IRC logs for #openttd on OFTC at 2022-12-15
⏴ go to previous day
00:11:53 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
00:12:07 *** WormnestAndroid has joined #openttd
00:20:25 *** Wormnest has joined #openttd
00:51:09 <pickpacket> petern: in Sweden we have the expression "villhöver" which is a portmanteau of "wants" (vill) and "needs" (behöver). It's when you want something so badly it feels like you need it. That might apply here
01:57:08 *** Wormnest has quit IRC (Ping timeout: 480 seconds)
02:02:26 *** Wormnest has joined #openttd
02:32:00 *** wallabra_ has joined #openttd
02:37:33 *** wallabra has quit IRC (Ping timeout: 480 seconds)
02:37:33 *** wallabra_ is now known as wallabra
02:52:15 *** wallabra has quit IRC (Remote host closed the connection)
02:54:22 *** wallabra has joined #openttd
03:55:06 *** debdog has quit IRC (Ping timeout: 480 seconds)
04:16:50 *** _aD has quit IRC (Quit: leaving)
05:40:59 *** D-HUND is now known as debdog
05:53:31 *** keikoz has quit IRC (Ping timeout: 480 seconds)
07:16:21 *** sla_ro|master has joined #openttd
07:24:10 <reldred> I bought an obnoxious hipster fixie today petern
07:24:27 <reldred> (Eh it’s just a single speed, not a fixed hub)
08:02:07 <andythenorth> petern: why not 48?
08:02:12 <andythenorth> was it an ebay listing?
08:08:42 <petern> -9°C outside apparently, that's cold for us right?
08:12:05 <andythenorth> that is....not ideal
08:12:06 <petern> PlanetX keep spamming me with bike sales 😦
08:13:12 <petern> I've got 4kW of electric heating on right now to take the edge off. 🔥💷🔥💷🔥
08:35:37 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
08:36:12 *** WormnestAndroid has joined #openttd
08:38:25 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
08:38:33 *** WormnestAndroid has joined #openttd
09:02:08 <petern> Right, now I'm settled it's 9:01, so no time to tickle any OpenTTD 😦
09:08:11 <Eddi|zuHause> 4kW on a household line is quite... challenging
09:16:07 <petern> It's not my phone line...
09:17:02 <petern> It's 16A, our circuits are normally rated for 32A.
09:17:43 <andythenorth> Pruple: when requesting a fix an oversight in the spec, you should also have requested a change to the way the game works
09:18:04 <petern> I've turned it all off now anyway.
09:18:45 <petern> Technically almost anything beyond a typo in comments is a change to the way the game works 😄
09:19:56 <LordAro> andythenorth: hang on, let me fix that
09:22:46 <andythenorth> the other one about terrain, I have never understood quite how that works
09:23:45 <andythenorth> original game does this
09:23:50 <andythenorth> so then grf things do that
09:24:04 <andythenorth> so then grf authors can 'fix' it for themselves if they wish to
09:24:14 *** JustANortherner has quit IRC (Quit: User went offline on Discord a while ago)
09:24:24 <andythenorth> not sure if thats weird, or 100% fine
09:24:27 <LordAro> original game didn't have foundations :p
09:25:05 <reldred> we'd have to fundamentally change how snow works
09:25:05 <andythenorth> well I meant the lack of partial snow tiles for rail 😉
09:25:10 <reldred> a set of overlays maybe
09:25:31 <andythenorth> basesets don't compose tracks iirc
09:25:36 <reldred> ground tile+snow stage tile+road/rail tile+optional road/rail overlay
09:25:41 <andythenorth> and probably don't have partial snow drawn?
09:25:58 <reldred> no, only things that have partial snow is ground tiles, nothing else in game does
09:26:14 <reldred> I only just did snow rocks for the new jgrpp rocks feature
09:26:28 <reldred> after abusing jgr's feature wot wasn't even designed for it 😛
09:26:50 <Eddi|zuHause> i vaguely remember some track set trying to show partial snow
09:26:51 <reldred> trees are only binary snow
09:27:41 <Eddi|zuHause> reldred: lots of newgrf is abusing features that weren't designed for it
09:28:28 <reldred> Eddi|zuHause: well aware 🙂
09:28:29 <andythenorth> although there's also something in "let's start with world peace"
09:30:00 <reldred> but yeah I've got snow rocks for my ttdx baseset alpine climate going, I've got desert rocks for my re-code of pineapple landscape working, I just need to do regular arctic snow rocks now.
09:30:12 <reldred> I need rock variations though
09:30:56 <Pruple> andythenorth: it's fine, original rail and road tiles are full-tile and are either snow or not
09:31:50 <Pruple> whatever lc wants to do is a huge change to that. the bug I reported is just a tile that should logically be snowy but isn't.
09:32:54 <reldred> yeah I mean who's gonna draw all the graphics,
09:33:03 <reldred> just usual "good idea fairy" crap
09:33:15 <Eddi|zuHause> andythenorth: that requires two basic things: a) composing ground tiles and overlay tiles, and b) map bits storing the partial snow level
09:33:40 <Eddi|zuHause> no idea which of those two is the harder problem
09:34:46 <reldred> you'd kill all existing base set compatibility.
09:34:46 <petern> You can compose already with the trackbit sprites, but it'll look ugly.
09:35:08 <andythenorth> why do we need map bits? we already know the snow line height
09:35:20 <andythenorth> grfs have this solved, it's just the baseset problem is quite intractable
09:35:54 <petern> But yes, the issue itself is quite clear.
09:36:12 <petern> Foundation doesn't affect the height check and should.
09:37:11 <andythenorth> goes it fix everything
09:37:18 * andythenorth should make more train variants
09:37:45 <Eddi|zuHause> andythenorth: the snowline height can change, and the map needs to be redrawn when that happens. with map bits, this can be relegated to the tileloop
09:38:18 <andythenorth> so we need them for performance caching reasons?
09:38:37 <reldred> I do enjoy what rivers can do in jgrpp now with the right heightmap. That's about 1000 tiles long. Maybe 1500.
09:39:03 <Eddi|zuHause> for "don't immediately flash everything from green to white" reasons
09:39:20 <reldred> still has a problem with rivers reaching the coast though
09:39:41 <petern> Rivers should be able to erode.
09:40:19 <Eddi|zuHause> no, rivers should be able to flow down halftile slopes
09:40:35 <andythenorth> escalate feature
09:41:26 <Eddi|zuHause> on a separate note: river slopes need better sun shading
09:41:42 <petern> We discussed that and andy tried. It's not pretty.
09:42:12 <petern> You can do it if you use non-water-animation colours for water.
09:42:32 <petern> But they're quite the right hue.
09:42:55 <andythenorth> we need more palette 😛
09:43:19 <andythenorth> we have the spare pinks, we could use them as offsets into a new palette?
09:43:40 <Eddi|zuHause> 32bpp water cycle?
09:43:57 <andythenorth> that was the more sensible proposal yes
09:46:00 <Eddi|zuHause> water waves can probably be done with some alpha overlay magic
09:46:57 <petern> You can manually animate tiles but it's way more CPU expensive than a palette animation.
09:47:44 <petern> Time to implement a modern rendering pipeline, ignoring the current blitter system.
09:48:14 <reldred> goes it throw out entire rendering system?
09:48:29 <reldred> time for vulkan implementation
09:48:42 <petern> Compiling a sprite atlas of 100,000+ tiny sprites is not going to be simple 😄
09:52:00 <petern> That's a big rewrite just to apply shading on rivers 😄
09:55:17 <reldred> we'll just get zorg to do it, he obviously is much much smarter than all of us
10:01:34 <andythenorth> I think I worked out we have enough unused palette to do more blue
10:01:44 <andythenorth> if we wanted a simple-ish solution
10:01:52 <andythenorth> or we could just delete rivers also?
10:03:15 <andythenorth> 10, I think. There are 12 pinks unused
10:03:15 <petern> Presumably separate darker and lighter sections.
10:05:05 <petern> Colour dark_water[EPV_CYCLES_DARK_WATER]; ///< dark blue water
10:05:11 <petern> Colour glitter_water[EPV_CYCLES_GLITTER_WATER]; ///< glittery water
10:06:40 <petern> Glittery water is 15 whole colours.
10:07:13 <kamnet> Speaking of rivers, was anything ever considered for docks on rivers?
10:07:26 <petern> Hmm, actually it's 15 cycles, not sure if it's 5 palette indexes.
10:07:56 <petern> I don't think table/palettes.h contains that info.
10:08:23 <andythenorth> I did count the palette indices, but I don't remember the conclusion now 😛
10:09:49 <petern> for (i = 0; i != EPV_CYCLES_GLITTER_WATER / 3; i++) {
10:11:01 <petern> kamnet: If I ever finished NewGRF Docks that would allow it.
10:20:25 <kamnet> Having rivers as a viable form of transport rather than decoration or obstacles would be quite a QOL improvement
10:22:50 <TallTyler> reldred: For once the vanilla implementation is ahead of JGRPP, unless my wide river terraforming got downstreamed (no pun intended) already 🙂
10:23:19 <reldred> TallTyler: not sure if it did or not
10:23:24 <TallTyler> I have a bugfix PR because I made rivers a little too good at flowing downhill 😛
10:23:49 <reldred> I'm also leveraging my other river related patches though
10:24:12 <reldred> such as removing the restrictions on where rivers can start, etc, which is why the chain can stretch 1000+ tiles long.
10:30:05 <petern> What's the normal limit?
10:31:39 <reldred> There's not a length limit as such
10:31:45 <reldred> Not that I could see,
10:32:03 <reldred> but, rivers often kinda just... end sometimes. Not sure why
10:32:28 <reldred> What I did was remove one crucial check around where a river is allowed to *start*
10:32:38 <reldred> being the 'tops of hills'
10:33:04 <reldred> and, with the right heightmap, that ended up resulting in rivers generating in long chains
10:33:52 <EmperorJake> Or we could make rivers lower the terrain so that they can have docks and flat bridges
10:33:53 <reldred> so in my typical brain-dead fashion because I can't actually *write* C++ I just bundled up a game setting to turn the darn thing off.
10:34:40 <reldred> 99% of my patch writing for JGRPP consists of 'find this arbitrary logic or number that I disagree with and parameterize it'
10:35:31 <reldred> and then JGR wipes my backside for me and makes whatever vomit I've raised as a PR shippable.
10:36:24 <reldred> EmperorJake: yes we should do that as well
10:36:56 <EmperorJake> was there ever a patch that tried that?
10:37:19 <reldred> Umm, there was a patch someone made that tried to do full river errosion
10:37:57 <reldred> I mean, river gen is already pretty slow in JGRPP, at least with the silly maps I play.
10:38:24 <Eddi|zuHause> there was a rainfall river generator, but i don't think it did erosion
10:38:47 <reldred> I used to use it quite a bit
10:38:47 <andythenorth> we could just start at a coast tile and walk back
10:38:55 <andythenorth> but that doesn't work for some reason
10:40:16 <Eddi|zuHause> that just runs into the same problems: not all slopes are suitible for rivers, and trying to terraform them to make them suitible breaks it in other places
10:40:18 <petern> EmperorJake: That's what I meant with erode.
10:40:47 <petern> River gen is already pretty slow in vanilla.
10:41:08 <andythenorth> Eddi had the breadth first patch which produced visually correct results
10:41:10 <andythenorth> but it went to die
10:41:16 <petern> I also would like deep water to exist, but maybe not in the form that the last patch did it.
10:41:31 <Eddi|zuHause> there's no real reason for river generation to be fast
10:41:56 <Eddi|zuHause> andythenorth: it had some... corner cases. plus the slope thing
10:42:00 <reldred> well if we end up with more water palette shades then deep water might be feasible.
10:42:02 <TallTyler> Rainfall did do erosion, I believe, and also many other things
10:42:04 <petern> Can we multithread it? 😄
10:42:43 <andythenorth> petern: how deep does deep water need to be? 😛
10:42:49 <reldred> Eddi|zuHause: *cries in 8k x 2k with two additional river density settings*
10:42:59 <reldred> my map gens are slow :'''(
10:43:58 <Eddi|zuHause> reldred: people who make 8k maps and expect anything to be not-slow deserve to cry
10:43:58 <TallTyler> Bug maps are slow, that’s a user choice 😛
10:44:18 <reldred> Eddi|zuHause: :''''''(
10:44:28 <andythenorth> much as I enjoy large-map hating
10:44:33 <TallTyler> Can we reduce maximum map size and implement rainfall patch? 😛
10:44:40 <andythenorth> pushing the edges is where good performance can be found
10:44:49 <reldred> Nah I normally never play more than 4k x 1k
10:45:01 <andythenorth> if I didn't push the edges of grf compiles, NML would still be dogshit slow
10:45:12 <reldred> But I also deliberately play verrrrry spaced out maps with lots of straight up impassible terrain
10:45:13 <petern> Wentbourne is where bad performance can be found.
10:45:21 <JGR> "Slow" is a whole spectrum of values
10:45:27 <petern> Spaced out is the best.
10:45:28 <reldred> forcing dense corridors
10:45:46 <petern> The default "scale quantities up by map size" is not good.
10:46:04 <petern> Yeah, allowing the terrain to actually be high.
10:46:47 <reldred> you know I was kinda surprised that my patch that untethered the 1 to 1 tile width to height step didn't *break* more than it actually did.
10:47:04 <reldred> and wow did some of the heightmaps produce interesting results.
10:47:34 <reldred> damn I'll have to find a screenshot
10:48:07 <TallTyler> Sounds like you’re talking about cliffs but that’s not possible despite someone on TT-Forums trying
10:48:17 <TallTyler> *not currently possible
10:48:18 <reldred> I basically removed all the asserts and all the checks and balances that prevents cliffs
10:48:30 <reldred> it was a horrible broken mess graphics wise
10:48:38 <reldred> but uhhhh, interesting experiment
10:49:21 <Eddi|zuHause> you can easily fix the graphics by adding foundations. but the big problem is the UI
10:49:31 <petern> It's probable that the algorithm that enforces it leans towards a lower map height.
10:50:11 <petern> So that you actually need to terraform or tunnel. Pipedreams eh?
10:50:16 <reldred> Eddi|zuHause: my thought was cliffs, and then a viewable height 'slicer'
10:51:07 <reldred> that edit wouldn't have come through on irc, I meant foundations not cliffs
10:51:08 <Eddi|zuHause> well making foundations that look cliff-y is an exercise left for the reader
10:51:14 <LordAro> someone find the gif from smatZ(?) that made tunnels
10:51:46 <Eddi|zuHause> i think i have that somewhere
10:52:19 <Eddi|zuHause> it might have been an actual video
10:52:21 <petern> We aren't running on 2005 computers anymore.
10:52:39 <LordAro> petern: except that guy who wanted an XP build
10:53:01 <petern> The tile slope code (and foundations) looks at all the neighbouring tiles to determine the slope.
10:53:03 <reldred> But yeah, during that experiment it basically didn't crash
10:53:22 <petern> If you introduce cliffs, it may be better to store the slope info instead.
10:53:22 <reldred> usually just asserts that I commented out for the sake of the experiment
10:53:28 <andythenorth> I thought if we did this, trains could carry on across the cliff
10:53:33 <andythenorth> because the track bits are there
10:53:46 <andythenorth> someone showed me a video of, might have been rubidium
10:53:47 <reldred> andythenorth: that's exactly what happened
10:54:01 <petern> Only because there's currently no checks to stop it because they're not necessary without cliffs.
10:54:06 <Eddi|zuHause> andythenorth: that can be solved by preventing rails to be built towards cliffs
10:54:27 <andythenorth> no thelma and louise moment?
10:54:33 <Eddi|zuHause> andythenorth: or the pathfinder checking the height difference
10:55:09 <reldred> but yes, it was a very silly experiment
10:55:11 <petern> I was going to say "does this track connect is not tricky", but then I remember bridge heads 😄
10:56:17 <reldred> my thoughts on UI was basically have a slicer UI in the transparency menu to hide everything above level X so that you could work on the diorama behind the mountain. Maybe make it transparent? Who knows.
10:56:49 <reldred> now we have those fancy sliders in the interface zoom 😉
10:57:02 <petern> Slicer sounds reasonable.
10:57:12 <petern> Simpler than making up rotations.
10:57:23 <petern> reldred: I await the PR for vanilla.
10:57:43 <reldred> Well, does the rotation need rotated sprites? Would it bother people *that much?*
10:58:07 <reldred> I've seen other games with rotations but only one view for buildings and such
10:58:19 <petern> For multi-tile buildings/industries it could be awkward
10:58:33 <reldred> Okay yes I concede on industries
10:58:49 <reldred> and yeah I guess 1x2 and 2x1 buildings would be shagged
10:59:04 <reldred> But yeah, slicer does solve a lot of construction issues.
10:59:15 <reldred> The one thing I didn't have a good answer for was terraforming.
10:59:24 <petern> Also all the shading would be wrong.
10:59:37 <reldred> RCT's style of terraforming always felt really clunky and slow and fiddly.
10:59:53 <petern> Is it like LoMo? Cos that was horrid.
11:01:41 <Eddi|zuHause> on the topic of that video, what happened to dev.openttd.org?
11:20:27 <dwfreed> reldred: it was so bad, I never bothered with terraforming
11:22:46 <reldred> It’s my one giant turn off about cliffs in TTD
11:24:33 <dwfreed> granted, I don't think I ever made it much further than Diamond Heights, so it's not like it was a huge issue
11:31:08 *** WormnestAndroid has quit IRC (Remote host closed the connection)
11:37:19 *** WormnestAndroid has joined #openttd
12:37:39 <Timberwolf> RCT terraforming is great for adding tiny landscaping details to a roller coaster, it's horrible for doing large scale landscape changes.
12:38:49 <Timberwolf> LoMo really needed some things like autorail and better area-effect terrain tools (smooth, progressive flatten, etc.)
12:40:28 <Timberwolf> iirc RCT also makes it fiddlier by locking away one of the more useful terrain options you get in the scenario editor (dragging one tile and having the land around it raised/lowered). OpenRCT2 enables it in-game, which helps a lot.
12:55:41 <pickpacket> TrueBrain: Last night I was asking about funding for, well, everything really. All the services in AWS etc. Who pays for it and what does it cost?
13:21:46 <dP> apparently using random water colour for each pixel doesn't look that well 😅
13:29:22 <petern> Make the river rocks *huge*, so that they can be shaded appropriately. Then sprinkle some sparkling water around them.
13:32:44 <dP> ok, it does work, I just need to randomize f5-f9 for opengfx water look xD
13:33:41 <dP> I have directional rivers though so I could try to use that...
13:35:42 <petern> Slopes don't anything extra for directions 🙂
13:37:13 <dP> i can try make it flow not only on slopes
13:39:31 <dP> tiling would be tricky though
13:39:45 <dP> whatever, have more important stuff to do rn xD
13:40:03 <petern> andythenorth: was I missing anything for variants?
13:40:40 <petern> I've got a load of minor interface scaling fixes to do as well.
13:45:13 <dP> how does min compatible version work in newgrf? if people with older but still compatible version join the server will they be required to download the new one?
13:46:03 <petern> No, you always need the exact same version.
13:46:27 <Brickblock1> I think it only applies when updating your grf lists
13:46:38 <petern> It means the NewGRF says that OpenTTD can upgrade the NewGRF to the newer version without breaking a savegame.
13:47:30 <dP> hm, I thought I've seen openttd using older versions to open saves
13:47:43 <petern> It will. Maybe I made that last bit up 🙂
13:48:04 <petern> But it's nothing to do with network compatibility either.
13:48:55 <dP> well, I'd like them to upgrade if they can when opening downloaded save too
13:49:14 <dP> but techically it's not required since I'm just changing sprites
13:58:51 <andythenorth> petern: I didn't see anything
13:59:03 <andythenorth> merge, merge the nml patch, needs other authors to test it
13:59:12 <andythenorth> caution is good and all...but
14:02:12 <dP> quite a decent river for not drawing a single pixel manually
14:07:55 <andythenorth> truly generative rivers? 😛
14:12:15 <dP> I'm too lazy so I write code to do the work for me 😆
14:16:13 <andythenorth> procedural rivers might make more sense than sprite rivers
14:24:28 <andythenorth> didn't we used to have nightlies or something?
14:24:33 <andythenorth> so we could merge stuff in
14:24:43 <andythenorth> hmm...we have WASM
15:09:35 <andythenorth> clearly the next big thing
15:18:29 <andythenorth> electrified or not
15:18:43 <andythenorth> whether the sun shines through on Brunel's birthday
15:19:01 <andythenorth> standardised tunnel label type scheme
15:19:10 <andythenorth> standardised tunnel label type scheme v2
15:19:29 <andythenorth> deletion of someone's entire forum history because they had their own label scheme for tunnels
15:19:52 <andythenorth> not putting grfs on bananas because you disagree with tunnels being a player choice, and it ruins your vision for the game
15:19:58 <andythenorth> the future is bright
15:20:16 <andythenorth> hmm what the tank game has taught me
15:20:20 <andythenorth> is that some people are very stupid
15:20:22 <andythenorth> sometimes it's me
15:34:59 <FLHerne> It's good of Zorg to have a name that can be so satisfyingly verbed
15:39:03 <LordAro> is your computer melting?
15:40:25 <FLHerne> ...where did Zorg's tunnel types screenshot come from?
15:40:42 <FLHerne> is that a patch that I missed, or did he painstakingly make one for the purpose?
15:41:01 <LordAro> i believe he's very adept at photoshop
15:41:11 <LordAro> same for all his UI designs
15:41:17 <LordAro> certainly he's never shared any code
15:41:53 <FLHerne> drawing an entire pixel-perfect fake UI for a semi-throwaway GH comment is some dedication
15:42:03 <FLHerne> even if it's a bit crazy
15:42:19 <FLHerne> tbh I'd use that if it existed
15:42:39 <FLHerne> it seems like the logical counterpart of bridges
15:43:14 <FLHerne> don't know about the speed restrictions though, that just seems pointless
15:43:33 <FLHerne> albeit technically realistic above a given speed
15:43:40 <andythenorth> we need tunnels that have speed restrictions in winter due to water ingress
15:43:58 <andythenorth> or that only allow trains on 15 minute headways to allow steam engine smoke to clear
15:44:05 <petern> LordAro: Maybe now the boiler is on.
15:59:06 <petern> Oops, accidentally deleted the wrong thing and automatically closed a dozen PRs :/
16:09:21 <andythenorth> can I stop having covid yet?
16:21:26 *** nielsm has quit IRC (Ping timeout: 480 seconds)
16:22:05 <petern> Silly YouTube keeps going to 360p 😦
16:32:42 <MnHebi> I've found youtube keeps using lower quality than auto quality claims
16:48:44 *** Flygon has quit IRC (Quit: A toaster's basically a soldering iron designed to toast bread)
17:08:13 *** HerzogDeXtEr has joined #openttd
17:26:40 <wyndbain> petern: i had this issue
17:26:53 <wyndbain> it magically went away when i got yt premium
17:38:09 <TallTyler> Visual Studio is complaining that my line endings are inconsistent (probably because I edited the code for the PR in Notepad at the library), but when I change the whole file to CR LF Git Gui thinks the entire file has changed - help?
17:41:18 <LordAro> (unless you've previously told git to translate them)
17:42:16 <TallTyler> Ah, that fixes it. Other files display as CRLF for some reason
17:42:24 <TallTyler> (at least in the bottom right corner of the code window in VS)
17:43:21 <TallTyler> Yeah, Git seems to be translating them for some reason
17:43:43 <petern> That's fine as long as it's consistently translating 🙂
17:44:25 <TallTyler> Seems to be fixed now 🙂
17:46:13 <petern> That code-duffing or "never nester" is nicer 😉
17:47:05 <TallTyler> Yeah, not sure how I wrote it so poorly originally 😛
17:51:24 *** Wormnest has joined #openttd
17:51:27 <petern> It's normal to iterate 🙂
17:53:28 <glx[d]> usually final version is simpler
17:55:06 <glx[d]> but each version is good for learning
18:47:18 <dP> oh, I thought he PRed them to vanilla
18:47:28 <dP> I can take a look on 18th, not sure about earlier
19:33:11 *** HerzogDeXtEr has quit IRC (Read error: Connection reset by peer)
19:45:47 <petern> All text height dimensions don't include the shadow, and never have done.
19:46:57 <petern> You could bodge it by adding 1 scaled pixel to the clipping area set up before the widget is painted.
19:47:50 <petern> You could change FONT_HEIGHT_NORMAL to include the shadow, but then EVERYTHING gets taller.
19:48:08 <petern> (And that doesn't help the right-most shadow)
19:48:33 <petern> (Not just F_H_N, it affects truetype fonts too)
19:48:36 <TallTyler> Right-most shadow is drawn properly here, though?
19:48:48 <petern> In that particular case, yes.
19:49:09 <TallTyler> Is this a very old bug that I've never noticed?
19:49:52 <TallTyler> That's why progress stalled on my Company Infrastructure GUI PR
19:50:10 <TallTyler> Still deserves fixing, IMO
19:51:18 <petern> The reason the right-most shadow is fine in that picture is due to another bug IIRC -- the text layouter wraps on the space, but the space is still included, so there's still space for the shadow.
19:51:40 <petern> (But it depends which layouter is used)
19:55:45 <TallTyler> Something something when AIs rule the world they can search my images for text 😛
19:59:38 <petern> With my 'suggested' workaround of adding 1 scaled pixel to the clipping area.
20:00:17 <petern> Most things that really need a specific clipping area already set it up.
20:02:03 <petern> There may be other bits where a clipping area is set up to draw text which would also need extra provision.
20:08:51 <petern> Ah, not the bug with the layouter isn't the space.
20:09:07 <petern> The bug with the layouter is it always includes the inter-character spacing even at the end of the line.
20:09:24 <petern> Which handily means that the shadow is accounted for.
20:09:33 <petern> (Even when no shadow is drawn)
20:13:47 <Wolf01> Is that related with my problem on the deck?
20:16:36 <petern> What's your deck problem?
20:17:08 <Wolf01> Text on top half on widgets
20:34:44 *** gelignite has joined #openttd
20:38:06 <petern> [19:45]peter1138: By design.
20:38:55 <TallTyler> Do we want to be bothered by text shadows overlapping their containing boxes? (Unrelated to this fix, also happens in window titles)
20:39:45 <LordAro> feels like it might react poorly with chunky/unchunky bevels
20:40:07 <petern> It's part of our UI design tbh.
20:41:12 <petern> And no, we shouldn't make captions taller to accommodate.
20:41:34 <Wolf01> I won't have this problem :>
20:43:44 <petern> LordAro: In fact it works better with chunky bevels because there's always space now.
20:43:47 <glx[d]> now to "redo" the delay start
20:44:13 <TallTyler> Is that justification to make chunky bevels mandatory? 😛
20:44:29 <petern> No, even with chunky bevels is turned off, the space is the same.
20:44:54 <TallTyler> I guess we have to let people be wrong then 😦
20:45:21 <petern> The only time that isn't the case is the inset viewports and the inset minimap.
20:50:40 <petern> Oh my Q2 RTX saves don't load any more 😦
20:53:28 <dP> is there any way to make wagons work with engine from another grf that has `can_attach_wagon` ?
20:54:42 <andythenorth> is the grfid of the vehicle available as a 40+ or 80+ var?
20:54:48 * andythenorth thinks it might be
20:54:48 <dP> yeah but will it work if I do action 3 for overriden engine?
20:54:51 <petern> I can't remember if it's under the lead engine's control or the wagon...
20:55:09 <andythenorth> oh you want to force it?
20:55:18 <andythenorth> /me unsure of the goal
20:55:21 <petern> If you override its action 3 you'll lose all of its act 3/2/1.
20:55:50 <petern> Q2 is too hard for me 😦
20:55:51 <dP> well, that's kinda pointless then
20:55:59 <andythenorth> oof I need a beer, this covid not-drinking is boring
20:56:12 <dP> I may as well redefine the whole engine
20:59:02 <andythenorth> goes it approve?
21:07:10 <TallTyler> If it’s feature-complete I can review it tomorrow
21:09:52 <TallTyler> We should probably do an RC1 build soon?
21:54:49 *** keikoz has quit IRC (Ping timeout: 480 seconds)
22:25:44 *** sla_ro|master has quit IRC ()
22:28:03 <andythenorth> was it done yet?
22:28:21 <andythenorth> I have not coded and covid horse variants
22:29:18 *** greeter has quit IRC (resistance.oftc.net larich.oftc.net)
22:30:19 *** greeter has joined #openttd
23:01:45 *** kstar892[m] has quit IRC (Quit: Client limit exceeded: 20000)
23:05:53 *** Wormnest has quit IRC (Ping timeout: 480 seconds)
23:12:08 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
23:16:25 *** WormnestAndroid has joined #openttd
23:22:33 <petern> Gah at lying documentation.
23:27:21 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
23:28:01 *** Tirili has quit IRC (Quit: Leaving)
23:41:23 *** Wormnest has joined #openttd
23:50:55 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
continue to next day ⏵