IRC logs for #openttd on OFTC at 2025-07-15
⏴ go to previous day
00:28:19 *** Wormnest has quit IRC (Ping timeout: 480 seconds)
00:40:44 *** Wormnest has joined #openttd
00:52:56 *** toktik is now known as Guest21963
00:58:54 *** Guest21963 has quit IRC (Ping timeout: 480 seconds)
02:01:05 *** Wormnest has quit IRC (Quit: Leaving)
03:02:59 *** gnu_jj_ has quit IRC (Ping timeout: 480 seconds)
04:00:37 *** tokai|noir has joined #openttd
04:00:37 *** ChanServ sets mode: +v tokai|noir
04:07:34 *** tokai has quit IRC (Ping timeout: 480 seconds)
04:27:14 *** Flygon has quit IRC (Remote host closed the connection)
04:42:54 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
04:57:49 *** WormnestAndroid has joined #openttd
06:40:25 *** cu-kai has quit IRC (Quit: Leaving for a bit.)
06:42:09 *** dwfreed has joined #openttd
08:24:54 *** SigHunter has quit IRC (Ping timeout: 480 seconds)
08:25:31 <LordAro> peter1138[d]: i have a lot of thoughts on the changelog as it is, do you mind if i take it over?
08:48:39 *** SigHunter has joined #openttd
09:05:29 *** SigHunter has quit IRC (Ping timeout: 480 seconds)
10:04:56 *** dh1 has quit IRC (Quit: My Mac has gone to sleep. ZZZzzz…)
10:17:24 *** SigHunter has joined #openttd
10:25:00 *** Smedles has joined #openttd
11:03:30 *** reldred has quit IRC (Quit: User went offline on Discord a while ago)
11:15:43 <peter1138[d]> Ooops, left a function lying around.
12:17:42 <peter1138[d]> My philosophy, do it badly enough and someone else will take over.
12:18:47 <LordAro> you must be quite bad at applying your philosophy :p
12:21:53 <locosage> he's waiting for someone to take that over too 🤭
13:12:47 <talltyler> Looking at the milestone for 15.0… Anyone want to review #13289 and/or #14408, or shall I approve the changelog?
13:13:26 <talltyler> I will update the website post either way 🙂
13:16:21 <LordAro> i'm not brave enough for either of those
13:16:59 <LordAro> still feels like freeform_edges & water_borders have enough overlap to be able to be reused
13:20:22 <talltyler> `water_borders` is already reused, but `freeform_edges` would only work for new games, since it has no void tiles.
13:21:13 <talltyler> I'm not opposed to only allowing this for new games, but the current design automatically upgrades existing saves if they are surrounded by water.
13:21:58 <LordAro> perhaps freeform_edges needs renaming to something more obviously "this is legacy"
13:24:23 <peter1138[d]> What should it be set to with this PR?
13:24:30 <talltyler> It's already hidden from the menu, it only exists because of old savegames. I suppose you could set it in the config, but I'm not sure renaming it would do much.
13:24:48 <peter1138[d]> There's a certain amount of doubling up going on with behaviours, mainly due to past changes.
13:26:15 <talltyler> The PR sets all bits of `water_borders`. It does not care whether `freeform_edges` is set or not, and does not change it.
13:27:26 <talltyler> (If `freeform_edges` is off, `water_borders` will have all bits set, by definition. But the inverse is not true.)
13:28:26 <peter1138[d]> How do I ensure I don't get infinite water?
13:29:02 <talltyler> Set one of the edges to freeform. There is no way to get water on all edges, without infinite water.
13:31:02 <talltyler> Somehow that spacebar heating change survived 6 months without pitchforks, although Zorg did suggest NewGRF map edges `to create graphics of elephants lifting the Earth and space with stars around, or the ice ring that some believe surrounds the flat Earth.` 😉
13:42:59 *** asasnat has joined #openttd
13:42:59 <asasnat> and the landmasses that some believe exist beyond the ice ring
14:21:23 <peter1138[d]> Elephants on top of a turtle, yes. pTerry was such a plagiarist...
14:31:37 <peter1138[d]> Hmm. Can't see how to turn it off for my test savegames.
14:33:17 <talltyler> Infinite water? It cannot be turned off.
14:52:36 <orudge> Hmm, an e-mail from Atari (which appears to be genuine). "We would like to find a way to work with OpenTTD so that our game can be supportive of the OpenTTD community while keeping our rights to the game protected." :-/
14:56:18 <kuhnovic> "15.0 postponed indefinitely"
14:57:10 <orudge> At the same time, somebody sent an e-mail - I think via a 'domain broker' type thing - asking if we would sell openttd.com for like $2000. I ignored it. I wonder if it's just a coincidence...
14:57:41 <orudge> (this was a couple of months ago though)
15:06:11 <talltyler> At the same time or a couple months ago?
15:09:07 <orudge> well, it wasn't literally the same time, no. It was a couple of months ago. Could be entirely unrelated, but who knows.
15:10:17 <talltyler> A couple of months earlier sounds unrelated to me, also, if they “want to work with” us, it wouldn’t make sense to buy our domain out from under us.
15:30:05 <orudge> Anyway, I guess I'd best reply in some way
15:34:04 <belajalilija> please keep us updated
15:45:31 <_zephyris> "Hi, how would you like to support us and what are your concerns..."
15:47:25 <peter1138[d]> I prefer not to speculate... so I won't.
15:47:30 <kale91> well that's interesting
15:49:58 <kale91> I mean they have to know that it's popular because it's open source
15:50:26 <kale91> It wouldn't have a playerbase if we were all still playing TTDLX
15:59:00 *** Wormnest has joined #openttd
15:59:38 *** gelignite has joined #openttd
16:02:13 <mnhebi> goddamnit Chris did you need money for new Land Rover or something
16:04:06 <belajalilija> mnhebi: wouldnt surprise me, isnt he on record saying stuff along the lines of that he doesnt like when people adjust what he has made and essentially calling any bugs features?
16:05:06 <belajalilija> i mean, that sound exactly like me, so i get it, but still
16:37:17 *** Smedles has joined #openttd
16:45:14 <peter1138[d]> Urgh, C# generics are pissing me off 😮
17:41:02 <pickpacket> orudge: I, maybe naïvely, thought the legal situation was sorted a long time ago. Would be very interesting to know what they mean by "protecting their rights"
17:44:28 <michi_cc> No legal situation is ever "sorted" until you have a judgment from the highest applicable court.
17:44:56 <michi_cc> And if you are POTUS, even that doesn't matter anymore.
17:48:45 <pickpacket> I only have a layman's view of things, but I can't see how they have any rights over anything relating to OpenTTD. But I don't know. As peter said it's probably best not to speculate
18:50:44 <andythenorth> it's definitely best not to speculate
18:50:56 <LordAro> what if we speculate about speculating?
18:51:36 <andythenorth> LordAro: sounds quite speculative
18:52:58 <LordAro> i think it was dinner
18:53:47 <andythenorth> I made sausage and mash
18:57:39 <andythenorth> ok let's see how badly my game breaks if I add cargos
19:01:35 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
19:01:37 *** WormnestAndroid has joined #openttd
19:01:50 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
19:01:51 *** WormnestAndroid has joined #openttd
19:01:56 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
19:01:57 *** WormnestAndroid has joined #openttd
19:02:37 <peter1138[d]> My patch to add more cargoes is somewhat out of date again.
19:03:10 <peter1138[d]> But, uh, what's the point eh.
19:03:30 <peter1138[d]> slowah did you say?
19:04:09 <peter1138[d]> Anyway, also trying to understand which bits of JGRPP's sprite cache I can steal.
19:04:32 <peter1138[d]> But as it's again vastly different from the work that we recently merged, I suspect... none.
19:05:00 <truebrain> `cp -R ../jgrpp/src/* src/`? 🙂
19:05:23 <truebrain> "I pulled this one string, and the rest came with!"
19:07:32 <peter1138[d]> Well, we moved to unique_ptr<std::byte[]>, but JGRPP sticks with malloc/free.
19:08:25 <peter1138[d]> So getting it wrong means leaking memory.
19:09:41 <truebrain> I hadn't checked survey reports in a long time, but really nobody uses 8bpp blitters anymore? Very nice. Can we now remove it? 😄
19:11:00 <peter1138[d]> Silly `Sprite` being variably sized.
19:11:20 <andythenorth> truebrain: my cargos are tomotoes and potatoes
19:12:15 <peter1138[d]> truebrain: Sure, and I guess remove the recently added 32bpp-to-8bpp conversion code too.
19:12:40 <truebrain> I like that artstyle too much 🙂
19:15:59 <peter1138[d]> > allocating the structure and the data with a single allocation is a good thing. You halve the number of allocations that way, and you improve the locality of data as well. This can improve the performance quite a bit if you allocate lots of packages.
19:15:59 <peter1138[d]> > Unfortunately c++ does not provide a good mechanism to do this,
19:16:52 <peter1138[d]> Does JGRPP still encode zoomed out sprites?
19:17:11 <peter1138[d]> At 8x zoomed out, some of those are just a couple of pixels...
19:17:31 <peter1138[d]> For vanilla it's "not too bad" because it's packed into the single byte array.
19:20:24 <_jgr_> Yes, I'm still doing the encode for each zoom level on demand. The main rationale for separating them out is to not have to unnecessarily encode and store the really big fully zoomed in sprites. Very small zoomed out ones aren't such an issue.
19:20:54 <andythenorth> do cargo abbreviations still have to be 2 characters?
19:21:00 <andythenorth> dare I call them 'silly'? 😛
19:22:58 <peter1138[d]> I know why they're separated, but that seems to add a bit of overhead as each sprite for each zoom level has 32 bytes of metadata surrounding it. Outweighed by not encoding 4x 32bpp, but still.
19:24:13 <peter1138[d]> I... really wonder if we could/should go to runtime scaling, so things like 1x 8bpp sprites aren't needlessly upscaled to 4x 40bpp, but I guess performance suffers.
19:24:39 <peter1138[d]> Even better if we can offload it to GPU but that is a bit of a lost cause.
19:25:12 <truebrain> At least the endless stream of bug reports about crashing Intel drives stopped, I think? 🙂
19:26:05 <andythenorth> adding a cargo does break my savegame a lot 😄
19:27:12 <peter1138[d]> The current version of my fractional sprite scaling is very hack and without something like the optional on-demand zoom levels that JGRPP has it's a bit of crap idea.
19:28:14 *** Smedles has joined #openttd
19:28:21 <truebrain> as in, fluent zoom-in/zoom-out, or more a fixed 1.25x zoom level?
19:28:42 <peter1138[d]> Eh, properly fractional, but only for interface sprites, not viewport sprites.
19:28:59 <truebrain> so you can set your interface at 1.25x?
19:29:06 <truebrain> or what-ever value you fancy, ofc
19:29:23 <peter1138[d]> Yeah. You can do that right now, but the sprites in the interface snap to 1x, 2x or 4x as nearest.
19:29:34 <truebrain> How often do people change that value?
19:29:40 <truebrain> and how long does pre-calculating them take?
19:30:06 <peter1138[d]> Eh, they're encoded on demand just like normal scaling, so that's not really noticable.
19:31:02 <peter1138[d]> Being able to do entire viewport would be a more useful feature though.
19:31:16 <peter1138[d]> But for that, it really wants runtime scaling.
19:31:35 <peter1138[d]> 3D models when 😮
19:33:41 <peter1138[d]> > opengl18.diff 2008-11-09 16:18 21K
19:34:56 <peter1138[d]> (Not the current "opengl" blitter, my ancient experiment that actually blitted with opengl)
19:35:11 <truebrain> on a surface? So you can scale it? 😄
19:36:03 <peter1138[d]> I guess? I was all immediate-mode, so to do anything it had to redraw it all every frame.
19:36:08 <peter1138[d]> That's what make its performance suck.
19:36:28 <truebrain> 0.8% still have a 32bit system .. how?!
19:36:30 <peter1138[d]> That an in 2008 I was using a Geforce 8800GT with 512MB, I think.
19:37:11 <truebrain> Still 23% play without hardware acceleration .. oof.
19:37:21 <peter1138[d]> Yes, it often doesn't add anything.
19:37:23 <truebrain> MacOS being the #1 ofc
19:38:01 <peter1138[d]> Funnily enough it requires the CPU to move more data about...
19:38:24 <andythenorth> hmm maybe a dirty protest about 2 char cargo abbrevs 😛
19:38:30 <andythenorth> I just make them numbers 1...99
19:38:51 <peter1138[d]> If you can't abbreviate to 2 letters, you have too many cargo types.
19:39:07 <peter1138[d]> But also, the game doesn't actually care.
19:39:34 <andythenorth> I've added 64 types of Steel
19:39:40 <peter1138[d]> Sounds about right.
19:39:52 <andythenorth> and I think 4 chars works
19:40:04 <andythenorth> major feature, very important
19:40:18 <truebrain> Why do you hate players? 😛
19:40:27 <truebrain> 64 types of Steel ...
19:40:37 <truebrain> "99% pure", "98% pure"? 😄
19:40:41 <andythenorth> hate the players, don't hate the game
19:40:50 <andythenorth> I like the purity idea
19:41:02 <truebrain> if you mix 95% and 99% in a single wagon, does it becomes 97%?
19:41:04 <peter1138[d]> 01...99% pure, infact.
19:41:33 <andythenorth> if you wait long enough it just becomes entropy
19:41:36 <peter1138[d]> (What is "pure" steel anyway, given what steel is...)
19:41:37 <andythenorth> how long do we have?
19:41:37 <truebrain> - Cow (1 month old)
19:41:37 <truebrain> - Cow (2 months old)
19:41:37 <truebrain> - Cow (3 months old)
19:41:37 <truebrain> Can you just imagine? 😄
19:41:48 <truebrain> peter1138[d]: No signs of uranium ofc
19:42:30 <andythenorth> it's ok, devs don't play 😛
19:42:52 <peter1138[d]> Honestly, I used to love playing with NewGRFs back when UKRS was the best thing since sliced bread.
19:43:06 <peter1138[d]> I tend to just do vanilla these days.
19:43:11 <andythenorth> it's all about being able to then choose the appropriate train wagon
19:43:22 <truebrain> who has time for that? 😄
19:43:25 <peter1138[d]> I made Iron Horse too big 😦
19:43:40 <peter1138[d]> FIRS is not my fault though.
19:43:42 <truebrain> I always envy those people that spend hours on picking the right outfit for their character
19:43:44 <andythenorth> obviously the perfect wagon is
19:43:47 <truebrain> this is OpenTTD's equivilant 🙂
19:44:11 <truebrain> I like how sharp the text is 🙂
19:44:27 <peter1138[d]> The one bit that is variably scalable.
19:44:40 <andythenorth> did I unbreak my font?
19:44:54 <andythenorth> we improved so much
19:45:07 <peter1138[d]> Yes, I think you'd managed to set it to the string "default" instead of just nothing.
19:45:17 <peter1138[d]> And then proclaimed that fonts just don't work on your system.
19:45:31 <andythenorth> it was technically true
19:45:38 <andythenorth> they don't work if I set the settings wrong
19:45:45 <peter1138[d]> Leading me to wonder what things I broke on OSX... when it turns out that, for a change, I didn't.
19:45:53 <truebrain> andythenorth: Why is #44's engine going the other way?
19:45:55 <andythenorth> what you don't break, I will
19:46:13 <andythenorth> truebrain: this is like "why does FIRS have 5 parts"
19:46:22 <andythenorth> it's in the top 3 reddit topices
19:46:35 <truebrain> but it is going backwards (compared to the rest)
19:46:39 <truebrain> how are they getting home?
19:46:53 <andythenorth> I'll change the setting for you 😉
19:47:16 <truebrain> Why is it a setting in the first place?! Engines should go forward!
19:47:29 <andythenorth> the back cab is just for aesthetics
19:47:39 <truebrain> But it is in the front!
19:47:44 <andythenorth> also trains just reverse in stations 😮
19:47:49 <truebrain> Now the back cab is in the front .... nooooo
19:47:59 <truebrain> andythenorth: But that is just teleportation
19:48:01 <truebrain> nothing wrong with that
19:48:01 <andythenorth> I have missed our chats 🙂
19:48:07 <andythenorth> you worked too much
19:49:18 <andythenorth> now I'm distracted
19:49:29 <andythenorth> what were we doing? New css for bananas?
19:49:35 <truebrain> no more talk about abbreviations \o/
19:51:43 <peter1138[d]> Some of these are dumb by themselves.
19:54:34 <truebrain> Browser: open new tab
19:56:01 <andythenorth> "Great Idea! How thoughtful! Let's open that browser tab for you ❤️ "
19:56:13 <andythenorth> [opens calculator]
19:58:10 <peter1138[d]> Who knows what he's doing.
19:59:31 <peter1138[d]> orudge, are you aware of zernebok performance?
20:18:36 <truebrain> Maybe throw him an email; he tends to be quick reading those
20:37:03 <peter1138[d]> Oof, 1147 issues/PRs ago 😮
20:45:11 <peter1138[d]> I wonder if anyone has reported the scenario editor issue yet.
20:45:21 <peter1138[d]> I don't know why I'm asking, because I know the answer.
20:48:21 <_zephyris> I'm assuming no... which one?
20:48:39 <_zephyris> The 'you get dumped into the scenario editor with no sensible way to set things like newgrfs' issue?
21:22:19 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
21:24:09 <peter1138[d]> Hmm. This does so much at one 😮
21:44:46 *** dh1 has quit IRC (Ping timeout: 480 seconds)
22:15:50 *** keikoz has quit IRC (Ping timeout: 480 seconds)
continue to next day ⏵
