping timeout: 480 seconds)
06:59:22 *** wallabra_ is now known as wallabra
07:26:46 *** sla_ro|master has joined #openttd
07:38:26 <Eddi|zuHause> you know this feeling when something breaks but you can't imagine how it could even break this way?
07:39:04 <Eddi|zuHause> my kde taskbar broke in that it overlaps the widgets over each other
10:50:14 <frosch> "Hi, I'm wondering where I should go, or who I should contact for help/assistance with updating a Dell motherboard BIOS. I get the message: "opensfx err: signature sfx mismatched" when I try to run the file through a MS-DOS USB created by Rufus."
10:53:15 <Rubidium> amazing... making OpenSFX executable by pure luck I guess. I didn't account for it to be executable ;)
12:25:19 *** tokai|noir has joined #openttd
12:25:19 *** ChanServ sets mode: +v tokai|noir
12:32:04 *** tokai has quit IRC (Ping timeout: 480 seconds)
12:57:19 <glx[d]> IIRC a flag exists to allow different cost in test and exec
13:24:06 *** Flygon_ has joined #openttd
13:27:16 *** Flygon__ has joined #openttd
13:31:27 *** Flygon has quit IRC (Ping timeout: 480 seconds)
13:32:11 *** Flygon_ has quit IRC (Ping timeout: 480 seconds)
13:53:07 <petern> > function was explicitly deleted
13:53:07 <petern> No idea that was a thing 😄
13:56:56 <glx[d]> And of course mingw is also slow to fail
14:09:48 <glx[d]> Every target failed as expected
15:57:53 *** Wormnest has joined #openttd
16:14:16 <dP> glx[d]: Any ideas what to do with it?
16:16:09 <dP> michi_cc[d]: maybe you can take a look?
16:38:40 <TallTyler> Two bugs created, two bugs squashed 🙂
16:40:15 <TallTyler> Need to uphold my pinky-promise to michi_cc[d] 😛
16:46:11 <michi_cc[d]> dP: I don't see anything obviously wrong in the code, and as it is only this single compiler that's failing, there might not be anything wrong.
16:46:48 <michi_cc[d]> Don't have an 32-bit mingw environment setup to test any further right now.
16:48:15 <michi_cc[d]> Well, I don't have a 64-bit environment setup either 🙂
16:50:39 <Merni> are you running 16 bit dos :p
17:04:05 *** Flygon__ has quit IRC (Quit: A toaster's basically a soldering iron designed to toast bread)
17:54:07 <petern> 10134 reads as "I don't care what NewGRFs/GS want to do"
18:00:21 <TallTyler> To be fair, I tend to disagree slightly with the "if it can be done in NewGRF it has no place in the base game" sentiment. That would require everyone to roll their own NewGRF for every combination of features they want, which I don't think is realistic.
18:02:05 <JGR> A large subset of players simply do not use GRFs or GSs
18:02:13 <JGR> Presumably the author is one of those
18:04:45 <frosch> TallTyler: "every combination of features" is unrealistic in any case, no matter whether built-in or as add-on
18:05:17 <frosch> the advantage of the add-on is that you can implement a subset that fits with itself, and does not need to address every combination
18:06:20 <TallTyler> For sure, I get the practicality of outsourcing everything to NewGRF, but I don't think that answer is particularly helpful to players
18:07:03 <frosch> sure, but it's the only way that works
18:07:19 <frosch> in closed-source games you have a game designer who designs everything
18:07:31 <frosch> in open-source games you have N people wanting N+1 things
18:07:46 <JGR> Pushing too much stuff to NewGRFs results in nonsense like that fellow who uploaded 64 near-identical config GRFs
18:07:55 <frosch> which will never work, so in the end you need a dictator to make the decision
18:08:25 <frosch> and i vote for making CS the dictator who created a divine game, and everything else has to use add-ons
18:09:52 <frosch> JGR: how could anything be sane, which tries to make 64 cargos adjustable?
18:10:17 <frosch> the problem is what it tries to do, not that the method of newgrf is unsuitable
18:25:00 *** virtualrandomnumber has joined #openttd
18:25:15 *** virtualrandomnumber has quit IRC ()
18:29:17 <TrueBrain> factorio did learn that lesson from us, I guess 😄
18:36:36 <frosch> TrueBrain: anything in particular they did learn?
18:36:50 <TrueBrain> base game is opinionated, and everything else is modifiable via lua
18:36:59 <frosch> i always only see them doing the same mistakes as ottd, like different length scales in differen screen directions
18:38:34 <TrueBrain> glass half full vs half empty
18:39:27 <frosch> does anyone want to be the bad cop and close 10087 or 10134 or team#341 ? or does it always have to be me? :p
18:39:53 <TrueBrain> not having the best mood there are we frosch? 😄 You sound like marvin 😄
18:41:19 <frosch> i read the books 3 years ago, i already forgot most of it :p
18:42:31 <michi_cc[d]> I'll take one for the team 🙂
18:42:48 <petern> Can GS create industries?
18:45:34 <frosch> even when disabled for players (raw construction/prospecting)
18:45:52 <petern> I would assume that, GS are not players 🙂
18:46:31 <frosch> well, gs cannot overrule newgrf slope / in-town / whatever checks
18:46:46 <frosch> so if you use ECS, I guess GS can build industries just as little as players can
18:48:18 <petern> Well that's ECS' fault 🙂
18:49:23 <petern> And 10087 is redundant with 10001
18:50:08 <petern> Even if dP's screenshots look terrible with those massive weird icons 😄
18:50:22 *** Wormnest has quit IRC (Ping timeout: 480 seconds)
18:50:53 <dP> Btw, I found out GS can't always refund the industry in the same place. Because players can dig under industries but industries require that land.
18:52:05 <TallTyler> Re-found? After closing?
18:52:29 <petern> Oh. I wouldn't expect that anyway.
18:52:42 <petern> The industry closed, why should it be recreated in the same spot?
18:53:18 <TallTyler> There's a GS that does that, trying to fix the "oh no the industry closed as I was about to serve it" that some people are strangely incensed by
18:53:21 <bigyihsuan> for the scenario editor folks in the house, it possible to generate fewer random industries? i set the setting for industry density to "minimal" and clicking the "create random industries" button gives the same industry density as the other settings
18:53:50 <petern> Because of industry creation rules 😄
18:56:08 <dP> Idk, just does seem logical to me when it's not allowed to be built on the same spot
18:56:47 <dP> I could understand if that was because of town zones or smth but this is just landscape check
18:58:17 <dP> petern: How is it even related to 10001?
18:58:38 <JGR> NewGRF industries are allowed to be arbitrarily precocious, it's not really about logic
18:59:09 <petern> Could even be a date check or something...
18:59:37 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
18:59:49 *** WormnestAndroid has joined #openttd
18:59:55 <petern> dP: Oh you're right, I was thinking the built-in league table rather than station rating.
19:00:26 <andythenorth> Hmm such industry stuff
19:00:55 <andythenorth> So reddit can report FIRS broken even more
19:00:55 <petern> My docks go the wrong way 😄
19:00:59 <dP> JGR: Yeah but that was vanilla industries
19:01:10 <andythenorth> Maybe it’s the right way
19:01:50 <JGR> A lot of the problems with vanilla industries are solved by using a NewGRF industry set with the desired behaviour instead
19:03:34 <dP> JGR: So by not using them 🤣
19:04:11 <petern> Hmm, saveload works but I'm not sure how. Fun.
19:04:23 <JGR> You can re-use the vanilla industry sprites and tile layouts and then add whatever logic you want
19:04:27 <petern> I bumped the version then unbumped it.
19:06:12 <petern> I wonder if I should use NewStations-style ID manager, or if the OverrideManager is a better system.
19:16:41 <frosch> as we know newgrf, there will be thousands of near identical things, 256 docktypes is never enough
19:17:09 <frosch> overridemanager works for things which have a poolitem, since you can store the specid in the poolitem
19:17:25 <frosch> newstation stuff does not have the luxury of a poolitem for every stationpart
19:17:44 <frosch> so it creates a local subset of up to 256 parts, which can then be stored in the map array
19:18:38 <petern> Yes, looks like I stored the ID as the tile's StationGfx , which is normally 0-5 for docks, but with this it extends it to 255. Which as you say, isn't enough.
19:19:38 <petern> NewStations doesn't really extend the StationGfx part, and uses CustomStationSpecIndex instead.
19:19:45 <JGR> For road stops I added another map to larger ID mapping array, similar to with rail stations
19:19:47 <petern> I should know this as I wrote that.
19:20:07 <petern> JGR: yes I think I saw that. roadstopspecindex
19:20:14 <JGR> If I get the time I'll see if I can port it upstream, it should just be a case of removing stuff
19:20:31 <petern> And infact I did write that for docks originally, then trashed it to try out the OverrrideManager system.
19:21:41 <petern> For docks I also store 4 bits of adjency, so that determining multi-tile parts can be done by iterating the map.
19:25:38 *** Wormnest has joined #openttd
19:27:19 <petern> frosch: I guess a combo approach could use the overridemanager alongside the spec index array, instead of storing grfid and localidx in the *SpecList.
19:28:10 <petern> Would probably have done that if the override manager existed when I created newstations.
19:28:45 <petern> Funny though, that system was designed to behave nicely when NewGRFs were modified/removed/added again, and then we later said Nope, you can't do that any more.
19:48:34 <petern> Is it me or is `DrawRailTileSeq` actually used for all station tiles, not just rail?
19:51:45 *** THEONEPIECEISREAL has joined #openttd
19:51:45 <THEONEPIECEISREAL> petern: Wait what is this add-on for these docks
19:57:25 <THEONEPIECEISREAL> it will be modulable docks plugin?
20:00:54 <THEONEPIECEISREAL> Too much Theo Town so i have told plugin
20:01:01 <THEONEPIECEISREAL> anyways add-on
20:01:57 <petern> I wouldn't get any hopes up, this code has been lying around for 3.5 years!
20:03:57 <THEONEPIECEISREAL> this is pretty old
20:07:37 <petern> THEONEPIECEISREAL: just like me
20:08:43 *** wallabra has joined #openttd
20:09:13 <petern> LordAro, fun today, my rear brake failed (fluid related I think), and I had some descents. Never taken them so slow before!
20:16:57 *** wallabra has quit IRC (Ping timeout: 483 seconds)
20:17:52 *** Wormnest has quit IRC (Ping timeout: 483 seconds)
20:23:09 *** wallabra has joined #openttd
20:29:15 *** wallabra_ has joined #openttd
20:35:07 *** wallabra has quit IRC (Ping timeout: 480 seconds)
20:35:07 *** wallabra_ is now known as wallabra
20:44:11 *** m3henry has joined #openttd
20:44:15 <frosch> petern: it differs from the other Draw*TileSeq by supporting that weird railtype-to-spriteoffset thing. but looks like non-rail stations just set that to zero 🙂
20:44:54 <m3henry> I assume I need to rebase before approval, but not sure if that actually is the case
20:45:39 <petern> Yup, different stations for different railtypes.
20:46:33 *** wallabra_ has joined #openttd
20:47:55 <glx[d]> m3henry: rebase will discard any approval 😉
20:48:57 *** wallabra has quit IRC (Ping timeout: 480 seconds)
20:48:57 *** wallabra_ is now known as wallabra
20:49:26 <m3henry> Ok, so then I'm doing the right thing then
20:49:53 <frosch> if only I knew what nerd added that "SI" units system to ottd
20:50:02 <frosch> i can understand using kW instead of hp
20:50:23 <frosch> I guess "Mg" instead of "t" fits though
20:50:56 <m3henry> I'm still salty that kg is a base unit
20:51:18 <glx[d]> took me too much time to get what "Mg" was
20:51:58 <Eddi|zuHause> Magnesium obviously :p
20:54:25 <m3henry> (svn r4126) peter1138 - Add SI as option for measuring units.
20:55:48 *** Wormnest has joined #openttd
21:02:34 <frosch> ah by an englishman. probably to mock the SI system then 🙂
21:28:09 <TallTyler> m3henry: Is it ready for review and approval? I’ve been waiting for you to finish updating before taking another look.
21:35:19 <petern> How do you rename a variable and maintain compability with the "new" saveload system...
21:45:15 <glx[d]> objdump -t snippet for #10001
21:45:15 <glx[d]> [85159](sec -32108)(fl 0x00)(ty 0)(scl 2) (nx 0) 0x0000000000000000 .refptr._Z19CmdClearOrderBackup13DoCommandFlag9TileIndex8ClientID
21:45:15 <glx[d]> objdump -t snippet for master
21:45:15 <glx[d]> [80970](sec 31671)(fl 0x00)(ty 0)(scl 2) (nx 0) 0x0000000000000000 .refptr._Z19CmdClearOrderBackup13DoCommandFlag9TileIndex8ClientID
21:45:50 <glx[d]> I see a lot of negative section in #10001 dump, I'd blame the compiler
21:47:11 *** wallabra has joined #openttd
21:48:32 *** Wormnest has quit IRC (Ping timeout: 480 seconds)
21:49:11 <glx[d]> and `nm` says rdata section for master, and undefined for #10001
21:49:56 <frosch> does debug/release build make a difference?
21:51:11 <frosch> looks like both i686 and x86_64 use gcc 12.2.0, but only x86_64 fails
21:52:18 <glx[d]> we already have a workaround for mingw 64bit debug builds related to ASLR
21:52:53 <glx[d]> so maybe another mingw gcc "bug"
22:07:55 <m3henry> TallTyler: I believe it's ready
22:17:46 *** wallabra has joined #openttd
22:25:58 *** Wormnest has joined #openttd
22:38:48 *** HerzogDeXtEr has quit IRC (Read error: Connection reset by peer)
22:44:58 *** wallabra has joined #openttd
22:53:40 *** supermop_Home has joined #openttd
23:07:19 <glx[d]> ok it seems I found a solution
23:19:12 *** Wormnest has quit IRC (Ping timeout: 480 seconds)
23:36:14 *** wallabra has joined #openttd
23:40:37 *** bryjen has quit IRC (Quit: Leaving)
23:40:41 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
23:56:10 *** Wormnest has joined #openttd
23:56:20 <TallTyler> m3henry: We’ll have to wait for another developer to merge it. You wouldn’t have the rights to do that (and I’m not ready to approve and merge on my own).
continue to next day ⏵
