IRC logs for #openttd on OFTC at 2025-03-06
⏴ go to previous day
01:50:45 *** nebulabc_ has quit IRC (Server closed connection)
01:50:56 *** nebulabc has joined #openttd
03:00:05 *** herms has quit IRC (Quit: bye)
03:01:09 *** Wormnest has quit IRC (Quit: Leaving)
03:20:49 *** debdog has quit IRC (Ping timeout: 480 seconds)
04:45:00 <DorpsGek> - Update: Translations from eints (by translators)
06:38:51 *** keikoz has quit IRC (Ping timeout: 480 seconds)
07:09:09 *** fairyflossy has quit IRC (Quit: User went offline on Discord a while ago)
07:11:32 <locosage> PeterNviaGitHub: Why close? Seems useful.
07:11:55 <locosage> What it needs even more is comparison operators though
09:52:10 <xarick> hi, sorry I went to bed
09:54:22 <xarick> wow, the world is not enough for these 2 AIs
09:54:24 <peter1138> Hmm, well, at least the crashes are in 13.0, which means it's not caused by the cargo-related changes since 14.0...
10:25:44 <peter1138> How is OpenGFX 2 Hi Def getting on?
10:25:45 <xarick> _glx_: looks incorrect
11:11:05 <xarick> wondering if I can get by without the for line
11:13:12 <_zephyris> peter1138: All terrain/trees/infrastructure/bridges are decent.
11:13:32 <_zephyris> Most houses/industries/vehicles are 1x zoom only.
11:14:33 <peter1138> What I've seen looks really good.
11:16:15 <peter1138> There's a fair bit of 'dislike' for extra zoom graphics in the Discord community, but actually in the real world, new players do seem to prefer the higher detail.
11:17:14 <reldred> I must admit I am looking forward to OpenGFX2 Hi Def
11:17:34 <reldred> I don't know how much I'd use it, as I am far too attached to my .grf's
11:18:11 <xarick> this looks like unintended
11:18:53 <peter1138> My thouht is that having a good baseset that looks good by itself is enough for most players.
11:20:53 <reldred> do we have good numbers on what percentage play with .grf's or not? My gut feeling says it's probably a good 95% of players who'd be playing stock as a rock.
11:21:22 <reldred> Those players using something like Ogfx2 Hi Def would be a good look for the game as a whole.
11:21:31 <peter1138> It's less than that actually.
11:21:41 <peter1138> Well, play hours rather than players.
11:21:58 <peter1138> For 14.x, 51% of played hours have zero NewGRFs.
11:21:58 <reldred> Yeah righto, wonder what the split looks like per player.
11:22:05 <xarick> question, is it okay a conditional order to jumpt to itself?
11:22:34 <reldred> Yeah I'd expect high playtime players would branch out into .grf's
11:22:39 <peter1138> Because we don't track individual players, we don't really have player-count stats.
11:22:46 <reldred> that said, first impressions count.
11:23:12 <peter1138> When I tried getting friends into it, the first thing they did was download zBase because it's high resolution and 32bpp, and therefore must be better.
11:23:14 <reldred> and a new default base set especially for steam installs would be a very big visual drawcard
11:23:19 <peter1138> They are no longer my friends.
11:24:38 <peter1138> For JGRPP, only 2% of hours have no NewGRFs, but that's to be expected.
11:24:49 <reldred> Yeah that would make sense
11:25:41 <peter1138> Hmm, I should revisit snow line interpolation.
11:26:00 <reldred> I should revisit my bed
11:26:12 <reldred> for I am eeby and neebies of sleebies
11:26:35 <xarick> can a human player actually set a conditional order to jump to itself?
11:27:09 <peter1138> Only if you can play OpenTTD on your uConsole.
11:27:28 <_zephyris> I'm defo keen on the overall Ogfx2 High Def concept, I think it works as a style
11:27:49 <_zephyris> I think I might switch it to a 4x zoom 8bpp, but that's a really subtle stylistic change
11:28:32 <_zephyris> It's down to the slog now, need to stop tweaking other bits and just draw a house a day
11:29:01 <reldred> _zephyris: I dunno, I think it looks great in 32bpp.
11:29:39 <reldred> I think the importance of a palette is defining a ‘look’, but a good artist who understands that doesn’t then necessarily need the 8bit palette to constrain them to it
11:30:35 <reldred> I should play with my uConsole more
11:30:43 <reldred> See if a CM5 works in it
11:30:51 <peter1138> The workflow for 32bpp, when you need to allow for palette-based colour remaps and palette-based animated pixels, is a bit of pain.
11:31:13 <_zephyris> I've got a good workflow pinned down
11:32:27 <_zephyris> The workflow internally is 32bpp anyway, the 32bpp variant won't go away, jut might not be my preferred variant
11:32:52 <reldred> I wonder if my 3D printer is going to behave tonight. I’d like to get one of these stacking gridfinity cases done by tomorrow evening, see if they’re viable for stowing all my random things
11:33:03 <_zephyris> No CC on docks is annoying btw
11:35:29 <xarick> i kinda like the original
11:46:50 <xarick> testing `if (order == v->GetOrder(data)) return CMD_ERROR;`
12:01:19 <ahyangyi> You can choose your target branch when opening your PR
12:02:45 <_zephyris> Hmm, there should be nicer docks somewhere... I wonder what happened
12:03:52 <peter1138> Well, does enabling 1CC for docks have a chance of breaking any existing dock graphics...
12:05:03 <_zephyris> Haven't thoroughly checked
12:05:49 <peter1138> Are you sure it's not supported?
12:07:20 <_zephyris> Working from memory, but I think I tested CC on the docks and the locks
12:07:22 <_zephyris> And neither recoloured
12:07:47 <_zephyris> I'm fine with locks - shared infrastructure probably shouldn't have CC
12:08:49 <peter1138> Just, PALETTE_MODIFIER_COLOUR is set for docks.
12:09:54 <_zephyris> I'm doing a big ogfx2 rebuild right now, I'll put cc on the docks next build and see what happens
12:12:17 <_zephyris> Completely redoing all the 32bpp->8bpp conversions, painful
12:12:48 <peter1138> Remember that OpenTTD can do that itself, if 32bpp is the primary focus.
12:13:21 <peter1138> (The result might not be exactly ideal, but most players are not using 8bpp blitters.)
12:41:40 <peter1138> Hmm, I wonder if I should update OpenSFX to use Opus.
12:41:57 <peter1138> I mean, I have all the changes here, but it's not in a PR.
12:43:35 <peter1138> Was 13.0 the release that we forgot to disable asserts in?
12:47:54 <_glx_> According to crashlog it seems to be
12:48:44 <peter1138> Pom te pom. I don't know why they came back to OpenTTD and decided to use an old version.
12:49:51 <_glx_> Because they had it somewhere when reinstalled windows
13:04:12 <peter1138> Hmm, can I buy new inner tube valves... not used my road bike for too long and they've sealed shut :p
13:39:42 <LordAro> peter1138: what did you do to them?
14:08:10 <xarick> I gave them a 4k map to max trains
14:21:20 <xarick> hmm there is a flaw with my variable opcodes algorithm
14:21:38 <xarick> I don't know how to deal with it
14:22:50 <xarick> basically, at 250k ops, a large spike detected will get a lower penalty to the allowed max ops
14:23:11 <xarick> the spike is measured in fewer frames
14:23:52 <xarick> this same spike, at a lower value than 250k ops will get spread over more frames
14:24:26 <xarick> however, the algorithm seems to penalize more now because the spike lasted longer...
14:25:02 <xarick> it's a smaller peak, but since it's now spread into more frames, the calculations apply a larger penalty
14:25:31 <xarick> AAAHogEx is being heavily penalized for doing this
14:25:59 <xarick> I don't know how to fix this
14:27:42 <peter1138> LordAro, nothing. The bike is set up for tubeless so it has sealant in it. When regularly riding they tyres generally need topping up every couple of weeks, so the valves don't gunk up. Stop riding it for 3-4 months, and...
14:28:30 <peter1138> Also, god I'm unfit.
14:35:11 <peter1138> Yes, when I wear these tyres out (if I ever do) I'll probably not bother again, on the road bike at least.
14:35:16 <peter1138> Mountain bike makes sense.
14:44:09 <_jgr_> xarick: No, it's not useful and really should be rejected by the command, I can probably PR this later
15:21:05 *** HerzogDeXtEr has joined #openttd
15:22:10 <xarick> I thought glx was on it
15:38:00 <peter1138> You could just make a PR yourself.
15:38:13 <peter1138> It's not actually related to the script issue.
16:00:19 *** ChanServ sets mode: +v tokai
16:07:25 *** tokai|noir has quit IRC (Ping timeout: 480 seconds)
16:07:42 <xarick> only scripts have access to `case MOF_COND_DESTINATION`
16:13:23 *** nielsm has quit IRC (Remote host closed the connection)
16:13:26 *** kuka_lie has joined #openttd
16:34:39 <_zephyris> hmm, nml does get slow for big grfs...
16:35:29 <gwyd4016> I'd consider RUKTS to be maybe a medium sized grf, and without a cache that takes like 2 minutes tops
16:35:53 <gwyd4016> Much slower than nfo of course, which has never felt much slower than "instant"
16:38:04 <_zephyris> Yeah, that's a tiny grf compared to what I'm wrangling
16:39:12 <_zephyris> OGFX21 makes 635Mb of grfs now...
16:57:12 <_zephyris> peter1138: peter1138[d] I'm dumb... Docks are CC. Must've thrown the baby out with the bathwater when I realised locks aren't
17:13:05 <peter1138> A town made out of Mac Minis.
17:51:15 *** Wormnest has joined #openttd
18:22:26 <andythenorth> gwyd4016: Horse is big-ish and takes 35 seconds
18:22:34 <andythenorth> I want to make it faster, it's annoyingly slow
18:31:12 <gwyd4016> andythenorth: Rewrite in pure NFO
18:58:53 <truebrain> LordAro: did you have a chance to check why you didn't upload OpenMSX 0.4.2 to BaNaNaS all those years ago? I don't have permission, so I even can't 😛
19:00:14 <xarick> visual studio is complaining settings_table.cpp has many includes not in use
19:27:11 <xarick> visual studio is wrong
19:27:21 <xarick> seems to fail at this include stuff
19:42:25 <xarick> what is safeguards file for?
19:45:47 <xarick> I could remove 4 includes from settings_table
19:45:59 <xarick> or 3 if not counting safeguards.h
19:48:43 <xarick> is this worthy of a PR?
19:51:53 *** greeter has quit IRC (Remote host closed the connection)
19:56:34 <xarick> look! it's our corrupted ex-prime minister on the left! amazing how he got in such a position there... zero credibility
19:58:41 *** xarothbrook has joined #openttd
19:58:41 <xarothbrook> I don't think politics is a good topic for Discord channel #openttd-development .
19:59:29 <peter1138> IndustryGfx and IndustryGraphics, eh?
19:59:42 <peter1138> Anyone fancy sorting that out?
20:00:22 *** greeter has joined #openttd
20:17:17 <peter1138> Out of memory: Killed process 79496 (clangd.main) total-vm:54275956kB, anon-rss:43843388kB, file-rss:0kB, shmem-rss:53616kB, UID:1000 pgtables:103552kB oom_score_adj:100
20:19:23 <Rubidium_> that's quite a bit of memory :D
20:34:03 <xarick> hmm glx doing glx things
20:34:09 <LordAro> truebrain: pretty sure i don't have permission either
20:35:07 <truebrain> Sorry, but the system is ratting you out. You do have permissions 🙂
20:35:40 <_glx_> xarick: I'm just deduplicating code
20:37:37 <_glx_> ScriptOrder is "ugly" enough to not do the same thing in multiple places
20:40:30 <xarick> hmm, Set order jump to 0 should fail
20:44:01 <xarick> how are you going to fix it?
20:53:16 <_glx_> I won't, it's not related to this PR
20:57:57 <_jgr_> I had a look and the code deliberately allows scripts to insert conditional orders even when the order list is empty, in which case they must point to themselves. This seems pretty useless to me, but it's probably not worth breaking script compatibility over.
21:06:33 *** kuka_lie has quit IRC (Quit: Lost terminal)
21:20:47 <xarick> cannot insert conditional order when there's no orders
21:23:48 <xarick> oh, humans can also insert conditional orders when there's no orders, just tried
21:26:37 *** nielsm has quit IRC (Ping timeout: 480 seconds)
22:09:03 *** HerzogDeXtEr has quit IRC (Read error: Connection reset by peer)
22:34:25 *** keikoz has quit IRC (Ping timeout: 480 seconds)
23:26:08 <peter1138> Hmm, okay. The VS Code extension "Error Lens" makes those little tiny errors much easier to spot.
23:27:31 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
23:30:15 <peter1138> Hmm, PoolID is unfortunate naming. It's useful for non-pool IDs too :-)
continue to next day ⏵
