IRC logs for #openttd on OFTC at 2022-10-21
⏴ go to previous day
00:02:09 *** WormnestAndroid has joined #openttd
00:18:04 *** Tirili has quit IRC (Quit: Leaving)
01:38:26 *** HerzogDeXtEr has quit IRC (Read error: Connection reset by peer)
01:44:06 *** Wormnest has quit IRC (Quit: Leaving)
02:09:31 *** debdog has quit IRC (Ping timeout: 480 seconds)
03:06:41 *** D-HUND has quit IRC (Ping timeout: 480 seconds)
06:13:15 *** sla_ro|master has joined #openttd
06:20:26 *** gelignite has joined #openttd
07:16:05 <LordAro> black text would be an improvement
07:42:56 <petern> Hmm, shame that sprite font and truetype font ascenders/descenders don't match :/
08:38:21 <petern> Oh, the framerate window uses text spaces for indenting. Different.
08:51:57 *** HerzogDeXtEr has joined #openttd
09:32:00 <andythenorth> we should add
09:32:04 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
09:32:05 <andythenorth> my favourite entity
09:32:24 *** WormnestAndroid has joined #openttd
09:33:46 <Eddi|zuHause> pretty sure we a) already have that and b) doesn't matter if you're not doing automatic text wrapping
09:35:21 <Eddi|zuHause> src/string_type.h:#define NBSP u8"\u00a0"
09:37:11 <andythenorth> it's even used in FIRS 😛
09:39:21 <Eddi|zuHause> most occurances i found were with units
10:00:59 <petern> Yeah that's not what I meant. Most windows with indentations use offsets when drawing text, rather than leaving it up to the text.
10:02:02 <petern> I replaced those random offsets with a standard scaled offset, but that won't work here unless I rework it 😄
10:40:22 <glx[d]> Nice I can't see commit comment on GitHub mobile
10:55:17 <petern> Hmm, would be nice if PIP spacing would take account of zero-sized planes.
11:17:47 *** wallabra_ has joined #openttd
11:21:04 *** wallabra has quit IRC (Ping timeout: 480 seconds)
11:21:04 *** wallabra_ is now known as wallabra
11:24:32 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
11:24:34 *** WormnestAndroid has joined #openttd
12:19:53 *** Flygon_ has joined #openttd
12:22:24 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
12:22:40 *** WormnestAndroid has joined #openttd
12:22:44 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
12:22:47 *** WormnestAndroid has joined #openttd
12:22:57 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
12:22:58 *** WormnestAndroid has joined #openttd
12:24:57 *** Flygon has quit IRC (Ping timeout: 480 seconds)
13:29:33 *** ChanServ sets mode: +v tokai
13:36:14 *** tokai|noir has quit IRC (Ping timeout: 480 seconds)
13:46:41 <petern> If only comment and close was one op...
13:55:15 <LordAro> TallTyler: how about removing river generation from original landscape generator entirely?
13:55:46 <TallTyler> They can always disable rivers in original landscape generator
13:56:20 <TallTyler> If it were up to me the landscape generator setting would live in the config file, not in the GUI at all
13:56:38 <LordAro> wouldn't be opposed to it
13:56:49 <TallTyler> Out of scope for this PR though 🙂
13:56:51 <LordAro> wouldn't be opposed to removing the original entirely, tbh
13:56:52 <JGR> Why not just remove the original generator outright?
13:59:04 <TallTyler> That's not a hill I care enough about to die on; removing from the GUI is enough for me
14:03:54 *** gelignite has quit IRC (Quit: Stay safe!)
14:15:50 <camthesaxman> I prefer the original generator. Don't hurt me.
14:16:13 <LordAro> original generator works quite well for 256x256 maps
14:16:19 <LordAro> anything else, not so much
14:16:38 <LordAro> but that's also the case for the terragenesis tbf
14:16:47 <camthesaxman> I mostly play 256x256, so I guess that's why.
14:17:15 <TallTyler> I'll open a PR to remove it from the GUI and we can debate it there 🙂
14:17:35 <TallTyler> Just waiting for language files to build before a final test
14:47:18 <petern> Especially "variety", that should be replaced with something much better.
14:47:55 <andythenorth> I made my peace with it 🙂
14:47:57 <petern> Perlin noise alone is rubbish.
14:48:13 <andythenorth> just don't use the settings that produce blah-blah noise 😛
14:48:27 <andythenorth> alternatively, throw out it goes 🙂
14:49:27 <andythenorth> these settings are adequate
14:51:05 <camthesaxman> TerraGenesis isn't always easy to get working for the sub-arctic and sub-tropical climates. I keep having to fiddle with the settings to make it generate farms.
14:51:40 <petern> Original is better for those, but only for 256x256 :/
14:52:16 <LordAro> someone should make a new one
14:52:29 <petern> Voronoi always pops up, but I've never understood it enough to code it.
14:52:50 <camthesaxman> What happens in larger maps?
14:52:55 <petern> Also, erosion simulation.
14:53:11 <petern> It gets very repetitive.
14:56:56 <petern> Just knock that out in C++ 😉
14:58:40 <LordAro> probably not actually all that complicated
14:58:59 <LordAro> easy to mess up though
15:01:05 <petern> Better not do it then 🙂
15:02:42 <LordAro> hard bit would be converting it to OTTD world representation, i'd say
15:03:56 <LordAro> still got to start with some sort of "interesting" landscape as well
15:08:48 <petern> Start with perlin noise, yeah.
15:08:59 <petern> But there are tricks to improve that, folding etc.
15:09:21 <petern> Mainly though you need to not apply it uniformly.
15:10:47 <TallTyler> (Forgot to remove a string)
15:11:33 <petern> Cue people with terrain generation stuck on Original and missing options.
15:14:13 <TallTyler> Wouldn't they have to have chosen Original, used it, and then closed the game to save it in their config?
15:15:12 <petern> This is the problem with hiding settings, even from advanced settings. If they've changed it, there's no indication that it even exists as a setting...
15:16:28 <TallTyler> One possibility would be for saveload to set it to TerraGenesis, assuming people who prefer Original would know there's a choice
15:16:53 <TallTyler> Depends on how serious you think the problem is 🙂
15:17:14 <petern> Remove original then it's not a problem 😉
15:17:29 <petern> Cue people complaining that TGP sucks (wait, that's me :D)
15:18:16 <TallTyler> That sounds like it would generate a lot more complaints
15:18:20 <petern> Unix approach: Remove all terrain generation and relying 3rd party heightmaps only 😄
15:19:20 <LordAro> making the terrain generator actually generate heightmaps does sound easier
15:20:35 <petern> Well, if you ignore that the heightmap loader also has its own issues
15:21:17 <petern> There was an issue with a step from sealevel to the first layer of land.
15:31:47 <camthesaxman> What's the advantage of removing the original generator?
15:39:03 *** henry is now known as m3henry
15:42:13 *** m3henry has joined #openttd
15:48:23 <andythenorth> petern: Apple macOS Unix approach: remove height generator, offer 65536 different heightmaps, carefully curated and made from reyclable Tofu
15:48:31 <andythenorth> then make a glossy video about it
15:48:56 <andythenorth> but also ship the ability to log in as root without a password in the same release
15:49:17 <andythenorth> taking with one hand, giving with the other
15:50:51 <andythenorth> faking random is a thing I've done a lot in games
15:51:22 <andythenorth> Warcraft 1 'random maps' are actually just a bunch of pre-defined maps, with some randomisation of parameters
15:53:13 <TallTyler> camthesaxman: OpenTTD has many settings which are confusing and many which are not used but still exist for legacy reasons (there’s often but not always overlap between these two groups). There’s been an ongoing effort to simplify where possible to keep the code maintainable and to avoid overwhelming and confusing players, especially new players.
15:53:28 <petern> I did have a separate program I used to test different generation algorithms.
15:53:37 <petern> I had some "ok" results but it was a lot slower than TGP.
15:54:24 <TallTyler> See the move to hide block and pre-signals which consistently confused and frustrated new players, in favor of showing by default only the path signals which are easier to use and more effective in nearly all situations
15:54:39 <camthesaxman> How about have an "Advanced" button that opens a dialog with more options?
15:55:06 <TallTyler> Well, that’s just the config file 🙂
15:55:50 <TallTyler> At some point I want to add an advanced menu to the world gen window with more options they currently only exist in JGRPP
15:56:14 <TallTyler> It would most likely be a custom climate
15:56:17 <TallTyler> But that’s a ways off
16:04:15 * petern fiddles with dimensions
16:08:38 <m3henry> Speaking of confusing new players, I noticed that airport infrastructure costs often catch people out. I've created a patch for it and will submit tonight.
16:12:13 <FLHerne> m3henry: that they're too high?
16:12:42 <m3henry> No, just displays the costs in the construction GUI.
16:13:04 <FLHerne> oh, that seems like a good idea
16:13:59 <FLHerne> was going to say, aircraft are already unbalanced on the "far too profitable" side so reducing their costs would be an unpopular suggestion ;-)
16:14:27 <m3henry> Other GUIs might benefit from the change too, but airports are the one which stand out as needing it
16:22:34 *** Eddi|zuHause has quit IRC (Remote host closed the connection)
16:24:26 *** Eddi|zuHause has joined #openttd
16:33:11 <m3henry> I suppose another thing that catches people out is not knowing that large aircraft crash more often when landing at small airports.
16:34:40 <TallTyler> A screenshot in the PR description would be nice 🙂
16:35:19 <m3henry> I suppose, I'll cook one up
16:43:03 <m3henry> IMO, when dragging out track/road/landscaping, rather than holding shift to get an estimate, the cost should just be drawn near the cursor.
16:47:30 <TallTyler> That is how I've seen it done in more modern games
16:47:36 <m3henry> It would have veen nice if instead of {CURRENCY_SHORT} rendering £138,518k it did £138.5M. 6 significant figures seems excessive.
16:50:42 <TallTyler> We need {CURRENCY_SHORTER} 😛
16:51:55 <m3henry> Aight' you've pulled my leg
16:52:42 <m3henry> It's probably one of those things where the rabbit hole it much deeper than you initially think
16:53:36 <petern> (Absolutely no problem for displaying things, but we've always shied away from them
16:55:42 <Bouke> LordAro: Less code, less bugs!
17:05:22 <m3henry> I think the hard part is hysteresis around the point where scales change
17:05:56 <m3henry> Now you have state, eww
17:08:24 <m3henry> if values are changing quickly between 999k and 1M, this is going to be hard to read
17:09:37 <m3henry> Ideally you keep the previous scale until you really need to switch scale.
17:11:27 <petern> What about at 5 digits, not 999-1000 boundaries?
17:12:29 <m3henry> you still want hysteresis, no matter where your switchover point is
17:12:51 <petern> No, keeping previous state is next to impossible.
17:14:23 <m3henry> next to, but not actually. Definitely more effort than I'm willing to put in for now.
17:14:25 <petern> (Not to mention undesirable)
17:15:01 <camthesaxman> RCT displays the cost of anything before you click to put it down.
17:15:05 <m3henry> Perhaps a simper change would be to just add Billion/Trillion suffix
17:20:59 <Wolf01> andythenorth did you the cat?
17:21:28 <Wolf01> How is going with objects?
17:21:56 <frosch> oh, somone replaced the stupid cats with robots
17:22:09 <frosch> may have been days, but i only noticed now 🙂
17:22:33 <Eddi|zuHause> so only the non-stupid cats are left? great!
17:23:15 <frosch> you are the expert on empty-set-theory
17:23:50 <Eddi|zuHause> that is correct.
17:24:18 <Eddi|zuHause> the members of the empty set have all the coolest properties
17:29:00 <petern> Oh, I see a cat thing if I right click on the bot's "name" but I have compact mode on.
17:29:56 <petern> Hmm, some of them are cats.
17:54:42 <andythenorth> also how is objects?
17:54:45 <andythenorth> what is objects?
17:54:48 <andythenorth> goes it throw out?
18:22:24 <andythenorth> more objects you say?
18:22:32 <andythenorth> goes it throw out limitation?
18:30:16 *** Wormnest has joined #openttd
18:44:13 <Bouke> TrueBrain: just some random numbers to put out here, but I've found significant differences when turning on HW acceleration. In one particular savegame fast forward would increase from 8x to 20x when HW acceleration was enabled. In another savegame from 8x to 15x (similar ratio). So I think this is very interesting to further look into. Attaching the profiler shows that 63% vs 34% of the time is spent in `MainWindow::OnPaint`.
18:44:46 <Bouke> So "Viewport output times" is not the relevant metric to look at?
18:53:20 <andythenorth> Bouke: is this M1 or Intel?
18:53:56 <Bouke> Intel. 2015 model I believe.
18:54:15 <andythenorth> discrete or integrated GPU?
18:55:04 <andythenorth> is the savegame easily shareable?
18:56:06 <andythenorth> don't worry if not
18:56:12 <andythenorth> I could test on M1 if it's helpful
18:57:22 <Bouke> We’re watching a movie now, will share details later.
19:00:53 <DorpsGek> - Update: Translations from eints (by translators)
19:03:25 *** gelignite has joined #openttd
19:04:21 <TrueBrain> Bouke: sounds like a different blitter?
19:05:17 <TrueBrain> anyway, I remember why MacOS is disabled by default, it is a bit more sad: since 10.14 OpenGL is deprecated on MacOS
19:08:11 <andythenorth> FWIW, for an arbitrary empty-map savegame, on 12.2 official binary
19:08:23 <andythenorth> * hardware acceleration on: 95x FFWD average
19:08:34 <andythenorth> * hardware acceleration off: 80x FFWD average
19:11:17 *** m3henry has quit IRC (Remote host closed the connection)
19:11:27 *** m3henry has joined #openttd
19:24:23 <andythenorth> 18.75% faster on hardware?
19:36:49 *** m3henry has quit IRC (Remote host closed the connection)
19:40:56 *** virtualrandomnumber has joined #openttd
19:41:47 *** virtualrandomnumber has quit IRC ()
19:42:58 *** ChoHag_ has quit IRC (Read error: Connection reset by peer)
20:06:48 *** wallabra_ has joined #openttd
20:11:59 *** wallabra has quit IRC (Ping timeout: 480 seconds)
20:15:52 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
20:16:23 *** WormnestAndroid has joined #openttd
20:18:22 *** wallabra_ has quit IRC (Ping timeout: 480 seconds)
20:27:41 <Eddi|zuHause> TrueBrain: maybe we shold petition the EU parliament, now after they fucked apple over with the charging cable law, maybe they can bring them to support other universal standards
20:29:17 <petern> `git rebase --continue`
20:30:58 *** gelignite has quit IRC (Quit: Stay safe!)
20:31:29 <camthesaxman> I really don't give a damn what Apple does. I never buy any of their products anyway because they are inferior.
20:32:41 <camthesaxman> Apple cripples their products by not following standards.
20:32:46 <Eddi|zuHause> camthesaxman: that's fine for you. but that wasn't the question
20:34:00 <camthesaxman> There's MoltenGL which implements GL on top of Metal.
20:34:30 *** Flygon_ has quit IRC (Quit: A toaster's basically a soldering iron designed to toast bread)
20:44:45 <andythenorth> Eddi|zuHause: 'fucked Apple over' => forced them to drop a stupid non-standard cable
20:44:58 <andythenorth> which incidentally the US Mac forums are enraged about
20:45:15 <andythenorth> that and the requirement to not provide chargers
20:46:07 <Bouke> TrueBrain: But deprecated doesn’t mean unsupported. Given that HW acceleration yields better performance, why should we opt-out?
20:46:21 <Eddi|zuHause> the point of that was: nobody needs dedicated chargers anymore after everyone switches to USB
20:47:33 <Bouke> Apple was a major driver of usb-c, it’s just that they were already invested in lightning.
20:48:58 <andythenorth> Bouke: I can't be arsed to find the logs, but I think at the time it was just slower in tests with hardware on
20:49:04 <andythenorth> but today's results are different
20:49:17 <Eddi|zuHause> and even if you have a dedicated charger, you can switch your phone 5 times before you need to think about changing your charger
20:49:25 <andythenorth> all of the ^ above
20:49:33 <andythenorth> but the forums smell a conspiracy
20:49:36 <andythenorth> oh wait, 'forums'
20:49:38 <Bouke> andythenorth: this is the save game of before shown in a window of 1216x818 (virtual pixels0
20:51:54 <TrueBrain> Bouke: Bit weird reasoning tbh 🙂 but write a metal backend and we can talk 😄
20:52:03 <Bouke> andythenorth: Maybe that was before the Retina-support shipped? After that change we need to render 4x as many pixels. That's a significant change.
20:52:18 <andythenorth> hardware on, consistent game speed 24.99x / 25x
20:52:41 <Bouke> TrueBrain: How so? Just re-enable HW by default. OpenGL is still there, no need to not use it.
20:53:01 <camthesaxman> What does HW acceleration do in OpenTTD? Is it just for blitting sprites?
20:53:09 <andythenorth> less consistent, 80x-90x
20:53:28 <andythenorth> I think this is what we found before
20:53:43 <andythenorth> the hardware acceleration is faster for limited cases, but seems to have an upper limit
20:53:48 <Bouke> That simulation rate is .... well something
20:53:51 <TrueBrain> Bouke: Think about it longterm what you suggest .. at some point deprecation means removal .. and then .. we have to sell our users their performance is going to degrade? Sorry, I am not sold on that reasoning 😄
20:54:00 *** sla_ro|master has quit IRC ()
20:54:06 <TrueBrain> Users van opt-in if they like 🙂
20:54:27 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
20:54:53 <petern> I need to get bananas but it's raining and it's 10pm. Boo.
20:55:02 <camthesaxman> The game still runs fine without HW acceleration, so they're not left out.
20:55:45 <Bouke> TrueBrain: So selling a degraded performance now is better? Even though performance could be better now? And when OpenGL is going to be removed we even might have a Metal backend, so the point would be moot anyway.
20:55:50 <andythenorth> I didn't find any discussion there about mac peformance
20:56:23 <andythenorth> I don't understand why the hardware acceleration is capping at 25x for the savegame
20:56:24 <TrueBrain> andythenorth: No, the deprecation was the reason, as I mentioned earlier 🙂
20:56:26 <petern> I really wish Discord would ask if you want to make an embedded URL thingy rather than just do it an allow it to be removed.
20:56:49 <andythenorth> it should be at least a user settting petern
20:56:54 <JGR> You can put angle brackets round the URL to suppress that
20:57:05 <TrueBrain> andythenorth: Maybe vsync?
20:57:38 <camthesaxman> You can use MoltenGL if you want to keep the current GL backend on toyOS. It implements OpenGL on top of Metal.
20:57:44 <andythenorth> TrueBrain: dunno, for an 'empty' newly generated map hardware will do 95x and is faster than software
20:57:57 <andythenorth> but for the savegame Bouke provided...fixed solidly at 25x
20:58:11 <andythenorth> I send you a mac, you try 😛
20:58:26 <TrueBrain> Well, measuring performance difference like this is always a bit tricky 🙂
20:58:37 <andythenorth> I am only using FPS window, it's hardly science 🙂
20:58:50 <andythenorth> if we cared we should run repeatable tests on CI
20:58:53 <JGR> Likely Metal itself with be deprecated sooner or later, it's probably not worth the bother using it
20:58:56 <andythenorth> we don't care enough for that 🙂
20:59:14 <TrueBrain> JGR: Somehow that doesn't surprise me 😄 what will replace metal?
21:00:52 <camthesaxman> Unless Apple decides to resurrect GL and Vulkan, they have nothing to replace Metal with.
21:02:00 <Bouke> The call trees into the blitters look identical for both HW enabled/disabled.
21:02:11 <andythenorth> iOS 32: now with Synthwave 3 graphics
21:04:06 <petern> Sneaky thing about the OpenGL blitters, they don't actually use OpenGL to draw pixels, it "only" assists with copying pixels to the display hardware.
21:04:12 <petern> If I remember correctly.
21:05:59 *** wallabra has joined #openttd
21:07:08 <petern> It's an improvement, and allows vsync, but it's not a "proper" opengl pipeline. (I tried that once, it was slow, but then again I was using immediate mode)
21:07:41 <andythenorth> how fast does it need to be 😛
21:07:49 <andythenorth> I used to complain about performance
21:07:54 <andythenorth> 'used' being the important word
21:07:58 <camthesaxman> OpenGL doesn't provide vsync.
21:08:07 <andythenorth> OpenTTD got an absolute shitload faster in recent years
21:08:18 <TrueBrain> Next big performance gain will be to make the viewport drawing multithreaded
21:08:46 <TrueBrain> Early drafts show it is possible, but so many corner cases ... so mannnnyyy
21:08:54 <camthesaxman> Can the vehicle simulation be run in parallel?
21:08:58 <petern> camthesaxman: it's exposed in a way that isn't by the win32 api (and SDLv2) that we otherwise used.
21:11:10 <TrueBrain> It is like GL has a command to enable/disable :p
21:11:37 <petern> Good job you can edit messages in Discord 😄
21:11:47 <TrueBrain> On mobile, too much effort 😄
21:11:51 <andythenorth> do nightlies have asserts on?
21:12:58 <andythenorth> strange results, some saves are faster with hardware, some are not
21:13:42 <camthesaxman> There's wglSwapInterval and similar functions for other platforms that can enable/disable vsync.
21:13:50 <JGR> This may depend on what sort of stuff you happen to have visible in the viewport
21:14:34 <petern> "wgl" would be... an OpenGL thing
21:15:02 <TrueBrain> andythenorth: Understand that using FWD to benchmarks influences results .. with that, the balance between game thread and draw thread is influenced
21:15:14 <TrueBrain> Might be related, who knows 😄
21:15:42 <petern> You can do vsync with DirectX, but we never used that.
21:15:54 <TrueBrain> We don't even use DirectX :p
21:16:57 <andythenorth> hmm FFWD not the most important benchmark? 😛
21:17:02 <andythenorth> are we not just doing speed runs here?
21:18:02 <camthesaxman> andythenorth: I mean, that's really the end goal... to speed up fast forward.
21:18:13 <JGR> Testing an overly busy save game like wentbourne or whatever is probably more useful than empty maps
21:18:24 <TrueBrain> andythenorth: Make your window as small as possible, and broom broom
21:18:47 <petern> Just test HW acceleration with the null blitter.
21:19:03 <TrueBrain> petern: Lol, yes, very fast 😄
21:19:18 <Eddi|zuHause> null blitter is like lossy sort :p
21:20:13 <andythenorth> how about we test for battery life instead? 😛
21:22:03 <petern> Does the battery used while recompiling count?
21:22:10 <petern> That's the main game I'm playing.
21:26:08 <andythenorth> are are compiling from clean
21:26:12 <camthesaxman> andythenorth: That's a good thing to test, especially on mobile.
21:26:26 <andythenorth> what -j settings are allowed?
21:27:52 <TrueBrain> Or what was the MacOS equivalent
21:28:26 <andythenorth> I usually just use -j 19
21:28:48 <andythenorth> hmm dunno how to measure battery though
21:28:49 <andythenorth> make -j 19 390.65s user 38.37s system 792% cpu 54.155 total
21:29:14 <TrueBrain> `-j $(sysctl -n hw.logicalcpu)`
21:30:35 <petern> Hmm, that's a lot quicker than mine 😦
21:30:55 <petern> I have a feeling MSVC or whatever I'm using is pretty slow...
21:31:30 <andythenorth> not sure how often the battery thing samples
21:31:41 <andythenorth> yeah it's just dropped to 3584
21:31:46 <andythenorth> think it's every 60s
21:31:57 <andythenorth> ok so let's say about 100mAh
21:32:23 <andythenorth> give or take that Chrome is open running discord, which slams the battery
21:33:39 <andythenorth> ok the next two drops were only about 10 mAh
21:33:45 <andythenorth> so the compile is about 100
21:34:11 <andythenorth> so I can only compile 75 times on full charge
21:34:40 <andythenorth> compiling Horse eats more battery
21:37:36 <andythenorth> 75 types of chunky bevel to test
21:38:14 <petern> Technically there are 127 levels of scaling.
21:38:34 <petern> But, uh, that's only because the volume slider widget deals only with int8 😄
21:44:43 <andythenorth> sorry you'll have to reduce it to 64 levels
21:44:50 <andythenorth> otherwise my battery will die 😛
21:46:17 <petern> Weekend, just stay up all night.
22:05:38 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
22:21:41 *** WormnestAndroid has joined #openttd
22:22:50 *** Kitrana1 has joined #openttd
22:28:19 *** Kitrana has quit IRC (Ping timeout: 480 seconds)
22:30:39 *** wallabra has quit IRC (Ping timeout: 480 seconds)
22:37:53 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
22:38:52 *** wallabra has joined #openttd
22:51:34 *** WormnestAndroid has joined #openttd
23:12:29 *** HerzogDeXtEr has quit IRC (Read error: Connection reset by peer)
23:24:08 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
23:26:26 *** WormnestAndroid has joined #openttd
23:41:38 *** Wormnest has quit IRC (Ping timeout: 480 seconds)
23:44:03 *** _aD has quit IRC (Quit: leaving)
23:54:53 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
23:57:19 *** WormnestAndroid has joined #openttd
continue to next day ⏵
