IRC logs for #openttd on OFTC at 2024-12-23
⏴ go to previous day
02:46:14 *** Markk has quit IRC (Ping timeout: 480 seconds)
02:49:45 *** reldred has quit IRC (Quit: User went offline on Discord a while ago)
02:53:08 *** Tirili has quit IRC (Remote host closed the connection)
03:16:41 *** gnu_jj_ has quit IRC (Ping timeout: 480 seconds)
03:34:50 *** Wormnest has quit IRC (Ping timeout: 480 seconds)
03:37:11 *** Wormnest has joined #openttd
03:37:51 *** debdog has quit IRC (Ping timeout: 480 seconds)
04:14:28 *** Tirili has quit IRC (Remote host closed the connection)
05:46:10 *** Tirili has quit IRC (Quit: Leaving)
05:47:21 *** jinks has quit IRC (Remote host closed the connection)
06:31:32 *** APTX has quit IRC (Ping timeout: 480 seconds)
07:16:00 <DorpsGek> - Add: summary for week 51 of 2024 (by OpenTTD Survey)
08:31:23 <truebrain> Random fact: every minute OpenTTD 14.1 is played for 250 game minutes, on average
08:32:11 <andythenorth> does that give a proxy for count of active player base?
08:32:35 <truebrain> Based on Survey result; actual value is higher ofc
08:33:13 <truebrain> andythenorth: No; active players is far higher..nobody plays 24/7
08:34:06 <johnfranklin> Why is week 50 and week 20 with larger font?
08:35:32 <truebrain> Our peak is 1k players every single day on Steam. Double that to include non-steamies. And that is just the peak, not count
08:36:21 <truebrain> johnfranklin: Ask your browser. As it isn't.
08:40:16 <truebrain> An average session last week for 14.1 was 3 hours. So that would mean roughly 8 times 250 active players (give or take those that start a new game) in a day that have the survey enabled
10:25:46 <peter1138> Hmm, maybe I should add a separate dedicated window for user's badge configuration.
10:55:17 <peter1138> It's a NewGRF town name, but no idea which.
10:55:43 <xarick> game.settings.gui.lost_vehicle_warn this setting is in the wrong category?
10:57:32 <xarick> there's a news_display category, it should be there
10:58:16 <peter1138> But it's not, it's under gui settings.
10:59:46 <xarick> how did that get in there?
11:00:00 <peter1138> That's where it was added, many many years ago.
11:01:56 <xarick> it's not the only one 😐
11:02:50 *** APTX_ has quit IRC (Quit: Farewell)
11:08:13 <xarick> E:\OpenTTD Visual Studio\SamuXarick\OpenTTD\src\industry_cmd.cpp:2385: Backed-up value was not restored!
11:08:29 <xarick> this happens when aborting map gen during world gen
11:09:28 <peter1138> It's basically fine, you're aborting world gen so it does not matter.
11:22:26 *** Flygon has quit IRC (Read error: Connection reset by peer)
11:30:08 <kuhnovic> Time to run lunch gen
11:40:55 <johnfranklin> Why I always link stevenage with stonehenge… they are in different directions toward london…
11:44:10 <belajalilija> i dont think stevenage would be a good place for a pagan pilgrimage spot
12:07:21 <xarick> just deleted 19 stashes
12:11:28 <michi_cc> peter1138: I have some late observations about PR #13082 (animated tile state in map).
12:12:28 <michi_cc> First is probably just a doc thing, but to be deleted animated tiles can be MP_CLEAR, were the bits are listed as free.
12:12:57 <michi_cc> Second things is that you moved some map bits for stations, but I don't see any Afterload() conversion for these?
12:13:42 <peter1138> Second one is fine, they are always 'recalculated' on load.
12:16:07 <michi_cc> Hmm, first thing is probably an acual problem though.
12:16:09 <peter1138> Hmm I wondering if it's possible for any other tile type to be placed on a tile in between it being deleted and the next loop.
12:17:36 <peter1138> Any other non-animateable tile type.
12:18:12 <michi_cc> `DoClearSquare` calls `DeleteAnimatedTile` which sets the state to `deleted`, but the call to `MakeClear` will set m6 to 0. So in practice, at least when clearing, the deleted anim state will never exists.
12:19:32 <michi_cc> It is fine in the sense that `AnimatedTileState::Deleted` is never tested for anyway, but it does mean a superfluous bit.
12:20:20 <peter1138> Well, it is used, because an animated tile can be "deleted" from animation, but not actually cleared.
12:20:57 <peter1138> It could potentially mean a tile is added to the list twice.
12:21:34 <michi_cc> The check in AnimatedAnimatedTiles is `!= AnimatedTileState::Animated` though, not for the deleted state.
12:22:42 <peter1138> It's mainly for AddAnimatedTile.
12:22:57 <peter1138> A tile that is marked deleted is still in the list, so it shouldn't be added to the list again.
12:23:13 <peter1138> But it's not clear because the test is implicit.
12:23:34 <peter1138> `if (state == AnimatedTileState::Deleted) don't add to the list`
12:24:13 <michi_cc> If the state is actually important, `MakeClear` / `DoClearSquare` have to be modified though, because the function will set the complete m6 to 0, so animation state None.
12:25:57 <peter1138> It probably doesn't matter in that case, but I think maybe it makes sense to reserve the bits to allow ground tiles to have the state.
12:26:30 <michi_cc> And deleting a rail station tile for example will do `DoClearSquare` + `MakeRailNormal`, which essentially means animation state has to be global for *all* tile types.
12:26:40 <peter1138> The main purpose of the extra deleted state is when callbacks change the animation state of tiles, rather than actually clearing tiles.
12:28:02 <peter1138> Yeah, that's what I was wondering.
12:28:24 <michi_cc> Are there objects than can be animated and overbuilt by the player? In that case if overbuilding with something animated itself it might result in the tile being twice in the animated list.
12:29:10 <michi_cc> One option could be that `DoClearSquare` really does remove the tile from the list.
12:31:10 <xarick> my 2 methods for searching closest ship depot side by side testing automatic service interval
12:32:37 <xarick> insert at the front probably slowing down
12:32:51 <xarick> must test alternatives
12:32:58 <peter1138> michi_cc: I can try this. The alternative, as the bits are fortunately not used, is to reserve the bits for all tiles.
12:33:24 <peter1138> (And animated ground tiles, and level crossings, animated... trees... are something which sounds nice, even if not currently possibe.)
12:36:16 <peter1138> Hmm, I wonder if the 4096 house type patch made things faster...
12:36:37 <peter1138> (Only has to read 12 bits of one value instead of combining bits from multiple values)
12:52:10 <peter1138> Seems to be costing about 3ms approximately every 15 seconds.
12:52:44 <peter1138> Although I'm doing an expensive delete instead of the lazy-delete. Hmm.
13:08:42 *** APTX has quit IRC (Read error: No route to host)
13:10:57 <peter1138> Lazy-delete reduces it to about 2ms.
13:13:56 <michi_cc> If anim state should become a tile global, `DoClearSquare` would probably have to check the proper animation state instead of unconditionally setting the state to `deleted` for any potentially animated tile. Or something would have to clear out the state from un-animated map tiles.
13:15:29 <peter1138> Well, it's already only set to deleted if it's currently animated.
13:16:23 <peter1138> DeleteAnimatedTile itself does that check.
13:17:55 <michi_cc> Eh, yes, didn't read far enough 🙂
13:24:20 <peter1138> Animated houses can be cleared and replaced with other houses in the same tick, so yeah.
13:54:26 <peter1138> And even that was ~18 months ago...
14:16:06 <peter1138> Oh, there's a firmware update for one of my synths that adds another voice.
14:35:11 *** APTX has quit IRC (Quit: Farewell)
14:40:23 *** murr4y has quit IRC (Ping timeout: 480 seconds)
14:59:35 *** asasnat has quit IRC (Quit: User went offline on Discord a while ago)
15:17:56 <xarick> michi_cc: #13056 existed 😦
15:23:24 <xarick> gonna try to salvage it
15:41:46 <xarick> gonna drop the changes about the water tiles, or at least make its own PR
15:45:32 <talltyler> peter1138: Interesting read, thanks for posting.
16:25:35 *** D-HUND is now known as debdog
17:11:28 <xarick> just noticed comments still mentioning m_open_node m_node and 'm_' alike
17:33:49 <michi_cc> talltyler: "Haven't looked at all the Fix things" 🤣
18:24:11 <peter1138> Do I reserve space for animated tile bits, or do manual removal...
18:46:45 *** tokai|noir has joined #openttd
18:46:46 *** ChanServ sets mode: +v tokai|noir
18:52:02 <johnfranklin> I found an issue, it may be some bots uploading some old GRFs to bananas
18:53:46 *** tokai has quit IRC (Ping timeout: 480 seconds)
18:53:59 <LordAro> i'd be amazed if there were any "bots" uploading anything
19:03:44 <johnfranklin> I don't know, but something is strange
19:12:48 <LordAro> "something is strange" is vague and unhelpful
19:15:43 *** gelignite has joined #openttd
19:37:55 <truebrain> Yeah, it is not getting any more clear
19:41:03 <michi_cc> Well, if it is indeed taken from the linked website, MonsterhunterIndonesia is almost certainly not the author.
19:41:21 <michi_cc> The site just seems to be an old NewGRF collection.
19:41:53 <truebrain> seems little related to "bots" or "strangeness"? But okay, that happens if someone talks in vague terms. It only makes things more vague 🙂
19:50:13 <truebrain> no worries; but please speak plainly 🙂
20:33:18 <_glx_> oh and it's one of the problematic grf
20:41:46 <peter1138> I think what johnfranklin was trying to say is "I think someone has uploaded someone else's NewGRF to bananas without permission, here <...> is the link"
20:42:20 <peter1138> But being vague and imprecise is the way?
20:43:58 <andythenorth> I would suspect, on the face of it, that ones’s a ToS violation
20:44:35 <andythenorth> That project is also one that tends to attract drama 😐
21:21:27 <xarick> I probably have a better solution to both of those solutions
21:23:26 <xarick> I'm in the process of reverting part of #10348 which is already merged, but I need to verify if it in fact works better than #8492
21:36:04 <Rubidium> ooh... a comment that's already outdated when it was posted
21:38:35 *** Wormnest has joined #openttd
21:51:53 <xarick> what does [[unlikely]] do?
22:23:04 <_glx_> search on cppreference ?
22:34:06 <Rubidium> _glx_: you forgot to add [[unlikely]] to that statement ;)
22:35:51 <xarick> where in town_cmd.cpp do I find the code that makes them expand?
22:49:31 <xarick> load town data, what is this? how
22:55:42 <xarick> what file extension does it look for?
23:08:43 <xarick> ah, TownTickHandler, found it!
23:09:08 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
23:15:49 *** nielsm has quit IRC (Ping timeout: 480 seconds)
23:21:36 <xarick> can't use this fix in the scenario editor
23:22:21 <xarick> generating towns in the scenario editor is... complicated
23:23:22 <xarick> town growth terraforming tiles at sea level when the town is built at level 0, could initiate a flood, I need to take a better look at this
23:23:53 <xarick> i rather just not do it for towns 😦
23:32:02 *** Wormnest has quit IRC (Quit: Leaving)
23:46:33 *** keikoz has quit IRC (Ping timeout: 480 seconds)
continue to next day ⏵