IRC logs for #openttd on OFTC at 2025-04-14
⏴ go to previous day
01:59:50 *** Wormnest has quit IRC (Quit: Leaving)
02:17:07 *** gnu_jj_ has quit IRC (Ping timeout: 480 seconds)
03:54:00 *** debdog has quit IRC (Ping timeout: 480 seconds)
04:44:54 <DorpsGek> - Update: Translations from eints (by translators)
06:05:18 *** kuhnovic has joined #openttd
07:17:59 <DorpsGek> - Add: summary for week 15 of 2025 (by OpenTTD Survey)
08:12:14 <peter1138[d]> Hole in the ground?
08:12:42 <andythenorth> alpha sorted badges by default, easier to read in debug, even if OpenTTD does other things elsewhere
08:28:18 *** mindlesstux has joined #openttd
08:31:44 <pickpacket> LordAro: was that a reply to me?
08:33:34 <andythenorth> these can't (and shouldn't) be badges? (would need numeric values translated to user unit settings)
09:17:17 *** D-HUND is now known as debdog
09:23:54 <peter1138[d]> andythenorth: They're shown in listed order because that makes a difference and might be something you want to debug.
09:24:29 <andythenorth> yes, it makes sense
09:24:58 <andythenorth> I might use that later to control order
09:33:55 <andythenorth> no way to check badges on a railtype currently? I'm not just being dense?
09:36:46 <peter1138[d]> Should be in the debug window.
09:40:56 <andythenorth> I suspect that setting a badge for the label is a Bad Sign
10:06:34 <peter1138[d]> Someone will hit the internal 65535 badge limit, won't they...?
10:07:55 <andythenorth> I suspect that nmlc is counting wrong `nmlc info: Badge translation table: 118/65536`
10:08:12 <andythenorth> this increases my suspicion 😛
10:08:29 <andythenorth> I am going afk, but I will look at that later
10:08:46 <andythenorth> Horse has 898 badges
10:09:26 <andythenorth> my bet would be object sets get near the limit first 😛
10:09:35 <andythenorth> especially ground tile replacement sets
10:12:24 <reldred> I need to stop doing other sidequests and play with badges
10:15:11 <reldred> but I have tiny tanks and planes and spaceships to print
10:15:27 <reldred> and a proxmox cluster to build just as soon as I order this first switch
10:15:39 <peter1138[d]> 65536 should probably be 65535 too.
10:18:38 <pickpacket> LordAro: okay. So it's in beta2 but I forgot something in my commit. Got it. Thanks 🙂 I'll figure out what I missed until next time
10:19:44 <peter1138> pickpacket, "no, and yes"
10:20:10 <peter1138> "Did I forget something in my commit?" no. "Is it in beta2?" yes.
10:20:59 <peter1138> The changelog document is manually put together. The reviewers (including me) missed it, because remembering every change is hard.
10:21:06 <pickpacket> Me no read can dotay
10:22:46 <pickpacket> Yeah, it's really hard. I thought it was auto-compiled
10:24:26 <LordAro> well, it starts with the git log, so it's more a case of remembering/working out which changes are user facing
10:24:34 <LordAro> rather than remembering every individual change
10:25:42 <andythenorth> `nmlc info: Badge translation table: 898/65535` works
10:27:31 <LordAro> my guess is the checkout that RB had when generating the changelog was out of date, so it slipped through the cracks
10:34:41 <peter1138> andythenorth, see, I knew breaking through 255 badges would be easy peasy.
10:37:34 <peter1138> iron-horse.grf 40,193,726, iron-horse.grf.gz 5,078,846
10:37:59 <pickpacket> LordAro: don't worry about it 🙂 at least not for my sake
10:39:30 <andythenorth> yes grf compresss a lot 😛
10:39:37 <peter1138> Hmm, grfcodec can't decode a stripped grf.
10:39:38 <andythenorth> all those runs of 00
10:40:28 <peter1138> Can we gzip the non-image part and have the game decompress while reading...
10:40:56 <peter1138> grf container version 3.
10:41:20 <peter1138> Does it matter? No.
11:02:40 <andythenorth> ok nearly run out of things to badge 😛
11:05:46 <andythenorth> it's cute that we measure vehicle length in depot as decimals
11:06:57 <andythenorth> and we round the total length, not each unit
11:09:37 <andythenorth> or 6/16 in old money
11:11:26 <andythenorth> fractions are hard
11:36:52 <_zephyris> Fractions would be nicer!
11:44:07 <peter1138[d]> Remember it's tiles though, so 16ths not 8ths.
11:45:01 <frosch123> Should use hex obviously
11:46:30 <peter1138[d]> Players will definitely understand this 🙂
11:47:42 <peter1138[d]> As someone who grew up with learning Basic was normal, I'm familiar with hexadecimal, but I don't think I've ever use hexadecimal decimals...
11:55:10 <peter1138[d]> So this makes sense, yeah? 1/16, 1/8, 3/16, 1/4, 5/16, 3/8, 7/16, 1/2, 9/16, 5/8, 11/16, 3/4, 13/16, 7/8, 15/16, 1
11:55:28 <frosch123> So. Who noticed that I miscounted? Should have been 0.6,0.c, 1.2
11:56:03 <frosch123> Isn't it easier to always use /16?
11:56:39 <peter1138[d]> Yes, much like we always like partial tile acceptance in 8ths.
11:57:04 <frosch123> It could also be some visual bar, which fills up
11:57:40 <peter1138[d]> (Do we have gyphs for superscript characters?)
11:58:07 <peter1138[d]> You could represent the width by a horizontal visualisation of the vehicle.
12:04:23 <_glx_> peter1138[d]: Looks like american wrenches
12:07:27 <peter1138[d]> And then they have thou.... thousandths of an inch, because metric and decimals are too much.
12:07:49 <andythenorth> when can we have 1/16 length increments?
12:07:56 <andythenorth> sub pixel pixels
12:08:06 <peter1138[d]> When I get around to me tile-progress system.
12:08:15 <andythenorth> we have too many side quests
12:08:49 <peter1138[d]> Remember the ancient suggestion of having signals between tiles instead of on tiles?
12:09:07 <andythenorth> can't tell if BADGE_FLAG_COPY_TO_RELATED_ENTITY is not working
12:09:24 <peter1138[d]> If it's copied it's working.
12:09:38 <andythenorth> it's railtypes so 90% chance I'm not understanding what my own vehicles do
12:10:28 <peter1138[d]> Problem with signals between tiles is you then have signals between tiles AND signals on tiles, because compatibility...
12:11:24 <andythenorth> seems the engine sets IHD_ for track_type, and the railtype is IHD_
12:11:49 <andythenorth> but the pff discord formatting
12:11:57 <peter1138[d]> That doesn't make much sense, as tracktype and railtype are the same.
12:12:25 <peter1138[d]> Ah, on IRC they are the same 🙂
12:13:12 <andythenorth> never understood why we have railtypes, but nml sets tracktype for engines
12:13:29 <peter1138[d]> Because the original grf spec only mentions track type.
12:13:32 <peter1138[d]> Railtypes came later.
12:13:38 <andythenorth> now I understand
12:14:00 <andythenorth> anyway, the railtype badge isn't listed in vehicle debug, so I did a wrong
12:14:03 <peter1138[d]> NML could've used the same word.
12:14:58 <peter1138[d]> Stick a breakpoint in `AppendCopyableBadgeList()` and see what happens.
12:18:35 <andythenorth> I need to change cmake flags to debug build I assume?
12:18:41 <andythenorth> never used the debugger....
12:19:22 <peter1138[d]> You don't need a debug build to have a breakpoint.
12:19:56 <peter1138[d]> Although because of optimisations some variables are not available, and some things are out of order.
12:20:24 * andythenorth learning what the debugger is on macOS
12:20:29 <andythenorth> thought it would be gdb, but I don't have it
12:21:59 <peter1138[d]> Very commit messages.
12:22:13 <andythenorth> "in which we discover I never learnt to debug"
12:22:52 <andythenorth> probably there's a var with the Badge instance in context?
12:23:28 <peter1138[d]> I run lldb from within VS Code, I don't need to know how to use it :p
12:24:15 <peter1138[d]> Maybe just debug by sticking in a Debug() mesage 🙂
12:24:35 <peter1138[d]> Oh but you'd probably need to `#include "debug.h"` as well
12:25:04 <peter1138[d]> `Debug(grf, 0, "Badge {}", badge->label);`
12:26:58 <_glx_> Yeah `Debug()` is often enough
12:44:32 <talltyler> peter1138[d]: Nah, forcibly move everybody's signals in Afterload 😛
13:23:54 <xarick> machine learning in 1992?
13:30:06 <LordAro> the idea's been around since the 60s
13:46:22 <xarick> "Carl"'s existance in Dark Fate finally making sense
14:46:34 <mnhebi> xarick: well to be fair the T-800 has the chip size of a chocolate bar which superconducts at room temperature...so it'd be quite a bit ahead even the best computer of today
14:48:06 *** Wormnest has joined #openttd
14:50:52 <mnhebi> also, at that size the chip would be a quantum chip so it'd be quite useful for AI purposes
14:52:00 <mnhebi> also, its done on a hypercube design so delays are small..add superconduction to the mix and you basically have a chip that thinks at the speed of light.
14:54:47 <LordAro> you seem to be trying to apply science to science fiction
14:56:46 <mnhebi> well you can't blame me, the genre contains the word science. :p
14:57:26 <mnhebi> and the idea of the T-800s chip follows some very real theories about the capabilities CPUs would be able to reach if they were able to superconduct
14:57:53 <mnhebi> Cameron didn't pull it all out of his ass
14:58:55 <andythenorth> `produce_tech_tree_badges took 0.2637 seconds` I have regrets
14:59:11 <mnhebi> and its the reason why theres always someone willing to rush to say "we figured out superconducting materials!" only to be discovered as a quack a week after.
14:59:53 <LordAro> i'm aware of the possibilities of room temperature superconductors, yes
15:36:02 *** Wormnest has quit IRC (Ping timeout: 480 seconds)
15:43:49 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
15:45:25 *** WormnestAndroid has joined #openttd
16:27:54 *** WormnestAndroid has quit IRC (Remote host closed the connection)
16:27:58 *** WormnestAndroid has joined #openttd
16:28:12 <michi_cc[d]> Well, the UK fraction hasn’t shown up the last two times, has it?
16:29:39 <michi_cc[d]> Oh, and from scrollback, GRF container version 2 is already prepared for a compressed data section. No need for a new container version 😛
16:31:46 <peter1138[d]> Uh oh. No. No side quest. Bad.
16:39:23 *** gelignite has joined #openttd
18:00:54 <michi_cc[d]> pickpacket: It wasn't picked up by the changelog script because the actual commit message (not the PR title) starts with "Codechange", so it was filtered out. Someone should have made a squash merge to rectify that, but humans be humans.
18:07:03 <michi_cc[d]> The number must grow 🙂
18:12:41 <peter1138> Hmm, pi-hole crashed and didn't auto restart. Time to blame DNS and it be accurate.
18:28:22 <michi_cc[d]> Three triangles, start a new design trend
18:35:30 <_zephyris> Dots normally means more, like elipsis, burger normally means menu
18:42:54 *** ChanServ sets mode: +v tokai
18:43:24 <andythenorth> 3 vertical dots now covers all sins for 'there is no useful icon'
18:45:38 <_zephyris> Are badges labels? Would a label icon work?
18:45:38 <peter1138[d]> Add both and animate it.
18:49:57 *** tokai|noir has quit IRC (Ping timeout: 480 seconds)
18:55:23 <peter1138> Hmm, right, need to switch to not-OpenGFX 2 because I'm not making a high-def default sprite.
18:58:54 <peter1138> Okay, can't use antialiasing either.
19:00:43 *** kuka_lie has joined #openttd
19:02:04 <peter1138[d]> Maybe 3 px wide, compromise between dots and lines 😄
19:02:20 <peter1138[d]> (3 thick lines doesn't work, 3 thin lines looks weird too...)
19:03:55 <peter1138[d]> I dunno. At least the button is the same width now.
19:09:20 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
19:09:22 *** WormnestAndroid has joined #openttd
19:21:13 <xarick> is there a delete all towns in scenario editor?
19:29:17 <andythenorth> peter1138[d]: "this is fine"
19:29:18 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
19:29:38 <andythenorth> we could overthink it to no benefit
19:29:45 <_zephyris> I think it's good too, only other suggestion is a label
19:29:55 *** WormnestAndroid has joined #openttd
19:39:09 *** Wormnest has joined #openttd
19:39:36 <frosch123> I consider it a "configure list/view" button, which may gain more options than just labels
19:40:37 <frosch123> So I prefer the ellipsis icon over a label specific one
19:49:43 *** Flygon has quit IRC (Remote host closed the connection)
19:50:52 <_zephyris> Left/right arrows instead of up/down?
19:51:03 <_zephyris> Presumably they're using the scrollbar icons...
19:55:17 *** kuka_lie has quit IRC (Quit: Lost terminal)
20:00:46 <peter1138[d]> _zephyris: Kinda, it's actually glyph versions of them, so that they get recoloured like text.
20:01:01 <peter1138[d]> SCC_UP_ARROW and SCC_DOWN_ARROW already exist.
20:01:09 <peter1138[d]> SCC_RIGHT_ARROW exists too, but SCC_LEFT_ARROW does not.
20:02:53 <_zephyris> Yeah, I thought one was missing
20:03:03 <_zephyris> Couldn't quite remember the details
20:03:49 <_zephyris> Presumably changes vertical ordering in the text representation as well as icon column order? So up/down is fine.
20:09:37 <frosch123> I though left arrow was added for rtl support
20:31:52 <peter1138[d]> So originally sprites 130 and 140 were the up and down arrow sprites.
20:33:46 <peter1138[d]> Hmm, and they still are for UP_ARROW and DOWN_ARROW. Used in loading indicators, and here.
20:34:42 <peter1138[d]> There's also two small up and down arrows.
20:35:22 <peter1138[d]> Not sure those are used 🙂
20:36:15 <peter1138[d]> Neither are the small left/right arrows, which are actually just < and >.
20:38:34 <peter1138[d]> The arrows were replaced with sprites at some point (by me.)
20:43:16 <peter1138[d]> So anyway, the up/down/left/right arrows are different, and are not in the range available to `SCC_SPRITE_xxx`. The old up and down sprites are still there of course.
20:43:36 <peter1138[d]> Now, I could just draw them as sprites instead of text.
20:44:40 <peter1138[d]> But then I can't recolour them for the few fractions of a second when the text is drawn white.
20:47:44 <frosch123> Could be intentional, the item is selected. Other buttons are not part of the item
20:50:16 <_zephyris> To be a UI consistency curmudgron, does anywhere else in the UI have non-framed flat buttons?
20:59:43 <peter1138[d]> There's no neat way to set the clicked state though.
21:01:26 <peter1138[d]> Hmm, unless I tweak the bodge I already added.
21:06:42 <frosch123> Clicked state probably looks weird, if the button moves under the cursor
21:07:57 <_zephyris> I've poked some other software, column order type ui elements are often clunky
21:08:21 <_zephyris> No clear gold standard to copy!
21:14:48 <peter1138[d]> Without buttons?
21:15:06 <peter1138[d]> This is like andy choosing livery icon shape.
21:16:35 *** gelignite has quit IRC (Quit: Stay safe!)
21:18:24 <peter1138[d]> I guess buttons make it more obvious you can press them separately to the dropdown item.
21:31:05 *** keikoz has quit IRC (Ping timeout: 480 seconds)
21:36:38 *** xXWHITEOUTXx has joined #openttd
21:36:50 <xXWHITEOUTXx> how do i play the game?
21:37:08 <xXWHITEOUTXx> I DONT UNDERSTAND
21:37:11 *** xXWHITEOUTXx has quit IRC ()
21:38:40 <peter1138[d]> Hmm, I'm missing a Rect centring function.
21:55:15 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
22:34:00 <peter1138[d]> > build finished successfully.
22:34:33 <peter1138[d]> I changed interface scale and the button is no longer the same width, cry.
22:35:37 <peter1138[d]> Side quest, perhaps.
22:48:30 *** nielsm has quit IRC (Ping timeout: 480 seconds)
23:28:40 *** SigHunter has quit IRC (Ping timeout: 480 seconds)
23:39:52 *** Wormnest has quit IRC (Ping timeout: 480 seconds)
23:41:47 *** SigHunter has joined #openttd
23:50:52 *** felix has quit IRC (Ping timeout: 480 seconds)
continue to next day ⏵
