IRC logs for #openttd on OFTC at 2024-11-30
⏴ go to previous day
00:30:08 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
00:37:11 *** de_schmitt has joined #openttd
00:37:12 <de_schmitt> Mail trucks cannot share order list with buses. That is very sad. And I see no reason why this is the case for situations where there are only stations which have bus stops and truck stops.
02:32:43 *** tokai|noir has joined #openttd
02:32:44 *** ChanServ sets mode: +v tokai|noir
02:39:29 *** tokai has quit IRC (Ping timeout: 480 seconds)
03:02:15 *** gadg8eer has joined #openttd
03:03:15 <gadg8eer> Even though they use roads, busses and trucks have very different unloading requirements and required it in the original game (TTD).
03:05:42 <gadg8eer> Also, you know that you can build road stations with both bus and truck bays, right? Lastly, iirc the only NewGRFs that restricted whether a road stop would be passenger or freight have a parameter to allow every stop to be used for both. If you're not using roadtypes, then just build a bus station and truck station next to each other.
03:12:16 <wensimehrp> gadg8eer: I don't think there are such GRFs
03:12:59 *** gnu_jj_ has joined #openttd
03:15:14 <gadg8eer> Hold on, I can prove there is but this laptop isn't cooperating...
03:16:06 *** gnu_jj has quit IRC (Ping timeout: 480 seconds)
03:16:28 <gadg8eer> wensimehrp: There. That's the one.
03:24:00 <wensimehrp> gadg8eer: I thought you mean there are newgrfs that restrict merging lorries and bus stops
03:24:20 <wensimehrp> gadg8eer: That certainly not the case 😐
03:37:18 <gadg8eer> wensimehrp: Fair enough. As far as I remember, the pictured GRF can provide bus and truck stops as truck stops, and all others don't usually make a distinction either.
03:38:57 <gadg8eer> But it is a bit confusing, I know. It's realistic, but restrictive. Normally that's a "bad feature" but this one came from Chris Sawyer's original version itself so it was grandfathered in.
03:58:55 *** godbed_ has joined #openttd
04:02:19 *** debdog has quit IRC (Ping timeout: 480 seconds)
04:02:24 *** D-HUND has quit IRC (Ping timeout: 480 seconds)
04:10:16 *** godbed is now known as debdog
04:46:25 <DorpsGek> - Update: Translations from eints (by translators)
06:35:24 *** gwyd4016 has joined #openttd
06:35:24 <gwyd4016> gadg8eer: You can see from the station signs that most stations have both a bus and truck stop
06:35:55 <gwyd4016> de_schmitt: I'd say most likely you've just missed a truck stop from the bus stops, I thought you could do this
07:00:31 *** Smedles has joined #openttd
07:15:24 *** emperorjake has joined #openttd
07:15:24 <emperorjake> You can copy orders but not share them
08:31:38 <de_schmitt> emperorjake: Exactly. This is what I currently do. But this way the buses and trucks dont share orders and thus I always have to handle them separately. For all cities. I have no problem with building bus stop+truck stop for each stop. But the "not share feature" is really annoying.
09:36:40 *** Flygon has quit IRC (Read error: Connection reset by peer)
09:37:39 <mnhebi> You could always just write a ChatGPT station naming patch yknow.
10:36:02 *** kuka_lie has joined #openttd
11:08:20 <xarick> I accidentally `if (current_length != 1 and other_length != 1) {`
11:13:42 <johnfranklin> I tasted some blue cheese for the first time.
11:13:42 <johnfranklin> It was just a little “strange smell”, but more delicious than I thought.
11:32:25 <mnhebi> Try blue cheese on kebab.
11:32:43 <mnhebi> works great on hamburgers and stuff
11:34:29 <mnhebi> haven't tried blue cheese lasagna cause I cannot be bothered to go to the lengths it takes to make lasagna..but I bet it would be great
11:43:34 *** kuka_lie has quit IRC (Ping timeout: 480 seconds)
11:51:39 <xarick> copilot understands my shenanigans, so awesome
12:05:27 <peter1138> Overlapping widgets, which order should be used, first on top, or last on top?
12:06:31 <peter1138> Although this is just a detail of the definition, rather than using it.
12:14:46 <mnhebi> makes it easier to arrange
12:15:01 <mnhebi> first on top is just an exercise in frustration
12:20:06 *** gelignite has joined #openttd
12:39:23 <peter1138> I'll need to shuffle widgets around then. The main thing is that click order is opposite to draw order.
12:42:57 <peter1138> Top to bottom I can just reverse drawing. Bottom to top I have to reversing clicking, and reorder the widgets. Hmm.
12:50:12 <peter1138> "reorder the widgets" meaning actually moving them in the nwidgetpart source code. Not std::sort 🙂
12:55:40 <mnhebi> ah yes, std::sort the source of all solutions.
12:56:10 <peter1138> Hmm, But I'm already shuffling widgets around, so it's no problem.
13:05:52 <mnhebi> what about std::shuffle tho have you thought of that ehhh
13:06:11 <peter1138> On source code? Hmm.
13:18:33 <kuhnovic> Is there a way to set string colors via the SetDParam etc commands? It's for debugging purposes, it doesn't have to be pretty
13:21:04 <peter1138> No, but you can give it as a parameter to DrawString.
13:21:50 <peter1138> Hmm, actually there is {COLOUR}
13:22:04 <peter1138> But of course your string needs to start with {COLOUR}.
13:23:37 <kuhnovic> I already created a new string so that's ok, I can add it. I will check that out, thanks!
13:30:12 <xarick> how do i use std::swap on a range of values
13:30:54 <xarick> std::ranges::swap does not exist, I am disappointed
13:31:41 *** kuka_lie has joined #openttd
13:33:35 <xarick> oh this kinda works std::swap(sloped_rivers[0], sloped_rivers[1]);
13:34:48 *** kuka_lie has joined #openttd
13:36:05 <_glx_> peter1138: it can start with any colour code, and you can also use {PUSH_COLOUR}{COLOR}{POP_COLOUR} to change then restore to original
13:36:51 *** kuka_lie has joined #openttd
13:37:18 <peter1138> Okay, I should've written "contain" rather than "start".
13:38:41 <_glx_> push and pop also work nice for "uncoloured" strings, ie when colour is defined at DrawString level
13:39:31 <peter1138> I am glad you find the codes I added useful 😉
13:40:34 <peter1138> (And nice it is too)
13:44:51 <peter1138> With Truetype fonts, centred text seems to not quite be centred. Presumably it's including the space.
13:45:35 <peter1138> Well, ICU/Harfbuzz layouter, I suppose 🙂
14:09:46 <peter1138> Is it time for a new andy yet?
14:22:55 <gwyd4016> peter1138: Anti aliasing? In your pixel game? It's might be more likely than you think
14:26:33 <peter1138> andythenorthviaGitHu: Are you sure?
14:27:06 <talltyler> When you force-push 0 commits it closes the PR, I think 🙂
14:34:19 <talltyler> What do we need to do to maintain backwards-compatibility with our current Andy? Maybe...this?
14:47:44 <_glx_> hmm yeah would be nice to prepare for a new nml release
14:49:22 <_glx_> but first let's test the release workflow on my fork
14:49:28 *** kuka_lie has quit IRC (Ping timeout: 480 seconds)
14:50:31 *** kuka_lie has joined #openttd
14:59:47 *** speeder has joined #openttd
15:01:17 <speeder> hello! can someone point me to what I would need if I attempt an "ancient start" gameplay? earliest date as possible the better. I need to know if any industry sets is working well for this. And also need to know if anyone found a solution to the problem where you end with 5000 (literally, hititng the patched limit) barges to connect one industry and it still closes because it doesn't get enough stuff on time.
15:01:30 <speeder> Also if possible I wanted to use some kind of citybuilder scripts.
15:03:22 <kuka_lie> manyally editing save files might work
15:06:26 <kuka_lie> them are in ~/.local/share/openttd/save
15:28:49 <peter1138> I now have some Stilton.
15:31:44 <peter1138> Edit and resave should fix it. Not sure why it breaks as... I've not looked.
15:32:15 <peter1138> Or something. It works if you are logged in.
15:32:52 <peter1138> Works now after editing and saving.
15:45:09 *** kuka_lie_ has joined #openttd
15:46:51 *** kuka_lie has quit IRC (Ping timeout: 480 seconds)
15:48:19 <peter1138> andythenorth: are you going to fix #343?
16:10:44 <xarick> how is -1 even returned
16:17:35 <peter1138> When NPF used AyStar, that could return -1.
16:20:14 *** Wormnest has joined #openttd
16:31:34 *** XYZ has quit IRC (Ping timeout: 480 seconds)
16:33:37 *** Smedles has joined #openttd
16:52:49 *** gelignite has quit IRC (Quit: Stay safe!)
16:56:44 *** Wormnest has quit IRC (Ping timeout: 480 seconds)
17:12:32 <xarick> I think I'm making rivers uglier, let's see
17:38:20 *** Wormnest has joined #openttd
17:55:16 <mnhebi> yeah early circuits were noticeably cheaper if you went with smaller memory limits. At a certain point it was cheaper to stack 8-bit memory circuits instead of making 16-bit or 32-bit ones.
18:46:06 <xarick> aystar.h mentions "user_path" that doesn't exist
18:47:12 <xarick> and a user_data[10] thought user_data can be whatever I want
19:00:30 <kuhnovic> Amount of bits for the open and closed node. It determines their capacity.
19:05:54 *** kuka_lie_ has quit IRC (Ping timeout: 480 seconds)
19:19:09 *** kuka_lie has joined #openttd
19:35:00 <xarick> I see, seems unrelated
19:41:20 <xarick> what's the difference between static_cast and reinterpret_cast
19:41:31 <xarick> how do i know when to use one over the other
19:55:16 <peter1138> Step 1, don't be using either.
19:55:38 <peter1138> Step 2, almost always static_cast.
19:55:55 <peter1138> Step 3, if you think you need reinterpret_cast, redesign.
20:21:06 *** virtualrandomnumber has joined #openttd
20:29:16 *** virtualrandomnumber has quit IRC (Quit: virtualrandomnumber)
20:31:33 *** speeder has quit IRC (Remote host closed the connection)
20:31:55 *** speeder has joined #openttd
20:52:39 <xarick> what's more expensive? IsTileFlat or DistanceManhattan?
20:52:51 <xarick> for my optimization purposes
20:54:22 <xarick> ```/* A river, or lake, can only be built on flat slopes. */
20:54:22 <xarick> IsTileFlat(lake_center, &height_lake) &&
20:54:23 <xarick> /* We want the lake to be built at the height of the river. */
20:54:23 <xarick> height_lake == begin_height &&
20:54:23 <xarick> /* We only want a lake if the river is long enough. */
20:54:24 <xarick> DistanceManhattan(spring, lake_center) > min_river_length) {```
21:00:40 *** kuka_lie has quit IRC (Quit: leaving)
21:05:36 <peter1138> Why don't you look at what the functions do, and then if you need to benchmark them.
21:46:56 <xarick> i hope I made rivers faster
21:48:02 <xarick> some tropic checks could be done before computing slopes
21:49:27 <xarick> I'm pretty sure I made FlowsDown faster
21:50:03 <xarick> some checks only require computing 1 slope to fail
21:50:24 <xarick> only compute the 2nd slope if the checks for the first slope pass
22:12:26 <xarick> apparently i failed, i am getting no rivers now
22:12:38 <peter1138> That could make it faster.
22:21:25 <xarick> oh yeah, I failed in FlowsDown 🙂
22:48:35 <xarick> yeah, gains confirmed!
22:50:32 <xarick> it doesn't generate the same rivers though, but oh well...
22:51:09 <xarick> it should end up with more rivers and faster
22:51:58 <xarick> is it possible to count how many rivers were build successfully?
22:52:57 <xarick> the code in master can return true that it built a river, even when it doesn't
22:54:04 *** keikoz has quit IRC (Ping timeout: 480 seconds)
23:10:10 <xarick> I think I failed at lakes...
23:14:38 <xarick> maybe it's fine, there's the randomness factor too
23:55:48 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
continue to next day ⏵
