IRC logs for #openttd on OFTC at 2025-05-22
⏴ go to previous day
00:27:57 <goddess_ishtar> Is there a reason we don't display the uploader in the Online Content window?
00:28:57 <goddess_ishtar> the information is available through the Bananas web view, so we clearly *store* it, but it's not shown in-game
00:50:34 <talltyler> That is strange, I don’t see a reason not to show it 🙂
02:02:34 *** Wormnest_ has quit IRC (Quit: Leaving)
02:47:54 <goddess_ishtar> We're still going the "Bananas should be entirely uncurated" route, right?
02:51:26 <goddess_ishtar> because I'm thinking of a way to solve the whole "people constantly coming into the Discord asking for overly general newgrf recommendations, only to be pointed to the same pinned list of vanilla-like grfs" issue, and I realized that one solution might be to enable people to upload and share their presets (in effect, allowing people to upload "modpacks")
02:53:49 *** gnu_jj_ has joined #openttd
02:57:25 *** gnu_jj has quit IRC (Ping timeout: 480 seconds)
03:27:24 <goddess_ishtar> was there any reason in particular it hasn't been implemented yet, or is it just that nobody got around to doing it?
03:32:33 *** Zathras has joined #openttd
03:35:39 *** godbed has quit IRC (Ping timeout: 480 seconds)
03:35:54 *** debdog has quit IRC (Ping timeout: 480 seconds)
04:44:36 <frosch123> That's the "latest" draft
04:47:20 <DorpsGek> - Update: Translations from eints (by translators)
06:16:33 <_zephyris> Uncurated mod packs just kicks the van down the road, what do you do when that's a big messy list too?
06:17:30 <_zephyris> Why not use download or usage data, and add a sort by popularity. Both exist...
06:19:19 <reldred> Yeah I live by those on NexusMods
07:11:13 <reldred> well, look like ids habbening
07:11:25 <reldred> off to brisvegas to make big contractor bucks
07:11:47 <reldred> how the fuck did I bullshit my way into this
07:12:52 <peter1138[d]> Good ol' imposter syndrome.
07:13:50 <peter1138[d]> Is there a word for when you suspect you have imposter syndrome, but actually it turns out you are actually just shit at your job...?
07:18:06 <reldred> I think it's more that I'm decently competent in a broad enough range of really weird things that now after 20yrs makes me a compelling package.
07:29:23 <andythenorth> peter1138[d]: no comment
07:33:29 <kuhnovic> peter1138[d]: The dunning-kruger effect
07:34:11 <reldred> I dunno what word you use, but I use this png
07:34:36 <peter1138[d]> kuhnovic: that's just the last bit.
07:34:42 <reldred> but they're buying a 12mo subscription to My Bullshit so, you know,
07:55:24 <andythenorth> people with DK don't have imposter syndrome
07:55:42 <andythenorth> except at the higher end of the performance curve, where true high performers under-judge their results
07:55:59 <andythenorth> because they can't form accurate peer benchmarks
07:56:12 <andythenorth> also DK paper contains errors, and may not be valid
07:57:02 <andythenorth> and everyone is susceptible to DK, because humans have judgement errors
07:58:19 <pickpacket> andythenorth: speak for yourself. I don't have judgment errors. I *am* a judgment error
07:59:40 <pickpacket> I heard somewhere that we all use the term DK wrong, so I decided to read up on it. And apparently we do 😅
08:01:15 <pickpacket> DK doesn't say anything about what people think they know. It states that you can't know what you don't know, basically. I can't judge the veracity about something above my level of expertise
08:01:37 <pickpacket> when people believe they can that's apparently something else than DK, but I don't remember what
08:09:25 <peter1138[d]> I believe that I **can** not know what I'm doing...
08:18:38 <kuhnovic> pickpacket: There's a good amount of irony in this haha
08:22:34 <peter1138[d]> Hmm, I wonder what the cost difference of iterating a bitset vs a for loop which tests each bit is.
08:22:56 <peter1138[d]> Probably not measurable in isolation.
08:26:02 <peter1138[d]> What overpriced smartwatch do I want?
08:26:22 <pickpacket> peter1138[d]: which features are most important to you?
08:26:35 <peter1138[d]> 1) telling the time
08:27:42 <peter1138[d]> (Fitbits fail spectacularly at that, as they will lose time if their bluetooth connection fails. Which it does.)
08:28:29 <kuhnovic> What!? If there is _one_ thing a watch should be capable of...
08:28:31 <reldred> lol I still don't even wear a regular watch letalone a smart one
08:29:18 <LordAro> i thought the new pebble watches looked mildly interesting
08:29:30 <LordAro> though i'm still quite happy with my 20 year old casio
08:29:32 <kuhnovic> I used to wear a regular watch with a steel wristband, but I got sick of it scratching against the side of my laptop
08:30:27 <reldred> Yeah I had one with a cloth nato strap but it broke and I never bothered doing anything about that
08:30:53 <peter1138[d]> I had an original Pebble Steel. Quality control was a bit lacking, and its buttons no longer work.
08:41:35 <pickpacket> peter1138[d]: I'd say any smartwatch which is actually a watch
08:41:42 <pickpacket> as opposed to a fitbit
08:42:14 <peter1138[d]> Fitbit got Googlified anyway, so no.
08:42:30 <peter1138[d]> Not that the rest are any better.
08:42:56 <peter1138[d]> The remains of the original Pebble were also glomped by them.
08:43:08 <peter1138[d]> Maybe the new Pebble is the way.
08:43:42 <pickpacket> It would actually be pretty cool
08:46:31 <pickpacket> Although I'd just terraform instead either way
08:51:23 <peter1138[d]> IIRC JGRPP has a hack to allow under-ocean tunnels.
08:51:36 <peter1138[d]> One day we should revisit the deep water patch...
08:56:11 <foodliker> peter1138[d]: buy a 50 year old mechanical watch from the former ussr for 25 quid
08:56:33 <foodliker> it will more or less tell the time
08:56:44 <foodliker> doesn't need a battery (you are the battery)
08:57:31 <foodliker> can't lose bluetooth connection (didn't exist)
08:58:07 <xarick> is there a reason why it was necessary to make a distinction between AI and GS in the name like in AIList, GSList regarding scripts?
08:58:35 <xarick> why not just have them have the same name
09:05:41 <_zephyris> Fingers crossed for new pebble
09:13:29 <xarick> testing yesterday's slow valuate
10:30:29 <xarick> searcing for .Valuate( in .tar filees
10:43:39 <xarick> 44435452-DictatorAI-183.tar (205 hits)
10:44:13 <xarick> 4f545649-OtviAI-423.tar (177 hits)
10:44:45 <xarick> 52776169-RailwAI-30.tar (168 hits)
10:45:35 <xarick> 41444d4c-AdmiralAI-25.tar (152 hits)
10:45:46 <xarick> 484f4745-AAAHogEx-112.tar (152 hits)
10:59:48 <_glx_> xarick: Some functions/classes are only available to GS
11:08:09 <xarick> oops, corona isn't sorted
11:17:31 <xarick> it's not apples to apples comparison
11:17:40 <xarick> slow valuate on the right
11:17:59 <xarick> they build differently
11:21:41 <xarick> I wanted to check how cpu intensive slow valuate was, compared to the original
11:23:41 <xarick> it's nice to have it not crash, but if it costs too much cpu usage or makes the AIs a slugfest... it might not be worth it 😦
11:24:32 <xarick> I'd like to one day, push this into master
11:24:50 <xarick> and get rid of cpu evaluator
11:25:24 <xarick> or have it as an optional setting
11:47:26 *** WormnestAndroid has quit IRC (Remote host closed the connection)
11:47:36 *** WormnestAndroid has joined #openttd
12:13:26 <peter1138[d]> Hmm, well, lots of useful feedback for default badge class conventions.
12:13:55 <peter1138[d]> Oh silly me. No there isn't.
12:14:13 <peter1138[d]> Guess I delete that PR.
12:23:25 <pickpacket> I know you've explained what badges are to me, but I've forgotten 😛
12:30:15 <peter1138[d]> From a player perspective, they are extra icons (with text) in purchase lists, long with the ability to filter for them.
12:31:06 <_glx_> And provides some kind of unification between all grfs
12:31:47 <peter1138[d]> From a GRF developer perspective they are that, and can also be used as some kind of metadata for graphics and callbacks.
12:35:13 *** Flygon has quit IRC (Read error: Connection reset by peer)
12:49:14 <talltyler> Your default badge classes make sense to me, but none of my NewGRFs would use badges, so my feedback is probably not that helpful 🙂
12:56:13 <peter1138[d]> talltyler: Why not?
13:08:21 <xarick> 13457 and 13408 is like bone and meat
13:08:45 <xarick> further changes will be required when one gets merged
13:09:28 <xarick> with the passage of time, I tend to forget
13:14:28 <andythenorth> peter1138[d]: I think they're stuck in usual place
13:14:40 <andythenorth> nobody will make a grf for them until they're released in vanilla
13:15:01 <andythenorth> but then the vanilla affordances will immediately produce lots of feedback
13:15:11 <andythenorth> and "why wasn't this discussed" 😛
13:15:29 <andythenorth> "what if they were in JGRPP?" 😛
13:31:52 <xarick> my changelogs had typos and nobody told me 🙂
13:32:06 <xarick> just found out after months
13:43:12 <_glx_> xarick: Usual vcpkg failures
14:08:43 *** k-man has quit IRC (Ping timeout: 480 seconds)
14:09:54 <xarick> 13971 had a premature closure
14:30:53 <peter1138[d]> I think you mean permanent closure.
14:33:39 *** toktik has quit IRC (Remote host closed the connection)
14:39:41 *** kuka_lie has joined #openttd
14:56:07 <xarick> > apps should not be relying on the value of uninitialized memory
14:56:07 <xarick> I read someone say this
14:56:22 <xarick> is that why you're initializing everything in openttd
14:58:36 <LordAro> how many years have you been writing C++ now?
15:00:37 <dwfreed> and yet, a patch fixing that in openssl caused Debian to have weak keys for several years :>
15:01:37 <dwfreed> (that said, openssl should have had better random than that)
15:16:04 <talltyler> Also, the docs page has some broken images 🙂
15:21:25 *** Wormnest has joined #openttd
15:32:27 <peter1138[d]> Yeah, my build process just doesn't create those as it's the same image reused.
15:33:52 <jfkuayue> Okay, finally got rid of windows... maybe
15:37:14 <peter1138[d]> Ubuntu is shit though.
15:37:44 <LordAro> peter1138[d]: baby steps
15:37:51 <peter1138[d]> Basically, "Let's take Debian and commercialise it without our own shitty flatpak thing" (or whatever the thing is they use now)
15:38:07 <peter1138[d]> with not without.
15:39:21 *** gelignite has joined #openttd
15:42:41 <talltyler> _starts typing a reply, then remembers to comment in the PR_ 😛
15:43:04 <f_> It's like they know very well how I should use my computer
15:43:38 <f_> I mean, it's like they know what's best for my computer *
15:44:27 <f_> and if you deviate from that, well, good luck
15:44:54 <f_> E.g. there's that annoying thing where it'll force you to install firefox from snap instead of using a native deb package for it
15:49:06 <LordAro> f_: increasingly annoying, yeah
15:49:35 <f_> I needed to forcibly blacklist snap from being installed and add additional repos for firefox
15:50:02 <f_> Flatpak is kind of ok, snap is the one that's just bad
15:51:10 <LordAro> sounds like you should stop using Ubuntu :p
15:52:34 <goddess_ishtar> peter1138[d]: true facts
15:52:41 <xarick> I read firefox is bad now for linux
15:53:02 <goddess_ishtar> Mozilla removed a bunch of their wording saying they're not going to sell your personal data
15:53:14 <goddess_ishtar> so enshittification is clearly ongoing
15:53:58 <goddess_ishtar> but yeah just use Mint if you want something Ubuntu based but not shit
15:58:25 <jfkuayue> Should I use Debian instead?
15:59:33 <goddess_ishtar> it's up to you and your preference tbh
15:59:55 <jfkuayue> Firefox must be installed through snap now on Ubuntu?
16:00:21 <LordAro> if you don't know/care about the difference, it's fine
16:00:27 *** gelignite has quit IRC (Read error: Connection reset by peer)
16:00:52 <peter1138[d]> Well, I will say yes 😉 but that's me.
16:00:56 *** gelignite has joined #openttd
16:06:38 <f_> <goddess_ishtar> Mozilla removed a bunch of their wording saying they're not going to sell your personal data
16:06:54 <f_> yeah sure, just said firefox because that was the example I had in mind
16:07:31 <peter1138[d]> Also that only applies to Mozilla's services, and the binary version of Firefox.
16:08:00 <peter1138[d]> So yes, I did delete my sync account.
16:08:03 <f_> Thunderbird is another example
16:08:25 <f_> firefox/thunderbird are clearly not the only ones affected by snap replacing
16:08:48 <f_> if you're a beginner I recommend linuxmint instead these days
16:11:35 <LordAro> PopOS is supposed to be quite good too
16:12:06 <peter1138[d]> LordAro, bike ride 🙂
16:13:32 <f_> I'm not familiar with pop!os
16:13:43 <f_> but COSMIC looks interesting
16:13:50 <goddess_ishtar> I'm just kinda leery about it
16:20:14 *** aperezdc has quit IRC (Remote host closed the connection)
16:21:19 *** aperezdc has joined #openttd
18:18:34 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
18:24:10 *** WormnestAndroid has joined #openttd
18:33:20 <jfkuayue> I always cannot connect to WiFi after installing Debian 12…
18:34:26 <xarick> just checking that 74k trains savegame
18:34:46 <dwfreed> jfkuayue: wifi driver require firmware? or missing wifi driver entirely?
18:38:45 <jfkuayue> dwfreed: missing iwlwifi-bz-a0-gf-a0-X.ucode, where X is between 39 and 72
18:40:09 <jfkuayue> I may try again with mint…
18:44:05 <dwfreed> jfkuayue: debian trixie has files named like that, just higher X, in firmware-iwlwifi
18:44:34 <dwfreed> (trixie will be debian 13 when released; it's in hard freeze now, so it's pretty stable at this point)
18:44:57 <jfkuayue> dwfreed: No, it specifically lists “max version 72”
18:49:32 <andythenorth> there we go, such simulated thoughts
18:49:42 <andythenorth> emperorjake: ^^ JPSet badges yet?
18:51:37 <dwfreed> jfkuayue: no file by that name has ever existed in linux-firmware O.o
18:52:03 <dwfreed> did you mean bz-b0, not bz-a0 ?
18:54:10 <jfkuayue> Sorry, I had to take photos of screens; no WiFi available on the PC
18:55:28 <jfkuayue> I may retry using Linux Mint
18:55:39 <dwfreed> that said, try debian trixie kernel, it might support a different firmware range
18:56:02 <jfkuayue> But would I have to compile JGRPP myself?
18:56:19 <dwfreed> the bookworm build probably still works
19:00:28 <_jgr_> The generic Linux build should work fine on any distro
19:02:15 <andythenorth> does Debian 13 also require Debian 1 through 12?
19:19:21 <jfkuayue> No, my computer would not start and enter into mint after installing it.
19:29:41 <xarick> visual studio own conflict editor is so clunky
19:32:26 <andythenorth> pff am I going to fix more Horse?
19:47:48 <goddess_ishtar> wensimehrp: currently I think Endeavour is favoured over Manjaro in terms of Arch-based distros
19:48:19 <goddess_ishtar> honestly I think Fedora strikes a good balance if you don't want to dedicate effort to maintaining your system
19:50:05 <peter1138[d]> LordAro, power up and heart rate down, finally getting fitter. Legs dead though.
19:51:00 <andythenorth> long way from the head
20:18:13 <peter1138[d]> Browser-Archie now? Oh my.
20:20:24 <xarick> Vehicle::UpdateViewport also uses some hash?
20:20:56 <peter1138[d]> Yes, there are two hashes, one for tile position, and the other for pixel positions.
20:21:02 <andythenorth> pff that WASM emulator is Safari-only for me
20:21:12 <peter1138[d]> Works in Firefox.
20:22:11 *** kuka_lie has quit IRC (Read error: Connection reset by peer)
20:22:40 <xarick> viewport hash needs increase too... maybe
20:23:13 <xarick> _vehicle_viewport_hash
20:24:16 <xarick> nah, it's sometihng else, im not really sure
20:25:10 <xarick> it's the 74k trains save
20:25:21 <xarick> 824108 vehicles in the pool
20:32:13 <xarick> is it not this->GetStates?
20:34:58 <xarick> i discovered the reason AAAHogEx has big stalls
20:35:11 <andythenorth> is it because of programming? 😛
20:35:37 <xarick> it's getting list of vehicles in groups
20:35:53 <xarick> but there's 824k vehicles to iterate
20:36:40 <xarick> so I'm reviving the idea of groups caching their vehicle lists
20:37:09 <andythenorth> what does this excess performance case of 824k prove? 🙂
20:37:19 <andythenorth> is it just stunts?
20:39:47 <peter1138[d]> It proves that Xarick has more patience than most people :p
20:43:46 <andythenorth> would it be solved by xarick getting an M4?
20:44:03 <andythenorth> hmm ARM chips eh
20:44:52 <xarick> that cache checking is bad
20:45:53 <peter1138[d]> andythenorth: Sadly there's no Linux-supported M4 machine.
20:46:13 <peter1138[d]> (Asahi is a shit-show with it's obsession with Rust)
20:47:20 <peter1138[d]> (Cue kernel maintainer melt-downs over it...)
20:48:04 <andythenorth> pff I wondered how hot my CPU was 😛
20:51:49 *** Smedles has joined #openttd
21:11:44 <andythenorth> why is role/passenger/express better than role/express/passenger?
21:11:58 <andythenorth> can I use it for mail?
21:13:04 <andythenorth> if these are primarily for filtering, we have cargo filter?
21:13:47 <peter1138[d]> Not primarily for filtering. But not flooding the filtering is useful.
21:14:04 <peter1138[d]> And the cargo filter does not have options for express, or heavy, or whatever.
21:14:34 <andythenorth> taxonomies are hard
21:14:58 <andythenorth> the displayed role badges in Horse appear to mostly be "General Purpose/Express" and "Freight"
21:15:08 <xarick> waiting for 74 ticks to pass
21:15:26 <andythenorth> oh "General Purpose/Freight"
21:16:22 <andythenorth> this should probably be changed 😛
21:18:41 <xarick> groups caching vehicle lists, over 74 ticks
21:19:14 <goddess_ishtar> andythenorth: because imo "passenger locomotive" is more general of a category than "express locomotive"
21:19:40 *** keikoz has quit IRC (Ping timeout: 480 seconds)
21:19:53 <goddess_ishtar> but honestly yeah, the way to do it would probably be to apply both `role/passenger` and `role/express`, and then you can filter for one or the other individually
21:19:58 <andythenorth> from a string filtering context, it's probably equivalent either way
21:20:13 <andythenorth> there's no display hierarchy orginating from this
21:21:41 <goddess_ishtar> really what I'm worried about is standardising "generations" or "eras" is going to be extremely tricky
21:21:59 <andythenorth> model trains have approaches to it
21:22:13 <andythenorth> german manufacturers have standard epochs I think
21:22:24 <andythenorth> UK ones, of course, all defined their own 'era'
21:24:08 <xarick> peter, do you have this savegame?
21:24:26 <andythenorth> Hmm Horse gen 1 to 5 accidentally fits the epoch system
21:24:53 <xarick> i'd like to see whether you could find more places to boost performance :!
21:27:16 <jfkuayue> Turned out it was a “bug” in older Linux kernel. Solved after updating to kernel 6.11.
21:28:49 <xarick> anyway that's the savegame
21:46:23 <peter1138[d]> xarick: gameloop doesn't show much difference.
21:49:20 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
21:56:46 <peter1138[d]> Than Ubuntu, yes.
21:57:05 <peter1138[d]> Best one is the one you get working 😄
21:58:00 <jfkuayue> I am comfortable now
22:58:16 <_glx_> pff, and MSVC is happy when I write broken code
22:59:52 <_glx_> because of course it has `std::format`
23:58:31 *** WormnestAndroid has quit IRC (Remote host closed the connection)
23:58:37 *** WormnestAndroid has joined #openttd
continue to next day ⏵
