IRC logs for #openttd on OFTC at 2025-06-09
            
00:53:24 *** D-HUND is now known as godbed
00:53:42 *** akimoto has joined #openttd
01:14:10 *** Tirili has quit IRC (Quit: Leaving)
01:53:36 *** akimoto has quit IRC (Remote host closed the connection)
01:59:19 *** Wormnest has quit IRC (Quit: Leaving)
02:40:17 *** gnu_jj_ has joined #openttd
02:43:44 *** gnu_jj has quit IRC (Ping timeout: 480 seconds)
03:04:20 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
03:04:22 *** WormnestAndroid has joined #openttd
03:04:28 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
03:04:44 *** WormnestAndroid has joined #openttd
03:11:56 *** D-HUND has joined #openttd
03:12:20 *** godbed_ has joined #openttd
03:15:29 *** godbed has quit IRC (Ping timeout: 480 seconds)
03:15:44 *** debdog has quit IRC (Ping timeout: 480 seconds)
04:12:47 *** keikoz has joined #openttd
04:43:31 <DorpsGek> [OpenTTD/OpenTTD] eints-sync[bot] pushed 1 commits to master https://github.com/OpenTTD/OpenTTD/commit/3908df5fbe8b85a7fc28237faea5a42eec2f65e8
04:43:32 <DorpsGek> - Update: Translations from eints (by translators)
04:50:46 <peter1138[d]> How is it possible to do tis for song long yet not understand...
04:57:38 *** WormnestAndroid has quit IRC (Remote host closed the connection)
04:57:40 *** WormnestAndroid has joined #openttd
05:39:25 <DorpsGek> [OpenTTD/OpenTTD] Release workflow was not successful https://github.com/OpenTTD/OpenTTD/actions/runs/15527253049
05:40:43 *** godbed_ has quit IRC (Quit: http://quassel-irc.org - Chat comfortably. Anywhere.)
06:46:54 *** keikoz has quit IRC (Ping timeout: 480 seconds)
07:02:57 *** toktik has quit IRC (Quit: Ping timeout (120 seconds))
07:03:14 *** toktik has joined #openttd
07:06:12 <DorpsGek> [OpenTTD/survey-web] survey-summary[bot] pushed 1 commits to main https://github.com/OpenTTD/survey-web/commit/d238150fe63bc18b70f1990ff87f54ac3b733537
07:06:13 <DorpsGek> - Add: summary for week 23 of 2025 (by OpenTTD Survey)
07:27:24 *** keikoz has joined #openttd
07:41:18 <peter1138[d]> https://cdn.discordapp.com/attachments/1008473233844097104/1381538635433902110/image.png?ex=6847e19e&is=6846901e&hm=39d95ed8b24617de785745f9d81b56b69650e62ab1ec281b6a586d242684f4b7&
07:41:20 <peter1138[d]> ๐Ÿ˜ฎ
07:44:08 <peter1138[d]> Somehow needs some tweaking to make fonts of the same "size" actually be the same size...
08:01:26 *** keikoz has quit IRC (Ping timeout: 480 seconds)
08:22:53 <LordAro> ooh
08:23:54 <_zephyris> peter1138[d]: Glyph height vs line height, etc? Good luck ๐Ÿ˜‰
08:25:14 *** keikoz has joined #openttd
08:25:21 <peter1138[d]> Yeah, when a load a font at "size 20", we override the height with ascender - descender;
08:26:53 <peter1138[d]> Or +. Depends on the font API.
08:27:06 <peter1138[d]> Windows is ascender + descender.
08:27:22 <_zephyris> Ah, yeah, that's what I was going to ask
08:27:36 <_zephyris> Last time I poked those metrics it seemed wildly inconsistent
08:27:40 <peter1138[d]> Freetype gives a negative value for whatever reason.
08:27:44 <peter1138[d]> It is, yes.
08:28:17 <peter1138[d]> This is visible in a standard font viewer. When you switch through fonts without changing the size... and the size changes wildly anyway.
08:32:35 *** keoz has joined #openttd
08:34:19 *** keikoz has quit IRC (Ping timeout: 480 seconds)
08:40:49 *** keoz has quit IRC (Ping timeout: 480 seconds)
08:46:50 *** Flygon has quit IRC (Read error: Connection reset by peer)
09:06:24 *** D-HUND is now known as debdog
09:21:28 <peter1138[d]> Hmm, why does font_unix require an InitFontCache() when setting up a fallback font, but none of the other font systems have the same?
09:27:17 <DorpsGek> [OpenTTD/OpenGFX2] zephyris merged pull request #197: Fix: Corrected spelling mistake in STR_OBJECT_LIGHTHOUSE https://github.com/OpenTTD/OpenGFX2/pull/197
09:28:21 <xarick> hi
09:33:18 <peter1138[d]> peter1138[d]: Hmm, seems they do it by calling `FindMissingGlyphs()` again.
09:54:18 <_zephyris> At what point does providing a standard font with CJK etc. become the best solution?
09:54:39 <_zephyris> I'd throw Noto at it... https://github.com/notofonts
10:00:31 *** wensimehrp has joined #openttd
10:00:31 <wensimehrp> _zephyris: Noto Sans/Serif is ass for zh_CN
10:03:22 *** gelignite has joined #openttd
10:09:49 <wensimehrp> Plus given the density of strokes in CJK characters, if you were to pack OpenTTD sans glyphs with CJK glyphs from noto, you'll have to take care of the font weights.
10:09:49 <wensimehrp> An easy way to illustrate this is by comparing a piece of latin text displayed in Noto Serif and Noto Serif CJK SC. Note that when both using "regular" font weight, the CJK SC text's characters are much thinner.
10:11:43 <wensimehrp> wensimehrp: IBM Plex Sans SC doesn't have this issue, though it's missing some important glyphs
10:12:37 <wensimehrp> ChillRoundF (which is based on Zen Maru Gothic) would be a nice solution.
10:14:18 <peter1138[d]> Hmm, my packaged version of IBM Plex doesn't include CJK variants.
10:16:05 <peter1138[d]> ChillRoundF is reasonable and fits in okay.
10:17:03 <wensimehrp> peter1138[d]: Too heavy lah
10:17:03 <wensimehrp> (It's a readability issue. For me all Chinese characters in the video are bold)
10:17:04 <peter1138[d]> I'm not really sure on the best solution to having a predefined list of fallback fonts.
10:17:37 <peter1138[d]> Yes, this is because our font searcher assumes everyone wants a bold font, because the original sprite font is kinda bold.
10:18:07 <peter1138[d]> Hence why it picks the most boldest font even if it's terrible.
10:18:18 <wensimehrp> ๐Ÿซฃ
10:18:35 <peter1138[d]> The main point in the video was that it notices when a glyph is missing and automatically loads a font for it.
10:18:54 <peter1138[d]> Making it choose a better font is part 2... or maybe part 1, depending on the order of things ๐Ÿ™‚
10:21:23 <peter1138[d]> Hmm, if I remove the bold requirement I get "LXGW WenKai Mono TC, Regular".
10:22:13 <wensimehrp> That is a, uh, _jiaokeshuti_
10:22:26 <peter1138[d]> You'll have to translate that for me ๐Ÿ™‚
10:22:33 <wensimehrp> Textbook font
10:22:41 <wensimehrp> Not Sans, not Sans-serif
10:22:46 <wensimehrp> Textbook
10:22:50 <peter1138[d]> Serif.
10:23:04 <peter1138[d]> Or rather the CJK equivalent of serif.
10:23:06 <peter1138[d]> https://cdn.discordapp.com/attachments/1008473233844097104/1381579353040883736/image.png?ex=6848078a&is=6846b60a&hm=368220df1c48d56628551855947f96584e73e8e3e4cb68bc927384ef26f437da&
10:23:22 <peter1138[d]> It's probably better than the bold stuff, but still. ChillRoundF looks nicer.
10:23:24 <wensimehrp> eww
10:24:11 <peter1138[d]> <https://github.com/Warren2060/ChillRound>
10:24:26 <peter1138[d]> So... can we include this in a GPL project...
10:25:27 <wensimehrp> But it's only designed for zh_CN
10:25:35 <peter1138[d]> As a simple British person, it's difficult to know what CJK stuff is "standard" and what is stylized.
10:25:40 <peter1138[d]> Yeah, that was the problem IIRC.
10:28:32 <wensimehrp> It's based on a zh_TW font, which is based on a ja_JP font.
10:28:32 <wensimehrp> So... Take the glyphs and mix our own? ๐Ÿ˜›
10:28:48 <peter1138[d]> I wouldn't know where to start.
10:28:59 <peter1138[d]> But if the licenses permit it, and it can be done...
10:29:12 <wensimehrp> ChillRoundF
10:29:12 <wensimehrp> Justfont ็ฒ‰ๅœ†
10:29:12 <wensimehrp> Kosugi Maru
10:29:49 <peter1138[d]> There's always unifont... ๐Ÿ˜ฎ
10:30:01 <wensimehrp> oh no
10:30:55 <peter1138[d]> Quite.
10:31:40 <_glx_> peter1138[d]: According to license it seems we can
10:32:11 <peter1138[d]> Hmm, now that we have "OpenTTD Sans, Bold" as a built-in default font. I think we can drop the bold requirement for fallback fonts.
10:32:47 <_glx_> The OFL allows the licensed fonts to be used, studied, modified and
10:32:47 <_glx_> redistributed freely as long as they are not sold by themselves
10:32:52 <peter1138[d]> A lot of the non-Latin specialised fonts don't really have a bold variant, so we are excluding them.
10:33:52 <peter1138[d]> Heh, for Traditional Chinese I get Microsoft JhengHei, Bold.
10:33:56 <wensimehrp> OpenTTD Sans, Regular?
10:33:56 <wensimehrp> Inconsistencies in font weights break visual balance
10:35:07 <peter1138[d]> Does mixing non-bold CJK and bold Latin break that balance even though they look very different anyway?
10:35:54 <wensimehrp> Yes.
10:36:34 <_glx_> But latin from cjk is usually ugly
10:37:00 <wensimehrp> You mean Latin glyphs from CJK fonts?
10:37:14 <wensimehrp> In this case yes.
10:41:22 <wensimehrp> https://cdn.discordapp.com/attachments/1008473233844097104/1381583951205892116/playground2.png?ex=68480bd2&is=6846ba52&hm=de63a17b4df5eb4ae8cc1ab2dfdaecf8ccfdb8f6820957fe1a552eb69d3e8eae&
10:41:22 <wensimehrp> https://cdn.discordapp.com/attachments/1008473233844097104/1381583951503822930/playground21.png?ex=68480bd2&is=6846ba52&hm=273de8525963526ee16fe43b518ff54fe7ad6381e4323151224feea3c4984b01&
10:43:32 <peter1138[d]> Two problems, 1) creating an OpenTTD Sans, Regular is a likely a huge amount of work and 2) making the game switch between Bold and Regular depending on what other fonts are loaded, when the CJK font could be loaded later on dynamically, seems painful.
10:44:51 <peter1138[d]> e.g. using the multiplayer server list can cause a CJK font to be loaded, just because there are servers with CJK in their name. It would be undesirable to switch the rest of the game from Bold to Regular just because of that.
10:45:37 <peter1138[d]> I suppose we should be able to determine "fallback due to language files" vs "fallback due to external text", though.
10:45:47 <wensimehrp> Or change based on current language?
10:51:25 *** keikoz has joined #openttd
11:16:05 <DorpsGek> [OpenTTD/OpenTTD] PeterN opened pull request #14347: Fix: Remove U+200C characters from Lithuanian translation. https://github.com/OpenTTD/OpenTTD/pull/14347
11:19:31 <DorpsGek> [OpenTTD/OpenTTD] glx22 approved pull request #14347: Fix: Remove U+200C characters from Lithuanian translation. https://github.com/OpenTTD/OpenTTD/pull/14347#pullrequestreview-2909765045
11:23:35 *** keoz has joined #openttd
11:26:26 *** keikoz has quit IRC (Ping timeout: 480 seconds)
11:30:36 <peter1138[d]> Heh, Github's diff viewer isn't great for that ๐Ÿ™‚
11:36:28 *** WormnestAndroid has quit IRC (Remote host closed the connection)
11:36:50 *** WormnestAndroid has joined #openttd
11:45:24 *** locosage has joined #openttd
11:45:24 <locosage> hm, I'm getting `error: redefinition of โ€˜struct fmt::v11::formatter<T, Char>โ€™` when compiling 15.0-beta2, is that a known thing?
11:47:00 <_glx_> Never seen
11:51:43 <locosage> hm, is gcc-11.4 recent enough to build it?
11:53:16 <locosage> arikover: nvm, apparently I need 12+
11:53:34 <DorpsGek> [OpenTTD/OpenTTD] PeterN merged pull request #14347: Fix: Remove U+200C characters from Lithuanian translation. https://github.com/OpenTTD/OpenTTD/pull/14347
11:54:43 <locosage> guess it's time to finally upgrade my ubuntu
12:07:07 <debdog> DON'T! upgrades bring sorrow
12:07:51 <debdog> (well, not uprgading does, too)
12:08:42 <locosage> yeah, have to balance
12:09:03 <locosage> kinda sad, I really liked this version, everything worked on it for once
12:09:54 <debdog> I feel you
12:23:58 *** _jgr_ has joined #openttd
12:23:58 <_jgr_> You don't have to upgrade to use gcc 12
12:24:58 <_jgr_> I haven't got round to upgrading either, but you can just apt install gcc-12 or whatever and reference it in your call to cmake
12:25:11 <LordAro> but you probably should, ubuntu 20.04 is EOL
12:25:53 <LordAro> oh no, 22.04 is 11.2
12:26:13 <LordAro> ...which means you must be on 22.10 or something, which is even more EOL
12:28:31 <locosage> I'm on 22.04
12:29:02 <locosage> but I'll have to upgrade eventually, so might as well
12:29:24 <LordAro> then you must have already done "something" to get 11.4
12:31:39 <locosage> hm, maybe I did
12:31:52 <locosage> I also have 12.3 installed apparently ๐Ÿ˜…
12:33:54 <locosage> lol, but only gcc not g++ apparently
12:36:15 <locosage> but I've already started the update so I'll go for the shiny new stuff xD
12:39:07 <_jgr_> In 22.04 the gcc-11 is version 11.4 but the top level gcc metapackage is still 11.2
12:39:52 <LordAro> aha
12:55:10 <pickpacket> peter1138[d]: this looks amazing: https://github.com/OpenTTD/OpenTTD/pull/14319 Who made the art so high def?
12:55:17 <pickpacket> it's beautiful
12:56:06 <LordAro> if in doubt, probably zeph
13:22:48 <_zephyris> The faces, from way back in the early 2010s, were Ben Robbins
13:23:37 <_zephyris> Nice to be able to finally use them in 32bpp high def easily in the base set
13:34:27 <xarick> another bug found ๐Ÿ™‚
13:37:10 <pickpacket> _zephyris: so you made the 32bpp adaptation? :)
13:37:14 <pickpacket> well done!
13:37:16 <xarick> forgot `[&vli](VehicleType type) { return type == vli.vtype; },`
13:37:30 <_zephyris> pickpacket: Reworked from the original source
13:37:49 <pickpacket> ๐Ÿ‘
13:39:17 <pickpacket> I need to get started on fixing that string stripping issue I found
13:55:38 <DorpsGek> [OpenTTD/OpenTTD] 2TallTyler commented on pull request #14319: Change: Company manager face definitions and dynamic UI. https://github.com/OpenTTD/OpenTTD/pull/14319#pullrequestreview-2910126192
14:27:12 <DorpsGek> [OpenTTD/OpenTTD] PeterN commented on pull request #14319: Change: Company manager face definitions and dynamic UI. https://github.com/OpenTTD/OpenTTD/pull/14319#pullrequestreview-2910249855
14:44:02 <xarick> iteration overhead is noticeable
14:44:37 <_zephyris> I'm weirdly excited by the idea of NewGRF faces
14:45:14 <xarick> my templaytes are probably bad too
14:45:47 <_zephyris> It's a good casual player feature - good for personalisation, good for humanising multiplayer players, etc.
15:07:32 <peter1138[d]> Well, we need to do the spec design before getting excited :p
15:09:25 <_zephyris> Well, yeah, but good to be motivated!
15:17:44 *** Wormnest has joined #openttd
15:24:02 <andythenorth> can we make them related to status?
15:24:07 <andythenorth> so you get a crown for Tycoon?
15:30:44 <peter1138[d]> Depends what varactions we have ๐Ÿ™‚
15:46:07 *** talltyler has joined #openttd
15:46:07 <talltyler> ~~lootbox~~ unlockable accessories would be interesting, although those would best be controlled by GS rather than the built-in (somewhat silly) scoring. ๐Ÿ™‚
15:46:46 <talltyler> โ€œDeliver 10k tons of coal to unlock the engineerโ€™s hatโ€
15:50:40 <talltyler> Or coal minerโ€™s helmet, more appropriately
15:52:56 <andythenorth> no we sell DLC ๐Ÿ˜›
15:52:59 <andythenorth> to fund infra
15:53:12 <andythenorth> buy a hat, pay for Bananas fees
15:55:18 <_zephyris> https://cdn.discordapp.com/attachments/1008473233844097104/1381662955178823870/image.png?ex=68485566&is=684703e6&hm=85a4d72ee1844e346672bf21bf3ff204bde266a9375e925a2de8a3fb14b46567&
15:56:08 *** WormnestAndroid has quit IRC (Remote host closed the connection)
15:56:09 *** WormnestAndroid has joined #openttd
(this day is still ongoing; reload to check for any updates)