IRC logs for #openttd on OFTC at 2025-04-17
⏴ go to previous day
01:06:10 *** emperorjake has joined #openttd
01:06:10 <emperorjake> This is a great feature, I've been wanting something like this for a long time 🙂
01:24:00 *** Wormnest has quit IRC (Quit: Leaving)
02:10:52 *** gnu_jj_ has joined #openttd
02:14:10 *** gnu_jj has quit IRC (Ping timeout: 480 seconds)
03:04:29 *** belajalilija has joined #openttd
03:04:29 <belajalilija> tabytacviaGitHub: 👀
03:12:42 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
03:12:43 *** WormnestAndroid has joined #openttd
03:35:36 *** Smedles has joined #openttd
04:45:33 <DorpsGek> - Update: Translations from eints (by translators)
06:01:53 *** akimoto has joined #openttd
06:12:28 *** k-man has quit IRC (Ping timeout: 480 seconds)
08:39:07 *** Extrems has joined #openttd
08:43:45 *** nielsm has quit IRC (Ping timeout: 480 seconds)
09:20:34 <peter1138[d]> > 25 changed files with 1023 additions and 7 deletions
09:20:39 <peter1138[d]> Sadface at work stuff.
10:31:47 <peter1138[d]> What if this worked?
10:32:46 <peter1138[d]> More simply, I suppose :p
10:34:00 <reldred> *has a seizure whilest ranting about PR8480*
10:34:10 <reldred> But yes more simply would suffice
10:34:23 <reldred> Drive through depots are basically mandatory for me in jgrpp
10:34:58 <peter1138[d]> I don't know how it works in JGRPP, I saw something about two depots back to back?
10:35:06 <reldred> Yeah needs min two back to back
10:35:29 <reldred> That said, I do like my depots to be as long as my trains
10:35:43 <peter1138[d]> Basically, if this worked, and depots could occupy multiple tiles, then "long depots" "just work". Or something.
10:35:43 <reldred> there's no enforcement of that in jgrpp though
10:36:04 <reldred> rather lovely actually
10:36:13 <peter1138[d]> But also, wouldn't need to all be done at once.
10:37:38 <reldred> so, 8480 did get some work to split itself out into multiple PR's, there's another PR that did some simplification, #9577
10:38:37 <reldred> but even that was a rather crunchy looking PR
10:38:52 <peter1138[d]> Yeah, but both times they've been long branches./
10:39:05 <peter1138[d]> And the second didn't refer to the first.
10:39:17 <reldred> Yeah just mentioning them incase there's anything useful in them
10:39:47 <peter1138[d]> Yes, there probably is, and I am aware of those.
10:40:34 <reldred> I’ve been pining for it for a while
10:40:52 <peter1138[d]> I was looking through "some source code" to see what could be backported.
10:41:06 <peter1138[d]> And well... nothing much with ease.
10:41:09 <reldred> I did quite enjoy my test games on it despite missing the rest of the jgrpisms I’m attached to.
10:42:04 <reldred> Regardless of the rest of the gameplay that 8480 added, just allowing them to be traversed from the backside and also connect into a single depot would make doing orders a lot easier
10:42:28 <reldred> Currently I have to do a depot order in one direction pointing at the entrance and one from the other side at the exit.
10:43:27 <reldred> The full 8480 experience of needing to build depots big enough for everything to *fit* is quite nice but I’d be happy to wait that one out.
10:44:32 <peter1138[d]> IMHO if you have these single-drive-through depots, then it is simpler to extend to long-drive-through depots, and then it's "just" a case of requiring a long depot to be long enough.
10:44:47 <peter1138[d]> That's my thought process on the steps involved :p
10:44:54 <reldred> that approach seems reasonable
10:45:01 <peter1138[d]> Without doing anything like actual coding.
10:45:47 <peter1138[d]> Other tweaks from those PRs, like making depot windows refer to the depot id instead of tile are basically things that can be done anyway.
10:47:14 <peter1138[d]> <obligatory-dad-joke>And that shalot</obligatory-data-joke>
10:48:05 <peter1138[d]> Thanks Ants. Thants!
10:48:49 <reldred> government spy drones, obviously
10:49:15 <peter1138[d]> I added my co-workers repository fork (yeah, I followed that workflow even for work stuff) as an origin, so I can just fetch it and rewrite his work without having to wait for understanding :S
10:49:19 <peter1138[d]> Hate it, but gets things done quicker.
10:50:44 <peter1138[d]> I'm not sure I actually like the disjoint depot behaviour of #9577.
10:51:12 <peter1138[d]> But on the other hand, long depots, and adjacent long depots, makes complete sense to me.
10:51:19 <peter1138[d]> Stupid realism getting in the way.
10:51:25 <reldred> I think it should be treated like stations, there's an option to enable/disable.
10:51:37 <reldred> to prevent walking and other exploits
10:52:04 <reldred> it does from a certain standpoint make more aspects of rail building more 'consistent' having depots behave a bit more like stations
10:52:32 <peter1138[d]> Hmm, with my drive-through idea, then it's workable: Vehicles can only exit on the tile they entered.
10:52:59 <reldred> a train using disjointed depots to teleport would be a potential short term exploit before the full drive through logic is implemented
10:53:22 <reldred> but then magic depots are an exploit as it is
10:53:23 <peter1138[d]> This "breaks" the design of the depot in the PR's picture, but makes a simpler 3-adjacent-depots system work nicely?
10:54:09 <reldred> 8480 does force drive through like you're describing, 9577 does technically allow it as an exploit
10:55:03 <peter1138[d]> I'm not sure if it that works for a single tile though.
10:55:32 <peter1138[d]> Whatever, I have badge filters to finish.
10:58:09 <reldred> starting to get a bit fallout 4 'who the fuck is shaun' esque
11:08:10 *** nielsm has quit IRC (Ping timeout: 480 seconds)
11:46:08 <_zephyris> A full implementation...
11:47:03 <_glx_> A lot less aggressive indeed
12:01:31 <_zephyris> Well, I did use a question mark 😉
12:01:40 <_zephyris> It's done and coded as an alternate style
12:02:09 <_zephyris> Easy to change the default
12:04:18 <peter1138[d]> Easier-on-the-eyes should be default, if you do it as an option.
12:04:32 <peter1138[d]> I'm not sure there there benefit to having an option here though.
12:04:44 <kuhnovic> I agree, just make this the default
12:09:28 <andythenorth> _zephyris: pretty epic
12:09:34 <andythenorth> actually tempted to switch
12:09:39 <andythenorth> Horse sprites will **pop**
12:14:11 <_zephyris> Aww, looks like your candyfloss is grey 😦
12:14:32 <_zephyris> Don't think I ever thought I'd see such an outpouring of positivity for toyland!
12:18:48 <andythenorth> _zephyris: nice eh 🙂
12:26:23 <_zephyris> There's bugs to fix, but looks like this will be the new default.
12:32:42 <talltyler> Title game voting opens next week (the 22nd) so if you get it done before then maybe Chris’ Toyland entry has a shot at winning 😛
13:00:53 <_zephyris> andythenorth: Yes please 😉
13:01:07 <_zephyris> Good thing with fictitious industries is you can do whatever you want!
13:06:45 <talltyler> Sounds like a UI nightmare to see all waypoints on the map
13:08:46 * pickpacket can't take part in the title game contest because he never plays without NewGRFs
13:08:52 <pickpacket> this is a me problem
13:26:28 <talltyler> You play a game specifically to build an entry, the map size is quite limited 🙂
13:27:15 <talltyler> (Although a perfectly reasonable size to play normally, you want to avoid building close to the edges in case people have high-dpi monitors and don’t adjust interface scaling)
14:00:00 <_glx_> Yeah 64² is clearly too small
14:03:36 <andythenorth> redstone via waypoint
14:17:19 *** akimoto has quit IRC (Remote host closed the connection)
16:05:11 *** Wormnest has joined #openttd
16:35:35 *** Smedles has joined #openttd
17:11:13 *** kuka_lie has joined #openttd
17:17:46 *** zanooda2000 has joined #openttd
17:17:46 <zanooda2000> _zephyris: eh, is it final design of trees? old-school unrealistic were way more stylish
17:17:46 <zanooda2000> still, colored are also cool
17:24:34 <zanooda2000> peter1138[d]: will not hurt probably
17:26:15 <zanooda2000> _zephyris: "default" is non-randomized previous version? I can suggest making the current one as default and leave original-like as "toyland legacy (epilepsy warning)"
17:28:12 <zanooda2000> _zephyris: (also would like to see possibility to use toyland grass in other climates, for _some reason_ it is not available now)
17:56:51 <andythenorth> oof silly way to write python
17:56:51 <andythenorth> `(print(self.formation_ruleset)) if self.formation_ruleset is not None else None`
18:22:54 *** gelignite has joined #openttd
18:24:42 <_zephyris> zanooda2000: Just happened to have the alternate (realistic but colourful) trees on. The original style ones are the default.
18:27:40 <_zephyris> zanooda2000: The problem is that toyland has different road/rail/etc, so you end up needing big extra sprite sets (normal infra on toyland grass) for an extremely niche use case.
18:43:07 *** tokai|noir has joined #openttd
18:43:07 *** ChanServ sets mode: +v tokai|noir
18:50:00 *** tokai has quit IRC (Ping timeout: 480 seconds)
19:05:55 <peter1138[d]> PR title styles eh? Hmm
19:24:41 <andythenorth> was it lunch though?
19:25:28 <peter1138[d]> Is it cheese o'clock?
19:27:18 <andythenorth> I am at the "much regret" stage of refactoring
19:27:36 <andythenorth> I cut all the string out in favour of a more rational design
19:27:51 <andythenorth> but now some of the remaining bugs require putting back string
19:54:35 <Rubidium> the commit checker is configured to just get the last 256 revision, and it tries to find the master commit the branch is based off. However, that's more than 256 revisions ago so it can't find them. So in short: just rebase the branch
19:55:57 <peter1138[d]> Beat me to it. It's out of date 🙂
20:03:10 <_glx_> and 256 revision is a lot 🙂
20:15:15 *** nielsm has quit IRC (Ping timeout: 480 seconds)
20:35:27 *** D-HUND has quit IRC (Ping timeout: 480 seconds)
21:05:37 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
21:05:58 *** WormnestAndroid has joined #openttd
21:21:36 <zanooda2000> _zephyris: Hmm, why then not to go by the easiest route and just use that toyland rails/roads/etc included? I think it will be even more... atmospheric. If a player _wants_ toyland everywhere, this certainly won't stop them
21:30:40 <zanooda2000> _zephyris: *yeah, about toyland infrastructure. Will «christmas candy» monorail be added?*
21:31:12 *** keikoz has quit IRC (Ping timeout: 480 seconds)
21:32:54 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
21:53:10 *** kuka_lie has quit IRC (Quit: Lost terminal)
21:59:14 *** Flygon has quit IRC (Read error: Connection reset by peer)
21:59:51 <andythenorth> getting the right badges in the right place is breaking my brain 😛
22:00:32 <andythenorth> is so much better than checking specific IDs
22:01:26 <andythenorth> makes end cars and the restaurant car appear
22:03:13 <andythenorth> my procedure for this used to be unrolled to approx 4096 switches, then whatever nmlc does to expand advanced varact2
22:03:21 <andythenorth> only one procedure per grf but eh
22:03:22 <_zephyris> zanooda2000: It's on my want list, feel free to add an issue on the repo to remind me.
22:03:49 <andythenorth> and I couldn't reasonably check more than 16 vehicle IDs, and I ran out 😛
22:04:08 <andythenorth> now it's ~256 switches, one badge, not 16 IDs
22:34:46 <andythenorth> num_vehs_in_badge_predicate_chain as a built in? 😛
22:34:51 <andythenorth> akin to num_vehs_in_vehid_chain
22:35:02 <andythenorth> maybe put the predicate in a register or something
continue to next day ⏵
