IRC logs for #openttd on OFTC at 2023-06-01
⏴ go to previous day
02:41:06 *** debdog has quit IRC (Ping timeout: 480 seconds)
02:50:00 *** D-HUND is now known as debdog
03:00:05 *** herms has quit IRC (Quit: bye)
04:30:52 *** Mek_ has quit IRC (Ping timeout: 480 seconds)
05:26:27 *** keikoz has quit IRC (Ping timeout: 480 seconds)
06:11:27 <pickpacket> a lot of vitriol in the thread about shares buying. From very few users.
06:11:59 <pickpacket> Does JGR have much changes that would be good to upstream to vanilla?
06:26:56 <pickpacket> Is there anything I can do to help upstream them?
06:28:16 <andythenorth> I think they tend to get upstreamed when they scratch someone's itch
06:28:39 <andythenorth> there's no systematic approach to it
06:29:33 <andythenorth> also some of the headline atttractions wouldn't necessarily be good for vanilla
06:29:53 <andythenorth> it's a completely viable strategy to have 2 tracks for the client
06:30:33 <andythenorth> there are a handful of forum posters who see JGRPP as a complete vindication of their view that vanilla is a disaster
06:30:48 <andythenorth> whereas actually it's a sign of the success of JGRPP
06:46:50 <LordAro> "For what it's worth, I've been playing OpenTTD since version 0.8.0"
06:46:56 <LordAro> ...that version does not exist
06:49:02 <pickpacket> meh. Memory isn't perfect
06:49:25 <pickpacket> I should be off to work soon... Just not very motivated
06:49:42 <pickpacket> if I'd had the work computer at home I would have wfh today
06:51:54 <pickpacket> Rubidium_: yes? The point you're making goes above my head :)
06:52:24 <LordAro> JGRPP 0.8.0 was a thing that existed
06:53:13 <Rubidium_> so, he never played OpenTTD just JGRPP since version 0.8.0 of it
06:53:28 <pickpacket> or he's misremembering the version number
06:53:56 <pickpacket> no big deal either way since the shares have worked the same way in both iiuc
07:01:38 <Rubidium_> s/worked/been broken/ :D
07:02:31 <pickpacket> Rubidium_: technically it's *worked*, just not in a good way
07:02:59 <TrueBrain> nothing beats the smell of splitting hairs in the morning 😛
07:04:17 <pickpacket> bah. I called in sick for the rest of the week. Feeling a bit queasy about work -- at this year's salary revision I was told by my boss that my previous boss had said that I didn't live up to expectations. Not only is it the first time in my career that I've heard that, but he also never told me when he *was* my boss
07:04:40 <pickpacket> kind of trying to figure out how to deal with it, honestly.
07:06:50 <LordAro> i searched GH for "shares"
07:07:03 <TrueBrain> owh wow .. I knew it was bad, just not this bad 🙂
07:07:10 <TrueBrain> that are a lot of tickets
07:07:35 <TrueBrain> the ticket / line ratio of this feature are a bit out of whack 😄
07:10:25 <EmperorJake> Hold up, there was PBS before 0.4.5? I never knew, my first version was 0.4.8 so that was just before my time
07:10:37 <TrueBrain> it was ..... not as good 😛
07:10:56 <TrueBrain> and we overrushed it in the codebase ... that was a mistake 😄
07:11:08 <TrueBrain> "fix the bugs when they show up" attitude
07:12:02 <EmperorJake> I know there was a version of PBS in MiniIN which I never got to work, I wonder if it's the same as that. might have to dig up the archives lol
07:12:24 <TrueBrain> owh, MiniIN ... like JGRPP, but smaller 😛
07:25:12 * pickpacket wants to understand pre-signals, but doesn't yet
07:38:57 <andythenorth> pickpacket: is it a tech company that had unrealistic growth projections
07:41:24 <TrueBrain> now .... how to actually test this in a sane way ... hmmm ... now that is the interesting question ...
07:47:47 <pickpacket> andythenorth: it's a tech company with realistic growth projections, actually. I started there a year ago today, and with my (then) 11 year career I for once decided to let go of my impostor syndrome and anxiety and relax at work. And well... that backfired
07:48:58 <TrueBrain> owh, meh, the website ofc has hooks to rebuild every night .. hmmm .. I kinda forgot 😄
07:58:31 <TrueBrain> right, in theory this should work .. but how do you find out! 😄
08:13:22 <pickpacket> alright, seeing as I have a couple of days off now I may have a look at how to add beards to the CEO faces
08:13:46 <pickpacket> where do I start? Adding things to the save file?
08:15:16 <dP> LordAroviaGitHub: pretty normal for any feature, half those "attemts" are merged, and lots of ideas were just rejected with little to no discussion
08:15:29 <dP> there are 830 issues mentioning trains, let's remove trains! :p
08:16:13 <LordAro> dP: some of them, by no means "half" though
08:16:34 <LordAro> but the point was to illustrate there's always been a lot of discussion around "this is bad"
08:17:01 <dP> same as any other feature :p
08:21:54 <LordAro> doubtful for a single small feature like that
08:22:21 <LordAro> "trains" is not a "feature" in the same way that "shares" are
08:22:47 <pickpacket> Okay! I have an offer for all of you: practice the process of onboarding a new dev :D i.e. I'd like a bit of hand-holding in trying to add this feature. Been a very long time since I wrote anything in C and I've never written C++. But that'll probably not be a problem. Getting to know the codebase may, though
08:23:05 <pickpacket> and figuring out how to run all the tests
08:24:31 <TrueBrain> pickpacket: follow the `COMPILING.md` 🙂
08:28:06 <pickpacket> TrueBrain: I've compiled it a lot of times but I can't see that COMPILING.md says anything about running tests
08:28:31 <TrueBrain> did Rubidium not update that? Pfff ... but yeah, `ctest .` or `make test` works for most of the time 🙂
08:29:10 <Rubidium_> TrueBrain: I haven't changed anything about the way to run tests
08:29:13 <LordAro> tests are extremely limited and almost certainly won't be relevant to anything related to faces
08:29:26 <TrueBrain> Rubidium_: except for the fact you ... added them? 😛
08:29:35 <TrueBrain> dunno .. I call that a pretty big change 😛
08:29:54 <LordAro> TrueBrain: pretty big add, not a pretty big change :p
08:29:54 <pickpacket> LordAro: that makes things both easier and harder for me!
08:30:38 <Rubidium_> TrueBrain: I added unit tests, but regression tests were already there and unit tests use the exact same way to call/run them as the regression tests.
08:31:00 <TrueBrain> that is fair too; just nobody actually ran regression tests locally .. we just waited for the CI 😄
08:32:24 <dP> CI needs a button "it's correct, replace the result"
08:34:00 <TrueBrain> still need a way to test if the release pipeline actually works ... let's first make the infra for it 🙂
08:44:20 <TrueBrain> lol, cool, so I am using Pulumi to provision all our infra, including Cloudflare and GitHub .. so I can rotate secrets easier etc 😛 But what I didn't know, that you can set a secret on GitHub by fetching the public key, encrypting the value with it, and send that to them .. so even if you look on the wire, you can't see the secret. That is nice.
08:44:44 <TrueBrain> first provider I have seen that allows that .. AWS, etc, doesn't seem to have that .. GitHub is ahead of its time? 😄
08:46:27 <TrueBrain> too bad the example in pulumi's documentation doesn't actually tell you how to encrypt it .. just how to get the key .. right. Let's find out!
09:04:40 <dP> hm, is tt-forums deleting private messages? I can't find where did they go
09:07:06 <JGR> pickpacket: If you see something you like, feel free to adapt and PR it
09:08:42 <dP> oh, lol, I have 2 forum accounts apparently
09:10:14 <pickpacket> JGR: haven't played your patch so far :) If/when I'm good enough at working with the code I may take requests to upstream things from it
09:12:02 <pickpacket> I need some pointers on save file format. Specifically where and how to add new things to a savefile
09:12:23 <TrueBrain> I suggest you first start with making the actual GUI things, that it is actually functional
09:12:26 <TrueBrain> the saveload comes after
09:15:22 <pickpacket> hmm. I looked at this some time ago...
09:18:35 <pickpacket> I guess the first part would be to add a beard or two to the OpenGFX
09:18:47 <pickpacket> because otherwise they won't be selectable anyway
09:21:01 <pickpacket> oh... shit. I can see how this may be a problem
09:22:25 <petern> The baseset sprite list is fixed, basically, although there is an extras section which for original graphics is included in openttd.
09:22:53 <pickpacket> I'll do more when I have spoons for it
09:23:11 <TrueBrain> starting with adding things to faces might not be the easiest start 🙂 Maybe look for something else that fancies you that is a bit easier 🙂
09:30:34 <pickpacket> But I want beeeeeaaaards! :D
09:31:35 <pickpacket> it motivates me and is thus a good start :) even if it's not the easiest. If you guys feel that helping me with this is too much then I'll probably have to pick something else, but that'll be less interesting for me 🤷
09:33:40 <TrueBrain> I wouldn't even know where to start; and I have been doing this for 19 years 😛 But have at it, go nuts, have fun; that is more important than anything else 🙂
09:36:34 <petern> Does a beard replace the chin sprite?
09:37:07 <pickpacket> I was thinking it would go with moustache and changing that to "facial hair"
09:37:53 <pickpacket> I see that moustache disables the nose choice... I guess beard may have to disable the chin choice
09:38:23 <pickpacket> Might depend on what the beard graphics look like
09:38:32 <pickpacket> well, and the chin graphics
09:38:55 <petern> moustache also replaces lips choice, presumably because it hides them, but a beard may not?
09:39:06 <dP> As I understand adding beards would mean adding some way to define more face sprites that would only be used when available. One route would be to add them to the baseset but I'm not sure how to make them conditiotally available. Another is to add a newgrf api for face sprites so there won't be beards by default but newgrfs can provide them. In that case you need to somehow solve the compatibility
09:39:06 <dP> issue with different face sets. Ideally in a way that would allow both extending existing set and overriding all the face graphics completely.
09:40:36 <pickpacket> Here's the big issue I see: when I've made a face and saved it as my favourite and then join a multiplayer server I want to be able to load and use that particular face even if the server has different NewGRFs than I usually play with
09:41:07 <pickpacket> And of course I'd love it if other players saw my face the same way regardless of the graphics pack they're using
09:41:22 <pickpacket> I have a feeling one or both of these issues may be impossible
09:41:23 <dP> providing custom user pictures is a whole other can of worms
09:41:30 <dP> there is nothing like that currently in the game
09:42:34 <pickpacket> no, but I can save the face I've created in the face editor and load that to use in any single or multiplayer game
09:43:10 <pickpacket> in order for that to be possible the choices must be available for everyone running the same version of the game
09:43:34 <dP> currently face is just a number that defines which parts are combined, what you want is to pass an actual image data
09:44:17 <pickpacket> or extend the format or number series to include more choices, which then must be available for all participating players
09:44:55 <dP> dP: yeah, that's what I was talking about here
09:44:58 <pickpacket> there's a non-zero possibility that the best way to solve this is by making a full-fledged face editor with sliders and stuff
09:45:30 <pickpacket> but I guess that would leave the same problem with stuff like clothes and glasses either way
09:47:24 <dP> Though I guess, UI aside, providing custom pictures may not be that hard. Finally some use for the new command structure xD
09:47:32 <pickpacket> sending a custom picture would probably lead to people creating all sorts of weird images (including swastikas and dicks)
09:47:53 <dP> they already do that with rails and usernames
09:48:19 <pickpacket> oh, you mean writing stuff on the map
09:48:33 <dP> yeah, swastikas are especially popular
09:49:00 <pickpacket> easy enough to make with straight lines
09:49:14 <pickpacket> harder to make a dick
09:49:43 <pickpacket> Anyways! Back to this. Sending custom images then, maybe?
09:50:34 <pickpacket> size restraint in pixels to the current size is nice. Keeping the current functionality for those who haven't made a custom image is also nice
09:50:52 <TrueBrain> I think here too again you get ahead of yourself; keep it simple. First make the functionality how you think it should work / look. Don't worry about shit like saveload, multiplayer, etc etc 🙂
09:51:18 <TrueBrain> getting the game to show anything remotely like a beard first is a challenge on its own
09:51:24 <TrueBrain> hack it in there if you like
09:51:45 <pickpacket> TrueBrain: the simplest thing to do is to change a moustache sprite in OpenGFX
09:51:57 <TrueBrain> get a feel for the landscape 🙂
09:52:38 <TrueBrain> what you want is incredibly difficult with many things to consider; but figure all that out without ever having touched our codebase .. yeah, get your hands dirty first; my opinion 😄
09:53:01 <dP> It nice to say "don't worry" but then you do 90% of work and realize 10% is impossible because of some limitation of the network layer
09:53:33 <dP> like, there was something like 1k limit for a network command, not enough for a picture
09:53:33 <pickpacket> Going for the second-simplest solution is probably more fulfilling :D And that's (probably) to add another sprite. Just one. And figure out how to handle that
09:53:53 <pickpacket> dP: ah, yeah. That's of course a risk
09:54:46 <petern> If you add just one sprite then you can fit in into the unused space for moustache.
09:54:55 <TrueBrain> hmm .. I can't make API tokens that are limited to a single Cloudflare Pages .. that is a bit annoying .. hmmmmm
09:55:04 <petern> But then you've added just sprite and there are a lots of different styles of beard...
09:55:16 <TrueBrain> there is only one approved style of beard
09:55:31 <TrueBrain> the rest are all fake-news
09:56:44 <pickpacket> petern: yeah, that one sprite should be added in a way that allows for more to be added later
09:57:07 <pickpacket> TrueBrain: I mean... the only approved style of beard right now is "none" :D
10:11:13 <TrueBrain> right, infra up-and-running .. now to check if the workflow actually works ..
10:11:37 <TrueBrain> I did spot a mistake .. too much copy&paste from survey-web .. but okay 😛
10:16:22 <TrueBrain> `Calling 'DidYouMean::SPELL_CHECKERS.merge!(error_name => spell_checker)' has been deprecated. Please call 'DidYouMean.correct_error(error_name, spell_checker)' instead.` .. a deprecation over a hint .. I dunno, I find that funny 😄
10:17:35 <dP> petern: any examples of "well thought-out concrete proposal" ever being implemented?
10:23:12 <petern> any examples of 'well thought-out concrete proposal' ever being proposed?
10:24:04 <dP> nothing is well thought if you're not the one thinking it? :p
10:24:20 <TrueBrain> right, now the question is .. does anyone dare to review #257, or shall I just hit the magic button?
10:25:08 <dP> LC had some well-thought proposals
10:25:33 <dP> not saying they were good but there was sure enough effort put into them
10:37:42 <dP> lol, even #10891 is technically well-though as I've been dealing with this thing for like 10 years :p
10:39:01 <dP> that is just the conclusion, nothing good can be done with current options
10:44:44 <TrueBrain> owh, you are right, they still haven't fixed E203 .. I was sure they did ..
10:44:54 <TrueBrain> flake8 is a bit weird there ... their example case is for something completely different
10:45:16 <TrueBrain> and it breaks on `line = line[line.find("return") :]`
10:45:47 <TrueBrain> but W503 is now disabled by default in flake, so that should go for sure
10:47:31 <Rubidium_> dP: network packet size limit is now a mere 32767 bytes with a description how to potentially extend the format to even bigger packets
10:49:19 <TrueBrain> right, time for some lunch 😄
11:21:15 <pickpacket> LordAro: > "fluidsynth dep missing, possibly?" Returning to this from a couple of days ago. Fluidsynth isn't mentioned in COMPILING.md. That said I apparently have it installed 🤷 Not sure why music doesn't work for me
11:24:40 <LordAro> dunno then, missing soundfont?
11:24:45 <LordAro> did it work in the past?
11:25:45 <glx[d]> Oh you can patch the doc if it's missing stuff 🙂
11:26:07 <pickpacket> it worked when I used the openttd debian package from the repo, but not when I've compiled myself
11:26:53 <glx[d]> Cmake should tell if it found fluidsynth or not
11:27:37 <pickpacket> I'll recompile. But not today, because my computer is sloooow and I want to spend the CPU cycles playing the game instead right now :D
11:31:12 <TrueBrain> now the question .. how do you enable cargodist? 😄
11:33:35 <TrueBrain> at least, I think it is active, but it isn't crashing like the bug report suggests it should ... annoying 😛
11:34:47 <TrueBrain> ah, it is .. but it is underflowing, instead of crashing .. also nice 🙂
11:36:13 <TrueBrain> yeah, I did that, and it didn't crash, so I was like: huh?
11:36:16 <glx[d]> In settings, it's not explicitly named cargodist, but it's the manual/symmetric/assymetric thing
11:36:20 <TrueBrain> but yeah, the bug is a tiny bit more annoying 🙂
11:36:37 <glx[d]> Anything other than manual is cargodiqt
11:36:48 <TrueBrain> as it is not like there is some clear feedback ingame when it is active or not, until you build some shit 😛
11:37:05 <TrueBrain> ah, look at those values .. no longer insanely large ..
11:41:28 <TrueBrain> right, what was I actually doing? Ah, yes, to see if I can bribe someone to approve #257 😄
11:43:14 <TrueBrain> other than that, I think I have everything setup via Pulumi now .. no more manual labour 😄
12:08:20 <TrueBrain> LordAro: anything else for 257, or can I go ahead? 🙂 (sorry for pushing, but it needs to be fixed before the next nightly :P)
12:09:22 <TrueBrain> Thank you! Now to roll this out properly 😄
12:10:13 <TrueBrain> meh; I can do it very complicated, or I just create a short downtime ... 😛
12:12:29 <TrueBrain> okay, that is longer downtime than expected ... "verifying" .. euh .. okay
12:14:34 <glx[d]> Next step is to reuse the new reusable workflows everywhere else ?
12:15:16 <TrueBrain> sort-of, kinda, but not really yet 😛
12:15:22 <TrueBrain> website (and survey-web) are different from the rest
12:15:37 <TrueBrain> so I now first need to template a Python application deployment 🙂
12:15:53 <glx[d]> Well I mean annotation check
12:16:15 <TrueBrain> yeah, parts can be done; but I plan on doing that while I also address the other things that need doing 😄
12:16:46 <TrueBrain> I have a bit of a hope that I can template 1 Python app, and we can reuse a single workflow everywhere .. as they are all scary similar 🙂
12:17:00 <glx[d]> Though splitting CI workflow like release one might be nice
12:17:00 <TrueBrain> so basically a reusing workflow that uses reusing workflows I just created 🙂
12:17:41 <TrueBrain> Look at that preview!
12:18:54 <glx[d]> Annoyingly the link opened in GitHub app so no preview:)
12:19:14 <TrueBrain> from the app you should see a "View deployment" button
12:19:17 <TrueBrain> that should open a browser?
12:20:22 <TrueBrain> how shall I make clear we no longer have to tag .. hmmm
12:20:26 <dP> so simple with flexbox now
12:20:38 <dP> remember when sticking footer was quite a pita
12:23:08 <TrueBrain> if anyone is interested in picking ^^ up, would be appreciated 🙂
12:23:31 <TrueBrain> so silly, to have old links in our page 😄
12:23:57 <TrueBrain> glx[d]: wow, that I did not expect from GitHub .. they do mention the deployment, but don't give the link ... silllllyyyyyyy
12:29:59 <pickpacket> right. Back from a nap. Back to beards... hmm. Where to start...
12:33:01 <TrueBrain> anyway glx[d], my intention otherwise is indeed that most of the workflows are in a single place, so updating them isn't such a hassle next time 🙂
12:34:34 *** Ahyangyi has joined #openttd
12:34:34 <Ahyangyi> I have seen quite a few proposals about heightmaps with rivers, is there a consensus about this one? 😛
12:35:28 <TrueBrain> I haven't seen any proposal related to that in any recent times; so I guess not 😦
12:39:33 <TallTyler> Heightmaps with rivers are going to be problematic if you don’t load them at exactly the same size they were saved at, because rivers can only flow down certain slopes. You’d need to terraform other slopes that appear on a resized heightmap, which introduces all sorts of fun edge cases. I encountered most of them when adding wide river slope terraforming, but could get away with “eh, just let the
12:39:33 <TallTyler> river be narrow” if I couldn’t solve a conflict. You can’t do that when placing the base river or you get a disconnected river.
12:40:08 <TallTyler> I suspect a lot of the proposals (none of which I’ve seen) were “wouldn’t this be cool” not “here’s how we’d solve edge cases” 😛
12:42:07 <TallTyler> Not that it’s impossible, if you want to try go for it. It’s just not quite as easy as it sounds.
12:42:13 <TrueBrain> ugh, finding bugs in AWS Launch Templates .. and not a real way to report those to AWS .. mmmeeeeehhhhhhh
12:47:31 <Ahyangyi> TallTyler: Alright, that sounds like something that could really use a proof of concept 😛
12:48:06 <Ahyangyi> I was worrying about the interface more than the implementation, so if the answer is "actually the implementation is the concern", then it's quite informative in itself...
12:54:00 <JGR> TallTyler: Isn't the simple solution to that, that height maps with attached river information are not resizable?
12:55:35 <JGR> It wouldn't surprise me if some height map authors would like a flag for that even without rivers
13:02:59 <TallTyler> Yes, I suppose that would solve it
13:04:09 <TallTyler> The traditional way is just to make rivers at sea level, so they flood automatically
13:05:54 <dP> if your heightmap isn't resizable it's almost like scenario
13:06:29 <TallTyler> Yes, I'd prefer resizable too, but 🤷
13:08:10 <TallTyler> #10409 will add a town import feature, although the interface will change from its current design to a `Load from file` button in the Scenario Editor town GUI
13:09:05 <TallTyler> I should really work on that one sometime, although on Saturday I leave for three weeks of new job training in Germany so I won't be writing any code any time soon 🙂
13:12:13 <JGR> dP: Isn't that pretty much the point of height maps?
13:12:36 <dP> yeah, but with more flexibility
13:13:03 <dP> though I guess scenarios bake in industries and towns on top of that
13:13:44 <dP> don't think you can make an "empty" scenario
13:16:41 <Ahyangyi> I personally think that since heightmap itself is already subject to slope restrictions upon importing, the rivers having *more* slope restrictions should mean a quantitative change instead of a qualitative one.
13:17:22 <Ahyangyi> And as for height map vs scenario: to me, if something should work with different choices of newGRF, then making it possible via heightmaps have value
13:20:29 <dP> but it can be made possible with scenarios as well
13:20:49 <dP> cleaning up industries from scenarios is already a thing some people did manually
13:23:46 <pickpacket> CompanyManagerFace is a uint32. No more bits can be added. If I want to add additional sprites I could refactor it to a struct that contains the original uint32 and some kind of data structure with additional sprites which can be used iff the gfx pack has them
13:24:39 <dP> shouldn't be that hard to convert u32 to u64 if you just need a few more bits
13:25:05 <LordAro> an actual struct may not be the worst thing in the world though
13:25:41 <LordAro> although "iff the gfx pack has them" - gfx will always have them
13:25:52 <LordAro> there's no "optional" with base sets
13:26:39 <pickpacket> LordAro: in that case adding more sprites (now or in the future) will break all existing gfx packs don't have them yet
13:26:51 <dP> well, technically, there is, original doesn't contain any new sprites
13:26:58 <dP> though I guess openttd just provides defaults for those
13:27:37 <glx[d]> If you add them they will be in openttd_extra by default
13:27:37 <pickpacket> I've never used another gfx than OpenGFX so I honestly don't know what they usually add and how it affects manager faces
13:28:19 <dP> basesets just redefine existing sprites, nothing else
13:28:43 <glx[d]> Most likely with action5
13:28:59 <pickpacket> and all sprites that aren't redefined are taken from OpenGFX?
13:29:36 <pickpacket> because in that case it's easy, as long as OpenGFX is updated when one updates OpenTTD
13:29:55 <glx[d]> If not redefined '?' is shown unless openttd provides a default
13:30:12 <glx[d]> I use original graphics
13:30:20 <dP> yeah, iirc original sprites have to be defined and new sprites can be taken from openttd_extra
13:31:42 <pickpacket> But where is the default taken from? OpenGFX? I recall getting an error the first time I started the game with a question of whether I wanted to download OpenGFX
13:31:48 <pickpacket> I may remember wrong
13:31:59 <dP> openttd_extra, it's a separate grf that comes with openttd itself
13:32:05 <dP> opengfx is entirely optional
13:32:36 <Ahyangyi> I see, openttd_extra refers to the tiles OpenTTD uses but the original TTD doesn't, right?
13:32:52 <Ahyangyi> And OpenGFX is a replacement set for those stuff the original TTD also has
13:33:40 <glx[d]> OpenGFX can replace both
13:34:08 <Ahyangyi> yes, but still optional
13:34:36 <dP> actually, I think openttd_extra is the name of the project or smth, the file itself is called something else
13:35:19 <JGR> It's in media/baseset/orig_extra
13:36:38 <JGR> You can just add your new sprites in there in the same commit as your new feature
13:37:52 <dP> though I guess for just action5 it's not a big deal
13:37:57 <JGR> You can make NML or whatever write the NFO you
13:47:03 <pickpacket> I can't find any faces there. Does that mean one *has* to use a gfx pack?
13:47:18 <pickpacket> but yeah... I could just add stuff there :D
13:48:03 <LordAro> faces are very much part of the original base set
13:48:08 <LordAro> so won't be in orig_extra
13:53:56 <pickpacket> LordAro: where are they though? I can't find them :)
13:54:07 <LordAro> original base set? not in OTTD repo
13:54:19 <LordAro> we're not allowed to do such things :p
13:56:10 <pickpacket> well that would mean that any extra sprites I'd want to add would be in orig_extra
13:57:14 <pickpacket> where do I find the faces though? In OpenGFX? I mean there are faces there, but do all gfx include face sprites?
13:57:59 <LordAro> all base sets include face sprites because that's how base sets work
13:58:31 <LordAro> or rather, OTTD will display something (probably a '?') if they're not there
14:00:01 <TrueBrain> meh, stupid AWS ... why can I do something manually, but you can't do via a launch template
14:00:04 <TrueBrain> that makes no sense 😦
14:22:47 <TrueBrain> and I can't find a proper way t work around it .... network traffic .. it is the worst! 😦
14:24:22 *** gnu_jj has quit IRC (Ping timeout: 480 seconds)
14:32:06 <TrueBrain> well, if everything fails, just flip the interfaces around and that fixes the issue .. or something 😛
14:32:13 <TrueBrain> I understand why, but ugh, it is such a shitty solution 🙂
14:32:27 <TrueBrain> well, that took only .... too much time
14:39:02 * pickpacket has left the coding for a bit and is trying to understand how to make terminus stations efficient
14:50:16 <JGR> The unstated context is that this is pre-PBS
14:51:44 <JGR> Nowadays you should probably ignore any suggested track/junction/station layouts on the wiki
14:51:53 *** Flygon_ has joined #openttd
14:53:22 <Rubidium_> though with PBS with two terminus platforms half of the times there'll be a conflict between the paths when entering the platforms at the opposite side
14:54:41 <Rubidium_> but with PBS that'll not have such a big effect than with the non-path signals
14:54:54 <TrueBrain> so ... I provisioned our new stack in eu-central-1 on AWS, thinking nothing of it .. turns out .. that region doesn't support IPv6 access to AWSs internals ...... seriously, this IPv6 support is to cry about 😛
14:56:10 <TrueBrain> it doesn't help that certain internal AWS APIs have a public IP you need to reach
14:58:31 *** Flygon has quit IRC (Ping timeout: 480 seconds)
15:04:08 <pickpacket> Cloud services… hate’em
15:04:54 <dP> pickpacket: each platform has separate entrance and exit, it's basically a roro
15:05:12 <dP> masquerading as a terminus
15:06:11 <JGR> It's overkill for typical players
15:06:31 <dP> it's overkill for pretty much everyone
15:06:40 <dP> if you have space just make extra platforms
15:21:50 <pickpacket> I want to try out different designs to see if I can 1) increase throughput and 2) enable incrementally upgrading the track type without disturbing traffic too much
15:32:18 <JGR> Assuming that your track types are compatible you should be able to upgrade the track without needing to do anything special
15:32:30 <FLHerne> re. what Ahyangyi said, anyone have the link to frosh's (I think) description of a hypothetical better heighmap/scenario format?
15:33:07 <TrueBrain> one good thing about Infrastructure as Code .. you can just pick up everything you have deployed in one region, and deploy it in another .. silly stuff being silly ..
15:34:24 <pickpacket> JGR: they're not :) it's the upgrade from electric to mono or maglev, or upgrade from mono to maglev, that is the issue
15:35:09 <FLHerne> Ahyangyi: you might find that document interesting
15:36:12 <JGR> There are track and vehicle GRFs available which make the whole mono/maglev upgrade busywork unnecessary
15:36:58 <FLHerne> stupid idea: "scenarios" could just be distributed as a GS that builds the desired map
15:37:08 <pickpacket> JGR: of course, but what's the challenge with that? :D
15:37:20 <FLHerne> GS can build all the features there for any company
15:37:38 <JGR> There isn't anything interesting about bulk converting rail types
15:37:46 <FLHerne> downside is it would be slow and difficult to create
15:38:34 <JGR> It doesn't make sense outside of the vanilla vehicles
15:38:38 <FLHerne> it's a lot easier to just leave existing networks as they are and overlay the new type as a separate network
15:38:48 <FLHerne> learning from the problems of the existing ones
15:39:01 <pickpacket> JGR: I mean... yes... but so is station redesign and train purchasing
15:40:01 <JGR> For train purchasing there is the clone button
15:40:31 <pickpacket> I've considered making a GRF with some faster electrical trains for the later part of the game just to skip the monorail altogether
15:40:52 <pickpacket> does anyone here tips for existing GRFs?
15:41:24 <JGR> Just about every existing train set GRF would with for that?
15:42:44 <pickpacket> JGR: of course (about the clone button). I just mean that the entire point of the game is to solve problems as they arise, whether that means rebuilding a station, buying more vehicles, redesigning rail tracks... it's all of it "drudge work" more or less. But I love it. :)
15:43:15 <pickpacket> there's just so many train GRFs :( I don't know which one to pick that won't flood me with a billion options
15:45:59 <JGR> I understand that, and I like problem solving
15:46:37 <JGR> The vanilla track upgrade thing seems more like enforced busywork rather than any kind of problem that can be creatively solved to me
15:47:29 <FLHerne> "Hi guys, recently got into this game and I like all the grfs but im really struggling with finding trainsets that aren't exhaustive lists of all engines that ever existed." <- reddit comment I loved
15:47:38 <JGR> That and maglevs running around with open-top coal hoppers is just silly 😛
15:48:12 <FLHerne> Iron Horse is ok, it has more options than vanilla but not totally ludicrous
15:48:47 <pickpacket> JGR: I can see that point of view. To me the eventual conversion to maglev is pretty much the only way to have a chance to keep up with production increases :D
16:05:01 *** Ttech has quit IRC (Ping timeout: 480 seconds)
16:06:22 <JGR> Until recently the default vehicles didn't work with industry GRFs
16:06:42 <glx[d]> fixed around 12.0 IIRC
16:06:47 <JGR> Most likely that set is to work around that
16:09:10 <pickpacket> FLHerne: I totally agree with that comment...
16:09:22 <pickpacket> I'd like 3-4 more trains, at most
16:55:11 *** HerzogDeXtEr has joined #openttd
18:02:31 <TrueBrain> did you just reference commits from 2013? 😄
18:02:59 <TrueBrain> You just also wanted to add some cleanup commits, didn't you? 😄
18:03:31 <JGR> I just happened to notice it when doing some other changes
18:05:19 <JGR> std::map of tiny PODs is something that I've been replacing with other containers where it makes sense
18:13:28 <petern> Mode 7 is where it's at.
18:22:10 <Eddi|zuHause> i still have no idea how i'm going to get to brussels
18:22:26 <Eddi|zuHause> actually, several ideas which are all terrible
18:26:17 <Eddi|zuHause> probably not worse than the other options :p
18:28:46 <TrueBrain> right ... next victim ... sentry doesn't have IPv6 support .. this is just one big disappointment ...
18:29:09 <Eddi|zuHause> ipv6 might be a fractal
18:29:48 <TrueBrain> but, given I already proxy 3 domains .. why not add a fourth?
18:33:22 <pickpacket> Eddi|zuHause: I read this book where someone wanted to travel quickly across Europe and stole a sports car to do it. In the book that was a very fast way to travel. Not sure how realistic it is
18:33:41 <pickpacket> The book also had vampires, but I'm sure that's just a coincidence
18:34:21 <Eddi|zuHause> right... but even in a stolen car, i have to pay the fuel
18:35:21 <Eddi|zuHause> apparently, from the german border to brussels is 8h by bike
18:37:19 <TrueBrain> still would like to redo how the wiki is stored, with better use of Cloudflare, but 1 project at the time 😄
18:38:29 <Wolf01> I also wanted to travel quickly but I got highway closed for crashed trucks and a lot of people trying to get to the sea
18:39:16 <Eddi|zuHause> somehow all the people always want to go to the sea at the same time :p
18:39:52 <Wolf01> Specially Germans these days, I wonder why :P
18:40:39 <Eddi|zuHause> on the german autobahn, it is customary to complain about dutch people with caravans blocking the lanes
18:41:26 <jfs-> I'm just taking the cheap bus departing friday at 18:00, one change in the morning, and then arrival in brussels at 11:30 (and hoping there won't be bad delays anywhere)
18:41:26 <jfs-> I hope I'll be able to sleep or at least rest on the bus...
18:42:48 <TrueBrain> I hope so too .. otherwise it is a long day 😦
18:42:52 <TrueBrain> TallTyler: will there be nametags? 🙂
18:43:45 <TrueBrain> or will there be badges? 😄
18:43:47 <Eddi|zuHause> bus would leave 23:25 and arrive 8:55 with no changing
18:44:36 <TrueBrain> if jfs- can do it, so can you!
18:44:45 <Eddi|zuHause> and i have a fear the space on the bus is too narrow
18:44:55 <Eddi|zuHause> i need a lot of leg room
18:46:20 <Eddi|zuHause> another option would be to take 7 or so "free" local trains and board the same bus at the last german stop at 6:00
18:47:24 <Eddi|zuHause> but you couldn't sleep on the trip that way
18:47:33 <frosch> i booked a train, but rail operators may be on strike...
18:48:27 <frosch> i stay in aachen for a few days more. so on saturday i only want to go aachen - brussel - aachen
18:48:39 <frosch> aachen - brussel is easy. the reverse is not
18:49:00 <Eddi|zuHause> i was also thinking a similar thing
18:49:07 <frosch> there are only very few trains in the evening, i booked the last one, which is the eurostar, and it cost as much as the other trains combined
18:49:13 <Eddi|zuHause> but go aaachen-brussels-luxembourg
18:49:21 <frosch> mostly because only 1st class tickets were left
18:50:20 <frosch> there was some altenative route via some belgian local train, and another local train from some belgian village to aachen
18:50:33 <frosch> but both would have been the last trains of the day (really, why?)
18:50:33 <Eddi|zuHause> i might just go to the bus station the next few days and ask the bus driver if i could test out the seats
18:51:07 <frosch> and i did not feel like walking 20 km from some belgian village to aachen in late evening in unknown territory
18:51:51 <Eddi|zuHause> a nice half-marathon for the evening? :p
18:53:45 <Eddi|zuHause> there's a flixbus from brussels to aachen at 22:10 and 1:20
18:53:56 <frosch> actually only 16km, but still unknown place 🙂
18:54:13 <michi_cc[d]> In other news, some problems can be partially solved by money. It turns out you can buy priority security at BRU airport. Which means I'm comfortable only arriving an hour before my flight back.
18:54:34 <michi_cc[d]> For a city-center station like Brussel-Centraal this works out to departing there at 19:30 the latest.
18:54:40 <Eddi|zuHause> an hour is still a lot
18:55:15 <frosch> my train is also 19:25-sh
18:55:16 <michi_cc[d]> An hour before departure, which is 30 minutes or so before offical boarding time.
18:56:15 <TrueBrain> and what time you arrive michi_cc[d] ?
18:57:07 <michi_cc[d]> Early, a bit before 9.
18:57:23 <TrueBrain> owh, so a relaxed day in Belgium 😄
18:57:28 <Eddi|zuHause> still, assuming starting from train world it's probably better for someone with a car to drop you off at the airport
18:57:40 <michi_cc[d]> Train world is long closed by that time.
18:58:22 <Eddi|zuHause> i don't know how far we plan to go away from there for whatever we do after, though
18:58:54 <frosch> we just eat waffles and fries
18:59:10 <Eddi|zuHause> typical belgian food, obviously
18:59:16 <frosch> though the combination is weird, so i consider doing one of them before train world, the other after
18:59:24 <TrueBrain> Steak & Ribs restaurant around the corner 😛
18:59:38 <michi_cc[d]> There would be a 19:16 train departure from Schaarbeek (Train world) with one change to the airport.
19:00:35 <TrueBrain> there is even a London Bar!
19:01:57 <TrueBrain> michi_cc[d]: how can there be a change in that short track? 😄
19:02:02 <TrueBrain> there are like 2 stations between there and the airport 😛
19:03:06 <Eddi|zuHause> i think you need to go back to brussels nord and then go from there to the airport
19:03:43 <Eddi|zuHause> which is why i said it's probably easier to go by car
19:03:58 <TrueBrain> but bringing a car by plane is so expensive
19:04:25 <Eddi|zuHause> i heard other people might be bringing their car :p
19:04:48 <michi_cc[d]> TrueBrain: Yes, but no train apparently 🙂
19:05:05 <TrueBrain> hmm ... the wiki preview is a bit weird ... on first load, it is terribly slow (2+ seconds) .. after that is is very fast .. but if you wait a bit, and try again, it is slow again
19:05:08 <TrueBrain> not sure what that is about
19:05:53 <frosch> oh, i thought that was only my connection
19:06:14 <TrueBrain> hmm, come to think of it, I might know .. and it is this bloody annoying IPv6 prefix that AWS doesn't allow me to setup properly via auto-scaling
19:06:27 <TrueBrain> and I cannot make a ticket for it, as you don't get techsupport with a basic subscription
19:06:33 <TrueBrain> but it is broken .. and I can't tell them ..
19:06:46 <TrueBrain> (and yes, it really is a bug on their side, as I can do it manually just fine)
19:10:53 <TrueBrain> hmm, I have some ideas to try out .. but not tonight; done with this crap 😛 almost there .. ALMOST 😄
19:29:05 <TrueBrain> I asked Bing Chat for suggestions where to eat nearby Train World. It keeps repeating itself: `Which cuisine do you prefer?` .. I keep asking for things nearby ... 30+ minute walk is not nearby
19:29:57 <TrueBrain> it now suggests a place to eat in .. Delft
19:30:00 <TrueBrain> that is a 2 hour drive 😛
19:30:38 <TrueBrain> `I’m sorry but it is not possible to walk from Train World to Delft in 5 minutes` .. NO SHIT ..
19:31:45 <TrueBrain> lolz .. now Bing Chat is giving me URLs to Google Maps for locations .. it has its own map ffs!
19:31:45 <frosch> when you find a place, don't post it here
19:31:59 <frosch> we need a escape strategy, if 29 people show up 🙂
19:32:24 <TrueBrain> owh, the links it produced don't even work .. lol
19:33:15 <TrueBrain> and now it tells me it cannot show a map .. it was showing a map earlier .. I dunno, the frustration with Bing Chat is really high ..
19:33:32 <TrueBrain> GitHub CoPilot Chat also is just ... well, shit. I guess I am incapable of asking the right questions or what-ever
19:33:59 <frosch> ask for a food tour through brussels :p
19:34:33 <TrueBrain> I closed it; enough annoyance towards an AI for 1 day 😛
19:35:04 <frosch> all the tourist tours are only about chcocolate with beer
19:38:28 <LordAro> TrueBrain: Delft does have some excellent places to eat, tbf
19:38:50 <michi_cc[d]> "Prompt engineering" is the new career.
19:38:57 <TrueBrain> LordAro: that is very true; but there are more excellent places between Brussels and Delft 😄
19:39:23 <TrueBrain> for some reason I always end up in arguments with AIs .. they are just lying to me .. so I tell them .. and they go all: OWH I AM SORRY, HERE IS ANOTHER LIE
19:39:44 <TrueBrain> guess being Dutch doesn't always help ... 😛
19:39:45 <frosch> ais are not allowed to argue 🙂
19:40:17 *** HerzogDeXtEr has quit IRC (Read error: Connection reset by peer)
19:40:41 <frosch> they are trained for marketing, always talk along the customer, provide nothing with substance
19:41:08 <TrueBrain> curious if the new Adobe Photoshop AI can do 8bpp images 🙂
19:41:19 <TrueBrain> sadly, it is not worth money to find out
19:43:48 <glx[d]> TrueBrain: it's silly how IPv6 is still WIP in many places while some ISP are natively IPv6 because they can't get IPv4 ranges
19:44:06 <TrueBrain> I am really surprised how many big services don't have IPv6 suppor yet
19:44:09 <TrueBrain> that is really dissapointing
19:44:14 <TrueBrain> OpenTTD did IPv6 .. in .. 2007?
19:44:19 <glx[d]> most of them are US based I guess
19:44:24 <TrueBrain> full dualstack support everywhere ..
19:44:56 <Eddi|zuHause> ipv6 was very niche and experimental back then
19:44:58 <TrueBrain> so if a small game in 2007 can do it, why can't a big corp in 2023?
19:45:17 <glx[d]> because they have enough IPv4 ?
19:45:40 <frosch> my ipv6 has not worked in a month
19:45:59 <TrueBrain> but take something like github .. or sentry .. you would think they have IPv6 because their customers ask for it
19:46:07 <glx[d]> if my IPv6 doesn't work I don't have internet at all (my IPv4 is a fake one)
19:46:33 <JGR> They may want it, but that doesn't mean that all their vendor boxes, network peers, bits of software, etc will actually support it correctly
19:46:38 <TrueBrain> anyway, Cloudflare is nice, Cloudflare can upgrade these shitty providers to become IPv6-enabled 🙂
19:47:03 <glx[d]> but they add enough time to work this out
19:47:04 <JGR> Probably they tried to turn it on and loads of random things fell over
19:47:07 <TrueBrain> JGR: that excuse worked in 2010 .. but in this day and age it is hard to find network gear that doesn't support IPv6 😛
19:47:37 <TrueBrain> we all know IPv4 is running out .. we know whole providers don't have IPv4 anymore .. so come on .. but okay .. no insentive, so yeah ..
19:48:11 <JGR> "Supports" and "battle-tested really works all the time without weird bugs" aren't the same thing at all
19:48:20 <glx[d]> not sure relying on hardware supporting only IPv4 is that safe 😉
19:48:36 <TrueBrain> and I expected some minor things to not work with IPv6 .. but the amount of shit I found ... it is a lot worse than I anticipated
19:49:45 <frosch> oh dear, i looked at the menu of some random restaurant in brussels. 17€ for tagliatelle bolognese
19:50:07 <glx[d]> looks like french prices
19:50:28 <Eddi|zuHause> can't we just get a Döner or something?
19:51:04 <frosch> there is a fast food place every 3rd house
19:51:16 <TrueBrain> for some reason I suspect we end up in a McDonalds 😛
19:51:39 <frosch> i have not been in a mcdonals since school
19:52:02 <TrueBrain> because the `d` ran away?
19:52:45 <frosch> there is so much superior junk food compared to burgers
19:53:07 <TrueBrain> I was mocking your spelling 😛
19:53:33 <frosch> i saw it while typing, did not bother to move the cursor back
19:53:58 <Eddi|zuHause> that's why i said Döner
19:54:08 <Eddi|zuHause> it's usually cheaper than McD
19:54:45 <frosch> anyway, there are so many restaurants. no point in planning one without a local. we should just pick one, which has a free table for our group size
19:57:41 <frosch> i found a walking tour from the train station to train world, passing all the landmarks noted on gmaps
19:57:57 <frosch> i gues i know what to do between arrival and party
19:59:38 <frosch> though going by the maps on google, a tour through my backwater-town is fancier
20:00:25 <frosch> so far brussel looks like a rather sad plac
20:04:08 <TrueBrain> Poor Brussel .. everyone talks shit about it :d
20:04:12 <Eddi|zuHause> there were a bunch of landmarks on the brussels local transport map
20:04:20 *** Flygon_ has quit IRC (Remote host closed the connection)
20:26:24 *** HerzogDeXtEr has joined #openttd
21:16:08 *** HerzogDeXtEr has quit IRC (Read error: Connection reset by peer)
21:18:59 *** keikoz has quit IRC (Ping timeout: 480 seconds)
21:33:04 *** nielsm has quit IRC (Ping timeout: 480 seconds)
21:57:49 <TallTyler> TrueBrain: If I can find name tags before my flight on Saturday! (Or can find them in Germany)
22:01:59 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
22:04:01 <michi_cc[d]> Two tips for Germany (which you might of course already know): 1: If you see a price, that is what you pay. Full Stop. No tax non-sense over here. 2: This includes restaurants. Service is already included. While a tip of a couple euros is appreciated, no 20% non-sense either.
22:18:05 <glx[d]> well there is a tax, but it's included in displayed price
22:18:37 <michi_cc[d]> Yeah, of course there's still taxes 🙂 You just don't have to guess thew checkout price.
22:19:30 <TrueBrain> Consumer-friendly countries, is what they call it right? 🙂
22:20:51 <TrueBrain> Can't remember the last time I tipped in a restaurant, or got the opportunity even .. all pay by card these days .. you used to be able to add some euros, but even that is gone
22:22:35 <glx[d]> yeah usually you left the change
22:23:35 <glx[d]> anyway they do rely on tip to survive
22:24:44 <glx[d]> I hear tiping culture is going crazy in US, they want tip for everything
22:25:24 <glx[d]> even on paid in advance food
22:46:25 <TallTyler> Yes, tipping is out of control here. Tipped wage is like $2/hour which is why 20-25% tip is expected, even for to-go orders
22:46:44 <TallTyler> I suspect my biggest habit to break will be jaywalking 😛
23:03:08 <glx[d]> ah yes it's not a thing here
23:23:59 *** _aD has quit IRC (Ping timeout: 480 seconds)
continue to next day ⏵
