IRC logs for #openttd on OFTC at 2025-04-19
⏴ go to previous day
00:45:04 *** Flygon has quit IRC (Read error: Connection reset by peer)
01:45:12 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
01:46:41 *** WormnestAndroid has joined #openttd
02:08:44 *** gnu_jj_ has joined #openttd
02:12:05 *** gnu_jj has quit IRC (Ping timeout: 480 seconds)
03:00:07 *** herms has quit IRC (Quit: bye)
04:11:06 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
04:11:07 *** WormnestAndroid has joined #openttd
04:11:22 *** WormnestAndroid has quit IRC (Remote host closed the connection)
04:11:25 *** WormnestAndroid has joined #openttd
04:11:28 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
04:11:29 *** WormnestAndroid has joined #openttd
04:44:10 <DorpsGek> - Update: Translations from eints (by translators)
08:09:49 *** fairyflossy has quit IRC (Quit: User went offline on Discord a while ago)
08:16:43 *** mindlesstux has joined #openttd
08:22:37 *** mindlesstux has joined #openttd
08:34:01 *** gelignite has joined #openttd
09:03:53 *** Emru has quit IRC (Quit: Gateway shutdown)
09:33:37 <andythenorth> advising me to put debug prints into my nml for display in the in-game console
09:33:55 <andythenorth> as though we could simply print values from varact 2 to a log, with formatting
10:36:41 *** WormnestAndroid has quit IRC (Remote host closed the connection)
10:36:46 *** WormnestAndroid has joined #openttd
10:54:41 *** kuka_lie has joined #openttd
11:00:48 *** gelignite has quit IRC (Quit: Stay safe!)
11:41:29 *** Christopher has joined #openttd
11:43:02 <Christopher> How Can I Create Routes
11:43:10 *** Christopher has quit IRC (Remote host closed the connection)
11:45:35 <merni> ...great way to get an answer
11:45:44 <merni> leaving immediately after asking
12:06:29 *** gelignite has joined #openttd
12:13:58 <exceptik> maybe got too accustomed to messengers that save all messages on server 😆
12:18:30 *** nielsm has quit IRC (Ping timeout: 480 seconds)
13:33:03 *** gelignite has quit IRC (Quit: Stay safe!)
14:26:48 <peter1138[d]> Hmm, without the extra frame.
14:28:23 <reldred> Finally consolidating those two bloody menus
14:43:16 *** gelignite has joined #openttd
14:43:46 <peter1138[d]> Hmm, perhaps those buttons should be drawn with `DrawBoolButton()` instead of wrangling button widgets to be approximately the same size.
14:57:18 *** Wormnest has joined #openttd
15:00:35 <peter1138[d]> How is it 4pm already.
15:11:20 <peter1138[d]> Hmm, this looks bad.
15:13:35 <peter1138[d]> Stuck in a resizing loop.
15:25:33 <peter1138[d]> Added a Debug() and... it stopped :S
15:27:16 <talltyler> The yellow volume slider is pretty bad, the other colours I don’t hate
15:27:46 <_zephyris> Certainly not red/green. Implies good/bad, and not colour blind suitable. Maybe neutral -> green.
15:28:12 <peter1138[d]> What about here?
15:28:21 <peter1138[d]> Where it's always been this way.
15:29:32 <_zephyris> It's not ideal design
15:30:12 <peter1138[d]> I have the same reservations, but thought I should try it at least.
15:30:25 <peter1138[d]> But actually I meant the rest of the colours, text, background, labels, etc.
15:31:48 <_zephyris> Yellow/dark purple ish is a decent contrast
15:32:04 <_zephyris> Light blue on purple ish isn't so good
15:32:28 <peter1138[d]> I didn't choose these colours either, just taking from existing windows, and frosch's suggestions.
15:34:01 <peter1138[d]> AI, GS, NewGRF and Advanced Settings are all (mostly) this scheme, except Advanced Settings goes with mauve buttons.
15:36:35 <_zephyris> I'd err on windows with lots of text sticking to grey background, and save strong colour backgrounds for less text heavy windows.
15:39:13 <peter1138[d]> Play TTO for a bit 🙂
15:39:35 <peter1138[d]> Grey windows are boring, and we have way too many of them already.
15:47:25 <talltyler> Agreed. I used to like grey windows but I now live in a Millennial Grey apartment and crave colour 😄
15:48:07 *** Wormnest has quit IRC (Quit: Leaving)
15:54:25 <_zephyris> Oh, totally. _sensible_ answer is neutral, fun/interesting and still readable answer is keep the purple/yellow
15:57:39 <peter1138[d]> I looked at that but did it with as templated function.
15:58:28 <peter1138[d]> Making it an iterator is nice.
15:59:52 <peter1138[d]> Okay, weirdly, this infinite loop depends on the game's window size.
16:08:44 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
16:11:58 *** WormnestAndroid has joined #openttd
16:16:20 <frosch123> About the setting buttons. IMO using the bevel lowered/raised status is too little visual indication to distinguish on/off.
16:16:20 <frosch123> If people dislike the green/red, my next preference would be on/off captions.
16:16:20 <frosch123> Though I guess you could already make that point for colourblindness.
16:20:50 <peter1138[d]> On / Off might be problematic for translations.
16:23:09 <andythenorth> BODMAS is so stupid
16:23:13 <andythenorth> just use brackets
16:23:34 <andythenorth> (had a bug because I didn't use brackets)
16:23:58 <frosch123> It's also a lot of work. Bool buttons are currently fix size to match the < > buttons
16:24:58 <_glx_> 1/0 drawn on top of green/red button ?
16:24:59 <peter1138[d]> It's a tiny amount of work if you don't change that 🙂
16:32:43 <peter1138[d]> This isn't good.
16:36:09 <_zephyris> You're probably better with neutral (yellow) for off and green for on
16:36:22 <_zephyris> Red for off, implying bad, isn't great
16:38:05 <_zephyris> They're actually checkboxes anyway, why not box and box with tick?
16:38:49 <_zephyris> peter1138[d]: Do I see three different shades of almost yellow here? Line colour, text colour and button colour?
16:39:22 <peter1138[d]> The text is orange.
16:40:18 <peter1138[d]> Arguably, I'd say the yellow is almost orange.
16:42:04 <exceptik> _zephyris: red for ON better? 😆
16:43:17 <_zephyris> You'll find a red led which is off for off and on for on is pretty typical 😉
16:44:01 <peter1138[d]> Also the same for green.
16:44:19 <peter1138[d]> Until that phase in the early 2000s, when every LED suddenly became a blindly bright blue.
16:45:40 <_zephyris> OTTD not using checkbox and radio buttons is a long-standing oddity
16:47:54 <peter1138[d]> Also, inactive buttons are not done the same way as all other inactive buttons. Fix it?
16:49:05 <peter1138[d]> The faded look is not bad, mind you. But everything else uses the chequer pattern.
16:49:07 <_glx_> keeping red/green button size we could do empty/filled rectangle
16:50:57 <peter1138[d]> I think they are no longer used anywhere.
16:51:08 <peter1138[d]> I also think they do not fit in here.
16:51:47 <peter1138[d]> Oh I'm dumb. They're used in the content selection window.
16:52:15 <peter1138[d]> (They used to be used in the Livery window, and also the NewGRF window at one point, IIRC.)
16:52:17 <_glx_> yeah they fill too small for the settings
16:52:47 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
16:53:48 *** WormnestAndroid has joined #openttd
16:54:14 <_glx_> hmm what about half red/half green and a yellow slider button ?
16:54:41 <peter1138[d]> Moving from left to right? Hmm.
16:54:47 <_glx_> same size as arrow buttons, moving left or right yes
16:55:44 <andythenorth> is it vibe coding if I don't edit the result at all?
16:55:48 <andythenorth> what even is vibe coding?
16:56:14 <exceptik> andythenorth: you ask ai, use whatever it spits out, profit
16:56:35 <frosch123> There is also a checkmark font character
16:56:44 <_glx_> mostly loss with AI and nml
16:56:58 <andythenorth> I mean...I now have 4 different functions from GPT for generating randomised vehicle maps as python dicts of lists
16:57:06 <andythenorth> all of them are exactly the pattern I want
16:57:17 <andythenorth> but rather than asking GPT to unify the method names, params, output
16:57:30 <andythenorth> I've just written 4 different nml templates to use the output
17:05:21 <peter1138[d]> Pretty meh also.
17:05:35 <_zephyris> Better than anything so far IMO
17:05:36 <peter1138[d]> And that is, apparently, centred.
17:05:51 <_zephyris> It's pushing towards standard design conventions
17:14:24 <peter1138[d]> I'm tempted to just not touch it.
17:15:39 <peter1138[d]> It's worked for 20 years... 🙂
17:18:38 <andythenorth> NoMoreSideQuests? 😮
17:18:57 *** WormnestAndroid has quit IRC (Remote host closed the connection)
17:22:32 *** WormnestAndroid has joined #openttd
17:24:58 <talltyler> Save the fix for a future PR? (kicking the can down the road for somebody else to do later 🙂 )
17:54:08 *** Emru has quit IRC (Quit: Gateway shutdown)
18:01:07 <peter1138[d]> Same with the suggestions to change what is shown when it's opened 🙂
18:43:26 *** tokai|noir has joined #openttd
18:43:26 *** ChanServ sets mode: +v tokai|noir
18:44:39 *** greeter has quit IRC (Remote host closed the connection)
18:47:29 *** greeter has joined #openttd
18:50:20 *** tokai has quit IRC (Ping timeout: 480 seconds)
19:06:25 <_zephyris> I've got some mockups somewhere of some more standard checkbox/radio elements, I'll see if I can find it
19:26:58 <peter1138[d]> /me grumbles about signed/unsigned comparisons.
19:30:08 <cu-kai> i like the look of the current red/green buttons
19:30:19 <cu-kai> i mean, the discussion about it
19:31:02 <talltyler> Red/green buttons are bad for colour blindness 🙂
19:37:03 <andythenorth> funny how often UI gets dismissed as bike shed 🙂
19:37:17 <andythenorth> "design is how it works"
19:37:19 <cu-kai> not dismissing it as anything :D
19:37:39 <cu-kai> i always considered it obvious if the button was "pressed" or not
19:37:42 <cu-kai> perhaps that's me being ableist
19:37:52 <cu-kai> (regardless of the colour, that is)
19:38:41 <talltyler> Sideways traffic lights, what is this, New Mexico?
19:44:40 <peter1138[d]> cu-kai: Does it ever confer "good" or "bad" to you?
19:45:09 <peter1138[d]> "This button is red, it would be better if it was green" or somesuch.
19:45:23 <cu-kai> not really, just "on" or "off"
19:45:33 <cu-kai> that is my opinion though, others may think good/bad
19:49:18 <peter1138[d]> Hmm. Kinda weird.
19:49:55 <cu-kai> can you add "lines" to the yellow part
19:49:58 <cu-kai> make it obvious that it's some sort of slider
19:50:03 <cu-kai> i don't know how to describe what i mean, sorry
19:50:35 <peter1138[d]> Scrollbars have "lines" on the non-button part.
19:51:09 <cu-kai> probably some 1px dark lines would be enough, not like the scrollbar
19:51:17 <talltyler> Trying to do this? The size of the “knob” is larger to help show its position.
19:51:52 <cu-kai> ^ with that in mind, you could make the green/red part narrower
19:52:10 <talltyler> Although our design is more inspired by 1990s computer design than React 😛
19:52:55 <_zephyris> Still don't know why checkboxes aren't the right answer
19:53:25 <_zephyris> Switch/knob slider designs are OK, but don't clearly convey whether left or right is the "on" state without an additional colour cue or checkmark
19:53:51 <talltyler> Checkboxes are very fitting for our graphics style
19:53:55 <cu-kai> to be fair, the "current" state is displayed anyway, in text format
19:54:20 <_glx_> in settings yes, but not everywhere the bool button is used 🙂
19:59:42 <peter1138[d]> I prefer red/green over this, tbh.
19:59:48 <_glx_> yeah doesn't fit too well
20:01:10 <_zephyris> The icon is designed for a different type of checkmark usage
20:01:32 <_zephyris> (which, as an aside, would probably work better with an updated design)
20:02:25 <exceptik> _glx_: what about having "recessed" yellow instead of "inactive" part, and main yellow for "active", that way you don't have weird color combos, still ahve the rectangle
20:02:25 <exceptik> like classical windows theme had scrollbars
20:03:13 <_zephyris> Better, but off really, really should be neutral - in this case yellow.
20:04:39 <_zephyris> iOS, off is neutral
20:05:09 <_zephyris> You'll also see lots of examples where on and off are shown in the background, or a checkmark is added to the knob in the on state.
20:05:39 <_zephyris> Because left/right on/off ambiguity
20:06:09 <peter1138[d]> Honestly, at this point I prefer the existing button, and I'm feeling less inclined to change it for the sake of changing it.
20:06:09 <xarick> if I enable magic bulldozer, will the AIs also benefit from magic bulldozer?
20:07:20 <_zephyris> Good to me too, but not because it shouldn't change...
20:09:53 <_zephyris> But because it should be trying to unify these four totally different-looking checkmark-like cases
20:10:11 <_zephyris> Informed decisions 😉
20:11:08 <_zephyris> Hmm. Good basic fix would be to left align these buttons, so they're not orphaned from their labels.
20:12:40 <frosch123> Fun fact, nlohmann/json.hpp makes clang-format consume 8GB of ram, until I kill it
20:15:41 <peter1138[d]> The content download toggle is different.
20:15:59 <peter1138[d]> It's not a boolean control.
20:19:06 <_zephyris> It's a boolean control for what happens when you click download...
20:22:09 <_zephyris> No need to split hairs though, just trying to point out an inconsistency
20:25:12 <talltyler> What causes sprite sorting issues like this? Bounding boxes? (Trying to write the bug report accurately 🙂 )
20:26:12 *** andybiotic has joined #openttd
20:26:12 <andybiotic> talltyler: Yeah, it’s bounding boxes
20:26:17 <talltyler> Absolutely beautiful sprite work besides this though 😄
20:27:17 <talltyler> That's widget outlines, but I know what you meant 🙂
20:28:42 <peter1138[d]> My problem with these stations: no automatic tile layouts, you gotta pick out of dozens of different station types.
20:28:51 <_glx_> most likely a full building that should actually be 2 or more sprites
20:35:26 <talltyler> All platforms seem to do it, but the red brick is especially bad.
20:36:06 <peter1138[d]> Is the shelter a child sprite, or is it the exact same bounding box.
20:36:17 <talltyler> Even ones without canopies; fences have z-fighting
20:38:45 <honza_> seems like "Wagon removal" button has wrong string in 15.0beta2
20:49:31 <xarick> how does AIPriorityQueue work again?
20:50:03 <xarick> Pops the item with the lowest priority value? I forgot
20:50:58 <xarick> > Remove and return the item with the lowest priority.
20:51:09 <_glx_> yeah the doc seems clear
20:51:57 <xarick> I have this problem: understanding my code
20:54:33 <xarick> imagine we have 2 dock stations with both cargo_production = 50, but one is 3 tiles away from town, the other is 6 tiles away. How to prioritize the 3 tiles station to Pop first?
21:00:22 <_glx_> xarick: production * something + distance ?
21:01:06 <xarick> `0x1FFF - 3 = 8188`,`0x1FFF - 6 = 8185`
21:01:06 <xarick> `distance 3 gets -8188 priority`,`distance 6 gets -8185 priority`
21:01:06 <xarick> lowest priority Pops first, -8188 < -8155, distance 3 is first
21:01:21 <peter1138[d]> Does the distance from the town matter to anything?
21:01:28 <_glx_> don't use negative values ?
21:01:53 <xarick> bigger houses spawn closer to center, better chances of high cargo going to station
21:02:22 <xarick> okay, my code makes sense now
21:02:43 *** alpapilus has joined #openttd
21:02:43 <alpapilus> talltyler: What grf is that?
21:02:48 <peter1138[d]> In that case if you're thinking about what might happen in future, the current production is not that relevant.
21:04:15 <_glx_> if you want higher production to have lower priority use something like `some_big_value - production`, so it's still positive but ordered in reverse
21:04:46 <_glx_> using plain negative is asking for troubles
21:05:50 <xarick> hmm cargo production is wrong?
21:07:08 <frosch123> Yoffset is wrong, yoffset+yextend > 16
21:07:11 <xarick> cargo 50, and cargo 30:
21:07:11 <xarick> -50 < -30, 50 pops first, it's fine
21:07:38 <peter1138[d]> I had patch that warns and clamps about out-of-bounds offsets.
21:07:47 <peter1138[d]> But way too many objects abuse this deliberately.
21:08:05 <alpapilus> talltyler: Oh that one.
21:10:33 <frosch123> If you bulldoze some tiles in front, you can probably see the overlap
21:10:49 *** gelignite has quit IRC (Remote host closed the connection)
21:14:10 <talltyler> I'll let andybiotic deal with their own bounding boxes, I'm just having fun with early stations 😄
21:14:57 *** gelignite has joined #openttd
21:15:17 <andybiotic> talltyler: I think the bounding box height is too low. I will have a look over the next few days. Love that station!
21:23:39 <andybiotic> frosch123: Oh, interesting. Thank you.
21:23:39 <andybiotic> I will likely remove all the y-offsets and set them to zero. I seem to remember it looked a bit wonky when I tried it originally, hence why I set it to 1, but evidently that might not have been a great idea…
21:23:58 <_glx_> xarick: but inverting via a big value and keeping everything positive would be better, like with `100 - prod` for 50 and 30 you get 50 and 70, so 50 is poped first
21:24:56 <_glx_> but if you add the distance a plain negative value will prefer long distance
21:26:59 <_glx_> it's easier to work with positive values (at least for human brain)
21:27:58 *** keikoz has quit IRC (Ping timeout: 480 seconds)
21:30:40 *** zanooda2000 has quit IRC (Quit: User went offline on Discord a while ago)
21:34:40 *** kuka_lie has quit IRC (Ping timeout: 480 seconds)
21:43:37 <peter1138[d]> So many things would be a lot easier if we just used int...
21:52:18 <peter1138[d]> Dimension<uint> Dimension<int> 😒
22:34:32 *** normalperson8046 has quit IRC (Quit: User went offline on Discord a while ago)
22:36:13 <xarick> end and tile_bot don't match, i fail at orthogonal areas
22:41:50 <xarick> CreateArea sends the correct parameters
22:42:06 <xarick> constructor fails at reconstructing the area 😦
22:45:08 <xarick> darned freeform_edges is difficult
22:47:09 <xarick> with freeform_edges off, it's wrong, with freeform_edges on, it's correct
22:52:23 *** nielsm has quit IRC (Ping timeout: 480 seconds)
23:04:48 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
23:10:07 *** zanooda2000 has joined #openttd
23:10:07 <zanooda2000> _zephyris: then the difference between On and Off, and also difference between [ ↓ ]drop-down lists / [←|→ buttons will be way less noticable.
23:10:07 <zanooda2000> If changes are really needed, then maybe make Off as purple-gray and On as green?
23:10:07 <zanooda2000> Although the standard red Off and green On still a simple and clear, which isn't really needed to be changed
23:17:25 <zanooda2000> peter1138[d]: hmm, yellow strip on ON button seems slightly shorter than on Off button, it's just an illusion, but still, if you don't look closely, the difference is here. Maybe it makes sense to make it 1 pixel wider (or yellow on Off 1 pixel shorter)
continue to next day ⏵
