IRC logs for #openttd on OFTC at 2026-04-05
⏴ go to previous day
01:03:06 *** gelignite is now known as Guest6596
01:03:09 *** gelignite has joined #openttd
01:09:30 *** Guest6596 has quit IRC (Ping timeout: 480 seconds)
01:50:18 *** Wormnest has quit IRC (Quit: Leaving)
01:55:11 <reldred> I'd get the increase in new contributors after your interview tyler but they obviously missed the bit about AI slop
02:26:47 <__abigail> Does anyone know of any decent resources for MFC applications? I know there are newer frameworks but the MSFS SDK docs recommend it
02:34:25 <reldred> __abigail: you'd probably have to wait for them to wake up again, they're all tuckered out from pushing out a release 😛
02:35:56 <__abigail> How about they do something useful instead
02:35:56 <__abigail> Like link to documentation that will help with my incredibly niche application that no one will ever use
02:39:52 *** tokai|noir has joined #openttd
02:39:52 *** ChanServ sets mode: +v tokai|noir
02:46:06 *** MinchinWeb[m] has quit IRC (Ping timeout: 480 seconds)
02:46:41 *** MinchinWeb[m] has joined #openttd
02:46:45 *** tokai has quit IRC (Ping timeout: 480 seconds)
02:57:18 *** Tirili has quit IRC (Quit: Leaving)
03:17:09 *** Phileman has joined #openttd
03:19:19 <rito12_13> __abigail: The answer to your question is 'Yes, someone do know a decent resources for developing MFC applications'. ~~It isn't me~~
03:20:28 *** Philemon has quit IRC (Ping timeout: 480 seconds)
03:36:26 <greeter> i have to admit, i like that answer
04:21:45 <__abigail> __abigail: For whatever reason the only SDK example with a UI seems to be made with WinUI, so it will at least work
06:43:05 <andythenorth> lol shall we just pull it from Steam and be done with it?
06:56:43 <LordAro> ban that guy specifically
06:57:18 <LordAro> 15.4 release just with user idrntification built in just for him
07:15:22 <peter1138> How large is your savegame directory?
07:29:56 <emperorjake> Worth a try, might be more of a JGRPP thing
07:37:59 <peter1138> Would it be silly to (conditionally) filter the server list of incompatible servers?
07:41:21 <peter1138> It's probably not so much of a big list to be worth bothering with I suppose.
07:44:09 <reldred> oooh new grindhard video
07:44:22 <reldred> perfect for dinnertime watching
08:10:02 <reldred> yolo merge reversing trains yet?
08:10:47 <reldred> yolo merge drivethrough depots next?
08:36:45 <andythenorth> yolo merge Horse?
08:57:15 *** gelignite has joined #openttd
11:14:40 <pickpacket> iirc there's some pretty big obstacle to having diagonal roads in the game, but I can't recall *what* that obstacle is
11:15:17 <emperorjake> It's not that big, there's an almost-functional diagonal roads patch for JGRPP
11:16:18 <andythenorth> vehicle routing is the main one
11:16:25 <andythenorth> RVs don't move in a suitable way currently
11:16:48 <emperorjake> Graphically, some people dislike how the diagonal roads become narrower, but old SimCity did it so I don't see the issue
11:20:00 *** yiffgirl has joined #openttd
11:20:00 <yiffgirl> peter1138: peter1138: "mastodon, ok" oh is it you who runs that account? nice
11:37:27 <belajalilija> peter1138: weeks or months?
11:39:13 <mmtunligit> dont you know how this works? every request tacks on an extra week :)
11:39:34 <mmtunligit> law of the universe
11:40:08 <belajalilija> that just gives me more time to draw))
11:41:48 <peter1138> Cool, servers slowly updating to 15.3
11:52:46 <__abigail> emperorjake: #notmysimcity
11:55:29 <emperorjake> Well that's a cherry-picked screenshot which doesn't show any diagonal roads 😛
12:02:05 <peter1138> IMHO, there's a graphical issue with diagonal roads, they have to become much narrower to fit. Rail tracks skimp around that my having lots of space either side of the track in the X/Y direction, but even then the joins between X/Y and N/S/E/W can be a bit wonky.
12:06:11 <peter1138> Oh, you did say that :)
12:06:19 <emperorjake> I think the right way to do diagonal roads would be to add a roadtype flag for diagonal movement, and let roadtype NewGRFs add diagonal roads if they want. Let them worry about the graphics, the base game doesn't need to be shipped with diagonal roads.
12:12:56 <will_marshall_> i should dig up the diagonal level crossings patch and see how bad the bitrot is
12:13:57 <emperorjake> That thing is 20 years old, most of the game's been rewritten since then lol
12:14:10 *** redlink_kam has joined #openttd
12:14:10 <redlink_kam> __abigail: beautiful game
12:14:14 <will_marshall_> oh crap it is gonna literally be old enough to drink
12:18:45 <will_marshall_> I just read the phrase "Mini Integrated Nightly" and had a little proustian reverie. Teleported back to the family computer in the front room of the house in 2005 or whatever.
12:20:57 <LordAro> MiniIN was dead by the time i came along
12:23:11 <__abigail> Back before my sibling was born... heaven on earth
12:25:12 <will_marshall_> Conceptually, diagonal crossings are simple, road and rail already coexist happily in the map array, there's already special logic to handle adjacent crossings?
12:25:13 <emperorjake> MiniIN was my first patchpack haha
12:28:10 <will_marshall_> Well it's 15 years newer.
12:28:42 <will_marshall_> But it's based on the old patch by the looks?
12:28:44 <talltyler> I don’t suppose anyone has heard from Eddi in the years since he disappeared? I am left to assume the worst, unfortunately. 🙁
12:32:52 <will_marshall_> I'll pull it down and engage 100% of my brain to try and bring it up to date with current trunk because holy moly I am not an expert git user.
12:46:19 <peter1138> Probably ought to squash that :)
12:46:46 <mmtunligit> we should update CODINGSTYLE.md to reflect the comment alignment change
12:47:34 <peter1138> TBH I wasn't going to worry about it there because it's already so inconsistent, a separate PR to fix it all up is not unreasoable :)
12:47:37 <talltyler> Haha, emperorjake fell into the trap of assuming that literally any of the buttons that GitHub provides are useful, what a n00b mistake 😛
12:47:49 <talltyler> (I am making fun of GitHub, not you, in case it's not obvious)
12:48:21 <will_marshall_> Using any tools github has touched is like kicking yourself in the head.
12:48:26 <emperorjake> What do you mean, they are useful 😛
12:49:03 *** telumendur has quit IRC (Quit: User went offline on Discord a while ago)
12:49:12 <emperorjake> squashing is so easy now, you can just drag a thing in the desktop app, no complicated commands to remember 😉
12:49:32 <will_marshall_> Github desktop quietly breaks anything it touches, yes.
13:03:42 <_glx_> unless you also use it to sync your clone
14:05:34 *** asymptotically2 has quit IRC (Remote host closed the connection)
14:05:34 *** nebulabc_ has quit IRC (Write error: connection closed)
14:08:08 *** andythenorth has quit IRC (Quit: Connection closed for inactivity)
14:10:44 *** asymptotically2 has joined #openttd
14:24:49 *** nebulabc has joined #openttd
14:26:24 *** killingfrenzy8522 has joined #openttd
14:59:57 <emperorjake> Merged already? That was fast ❤️
15:00:43 <belajalilija> some good shit happening here recently
15:02:06 <peter1138> Small and to the point.
15:07:18 *** Wormnest has joined #openttd
15:07:41 <emperorjake> That's not a junction, it's a jam factory
15:11:38 <will_marshall_> I'm embarking on reimplementing diagonal road crossings, aiming to implement it for RoadTypes/RailTypes rather than for basesets. I'll use Eddie's PR as a starting point. Is there any reason not to extend the scope of the Action3 'cargo' 07 (like was done for Fences) to support returning more sprites vs defining a new 'cargo'? (for context: https://newgrf-specs.tt-
15:11:38 <will_marshall_> wiki.net/wiki/Action3/Railtypes#Level_crossing_sprites_.2807.29)
15:14:29 <will_marshall_> Appreciate I'm probably pre-emptively worrying about little details, but it feels like a place where I could easily find out there's some limitation or backwards compatibility implications to interpreting the result differently if the grf returns 20 sprites rather than 10.
15:23:16 <peter1138> 07 isn't used in vanilla, but is "reserved" for level crossing sprites.
15:23:42 <peter1138> I would say it's fair game to use it as makes sense.
15:31:57 <peter1138> Though I think we'd generally prefer if these sorts of features can be supported without NewGRFs too, but deal with that later.
15:57:44 <peter1138> 283 railtypes. Too damned many.
16:18:57 <_glx_> ok, removed all preview labels and re-enabled preview workflow
16:32:27 *** Flygon has quit IRC (Remote host closed the connection)
16:49:37 *** overjay has joined #openttd
16:49:38 <overjay> huh, so trains reversing feature got merged? nice nice
17:02:51 <_zephyris> What's the previous diagonal level crossing sprite implementation?
17:03:36 <_zephyris> Track overlay and lights/gates overlay?
17:16:11 <overjay> guess I've read github incorrectly
17:19:32 <will_marshall_> _zephyris: As per X/Y level crossings but with a special case for the double-track tiles in X and Y, so 4(?) extra ground sprites and 8 extra light sprites (NE, SE, SW, NW, left and right)
17:21:25 <will_marshall_> or is it 6 extra ground sprites, X road left/right track and Y road lower/upper track, then the 2 double-track tiles for X and Y.
17:21:32 <will_marshall_> That'd make more sense?
17:49:20 <_zephyris> Yeah, if you're going for the one combined sprite solution, it's 6 tiles (two vertical, one left and one right, two horizontal, one tip one bottom, and two double track, verical and horizontal).
17:50:02 <_zephyris> Though, 6 tiles, times 4 climates, plus snow and desert, plus normal and toyland town, times 3 for rail, monorail and maglev 😉
17:51:29 <peter1138> "one combined sprite" sounds problematic, though I don't know what we're looking at :)
18:03:02 <will_marshall_> looking at how level crossings are delivered in railtypes, I don't think climate comes into it?
18:04:12 <brickblock19280> You can use variables for snow aware tiles but I am not sure about sand
18:04:29 <brickblock19280> The climate can also be used in the switch
18:06:17 <will_marshall_> porting the old PR seems successful, going to go and reimplement the graphics and then mod my newgrf to provide the right set of sprites
18:06:48 <belajalilija> street running trains when?
18:07:01 <will_marshall_> not possible without doing horrible things to the map array 😛
18:07:26 <will_marshall_> i was musing as I was writing this what you'd need to do for level crossings over junctions
18:07:32 <peter1138> Hmm, the 15.3 announcement didn't come through to my Discord. Weird.
18:23:11 <talltyler> Do multi-track level crossings "just work" for diagonal crossings? The PR was built atop an old implementation of that feature, which I later finished and got merged 🙂
18:31:00 <_zephyris> will_marshall_: Remember to think about the base set support - climate certainly does come into it.
18:35:43 <audigex> talltyler: From the screenshot, it looks like it - the closest track is showing a "closed" graphic
18:38:40 <will_marshall_> talltyler: Yes!
18:39:04 <will_marshall_> _zephyris: Yeah, I'm seeing the branches inside DrawTile_Road.
18:50:57 <Borg> howdy... I need a hand about YAPF.. again ;)
18:51:09 <Borg> I have a case where stuff should work.. but it its not.
18:51:47 <Borg> whats the difference between Node and Segment really?
18:52:08 <Borg> transitions between nodes are transitions between paths?
18:57:14 <Borg> there is also SegmentCache.. with I have no clue how it works
19:11:22 <Borg> hmm yapf debug confirm its weird..
19:11:37 <Borg> first, path is evaluated.. the better one, cost around 6000
19:12:05 <Borg> the when train gets closer its grabs second one, cost around 8000
19:12:10 <Borg> and that path is selected..
19:19:37 <wensimehrp> belajalilija: In simutrans yes
19:24:24 <peter1138> I keep reading distinguish as extinguish... :o
19:28:24 <_glx_> I'll abuse #15435 to test previem
19:30:50 <Borg> hmm whats the different between OpenNode and ClosedNode?
19:31:13 <_glx_> open is not tested yet, closed is done
19:32:31 <Borg> so closed means.. it does NOT reach destination.. hits some sort of EOL?
19:33:24 <_glx_> closed means it has been evaluated, and its neighbours are added to open
19:34:31 <Borg> okey, I need to plug DEBUG() into FindPath() to see when evaluator kicks in
19:35:30 <Borg> it might not be a bug in YAPF per se.. but rather combination of routing restrictions.. im trying to influence paths based on previous signal (look_ahead)
19:35:39 <Borg> and.. this might have side effects w/ caching
19:35:52 <_glx_> hmm as I kinda expected, deployment is not added to the PR
19:52:27 *** Borg has quit IRC (Quit: brb switching irc client)
20:03:07 <Borg> okey... paths seems equal.. with is weird
20:10:49 <peter1138> Someone™ should tell rubidium42 it's safe in here again ;)
20:12:23 <Borg> dbg: [yapf] [56]: Node[5]: 4D6E (3407)
20:12:23 <Borg> dbg: [misc] [56]: pst 536F ?= 536F
20:12:23 <Borg> dbg: [misc] [56]: pst 536F ?= 536F
20:12:24 <Borg> dbg: [yapf] [56]: Node[6]: 4C70 (7211)
20:12:33 <Borg> okey.. I think completly failed to understand how m_cost works
20:13:00 <Borg> it seems its per node.. and node cumulative (backpropagation)
20:13:28 <Borg> so if A -> B have cost 1000 and B->C have cost 1000
20:13:37 <Borg> and train is aproaching A..
20:13:44 <Borg> it sees cost 1000 only, not 2000
20:14:09 *** andythenorth has joined #openttd
20:14:55 <Borg> thats why my fancy stuff doesnt work ;)
20:15:44 <peter1138> Hmm, didn't we get rid of highscore difficulty levels?
20:20:39 <andythenorth> did I have lunch?
20:21:26 *** x1vi_zi has quit IRC (Ping timeout: 480 seconds)
20:21:41 *** truebrain has quit IRC (Ping timeout: 480 seconds)
20:21:44 <andythenorth> should we just stop releasing to Steam?
20:21:52 <andythenorth> the community truly are a disgrace
20:22:10 *** yiffgirl has quit IRC (Ping timeout: 480 seconds)
20:22:20 *** redlink_kam has quit IRC (Ping timeout: 480 seconds)
20:22:25 <andythenorth> and I believe there's now a distributor of a related paid product who could take over the distribution of OpenTTD
20:22:31 *** yozora3 has quit IRC (Ping timeout: 480 seconds)
20:22:31 *** masterhellish has quit IRC (Ping timeout: 480 seconds)
20:22:33 *** andythenorth[d] has quit IRC (Ping timeout: 480 seconds)
20:22:33 *** bfgiant has quit IRC (Ping timeout: 480 seconds)
20:22:33 *** emperorjake has quit IRC (Ping timeout: 480 seconds)
20:22:33 *** jfkuayue has quit IRC (Ping timeout: 480 seconds)
20:22:33 *** _glx_ has quit IRC (Ping timeout: 480 seconds)
20:22:33 *** _zephyris has quit IRC (Ping timeout: 480 seconds)
20:22:33 *** mmtunligit has quit IRC (Ping timeout: 480 seconds)
20:22:35 *** fairyflossy has quit IRC (Ping timeout: 480 seconds)
20:22:35 *** belajalilija has quit IRC (Ping timeout: 480 seconds)
20:22:35 *** scrubbles3587 has quit IRC (Ping timeout: 480 seconds)
20:22:35 *** squirejames has quit IRC (Ping timeout: 480 seconds)
20:22:35 *** locosage has quit IRC (Ping timeout: 480 seconds)
20:22:36 *** talltyler has quit IRC (Ping timeout: 480 seconds)
20:22:38 *** wensimehrp has quit IRC (Ping timeout: 480 seconds)
20:22:38 *** will_marshall_ has quit IRC (Ping timeout: 480 seconds)
20:22:38 *** DorpsGek_vi has quit IRC (Ping timeout: 480 seconds)
20:22:38 *** ezot_ind_87922 has quit IRC (Remote host closed the connection)
20:22:38 *** notluke2578 has quit IRC (Remote host closed the connection)
20:22:38 *** hugolarsson0672 has quit IRC (Write error: connection closed)
20:22:38 *** overjay has quit IRC (Read error: Connection reset by peer)
20:22:38 *** _jgr_ has quit IRC (Read error: Connection reset by peer)
20:22:38 *** quast65 has quit IRC (Read error: Connection reset by peer)
20:22:38 *** brickblock19280 has quit IRC (Write error: connection closed)
20:22:38 *** davidxn has quit IRC (Write error: connection closed)
20:22:38 *** audigex has quit IRC (Write error: connection closed)
20:22:38 *** kaibaneddy has quit IRC (Write error: connection closed)
20:22:38 *** definitelynotcheese_ has quit IRC (Write error: connection closed)
20:22:38 *** rito12_13 has quit IRC (Write error: connection closed)
20:22:38 *** northalicious has quit IRC (Write error: connection closed)
20:22:38 *** __abigail has quit IRC (Write error: connection closed)
20:22:38 *** mnhebi has quit IRC (Write error: connection closed)
20:22:38 *** killingfrenzy8522 has quit IRC (Remote host closed the connection)
20:22:38 *** teahigh has quit IRC (Remote host closed the connection)
20:22:38 *** tabytac has quit IRC (Remote host closed the connection)
20:22:38 *** michi_cc[d] has quit IRC (Write error: connection closed)
20:22:38 *** exceptik has quit IRC (Write error: connection closed)
20:22:38 *** spacemaniac[d] has quit IRC (Write error: connection closed)
20:22:38 *** kuhnovic has quit IRC (Write error: connection closed)
20:22:38 *** ahyangyi has quit IRC (Write error: connection closed)
20:22:38 *** reldred has quit IRC (Write error: connection closed)
20:22:57 *** DorpsGek_vi has joined #openttd
20:24:31 <andythenorth> RIP andythenorth[d]
20:28:02 <peter1138> Why do you think "we" care?
20:36:36 <_glx_> I can see IRC messages on discord so
20:38:55 <_glx_> ah but I missed everything was said after andy joined 🙂
20:40:19 <andythenorth> didn't miss much :)
20:57:04 <peter1138> `SettingsProfile` is a bit of a legacy.
21:08:27 <Borg> m_first = (parent) ? parent->m_first : this;
21:08:36 <Borg> so, this in templated struct is now allowed?
21:09:04 <Borg> oh screw it.. NULL will be fine
21:30:01 *** will_marshall_ has joined #openttd
21:30:01 <will_marshall_> Easy to see why previous attempts to get this patch over the line stalled out now.
21:31:06 <will_marshall_> Graphics are hard on the diagonal entries, there's some glaring UX issues caused by the fact that there's a lot of configurations that are just invalid
21:31:55 <will_marshall_> Needs a bit of thought in how the drawing code is written to address things like this.
21:32:52 <will_marshall_> There's no reason to build this but it will happen: towns can build roads after all.
21:34:20 *** talltyler has joined #openttd
21:34:20 <talltyler> Hmm, we could prevent towns from building roads over diagonal rails. Or build multiple road tiles at once until a valid stopping point is reached.
21:34:40 <talltyler> The latter could let players cheese town growth though, so maybe that's a bad idea.
21:34:50 <will_marshall_> Better to provide graceful graphics somehow. At least stopping the 2nd set of lights/gates being drawn isn't too bad.
21:34:56 <talltyler> (Although the player can always just build roads for towns...)
21:35:27 <will_marshall_> There's already an adjacency check for lights/gates, it just needs to understand what it means if there's bare track on that tile instead of another crossing.
21:35:48 <talltyler> Oh, definitely the graphics need to allow for this to look ugly but not broken. Lord knows we see enough roads halfway across multi-track railways in Discord channel #openttd-screenshots
21:36:28 <will_marshall_> Here's a crazy thought...
21:37:12 <will_marshall_> the 90 degree road graphics and the track graphics only *barely* overlap and only because of fences.
21:37:40 <will_marshall_> Tantalising (yes I know these aren't vanilla graphics don't yell at me)
21:37:58 *** mmtunligit has joined #openttd
21:37:58 <mmtunligit> if you have a lot of diagonal tracks in a row and try to draw two roads that intersect in the middle what happens? i could see that being a weird edge case
21:38:38 <will_marshall_> that's the thing right
21:38:59 <will_marshall_> All of these are valid (just I've not made the graphics work)
21:39:07 *** reldred has joined #openttd
21:39:07 <reldred> Time to post these bad boys again
21:39:15 *** tabytac has joined #openttd
21:39:15 <tabytac> will_marshall_: The only issue is also many road sets are wider and there isn't enough space for a diagonal rail on the same tile
21:39:42 <will_marshall_> tabytac: yeah it'd need to be a cooperative thing anyway since it'd still be road and rail occupying the same tile and therefore it'd be a level crossing etc.
21:39:57 <will_marshall_> that's a distraction
21:40:03 <mmtunligit> lets get the basic case out the door first
21:40:05 <will_marshall_> let's forget i said anything 🙂
21:40:52 <will_marshall_> i guess any situation where you have 2x rails on a level crossing you never have a need to draw lights
21:40:54 <will_marshall_> so that's easy
21:43:13 <Borg> holy moly.. I start to get it..
21:43:14 <Borg> dbg: [yapf] [56]: Node[4]: 5C6F (3307)
21:43:15 <Borg> dbg: [yapf] [56]: Node[5]: 4D6E (3407)
21:43:15 <Borg> dbg: [misc] [56]: pst 536F ?= 536F
21:43:15 <Borg> dbg: [misc] [56]: pst 536F ?= 536F
21:43:56 <Borg> so.. Node contain Segment. .and segment is a segment of rail tiles w/ single path..
21:44:27 <Borg> joins allowed.. but not junctions..
21:44:48 <will_marshall_> Oh it's an optimization of plain line path segments?
21:45:22 <Borg> dunno.. Im trying to understand it.. because my fancy stuff doesnt work
21:46:41 <Borg> dbg: [yapf] [56]: Node[3]: 526F (3307)
21:46:44 <Borg> dbg: [yapf] [56]: Node[4]: 5C6F (3307)
21:46:44 <Borg> dbg: [yapf] [56]: Node[5]: 4D6E (3407)
21:46:44 <Borg> dbg: [misc] [56]: pst 536F ?= 536F
21:46:44 <Borg> dbg: [misc] [56]: pst 536F ?= 536F
21:46:46 <Borg> dbg: [yapf] [56]: Node[6]: 526F (10891)
21:46:49 <Borg> dbg: [yapf] [56]: Node[7]: 526F (11175)
21:46:51 <Borg> dbg: [yapf] [56]: Node[8]: 4D6E (12675)
21:46:54 <Borg> sorry for little spamming
21:47:58 <Borg> im not really sure why I see 526F node 2 times..
21:48:14 <Borg> 526E would be fine.. its second path
21:48:52 <mmtunligit> will_marshall_: Sure my you only ever need to draw lights on a side without a diagonal
21:50:30 <Borg> okey I see it 3 segments (2 paths)
21:51:09 <Borg> so im totally puzzled.. why Node[5] isnt choosed
21:53:27 <will_marshall_> mmtunligit: yeah, i'm going to doodle a table to help me keep this straight
21:59:38 <talltyler> andythenorth: When did you fix FIRS industry animations? I'm playing GIST, which seems to be a FIRS fork created before you fixed it. Having industries drawn on-screen tanks my frame rate, and bulldozing them fixes it instantly. 🙂
22:00:46 <talltyler> Er, one of those seems to be paused somehow, try these
22:03:22 <will_marshall_> [side quest added]
22:03:34 <talltyler> Seems they're working in NML, not sure if it's actually your old code or if they've created the same bug from first principles 😄
22:04:28 <peter1138> They probably took the output of FIRS and decided that was the best way to edit it...
22:04:57 <peter1138> There's some kind of belief that FIRS is an industry framework, or something.
22:09:19 <talltyler> Yep, I have seen people discuss exactly that 🙂
22:09:31 <_jgr_> It doesn't look liek they've copied the FIRS output from what I can see. They have copied the PNGs though.
22:11:12 <talltyler> Hmm, probably worth reproducing in a stable build instead of my driving-backwards debug build 🙂
22:11:48 <_jgr_> Yeah, there's no point measuring anything in unoptimised builds
22:18:56 <peter1138> talltyler, I mean, if you also want to benchmark that build... :)
22:21:26 <talltyler> I need to isolate variables, but even after deleting the industries I started getting poor performance again. So I swapped out the station from WINS to CHIPS, and it's improved. How often can station tiles get redrawn...?
22:22:15 <talltyler> The right way to benchmark backwards trains is probably a no-NewGRF or NewGRF-lite build
22:22:31 <talltyler> (small train set, small industry set without animated tiles, etc)
22:22:51 <_jgr_> For redrawing issues first port of call is to turn on the dirty block colouring
22:27:58 <talltyler> It _really_ does not like multiple trains moving within this WINS station. I wonder if it's determining whether to draw the eyecandy dummy tracks each time? (which would likely be an NML switch that checks the adjacent tile)
22:31:53 <talltyler> wensimehrp: , have you gotten any reports like this before?
22:32:35 <peter1138> Can't really see anything from a screenshot, it's the changing that matters :)
22:36:13 <_jgr_> The sprite chain for that station does look pretty complicated but it shouldn't cause that much of a slowdown
22:38:24 <_jgr_> Did you do a normal build for this or is it still a debug/unoptimised one?
22:39:26 <talltyler> Still debug. I captured a video, it's uploading to YouTube currently
22:40:12 <talltyler> No idea if it's useful to anyone 🙂
22:41:21 *** Borg has quit IRC (Quit: leaving)
22:41:50 <_jgr_> The industry coal lifts are being refreshed even though they aren't visually moving
22:42:25 <_jgr_> That isn't that surprising as starting/stopping the animation is much more faff/complexity
22:44:06 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
22:44:13 <_jgr_> The performance problem is likely to mostly or completely disappear with a build with optimisation turned on
22:45:38 <talltyler> I will patiently await a YOLO approval then 😛
22:46:36 <talltyler> (although I don't imagine Nightly is much faster)
22:47:13 <_jgr_> Nightly builds are surely normal release builds?
22:47:49 <_jgr_> Albeit with asserts still enabled
22:48:17 <_jgr_> I am bit confused about what you're trying to measure here
22:55:38 <talltyler> Dunno, I am being reminded why I am a junior developer tonight 😉
22:56:28 <_jgr_> Personally, I would recommend not using the "Debug" cmake build type at all, certainly not for any kind of performance-related work
22:57:12 <talltyler> Yeah. I have not tried compiling any other type locally in years. 🙂
22:57:22 <_jgr_> Or at least make it usable by adding `-Og` or whatever the MSVC equivalent of that is
22:58:54 <_jgr_> The `RelWithDebug` build type should make a usable build out of the box
22:59:59 <peter1138> Is what I always use.
23:00:38 <peter1138> Well, unless there's a particularly awkward bit of debugging to follow that can't work without a debug build.
continue to next day ⏵
