IRC logs for #openttd on OFTC at 2026-09-16
⏴ go to previous day
00:21:22 *** MinchinWeb[m] has quit IRC (Ping timeout: 480 seconds)
00:21:45 *** MinchinWeb[m] has joined #openttd
02:34:20 *** MinchinWeb[m] has quit IRC (Read error: Connection reset by peer)
02:34:37 *** MinchinWeb[m] has joined #openttd
02:41:16 *** gebik4544 has joined #openttd
02:41:16 <gebik4544> karelmacha712: add a mechanism to go on other player tracks and you got me sold
02:41:59 <gebik4544> karelmacha712: also do you have a discord server for it?
02:43:11 <gebik4544> karelmacha712: also if you are going to make it work with CZTR, their discord might be interested
02:43:47 <Izzybee> What's the minimum OpenGL version for hardware acceleration to work?
02:55:56 <gebik4544> karelmacha712: beware it may be malicious as it is written by half crazy dude on weed
02:56:13 <gebik4544> actually gonna run full check now
02:58:59 <belajalilija> gebik4544: this is known
02:59:20 <belajalilija> its slop too iirc
03:11:17 *** MinchinWeb[m] has quit IRC (Ping timeout: 480 seconds)
03:12:10 *** MinchinWeb[m] has joined #openttd
03:13:45 <gebik4544> belajalilija: some may not know so best to say
03:18:53 <Izzybee> Ah, guess it's elsewhere in the stack, thanks.
03:19:49 *** MinchinWeb[m] has quit IRC (Read error: Connection reset by peer)
03:20:05 *** MinchinWeb[m] has joined #openttd
03:27:57 *** Phileman has joined #openttd
03:35:37 *** Philemon has quit IRC (Ping timeout: 480 seconds)
03:59:45 <Izzybee> been trying to make the game run better but I'm having to make SDL use libglx_amber which doesn't actually support the 7600GS in my G5... without, it bails out in mesa. will continue investigating.
05:09:56 <rito12_13> belajalilija: It is fun that forwarded messages don't cross the bridge
05:12:24 <Izzybee> oh there are actually missing messages? good I thought I was going insane
05:17:51 <rito12_13> Yeah it is long with a youtube link and two zip files
05:56:12 <andythenorth[d]> dunno about the names of these
05:56:24 <andythenorth[d]> their main product is 'process chemicals'
05:56:53 <andythenorth[d]> and there are 3 of them to allow many many combinations of cargos to make process chemicals
05:57:47 <andythenorth[d]> so they include the name 'chemicals' to show they're all similar
06:06:54 <peter1138> I've heard people (not here) complain about having to conform to 'trivial' things like code style and commit message style.
06:07:29 <peter1138> GitHub doesn't make it easy to tidy small things up.
06:13:50 *** jfs_ has quit IRC (Ping timeout: 480 seconds)
06:23:25 <reldred> If people are enthusiastic about actually contributing, taking ownership of their change (rather than just lobbing it over the fence yolo style), then they usually dont seem to mind conforming to the 'trivial' stuff
06:23:34 *** Phileman is now known as Philemon
06:24:44 <reldred> i think thats one of the things I find most offensive about the AI pr's, you couldn't be bothered to put your own heart and soul into it, i can tell you won't be bothered to fix bugs, clean up the code, answer questions, etc.
06:26:15 <reldred> (moral and ethical objections to AI asside)
06:26:33 <Izzybee> Ah, I love using cursed old computers. "Error: Unable to load driver 'sdl-opengl'. The error was: S"
06:26:47 <Izzybee> Thank you for the detailed error message, SDL.
06:26:58 <reldred> the S is for sugondese
06:27:15 <peter1138> Did we manage to convert a char to std::string instead of a char* to std::string...
06:28:00 <peter1138> Izzybee, before you spend too much effort "hardware acceleration" in OpenTTD isn't particularly game-changing.
06:28:42 <reldred> if anything I find on a lot of systems it doesnt work terribly well
06:28:58 <reldred> (but on others its mandatory for a smooth experience... go figure)
06:29:40 <Izzybee> I was hoping to get it semi-working at least to see if the framerate issues are to do with the speed of the system/GPU connection. Apparently AGP is entirely broken on PPC so it's downgraded to a fast PCI connection.
06:30:41 <peter1138> Mmm, that is the basically the only thing it accelerates, so it might.
06:30:45 <Izzybee> Figured if the sprites could live in GPU memory it might reduce the amount of bandwidth needed there.
06:31:17 <peter1138> THe thing that's accelerated is copying the rendered buffer to the display buffer.
06:31:53 <Izzybee> Hm. Well, maybe it'd help anyway, because the system isn't CPU constrained when it's struggling.
06:32:50 <Izzybee> But hey, the game no longer crashes with mainline mesa after I added an awful patch, so that's a win, I think
06:36:10 <emperorjake> Great, now I can finish the vactrain set
06:36:46 <peter1138> emperorjake, oh dear, can you rebase against update master too please? There's too many commits merged since the PR for our CI, which is why it's failing.
06:39:09 <emperorjake> Sure, lemme figure out how to do that
06:39:46 <peter1138> git fetch <upstream> master:master` then `git rebase master fix-dualhead-purchase-capacity`
06:39:55 <peter1138> Oops, I missed a backtick :(
06:40:11 <peter1138> `git fetch <upstream> master:master` then `git rebase master fix-dualhead-purchase-capacity`
06:59:45 <emperorjake> Git wasn't cooperating but I think I did it right
07:12:42 <peter1138> We'll now break all the NewGRFs that fudged the value when in a purchase list... :D
07:14:03 *** karelmacha712 has joined #openttd
07:14:04 <karelmacha712> gebik4544: You have no proof, just idle talk
07:14:40 <gebik4544> karelmacha712: your file in the zip about your story
07:39:47 <karelmacha712> What is going on here? I am doing a decouple—something that was considered too difficult to do for ten years. Take a look at my thread where I’m not even allowed to post. The alpha version in the developers' forum is being judged by people who don't even know that an alpha is just a showcase. They insult me, and I am not allowed to say anything back.Then a troll shows up there. I said goodbye to
07:39:47 <karelmacha712> him politely, but he keeps going and sending a year-old save file. I'm telling him to leave. When I report to the moderator that he is spamming questions, the moderator views it as me insulting the player. That thread is not my shame; it’s a showcase of how you treat me. A parade of malice. You’ve talked a bunch of nonsense there and silenced me. I couldn't even show the page of my decouple on tt-
07:39:47 <karelmacha712> forums and the official OpenTTD Discord to my friends. I don't have a "donate to me" button, so I don't have to put up with insults.
07:43:13 <karelmacha712> gebik4544: about malicious code . what is malicious, you have proof? hu is half crazy dude!!!!! how dare you?
07:43:58 <brickblock19280> andythenorth[d]: Agricultural supplies?
07:45:55 <gebik4544> karelmacha712: Discord Moderator
07:52:21 <peter1138> _zephyris, "yes". There's some fiddly bits going to make the OverrideManager stuff work with the split of trees and treetiles.
07:57:08 <emperorjake> karelmacha712: Look, I gave you a chance but if all you're going to do in this community is cause drama and strife, then that's not worth having a half-baked shunting patch.
08:05:03 <andythenorth[d]> we are not equipped to safely help people with mental health issue
08:05:12 *** MinchinWeb[m] has quit IRC (Remote host closed the connection)
08:05:25 *** MinchinWeb[m] has joined #openttd
08:10:40 * peter1138 waits for Teams to need me to login again.
08:24:14 <will_marshall_> Yesterday I had to log in to a vendor's dev tools site about 400 times and at the end of it I managed to get a spinning triangle on the screen like baby's first OpenGL 1.1 demo and I was so relieved that I went and napped on the sofa.
08:25:05 <will_marshall_> Now I just need to get the two key vendor's technologies to work together and that seems like it might be mission impossible.
08:25:44 <LordAro> also seems impossible to fix
08:26:47 <will_marshall_> I'm about to learn an uncomfortable amount of information about how MSVC projects/solutions are laid out in XML.
08:27:09 <LordAro> it's not uncomfortable until you have to learn XSLT to deal with it
08:28:23 <will_marshall_> There does come a point where I just write a whiny forum post to Epic Games about it.
08:28:28 <will_marshall_> Rather than try to fix it myself.
08:28:44 <will_marshall_> Being in the OpenTTD community truly prepared me for the world of work.
08:35:41 <peter1138> LordAro, it's DNS isn't it?
08:36:44 *** Philemon has quit IRC (Quit: Connection rest by beer.)
08:36:47 *** Philemon has joined #openttd
08:36:48 <LordAro> procmon didn't show any DNS networking events :p
08:37:00 <LordAro> but I suppose I can't conclusively rule it out
08:38:33 <Izzybee> From memory, project files aren't awful, fairly intuitive. Not too bad if you need to modify them to compile with mono and your only tool is vim, I'd say.
08:39:28 <will_marshall_> I have a project file that makes the weird vendor's debugging tool work and I have a project file that doesn't work and I need to sniff between them until I figure out what special sauce needs to be injected. Maybe.
08:39:36 <will_marshall_> I actually have like 3 other things to be doing today as well.
08:51:24 <will_marshall_> 3 things in parallel, the backlog is infinite.
08:52:59 <peter1138> Today I found my manager likes to delete the backlog because it's too cluttered.
08:53:56 <will_marshall_> We used to have a manager who would go to some lengths to hide incoming tickets from developers in the fear that they might go and fix them instead of the things that the manager had deemed hi-pri.
08:54:28 <will_marshall_> They sent all incoming tickets to a triage space, ostensibly for deduplication.
08:55:44 <will_marshall_> You see, developers are just like weird perfectionists and they can't possibly be trusted to fix root causes instead of nth-order effects.
08:55:53 <peter1138> I always argue that if a developer "needs a break" they should be able to pick up a small task that they will know won't take very long.
08:56:24 <LordAro> "that they will know won't take very long"
08:56:49 * LordAro looks at his... 15 open MRs
08:57:13 <peter1138> "weird perfectionists" eh.
08:57:57 <will_marshall_> Let's go with that instead of the real reasoning. Find me in a pub one day.
09:07:19 *** SigHunter has quit IRC (Remote host closed the connection)
09:07:21 <LordAro> the other day i spent far too long correcting various SQL strings to correctly use single quotes instead of double quotes
09:17:28 <Izzybee> Do vehicles not age in a depot?
09:23:52 *** mitchcz_81202 has quit IRC (Quit: User went offline on Discord a while ago)
09:39:54 *** SigHunter has joined #openttd
09:51:39 <kaibaneddy> iirc they are serviced (to max reliability) on leaving the depot, or possibly constantly while in the depot. which is a change from ttd where vehicles were only serviced on entering, and so would exit with 0 reliability if they'd been parked up for a while.
09:56:20 <peter1138> Nope, reliability is still only set on entering the depot.
10:22:57 <peter1138> Hmm, is it MIDI music day...
10:25:17 <peter1138> I probably ought to stop just bumping the minor version.
10:27:10 <andythenorth[d]> will_marshall_: oh you worked for me at some point?
10:56:43 <peter1138> Hmm, I can't mock the extension method, but maybe I can mock the methods the extension method calls...
11:00:48 <kaibaneddy> peter1138: Interesting re reliability, I thought that was something changed years ago
11:01:08 <kaibaneddy> Mind you, who even plays with breakdowns on?
11:04:32 <peter1138> I had/ve a patch to make that a percentage instead of off/reduced/on.
12:05:36 <talltyler> Time for work, but here's something to get started.
12:29:26 <kuhnovic> Oh wow, I didn't expect to double ended ships to make it into version 16 ❤️
12:42:06 <emperorjake> I need to draw a Sydney Freshwater class ferry now 😅
13:05:58 <kuhnovic> I guess we do need a new nml release
13:09:19 <peter1138> Not a blocker for a beta release
13:16:42 <peter1138> #16005 (or the other fix for #15824) would nice to have, that crash has come up a few times.
14:11:03 <rito12_13> Newmap is pretty fun
14:13:17 <belajalilija> rito12_13: this is in vanilla now?
14:14:34 <rito12_13> belajalilija: No. (at least not yet)
14:14:36 <LordAro> newmap has been "coming soon" for about 10 years
14:15:36 <andythenorth[d]> now we have cargo icons, could we have a map overlay that shows where cargos are being produced?
14:15:47 <andythenorth[d]> "if you code it andythenorth[d] we can"
14:19:23 <belajalilija> peter1138: "ive attached an ai generated mockup but it has done some weird things"
14:20:36 <belajalilija> the solution would be some like like waypoints
14:20:52 <belajalilija> or for more advanced players, maybe a type of signal that restricts certain trains
14:21:52 <peter1138[d]> Where's "Platform 1" anyway?
14:23:06 *** squirejames has joined #openttd
14:23:06 <squirejames> Wherever BR arbitrarily decides it is. And it might be Platform 1A or Platform 1B
14:24:15 <peter1138> Urgh, in Outlook, how the fuck do I view my own reply to a chain?
14:24:53 <belajalilija> peter1138[d]: isnt it the one closest to the middle of the town?
14:25:34 <emperorjake> Here in NSW it's the one that goes towards Central
14:26:42 <squirejames> I was going to say, it depends on the railway. Tends to be the "Up" platform, but "Up" can mean anything again depending on the railway, as sure most railways in the UK went "up to London" but, Scottish ones did not, branch lines did not etc. so, yeaaaaah
14:26:49 <squirejames> Asking railways to make sense?
14:27:17 <squirejames> belajalilija: is an AI generated mock up, a Slopup?
14:32:20 <Flygon> The amount of times I've seen people ask for clarification on Up/Down down here in Melbourne hahaha.
14:44:42 *** __abigail has joined #openttd
14:44:44 <andythenorth[d]> I'm very uncertain I could prompt GPT faster than I could photoshop an OpenTTD mockup
14:44:59 <andythenorth[d]> nearly all GPT images are shit
14:46:44 <belajalilija> andythenorth[d]: the best part is, it doesnt even need to look like openttd, it could just be something thrown together really quick in MS paint
14:47:45 <__abigail> I am looking forward to being in Sydney again
14:47:45 <__abigail> Albeit for an unfortunate occasion given the K sets will have their final run
14:48:59 <talltyler> Bonus points to anyone who can name the UK service pattern that can start and end at a Platform 0
14:54:12 <LordAro> Kings Cross to Leeds, I'd say
14:54:21 <LordAro> (I have the wikipedia page open)
14:54:48 <LordAro> though I'd imagine it'd be a very rare occasion that such a service used either platform
14:58:17 <emperorjake> __abigail: Ah yes, Lidcombe
14:58:49 <talltyler> Leeds is one of the endpoints I had in mind! Didn't realize Kings Cross had a 0 too. 🙂
14:59:47 <LordAro> talltyler: Then Doncaster or Bradford then :p
15:02:25 <talltyler> You have either an encyclopedic knowledge of the UK rail network, or there's a Wikipedia page I don't know about 😛
15:02:36 <talltyler> Bradford was what I had in mind, didn't know about Doncaster either
15:02:59 *** notluke2578 has joined #openttd
15:02:59 <notluke2578> Cardiff central comes to mind
15:04:56 <LordAro> talltyler: needs more Geoff Marshall
15:06:43 <talltyler> Conspiracy theory: LordAro is undercover Geoff Marshall
15:43:02 *** jinks has quit IRC (Read error: Connection reset by peer)
15:43:02 *** jinks_ is now known as jinks
15:58:50 *** jinks has quit IRC (Remote host closed the connection)
15:58:50 *** jinks_ is now known as jinks
16:00:43 <Flygon> I'm afraid to ask what happens when we need to go below platform 0 in numbering
16:00:55 <Flygon> Do we start going backwards into A, B, C, D etc? :P
16:08:57 <_jgr_> Works for Gare de Lyon, which has a rather fun numbering scheme
16:13:01 *** gelignite has joined #openttd
16:14:54 <Flygon> I didn't expect to accidentally actually discover the solution used IRL hahaha
16:18:21 <mmtunligit> also used for infill platforms sometimes, secaucus junction upper has platforms 2, A, B, and 3, and then ewark penn has 1-5, A, H, G
17:20:59 <peter1138> Hmm, how bad is a std::map...
17:30:18 <Rubidium> it depends on what you're looking for
17:35:26 <Rubidium> it ought to have a known/stable iteration order, which std::unordered_map obviously doesn't have
18:18:59 <mmtunligit> its good enough for collections
19:11:18 <peter1138> Which PR had the std::vector EndianReader/Writer...
20:24:15 *** Borg has quit IRC (Quit: leaving)
20:31:39 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
20:31:41 *** WormnestAndroid has joined #openttd
20:31:44 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
20:31:45 *** WormnestAndroid has joined #openttd
20:31:48 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
20:31:59 *** WormnestAndroid has joined #openttd
21:02:47 *** MinchinWeb[m] has quit IRC (Ping timeout: 480 seconds)
21:03:49 *** MinchinWeb[m] has joined #openttd
21:18:48 <peter1138> > free(): invalid next size (fast)
21:28:00 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
21:33:50 *** jfs_ has quit IRC (Ping timeout: 480 seconds)
22:33:12 <peter1138> Is there still a packet size limit?
22:34:09 <peter1138> Or more precisely, something that could/should limit the size of the buffer of EndianReader/EndianWriter?
22:59:05 *** Flygon has quit IRC (Remote host closed the connection)
continue to next day ⏵
