IRC logs for #openttd on OFTC at 2019-04-02
⏴ go to previous day
00:03:38 *** supermop_work_ has quit IRC
00:10:55 *** supermop_work_ has joined #openttd
00:16:53 <Wormnest> If I´m doing anything then it will be fixing bugs and unfinished features first
00:17:47 *** supermop_work__ has quit IRC
00:18:21 *** Supercheese has joined #openttd
00:23:02 *** supermop_work_ has quit IRC
00:29:51 *** andythenorth has left #openttd
00:33:52 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN approved pull request #7456: Fix #7440: Remove town sign when deleting town, not add it once more https://git.io/fjIUa
00:40:57 *** supermop_work has joined #openttd
00:49:10 *** supermop_work_ has joined #openttd
00:51:01 <DorpsGek_II> [OpenTTD/OpenTTD] eekee commented on issue #7447: Text for some railway station NewGRFs is missing/invisible https://git.io/fjIUX
00:57:12 *** supermop_work_ has quit IRC
01:27:31 <Samu> peter1138 authored and PeterN committed!
01:27:46 <Samu> definitely not the same guy
01:31:31 <Samu> gonna experiment script_max_opcode_till_suspend being editable during a running game with AIs
02:05:13 *** Thedarkb-X40 has joined #openttd
02:16:49 <Samu> wow, even SQVM uses the new vector stuff
02:29:42 *** tokai|noir has joined #openttd
02:29:42 *** ChanServ sets mode: +v tokai|noir
02:43:41 <Samu> seems that this can be editable during a game, at least for single player
02:45:02 <Samu> for multiplayer I suspect it wont cause issues, but have to test better than just the server
02:53:04 *** Supercheese has joined #openttd
04:10:42 *** Eddi|zuHause has joined #openttd
04:36:16 <Samu> how does openttd convert double into a int?
04:40:01 <dwfreed> probably the same way C converts a double to an int
04:44:08 <Eddi|zuHause> openttd doesn't use double (or float)
04:44:23 <Eddi|zuHause> not in a place that matters, anyway
04:48:25 *** Supercheese has joined #openttd
04:51:15 <Samu> framerate is using double
04:51:33 <Samu> must convert to either int or uint for something I'm working with
05:01:42 <Samu> working on something I'm calling variable script_max_opcodes
05:02:25 <Samu> takes the performance data gathered from framerate stuff and computes a max_opcodes value
05:02:52 <Samu> double to int, and then to uint
06:06:33 <DorpsGek_II> [OpenTTD/OpenTTD] SamuXarick opened pull request #7459: Feature: Self-adjusting AI/GS max opcodes https://git.io/fjILs
06:21:14 *** Suprcheese has joined #openttd
06:26:10 *** Suprcheese is now known as Supercheese
08:25:34 *** sla_ro|master has joined #openttd
08:39:28 *** andythenorth has joined #openttd
08:51:10 <peter1138> Hmm, I've got a bug where all road bits are removed but the road is still a road tile :p
08:52:06 <peter1138> I've got up to 32 road types o_O
08:52:41 <peter1138> Hmm, with road sub types, we had 63 road and 63 tram types.
08:52:51 <peter1138> Now it's just 63 road or tram types.
08:53:00 <peter1138> I wonder if that's a problem? :p
08:53:26 <LordAro> there are 63 railtypes, right?
08:54:00 <peter1138> Roads need to store INVALID_ROADTYPE in the map. I could fix that, maybe.
08:56:21 <peter1138> So we didn't backport #7449 cos, er, it wasn't marked :/
08:56:27 <peter1138> So 1.9.0 has a nice bug :(
09:06:02 <andythenorth> peter1138: Nobody Needs More Than 63 Road Types ®
09:09:50 <peter1138> I've 'halved' the amount.
09:10:49 * andythenorth reading the many nice comments about 1.9.0 release
09:10:54 <andythenorth> there are as many as 0
09:17:33 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN commented on pull request #7366: Add: List recently executed commands in crashlog output. https://git.io/fjIqu
09:18:25 *** Supercheese has joined #openttd
09:42:58 <andythenorth> bigger Steel Mill needed
09:43:01 <andythenorth> 3x5 is a bit small
09:43:07 <andythenorth> but will it blend?
09:45:54 *** WWacko1976-work has joined #openttd
09:46:37 <DorpsGek_II> [OpenTTD/OpenTTD] nielsmh merged pull request #7456: Fix #7440: Remove town sign when deleting town, not add it once more https://git.io/fjkpn
09:46:44 <DorpsGek_II> [OpenTTD/OpenTTD] nielsmh closed issue #7440: Crash when trying to delete town in scenario editor https://git.io/fjTxv
09:56:19 <DorpsGek_II> [OpenTTD/OpenTTD] nielsmh commented on issue #7447: Text for some railway station NewGRFs is missing/invisible https://git.io/fjImJ
09:56:59 <DorpsGek_II> [OpenTTD/OpenTTD] LordAro commented on pull request #7453: Remove AutoDeleteSmallVector and AutoFreeSmallVector https://git.io/fjImU
10:03:12 <peter1138> fdffssdfsdfwerwersdfsdfwerasdsdf
10:04:58 <planetmaker> err-not-a-number :P
10:10:41 *** blim[m] has joined #openttd
10:23:00 <andythenorth> V453000 now we have more cargos
10:23:05 <andythenorth> you should do small and large Yetis
10:23:11 <andythenorth> different production rates for each
10:30:12 <V453000> more cargo limit? :D how much? :D
10:32:13 <V453000> 16 in 16 out as a single industry?
10:34:20 <V453000> that's some significant madness
10:34:29 <V453000> I can't imagine what it would be good for right now, but eh, freedom :)
10:35:07 <andythenorth> yeti fashion shop
10:38:32 <V453000> I'm daunted as fuck by the though of touching anything in YETI so ... :D
10:38:50 <V453000> I'm pretty sure the train set I wanted to make is not feasible/possible though
10:39:00 <planetmaker> I totally love the yeti signals :)
10:39:22 <planetmaker> they look like yetis :)
10:39:24 <V453000> but yeah if anything I should probably extend that eventually
10:39:51 <planetmaker> you totally should
10:42:48 <V453000> train depots and train stations sound like a lot of reward for relatively little work ... compared to for example trains which I did start, but the amount of work on those is insane - and generally I don't see them :D
10:43:36 <planetmaker> additionally: trains are the pet of virtually 70% of the grf people... easier to get good ones
10:43:54 <planetmaker> and 70% is definitely a lower estimate :P
10:43:55 <V453000> well not for BRIX style 32bpp but yeah
10:44:26 <planetmaker> there's a new czech set in 32bpp. It's also quite good
10:44:49 <V453000> I know I helped them start with coding and stuff, a lot of their source is probably a copy of BRIX :)
10:45:11 <V453000> they do ok things, but it's realism which hate
10:45:32 <V453000> their philosophy is to turn OpenTTD into a modern 3D game basically
10:45:38 <planetmaker> reality is best when viewed offline ;)
10:47:53 <andythenorth> I am mostly designing it to make the chart look good :P
10:48:09 <andythenorth> irony: linux renders the chart differently, much lolz
10:49:04 <planetmaker> it's missing a legend :)
10:49:15 <planetmaker> green = primary, red = secondary?
10:49:23 <andythenorth> yeah it could use a key I guess
10:49:32 <andythenorth> not sure if that's easy to insert in graphviz
10:52:49 <planetmaker> SO tells me to use sub-graphs :P </so kiddie>
10:55:32 <V453000> I was talking to Borg yesterday, he did a weird X2025.grf not too long ago ... from the name and first look it seems to be just a monorail tweak, but actually it's mainly industry stuff ... First off, industries in a town area need powering by local power plant, and local banks somehow influence the cap of production as well, or something ... it's quite interesting
10:55:56 <V453000> I guess it's basically a less wtf version of YETI
10:57:02 <andythenorth> wonder how he does that
10:58:59 <V453000> I was surprised about that as well
11:01:19 <andythenorth> maybe town storage
11:01:36 <andythenorth> more newgrf town stuff pls
11:10:56 <andythenorth> economy is Steeltown, but I've added a plastics plant
11:10:59 <andythenorth> plastic isn't steel eh :P
11:11:13 <planetmaker> a glimpse into the future
11:58:01 *** Gustavo6046 has joined #openttd
12:39:23 <DorpsGek_II> [OpenTTD/nml] planetmaker opened pull request #26: Fix: Allow access to towns (feature 0x12) as parent object https://git.io/fjI3Y
12:47:48 <_dp_> damn, my evil plan to promote citymania client on that performance issue failed :(
12:47:59 <_dp_> ppl complain about macos version and I don't have any xD
12:54:01 <DorpsGek_II> [OpenTTD/OpenTTD] eekee commented on issue #7447: Text for some railway station NewGRFs is missing/invisible https://git.io/fjI3E
12:58:20 <DorpsGek_II> [OpenTTD/OpenTTD] LordAro commented on issue #7447: Text for some railway station NewGRFs is missing/invisible https://git.io/fjI3o
13:05:43 <DorpsGek_II> [OpenTTD/OpenTTD] eekee commented on issue #7447: Text for some railway station NewGRFs is missing/invisible https://git.io/fjI3M
13:26:49 <DorpsGek_II> [OpenTTD/nml] planetmaker opened pull request #27: Doc/Fix: Update the stated version requirements https://git.io/fjIsU
13:34:07 *** andythenorth has joined #openttd
13:56:44 <LordAro> planetmaker: i removed it from the issue
14:17:59 <Samu> hmm, it will always fail
14:22:58 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN commented on pull request #7459: Feature: Self-adjusting AI/GS max opcodes https://git.io/fjIGm
15:03:20 *** supermop_work has joined #openttd
15:09:57 <Samu> it works surprisingly well
15:10:24 <Samu> but has to be done in some other way
15:10:39 <Samu> I wanted to do it based on per AI basis
15:19:19 <LordAro> i don't like #7459 at all
15:19:58 <LordAro> opcode count was never supposed to be to help deal with AIs using all your CPU processing power, it was intended as a way to help regulate AI difficulty
15:20:07 <LordAro> i.e. the speed they can do things
15:20:47 <LordAro> i think it boils down to another "Samu runs AIs differently to everyone else" PR
15:23:06 <Samu> that's why I wanted to do it per AI :(
15:25:08 <Samu> there's settings profile for AI difficulty
15:32:22 *** andythenorth has joined #openttd
15:36:26 <planetmaker> Samu, what problem does that feature solve?
15:38:19 <Samu> AIs that would be really intensive would suspend earlier
15:38:34 <Samu> AIs that are lighter would suspend later
15:57:37 <andythenorth> V453000: can you make an MP3 of you saying porntptfs
15:58:54 <andythenorth> Samu: 7459, you get to learn about flapping, feedback loops, and runaway conditions :P
15:59:25 <andythenorth> when your measurement changes what you're measuring, that's a hard problem in control loops
16:06:20 *** sla_ro|master has joined #openttd
16:29:18 <andythenorth> I need a commodity export cargo :P
16:29:21 <andythenorth> for game balance
16:29:33 <andythenorth> also 'look at the time'
16:30:03 *** Wormnest has joined #openttd
16:39:17 <planetmaker> should be colonial goods :P
16:56:23 <Samu> pointer to incomplete class type is not allowed
16:56:43 <Samu> !Company::Get(this->elem - PFE_AI0)->ai_instance->IsDead()
16:56:47 <Samu> does not like me doing this
16:57:16 <nielsm> means exactly what it says
16:57:40 <nielsm> one of the class types you're dereferencing a pointer to is not completely declared
16:59:02 <Samu> class AIInstance *ai_instance;
17:01:15 <nielsm> that may be an incomplete type yes
17:01:35 <nielsm> if there isn't a complete declaration of class AIInstance somewhere else in the compilation unit
17:02:22 <Samu> #include "ai_instance.hpp"
17:04:07 *** dihedral has joined #openttd
17:14:41 *** GroovyNoodle has joined #openttd
17:18:16 <nielsm> okay time to downgrade to vs 2019 community
17:18:51 <nielsm> (on release channel instead of preview channel, where only enterprise was available, which then becomes unlicensed after the general release)
17:22:30 *** andythenorth has joined #openttd
17:27:51 *** supermop_work has joined #openttd
17:50:16 <Samu> that's such a small number
17:51:07 <peter1138> Don't forget to let the game do its own thing as well.
17:55:02 <Samu> debug build the measurements are always on the low side
17:57:11 <Samu> i need to test this on a release build
17:57:22 <Samu> but then I can't access these numbers
18:01:07 *** synchris has joined #openttd
18:13:01 <Eddi|zuHause> storks bringing babies
18:13:04 <Samu> lines 14 and 16 may need adjustings
18:20:15 <DorpsGek_II> [OpenTTD/OpenTTD] EgyLynx commented on issue #7430: Vehicle waiting at station affects cargo age rating even if it did not pick up cargo https://git.io/fjI8k
18:23:54 <Samu> it works, perhaps a bit too well
18:24:13 <Samu> i was hoping the total average was more time in the 30s
18:25:03 <Samu> makes me think I an subtracting too much opcodes
18:26:11 <Samu> then there's those suddem hiccups
18:26:19 <Samu> which may also ruin the averages
18:26:37 <Samu> and another big chunk of opcodes may be subtracted again :(
18:31:33 <DorpsGek_II> [OpenTTD/OpenTTD] EgyLynx commented on issue #7438: Music sets (including DOS music) broken after trying other music sets https://git.io/fjI8c
18:38:12 *** Progman has joined #openttd
18:43:05 *** HerzogDeXtEr has joined #openttd
18:47:44 <DorpsGek_II> [OpenTTD/OpenTTD] nielsmh commented on issue #7438: Music sets (including DOS music) broken after trying other music sets https://git.io/fjI8D
18:49:51 *** supermop_work has joined #openttd
18:52:41 * andythenorth assumed that would be pictures of birds
19:05:55 <LordAro> nielsm: think you can disable bounds.4 :)
19:06:24 <nielsm> I can't figure out where to disable those guidelines warnings...
19:07:45 *** frosch123 has joined #openttd
19:14:08 <frosch123> planetmaker: i reverted you spec change. towns do not have a feature number
19:16:20 <DorpsGek_II> [OpenTTD/nml] frosch123 commented on pull request #26: Fix: Allow access to towns (feature 0x12) as parent object https://git.io/fjI4G
19:25:17 * andythenorth has ideas about newgrf towns :P
19:28:00 *** Gustavo6046 has joined #openttd
19:33:21 <DorpsGek_II> [OpenTTD/OpenTTD] SamuXarick updated pull request #7459: Feature: Self-adjusting AI/GS max opcodes https://git.io/fjILs
19:41:33 <andythenorth> frosch123: have you got a git checkout of FIRS? o_O
19:42:43 *** gelignite has joined #openttd
19:46:04 *** Progman has joined #openttd
19:46:43 <DorpsGek_II> [OpenTTD/OpenTTD] nielsmh commented on issue #7438: Music sets (including DOS music) broken after trying other music sets https://git.io/fjI47
19:47:18 <Samu> oh, it's a || that should be a &&
19:48:01 <DorpsGek_II> [OpenTTD/OpenTTD] SamuXarick updated pull request #7459: Feature: Self-adjusting AI/GS max opcodes https://git.io/fjILs
19:52:19 <frosch123> andythenorth: i did not delete it after push
19:52:34 <frosch123> i did not pull either though
19:52:43 <frosch123> so it's in the hg->git transition state
19:52:45 <andythenorth> I wondered if cargoflow graphs could be improved for supplies
19:52:59 <frosch123> weren't they basically excludud :p
19:53:10 <andythenorth> input is excluded
19:53:15 <andythenorth> output they all go to one node
19:53:40 <andythenorth> I am curious if they didn't converge on one node, if the graph would be cleaner
19:53:50 <andythenorth> most cargoflow goes left -> right
19:54:09 <andythenorth> and producing supplies at both ends makes the graph horrible
19:54:21 <andythenorth> this might be fine though
19:54:53 <andythenorth> I think inserting fake cargos would do the truck, like FMSP_1, FMSP_2 etc
19:55:05 <andythenorth> but it might be TMWFTLB
19:55:06 <frosch123> we can not join them and give every supply producer their own output node
19:55:50 <nielsm> something like what you see for ground in electronics diagrams? gound connections just point to a symbol instead of all joining
19:56:14 <andythenorth> yes something like that
19:56:32 <andythenorth> frosch123: there's a v4-development-track branch
19:57:46 <andythenorth> I did try it myself first, but I miss some part of understanding graphviz :P
20:00:15 <Samu> for some reason regression test passed
20:00:33 <Samu> that wasn't expected to happen
20:10:57 <Samu> hmm GameScripts are Company 0, that's wrong
20:11:02 <nielsm> wow, Modern Motion sounds absolutely terrible on this Korg synth
20:13:39 <Samu> this is what i should be using
20:14:26 *** m3henry has joined #openttd
20:18:18 <DorpsGek_II> [OpenTTD/OpenTTD] SamuXarick updated pull request #7459: Feature: Self-adjusting AI/GS max opcodes https://git.io/fjILs
20:18:23 <Wolf01> <andythenorth> Wolf01: it's not an April fools :( <- I know, I already put 100 of them on the basket
20:20:21 <Wolf01> Mmmh, "new" android tablet, 300MB of free memory to install apps, google shit just updated... I can't install apps
20:28:56 <Samu> how to "deprecate" a game setting?
20:29:28 <Samu> like, when loading old savegames, how to ignore it
20:29:44 <Samu> and when saving how to not save it
20:29:57 <Samu> and make it unavailable in the openttd.cfg
20:32:43 <frosch123> oh, i need a new nml
20:34:17 <V453000> I'm on 1.8.0, is there any relation between goods pickup station's station rating, and goods being dropped(transferred) but not being re-transfered further?
20:36:26 <Eddi|zuHause> i'm never sure which version of that ended up in trunk, but cargodist experimented with a few methods to drop rating at the source station for cargo that is piling up on intermediate stations
20:36:59 <V453000> even if distribution is set to manual?
20:37:23 <Eddi|zuHause> manual doesn't disable all the linkgraph stuff
20:39:50 <V453000> wow I didn't know about that at all :D
20:40:23 <V453000> town stopped accepting goods and the factory went into shambles :
20:40:26 <DorpsGek_II> [OpenTTD/OpenTTD] SamuXarick commented on pull request #7459: Feature: Self-adjusting AI/GS max opcodes https://git.io/fjIBb
20:41:26 <frosch123> andythenorth: is there something like "else" in the templates?
20:41:36 <frosch123> or rather "else if"
20:41:53 <andythenorth> there's switch case
20:41:58 <andythenorth> or there's condition
20:47:46 <Samu> what can I do about regression?
20:48:23 <DorpsGek_II> [OpenTTD/OpenTTD] michicc commented on pull request #7453: Remove AutoDeleteSmallVector and AutoFreeSmallVector https://git.io/fjIRJ
20:49:45 <DorpsGek_II> [OpenTTD/OpenTTD] kiwitreekor opened pull request #7460: Fix #6222: Advanced sprite layout sometimes showed incorrect railtype… https://git.io/fjIRU
20:49:58 <frosch123> also, what's wrong with scrap yard and cryo plant
20:50:10 <andythenorth> the svg isn't loading for me
20:50:55 <andythenorth> oh x-frame options are deny
20:52:16 <andythenorth> frosch123: I think it's better :)
20:52:30 <andythenorth> if an economy has a lot of ENSP / FMSP sources, like extreme, it's quite silly
20:52:37 <andythenorth> but that's because the economy is silly
20:54:45 <DorpsGek_II> [OpenTTD/OpenTTD] michicc commented on pull request #7453: Remove AutoDeleteSmallVector and AutoFreeSmallVector https://git.io/fjIRG
20:56:20 <DorpsGek_II> [OpenTTD/OpenTTD] michicc commented on pull request #7453: Remove AutoDeleteSmallVector and AutoFreeSmallVector https://git.io/fjIRW
21:02:20 <Samu> question: how do i debug regression? follow it in visual studio
21:02:45 <Samu> it doesn't hit the breakpoints :(
21:03:37 <DorpsGek_II> [OpenTTD/OpenTTD] nielsmh commented on issue #7438: Music sets (including DOS music) broken after trying other music sets https://git.io/fjIRy
21:04:05 <Samu> hmm, maybe a better question: how to exactly reproduce regression as its tested in the azure?
21:04:22 <Samu> i need to be able to follow breakpoints
21:04:49 <glx> just run the regression project
21:06:08 <frosch123> andythenorth: how did you make a merge-commit for a fast-forward merge?
21:06:28 <andythenorth> I clicked 'merge' in GH pull request
21:06:57 <frosch123> gh is bonkers then :p
21:06:59 <DorpsGek_II> [OpenTTD/OpenTTD] TrueBrain updated pull request #7457: -Update: documentation updates from 1.9 branch https://git.io/fjkjD
21:07:16 <glx> avoid merge when possible
21:07:52 <frosch123> glx: gh made a merge of HEAD and ^HEAD
21:08:20 <frosch123> it merged a branch of 1 commit with a branch of 0 commits
21:08:38 <glx> because "merge" was selected
21:08:54 <andythenorth> I tried a rebase this week at work
21:09:02 <andythenorth> the result was 3 commits :P
21:09:12 <andythenorth> and then a merge commit because upstream had 'changed'
21:09:24 <andythenorth> we don't have git set up for rebase :P
21:09:43 <glx> everybody working in main branch ?
21:09:48 <TrueBrain> or you are not capable of dealing with it :P
21:10:33 <frosch123> i think my boss-boss also said once, that he is annoyed about always having to disable the push-hooks when he wants to push something
21:10:47 <michi_cc> Hmm, do we need a 1.9.1 miles? (To track e.g. #7449)
21:10:56 <michi_cc> s/miles/milestone/ of course :)
21:11:04 <TrueBrain> michi_cc: mark it as backport requested, I say
21:11:11 <andythenorth> we just don't care about history, so it's merge all the way
21:11:13 <TrueBrain> if that bulks up enough, we can create a 1.9.1 :)
21:12:23 <TrueBrain> adding a milestone with closed PRs is confusing ;)
21:14:30 <DorpsGek_II> [OpenTTD/OpenTTD] glx22 commented on pull request #7366: Add: List recently executed commands in crashlog output. https://git.io/fjI0O
21:20:01 <DorpsGek_II> [OpenTTD/OpenTTD] TrueBrain commented on pull request #7366: Add: List recently executed commands in crashlog output. https://git.io/fjI0Z
21:26:58 *** DiscordIRC has joined #openttd
21:27:48 <DorpsGek_II> [OpenTTD/OpenTTD] TrueBrain updated pull request #7270: Introduce CMake (and removing all other project-related code) https://git.io/fhbqc
21:28:21 <TrueBrain> right, OSX bundling was next
21:29:19 *** DiscordIRC has joined #openttd
21:30:38 *** DiscordIRC has joined #openttd
21:33:52 *** DiscordIRC has joined #openttd
21:33:52 <DiscordIRC> It was my fault, you don't need to be sorry.
21:33:56 <DiscordIRC> <@91800247117844480> Try again now
21:34:13 <DiscordIRC> Also, they are probably going to be pissed off at my bot for just going in and out all the time
21:34:40 <DiscordIRC> I want to apologize for getting so upset the other day. It was unfair of me to take out my anger at Pikkabird on all of you.
21:34:56 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN commented on pull request #7366: Add: List recently executed commands in crashlog output. https://git.io/fjI0V
21:36:46 <TrueBrain> well, that is just a weird conversation
21:36:46 <Samu> 2007-03-18 regression build date?
21:36:58 <Samu> shouldn't that be updated :p
21:37:36 *** Progman has joined #openttd
21:37:58 <TrueBrain> think about the children!
21:38:12 <Eddi|zuHause> did we ever have a not-weird conversation in here?
21:38:44 <glx> yeah but people were actually in the chat
21:38:52 <glx> not on the other side of a bot
21:39:24 <Samu> I think regression tests on Azure were "lucky"
21:39:30 <Eddi|zuHause> this discord split is going to become a problem at some point
21:40:01 <glx> Samu: regression works locally too
21:40:23 <Samu> lucky that the opcodes weren't adjusted so much that could cause differences
21:40:49 <glx> there is sometimes difference when we had stuff to regression
21:41:07 <Samu> i mean regression testing with my new PR
21:41:07 <glx> but we check and adapt result.txt in this case
21:41:53 <Samu> cpu load dictates the end result with it
21:42:13 <V453000> is there supposed to be some link between discord and IRC?
21:42:19 <andythenorth> if there is a regular bridge to here from random discord channels, I want op
21:42:23 <andythenorth> there will be kicks
21:42:53 <V453000> what could possible go wrong :>
21:42:57 <andythenorth> or someone sock puppeting SYL
21:42:59 <Eddi|zuHause> we discussed a discord bridge at least 3 times recently
21:43:16 <Samu> first time I uploaded the PR, the regression tests weren't so lucky
21:43:28 <TrueBrain> does this mean he bypassed a ban? :P
21:43:42 <LordAro> Samu: the issue is that you've introduced some nondeterministic behaviour
21:43:46 <LordAro> which is very very bad
21:44:04 <glx> everything must be deterministic in openttd
21:46:34 <glx> anyway if AI sleeps regularly it should not have a huge impact on the game, even with big opcode limit
21:49:32 <DorpsGek_II> [OpenTTD/OpenTTD] TrueBrain closed pull request #7459: Feature: Self-adjusting AI/GS max opcodes https://git.io/fjILs
21:49:33 <DorpsGek_II> [OpenTTD/OpenTTD] TrueBrain commented on pull request #7459: Feature: Self-adjusting AI/GS max opcodes https://git.io/fjI05
21:50:36 <TrueBrain> I wish I had any clue how to validate that PR :D
21:50:52 <LordAro> needs a frosch123 review :p
21:51:57 <DorpsGek_II> [OpenTTD/OpenTTD] TrueBrain commented on pull request #7460: Fix #6222: Advanced sprite layout sometimes showed incorrect railtype… https://git.io/fjI0F
21:52:05 <TrueBrain> I assigned him to it now :P
21:52:15 <frosch123> i can only say "why"
21:52:20 <frosch123> that change makes no sense to me
21:54:16 <TrueBrain> I see numbers and names .. I have no clue :P
21:56:29 <frosch123> hmm, maybe when the railtype defines monorail and maglev as fallback station type
21:56:47 <glx> and I think one of the main issue is valuate with native closures, because 5 ops per item doesn't seem very high (I guess non native closures have real ops count + 5, not really checked)
21:57:38 <glx> AI sometimes causing hangs
21:58:07 <glx> usually when many AIs running
21:58:18 <TrueBrain> ah, you continued where you left off 10 minutes ago
21:58:20 <TrueBrain> that was not confusing at all :P
21:58:41 <glx> was quickly checking while typing
21:59:32 <TrueBrain> your quickly and mine have vastly different meanings :D But now I am just teasing you :P
22:00:56 <frosch123> he, i commented on the original issue
22:11:14 <TrueBrain> LordAro: for next time, although I am 8 days late: feel free to remove stale label + add pinned label if you want to keep issues around
22:11:23 <TrueBrain> otherwise you will be saying 'bump' every 2 months or so :D
22:12:39 <Samu> but how would I be testing the building services of azure?
22:14:29 <Samu> i can't test clang or linux builds here :(
22:15:36 <DorpsGek_II> [OpenTTD/OpenTTD] TrueBrain commented on issue #7333: Libtimidity could be trivially enabled to play music through mixer.cpp https://git.io/fjIEZ
22:15:37 <DorpsGek_II> [OpenTTD/OpenTTD] TrueBrain closed issue #7333: Libtimidity could be trivially enabled to play music through mixer.cpp https://git.io/fhpqq
22:16:13 *** kiwitree has joined #openttd
22:18:17 <TrueBrain> if we all fail to see why, does this mean we can blame someone else? :D
22:23:30 <DorpsGek_II> [OpenTTD/OpenTTD] TrueBrain commented on issue #6884: Installer reports success even if not all files were copied https://git.io/fjIEE
22:24:11 <andythenorth> close everything
22:24:43 <glx> maybe detect_pkg_config doesn't set $with_libtimidity to 0 on failure, but that would also fail for other libs
22:27:40 <milek7> in configure it was originally `if ($0 == "LIBTIMIDITY" && "'$libtimidity'" == "" ) { next; }`
22:28:07 <milek7> but now it is reverse, it's not compiling it even when found
22:29:18 <TrueBrain> haha, sorry, I assumed the 'revert' commit was a plain 'git revert'
22:29:20 <TrueBrain> but you modified it :P
22:29:51 <milek7> it conflicted because of changed surrounding lines
22:30:18 <milek7> and i fiddled with it until it started to compile
22:30:35 <milek7> but original version is still broken, just in other way ;p
22:31:43 <TrueBrain> I still cannot compile for Linux on the current code ... as I have a space in my folder :P
22:31:51 <TrueBrain> we should finish CMake .. that solves so many issues
22:32:33 <glx> hmm detect_library does eval "with_$2=0" when not found, but detect_pkg_config does nothing like that
22:32:58 <glx> so I'd say there's an issue in config.lib ;)
22:33:02 <DorpsGek_II> [OpenTTD/OpenTTD] TrueBrain commented on issue #6682: menu bars disappear if enable windows screen scaling on high dpi monitors https://git.io/fjIED
22:33:03 <DorpsGek_II> [OpenTTD/OpenTTD] TrueBrain closed issue #6682: menu bars disappear if enable windows screen scaling on high dpi monitors https://git.io/fjIEy
22:33:21 <TrueBrain> right, enough cleaning up :)
22:33:31 <TrueBrain> finding issues that says: this is fixed
22:33:37 <andythenorth> only 5 more to go
22:33:43 <andythenorth> then PR > issues
22:35:11 <glx> milek7: you just managed to use something we never did before
22:37:32 <TrueBrain> ah, that makes sense glx
22:37:40 <glx> as other detect_pkg_config call are for "required" libs
22:37:41 <TrueBrain> libtimidity was never really used anymore when the PSP port died
22:37:56 <Samu> expermineting my stuff in a network game
22:38:12 <TrueBrain> well, proof and point why I want libtimidity used in the CI :D
22:38:51 <milek7> what about fluidsynth? it's too optional
22:39:51 <TrueBrain> it uses detect_library
22:39:58 <TrueBrain> which does set the with_ flag
22:40:05 <TrueBrain> so your PR is more correct than the reverted version
22:40:10 <TrueBrain> but .. the code was broken to start with :P
22:41:47 <glx> between line 2790 and 2791 you can add eval "$1=0"
22:42:00 <TrueBrain> that most likely works
22:42:08 <TrueBrain> glx: no, $1 is the resolved 'with' value
22:42:38 <Samu> testing variable op_codes in a network, and there seems to be no sync issues
22:42:42 <TrueBrain> milek7: I did not test it, but I believe in my patches :D
22:43:06 <glx> will not work correctly for liblzma ;)
22:44:32 <glx> but it's not an issue because the -D flag is handled correctly
22:44:43 <TrueBrain> we should finish CMake :)
22:44:46 <glx> and no lzma stuff is in source.list IIRC
22:44:49 <TrueBrain> well, biggest thing left is OSX Bundles
22:45:00 <TrueBrain> and validation of everything :D
22:45:17 <glx> I think some stuff is missing for windows installer, like registry key
22:46:06 <milek7> TrueBrain: thanks, it works
22:47:20 <glx> InstallDirRegKey HKLM "SOFTWARE\Microsoft\Windows\CurrentVersion\Uninstall\OpenTTD" "Install Folder"
22:47:33 <TrueBrain> do we really need that to exist? :P
22:47:49 <TrueBrain> please tell me OpenTTD doesn't use it :)
22:47:59 <LordAro> is it necessary for the uninstaller?
22:48:13 <TrueBrain> no clue how it works, but it works
22:48:53 <TrueBrain> it is what I love about automation :D
22:49:44 <TrueBrain> but I wonder why glx would expect that registry key to exist :)
22:50:30 <glx> I think NSIS used WriteRegDWORD HKEY_LOCAL_MACHINE "SOFTWARE\Microsoft\Windows\CurrentVersion\Uninstall\OpenTTD" "Version" ${INSTALLERVERSION} to detect installed version, but maybe cmake nsis does somethink like that too
22:50:51 <TrueBrain> yeah, well, no, I deliberately did not port that part over
22:51:00 <TrueBrain> it was kinda annoying me that NSIS refused to install if you had a newer version installed
22:51:07 <TrueBrain> by what ever definition it consider it to be newer
22:51:14 <TrueBrain> that is just ... bullcrap
22:51:43 <TrueBrain> basically, I modified as little as possible to how CMake NSIS works .. more standard, more better :)
22:52:34 <TrueBrain> I also installed a lot less files in the Start Menu
22:52:38 <TrueBrain> silly amounts were added there :P
22:53:01 <milek7> github display commit order is weird
22:54:03 <LordAro> it does get confused sometimes
22:54:10 <TrueBrain> reloading often helps
22:55:06 <TrueBrain> right; time to find a pillow to put my head on, close my eyes, and dream of an MMO version of OpenTTD played from the browser .......... *snorr*
22:56:53 <Samu> windows 10 links to multiple readme.txt files
22:57:49 <glx> any text editor can open .md
22:58:10 <Samu> it tried to find a program that can open .md files on windows store
22:58:18 <LordAro> hmm, it's not very helpful to put them in the start menu though
22:58:43 <Samu> so, they're not really readme.txt, i think they're readme.md
22:58:46 <LordAro> also there's 2 of them
22:59:48 <Samu> C:\Program Files\OpenTTD\README.md
23:00:17 <Samu> C:\Program Files\OpenTTD\scripts\README.md
23:00:57 <glx> yes the main readme and the AI/GS readm
23:02:12 <LordAro> glx: seems like something to improve for the cmake installer, anyway :p
23:02:24 <Samu> the scripts\README.md is actually blank
23:02:38 <LordAro> it'd be nice if it could set file association for .md files, if one isn't set already
23:02:58 <Samu> i also notice there's a readme.txt just next to them
23:03:30 <glx> readme.txt are from previous install I think
23:04:10 <glx> because since the move to GH we converted txt to md
23:55:53 <Samu> if i want to print a uint in the console, I use %d?
continue to next day ⏵
