IRC logs for #openttd on OFTC at 2020-12-22
⏴ go to previous day
00:01:22 *** Westie has quit IRC (Ping timeout: 480 seconds)
00:04:11 <michi_cc> Well, somebody might care. But seeing the rather lacklustre response, I wouldn't bet on that.
00:07:42 <TrueBrain> Sounds like you are better of focusing on other issues :D
00:10:19 *** Gustavo6046 has quit IRC (Quit: WeeChat 2.9)
00:10:34 *** Gustavo6046 has joined #openttd
00:42:44 *** tejanos has joined #openttd
00:42:58 <tejanos> hi! oh wow so many people!
00:43:41 *** Progman has quit IRC (Remote host closed the connection)
01:04:03 *** tejanos has quit IRC (Remote host closed the connection)
01:41:56 *** HerzogDeXtEr has joined #openttd
03:36:58 *** D-HUND has quit IRC (Ping timeout: 480 seconds)
03:46:28 *** Wormnest has quit IRC (Quit: Leaving)
03:52:02 *** Gustavo6046 has quit IRC (Quit: WeeChat 2.9)
03:53:02 *** Gustavo6046 has joined #openttd
07:50:43 *** andythenorth has joined #openttd
07:55:24 *** HerzogDeXtEr has quit IRC (Read error: Connection reset by peer)
07:57:50 *** jottyfan has joined #openttd
08:14:32 *** sla_ro|master has joined #openttd
08:16:05 *** DorpsGek` has joined #openttd
08:16:05 *** ChanServ sets mode: +o DorpsGek`
08:16:05 *** DorpsGek is now known as Guest8823
08:16:05 *** DorpsGek` is now known as DorpsGek
08:19:10 *** Guest8823 has quit IRC (Read error: Connection reset by peer)
08:44:53 <andythenorth> I had a crazy idea
08:46:12 <andythenorth> what if the economy part of the simulation was....just...CONTENT?
08:46:27 <LordAro> not original, certain
09:06:21 <andythenorth> got to clean algae off the fish tank
09:12:43 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain opened pull request #8411: Fix: next 67 savegame versions are used in PatchPacks; skip them https://git.io/JLKxX
09:12:46 <TrueBrain> LordAro: do you have any issue with ^^ ?
09:12:56 <TrueBrain> I kinda disagree with your opinion on: let's ignore the problem :D Sorry :P
09:13:23 <TrueBrain> and it sounds super simple to me to avoid it :)
09:13:39 <TrueBrain> but, something something, haven't coded for OpenTTD in 10 years, something something .. :D
09:15:20 <TrueBrain> owh, and an older Spring 2013 savegame (version 201) hits an assert, crashing debug builds :D
09:18:22 <TrueBrain> what are you doubts about?
09:18:23 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #8411: Fix: next 67 savegame versions are used in PatchPacks; skip them https://git.io/JLKxX
09:18:38 <TrueBrain> (honest question to be honest; to me this sounds like a simple solution, but there is a 50/50 chance I am missing something :P)
09:19:04 *** Progman has joined #openttd
09:19:15 <LordAro> i don't think you're missing anything, i just don't think we should be handling data from such versions
09:19:27 <LordAro> it's like handling a save game from minecraft or whatever
09:20:00 <TrueBrain> we are now not handling the data :D
09:20:20 <TrueBrain> a minecraft savegame can't load either, with an error the format is not understood :)
09:20:54 <TrueBrain> and to be clear, it is the patchpacks that shouldn't have been so close to the real revision .. but hindsight is 20/20 :P
09:21:21 <LordAro> perhaps we could add a "modified" flag to the savegame version somehow
09:21:29 *** WormnestAndroid has quit IRC (Remote host closed the connection)
09:21:34 <TrueBrain> JGR solves it nicely, just mask with 0x8000
09:21:36 <LordAro> which would (for us) only matter for 'M' versions
09:21:42 *** WormnestAndroid has joined #openttd
09:21:52 <TrueBrain> but okay, you are talking about solving it towards the future; which I am all in for
09:21:59 <TrueBrain> but we also have the here and now to solve :)
09:22:09 <LordAro> i guess it's probably fine
09:22:16 <LordAro> as you say, gonna take a while to get to 32k
09:22:41 <TrueBrain> I remember the days we thought we only had 255 versions .. that were .... interesting times :)
09:23:25 <andythenorth> hmm black keeps giving me cake
09:23:59 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on issue #8116: Old patchpack savegame versions will soon overlap https://git.io/JfZsT
09:25:51 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #8411: Fix: next 67 savegame versions are used in PatchPacks; skip them https://git.io/JLKjo
09:26:18 <TrueBrain> LordAro: what JGRPP does is so much better: feature toggles
09:26:34 <TrueBrain> would kinda remove the need for versions at all, and makes it compatible with patchpacks etc
09:26:53 <TrueBrain> well, future patchpacks, ofc :P
09:28:05 <TrueBrain> so we have 2 PRs that add GS commands for industry; they claim they are complimentary to each other, so that is good ..
09:28:11 <andythenorth> hmm, a super opinionated black would re-order function params alphabetically
09:28:31 <andythenorth> (named args only)
09:30:25 <TrueBrain> that would be a horrible idea
09:30:34 <TrueBrain> there is often logic in the order of parameters :)
09:30:40 <TrueBrain> (and the alphabet is not one of them)
09:30:50 <TrueBrain> it should, however, order imports .. which it does not
09:34:40 <andythenorth> does the import order ever ever matter in python?
09:35:28 * andythenorth hopes for black2: "now it writes your code"
09:35:52 <FLHerne> If they really want to, modules can have whatever stupid side-effects they like on import
09:37:03 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #8115: GS methods to let company have exclusive access to the industry https://git.io/JL6va
09:38:27 <TrueBrain> andythenorth: in practice, you should never run into an issue when ordering imports :) In my 10 years with Python now, that has never been the case :)
09:38:35 <TrueBrain> and at my old jobs we enforced them being on alphabet
09:38:57 <TrueBrain> (as otherwise the next developer with OCD changes them into alphabet, creating diffs which are completely unneeded :P)
09:39:15 * andythenorth would be the OCD one
09:39:24 <andythenorth> 'incoming, I fixed your whitespace' /me
09:39:54 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #8115: GS methods to let company have exclusive access to the industry https://git.io/JL6fW
09:39:55 <andythenorth> on my deathbed I can say 'well at least some of the html was indented sanely'
09:41:05 <TrueBrain> lot of the coding style I introduce in teams is to avoid silly conversations about "where this line should go"; just a waste of conversation :P
09:41:09 <TrueBrain> sorting imports is one of them :)
09:41:45 <FLHerne> I did run isort on nml just recently, need to add it to the tests
09:42:10 <DorpsGek> [OpenTTD/OpenTTD] LordAro approved pull request #8411: Fix: next 67 savegame versions are used in PatchPacks; skip them https://git.io/JL6fX
09:42:34 <TrueBrain> I so hate when people say: "sure", like: this is going to blow up in your face, but you will find that out later :P
09:42:50 <TrueBrain> going to wait for JGR, see if he sees a problem with it :)
09:43:38 <DorpsGek> [OpenTTD/OpenTTD] JGRennison commented on pull request #8411: Fix: next 67 savegame versions are used in PatchPacks; skip them https://git.io/JL6fF
09:44:12 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain merged pull request #8411: Fix: next 67 savegame versions are used in PatchPacks; skip them https://git.io/JLKxX
09:44:40 <TrueBrain> so now we can leave that horrible PatchPack savegame versions mess behind, and act like it doesn't exist :D
09:45:40 <TrueBrain> but I guess that is what NewGRFs can do, and GSes can't
09:46:57 <andythenorth> I don't think there's any harm in extending GS
09:47:03 <andythenorth> I can't see us removing it
09:47:18 <TrueBrain> I was not even suggesting that :)
09:47:44 <TrueBrain> it is more, if we take this route, it is unlikely someone is going to try find another
09:47:51 <TrueBrain> I am trying to balance that :D
09:48:03 <LordAro> TrueBrain: seems reasonable to me
09:48:47 <TrueBrain> guess I just create a new PR which stacks these two together, with a rebase and some fixes :D
09:50:04 * andythenorth considers reimplementing GS as content :P
09:50:18 * LordAro considers reimplementing andythenorth as content
09:50:27 <TrueBrain> can I join in on that party?
09:51:39 <andythenorth> it's all just a simulation right?
09:51:44 <andythenorth> and that sim is running in a sim?
10:01:16 <andythenorth> hmm slow compile is slow
10:01:23 <andythenorth> it's back to > 30s
10:01:36 * andythenorth seriously considers sharding the project into multiple grfs
10:02:44 <andythenorth> multi-grf container format for bananas?
10:04:29 <_dp_> TrueBrain, 7912 and 8115 both add a new network command, while that makes them (likely) to work together imo it would be better to merge it into one command
10:04:41 <TrueBrain> I am doing that as we speak :)
10:04:52 <_dp_> I even kinda prepared for it in mine with more general command (CmdSetIndustryFlags vs CmdSetIndustryFlags)
10:05:12 <TrueBrain> I just need to figure out the bitmask for p1 :D
10:05:35 <_dp_> I can do that too if you want
10:05:37 <TrueBrain> as you both took another approach there :)
10:05:44 <TrueBrain> appreciated, but almost there :P
10:06:00 <TrueBrain> sorry, didn't want to steal your work away from you :(
10:06:26 <_dp_> if you use mine as a base it should be trivial to integrate flags
10:07:00 <TrueBrain> the main issue is, the other uses a bitmask, and you use 0/1 .. which is rather incompatible :)
10:07:06 <TrueBrain> but that is easily fixed :D
10:08:58 <TrueBrain> just how to document it, lol
10:11:45 *** urdh has quit IRC (Quit: Boom!)
10:13:00 <TrueBrain> and now the big trick .. how to test this :D
10:14:14 <TrueBrain> src/lang/gaelic.txt:125: warning: String name 'STR_CARGO_PLURAL_CANDY' does not exist in master file
10:15:18 <TrueBrain> someone failed at a regex
10:15:22 <TrueBrain> and someone failed at review :P
10:16:03 <DorpsGek> [OpenTTD/OpenTTD] ldpl commented on pull request #8115: GS methods to let company have exclusive access to the industry https://git.io/JL6LT
10:16:06 <TrueBrain> Xaroth! You failed :P Would you mind fixing pretty please? (before 1800 tonight) :)
10:17:21 <TrueBrain> why did the CI not pick this up, I wonder ..
10:17:45 <TrueBrain> because they are not being picked up by the GCC-matcher :D
10:17:49 <TrueBrain> well, that explains
10:25:50 <_dp_> TrueBrain, I'll remove ->owner stuff from pr but with you working on it how should I do it, just force push as usual?
10:27:38 <TrueBrain> give me a few, and you can check if what I did is what you expect :)
10:28:17 <TrueBrain> again, sorry for taking over the work :P It is just a nice way to get my hands a bit dirty :P
10:29:00 <_dp_> I'm totally fine with someone else doing all the work ;)
10:29:29 <TrueBrain> don't get used to it :P
10:35:36 <TrueBrain> LordAro: we have a few hours to clean up that mess :P
10:35:41 <TrueBrain> just happy I noticed before eints hits :D
10:37:15 <TrueBrain> which means remembering how to ... to write a GS :D
10:38:34 <TrueBrain> we used to put new AI functions in the regression
10:38:38 <TrueBrain> but I guess GS never made it there :D
10:38:55 <_dp_> you can take gs from my pr as a base
10:39:33 <TrueBrain> totally forgot you attached one
10:41:30 <TrueBrain> I ... started in a year that has no engines
10:45:09 <LordAro> TrueBrain: can you still remember how to write squirrel? :p
10:45:41 <TrueBrain> I am shocked, honestly
10:46:06 <TrueBrain> okay, this is interesting ... I opened the game, made a savegame, loaded it again: invalid chunk
10:46:48 <TrueBrain> seems to be a local issue
10:47:03 <TrueBrain> WSL sometimes seems to be flushing incorrectly, losing data
10:47:38 <TrueBrain> ah, no, nevermind .. I did something stupid
10:47:43 <TrueBrain> saved between the 2 PRs
10:47:51 <TrueBrain> which they both use the same savegame version ... they store different data
10:47:56 <TrueBrain> LALALALALAAAa, nothing happened :P
10:51:25 <_dp_> these shifts in docommand always look suspicious to me... (((uint8)owner) << 16)
10:51:41 <_dp_> and I know it's like that in my pr as well :p
10:51:42 <TrueBrain> its your code :P I just change 8 into 16 :)
10:53:00 <_dp_> it will probably even work fine because 16 is int or smth like that
10:53:39 <_dp_> just can't help but think of possible ub when I see this stuff
10:53:51 <_dp_> also (control_flags & ::INDCTL_MASK) << 8)
10:54:40 <_dp_> I mean, enums are likely ints but aren't exaclty guaranteed to be afaik
10:54:56 <andythenorth> hmm 'writing the changelog' as form of QA
10:55:05 <andythenorth> found one bug already because I had to list detailed changes
10:55:43 <TrueBrain> GetLastProductionYear is a weird function to me
10:55:44 <_dp_> and it's all over the openttd code
10:55:47 <TrueBrain> but okay, I am sure there was a reason for that
10:56:36 <TrueBrain> the description suggests it only changes value if there was someone asking for the cargo
10:56:44 <TrueBrain> it is not .. just all output industries go up by 1 year every yerar
10:57:16 <TrueBrain> and GetCargoLastAcceptedDate can return "0", which also feels a bit off
11:02:30 <DorpsGek> [OpenTTD/OpenTTD] SamuXarick commented on pull request #8403: Feature: Reworked recessions setting to allow durations up to 10 years https://git.io/JL6Gn
11:03:33 <DorpsGek> [OpenTTD/OpenTTD] James103 commented on issue #8116: Old patchpack savegame versions will soon overlap https://git.io/JfZsT
11:04:15 <TrueBrain> well, found a bug during testing, I like that :)
11:05:26 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on issue #8116: Old patchpack savegame versions will soon overlap https://git.io/JfZsT
11:05:58 <DorpsGek> [OpenTTD/OpenTTD] JGRennison commented on issue #8116: Old patchpack savegame versions will soon overlap https://git.io/JfZsT
11:07:30 <TrueBrain> _dp_: do you happen to know why GetLastProductionYear was added?
11:07:59 <TrueBrain> is nielsm around? :) (I cannot see who is online on IRC :P)
11:08:05 <_dp_> my guess is just to expose some production-related vars
11:08:13 <TrueBrain> the var is really weird
11:08:20 <TrueBrain> it just basically outputs the current year :D
11:08:46 <_dp_> even for unserved indudstries??
11:08:54 <TrueBrain> especially for those
11:09:19 <TrueBrain> I have all industries reporting me the current year, as far as I can tell :P
11:09:56 <_dp_> that's weird, I don't see in a code why would it do that
11:10:14 <LordAro> isn't it used for the graphs?
11:10:38 <TrueBrain> it is used to consider if an industry should close, as far as I can tell
11:10:47 <_dp_> it's used for production changes and ind closures
11:11:30 <TrueBrain> code suggests only transported cargo should change the year
11:11:40 <TrueBrain> they are all btw initialized on the year they are created
11:11:45 <TrueBrain> which is also a bit surprising :)
11:11:46 <_dp_> and if it's always current nothing will ever close ;)
11:11:51 <TrueBrain> even for industries that do not produce anything :P
11:12:11 <_dp_> that's ok, it's a close protection for new industries
11:12:23 <TrueBrain> but reading it as GS is a bit weird
11:12:42 <_dp_> yeah, those vars are kinda internal
11:13:06 <TrueBrain> let me see if this only happens for the first year or something
11:13:21 <TrueBrain> nope .. 1982, 20+ industries change to 1982 too
11:14:09 <_dp_> TrueBrain, are those secondary industries?
11:14:13 <TrueBrain> code suggests it shouldn't ..
11:14:21 <TrueBrain> hmm, possible, I guess
11:14:24 <TrueBrain> how to check easily ..
11:14:32 <_dp_> primary industries produce stuff so it should be updated for them afaict
11:15:05 <TrueBrain> ah, yes, those are primary
11:15:11 <TrueBrain> so they always produce
11:15:16 <TrueBrain> bit counterintuative
11:15:51 <TrueBrain> but okay, it is not wrong
11:15:55 <TrueBrain> let's not try to fix the world in 1 day :)
11:16:00 <TrueBrain> _dp_: tnx, that was my missing link :)
11:18:34 <TrueBrain> let's see if I did not fuck this up ..
11:18:39 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #7912: Feature: Disallow industry production changes from GS https://git.io/Jej1Z
11:19:38 <_dp_> idk, they aren't wrong but they expose weird internals for no apparent reason
11:20:01 <TrueBrain> LordAro: if you have some time, would you be so kind to review ^^ ? I made minor changes, but enough that I would like a second opinion :)
11:20:02 <_dp_> I can't think of any use for those functions but I can see it becoming a bit of a problem if internals change
11:20:53 <TrueBrain> _dp_: I would guess the idea is that you can mimic what the game is doing, so if production year is 5 years in the past, close the industry
11:21:41 <DorpsGek> [OpenTTD/OpenTTD] SamuXarick commented on pull request #8403: Feature: Reworked recessions setting to allow durations up to 10 years https://git.io/JL6cK
11:21:53 <_dp_> oh, so like switch it to 10? set flags to no close and do it in gs if it's 10
11:21:55 <TrueBrain> we output more of these internal values via GS; cannot really think of a harm atm
11:23:39 <_dp_> well, pontential harm is the usual as it makes hard to change internal mechanic when some api locks it in
11:23:44 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #8115: GS methods to let company have exclusive access to the industry https://git.io/JfZt7
11:24:00 <_dp_> though I guess these can be replicated even if they become unused internally so shouldn't matter much
11:24:17 <TrueBrain> I pushed both commits in your branch
11:24:26 <TrueBrain> we have to merge one before the other, ofc :)
11:24:27 <andythenorth> GetLastProductionYear is...oh you figured it out :P
11:25:20 <TrueBrain> yes, in contrary to OpenTTD, the conversation doesn't pause if you leave :P
11:27:21 <andythenorth> so did frosch ship FML yet?
11:31:57 <TrueBrain> so that is 2 more PRs .. :)
11:56:22 <_dp_> was just thinking about making pr to disable it yesterday
11:57:26 <TrueBrain> no clue what I am looking at :P
11:58:44 <TrueBrain> I see words, but I cannot translate them into sanity
11:58:46 <andythenorth> it's a request for town rules to be content :P
11:58:47 <_dp_> new players first reaction to local authorities? :p
11:58:58 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain opened pull request #8412: Fix 86e08aa8: STR_CARGO_PLURAL_CANDY with cases was not renamed https://git.io/JL6Rz
11:59:32 <_dp_> sanity and local authorities don't mix well :p
11:59:38 <_dp_> and don't tell me it's realistic :p
11:59:55 <TrueBrain> I wasn't considering that, but now you mention it ...... :P
12:00:03 * andythenorth must to hair cut
12:00:11 <andythenorth> "for the livestream"
12:00:13 <andythenorth> and also because
12:00:22 <TrueBrain> I can't get a haircut :P
12:00:26 <TrueBrain> like .. shit closed
12:00:28 <LordAro> TrueBrain: need to remove a space to preserve alignment
12:00:38 <TrueBrain> LordAro: I spotted that too yes ... stupid space-alignment shit :P
12:00:38 *** andythenorth has quit IRC (Quit: andythenorth)
12:00:54 <TrueBrain> that sed is a lot more complex :D
12:02:25 <LordAro> let's see... s/STR_..CANDY(\.[a-z]+) /STR_..SWEETS\1/ ?
12:03:07 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #8412: Fix 86e08aa8: STR_CARGO_PLURAL_CANDY with cases was not renamed https://git.io/JL6Rz
12:03:15 <TrueBrain> sed -r -i 's/_CANDY.([a-z]) /_SWEETS.\1/' src/lang/.txt
12:03:19 <TrueBrain> `sed -r -i 's/_CANDY\.([a-z]*) /_SWEETS.\1/' src/lang/*.txt`
12:03:25 <DorpsGek> [OpenTTD/OpenTTD] LordAro approved pull request #8412: Fix 86e08aa8: STR_CARGO_PLURAL_CANDY with cases was not renamed https://git.io/JL60g
12:03:34 <LordAro> basically the same :)
12:04:04 <TrueBrain> I always have to fight with sed to get it in a mode that accepts \1
12:04:09 <TrueBrain> can never remember .. -r .. -e ... nothing ..
12:07:24 <_dp_> TrueBrain, you mean CmdIndustryCtrl? looks fine to me
12:12:51 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain merged pull request #8412: Fix 86e08aa8: STR_CARGO_PLURAL_CANDY with cases was not renamed https://git.io/JL6Rz
12:13:21 <TrueBrain> was hoping LordAro could give that a quick look too :)
12:15:02 *** HerzogDeXtEr has joined #openttd
12:23:47 <TrueBrain> guess you will have to find a way to make your OCD deal with that now :)
12:33:24 <DorpsGek> [OpenTTD/OpenTTD] Gymnasiast commented on issue #6922: Replace ICU ParagraphLayout with something else? https://git.io/JvfQ3
12:34:29 <_dp_> eh, now I'm having savegame issues on the current master...
12:34:34 <_dp_> can't open the game I just saved
12:34:52 <TrueBrain> _dp_: did you save with a custom build too? :D
12:35:08 <LordAro> ...the current savegame version wasn't bumped past the patchpack block, was it?
12:35:23 <TrueBrain> LordAro: that uses the latest enum value, doesn't it?
12:36:42 <TrueBrain> hmm, there is one boo-boo there, I guess
12:36:45 <TrueBrain> it is MAX_VERSION - 1
12:37:08 <TrueBrain> and 286 should -not- be used
12:37:16 <TrueBrain> well, with the other 2 PRs that will be fixed, so it is fine, I guess
12:37:19 <TrueBrain> I just need to merge them today :D
12:37:45 <TrueBrain> so yes, _dp_ , you cannot load a savegame created with current master atm :D
12:38:10 <TrueBrain> so if LordAro can tell me that function looks okay, I will merge them in :D :P
12:38:32 <_dp_> oh well, I'll save it in 10.3 I guess
12:38:50 <TrueBrain> alternative, I can make a PR that bumps the version to 287
12:40:18 <DorpsGek> [OpenTTD/OpenTTD] LordAro commented on pull request #7912: Feature: Disallow industry production changes from GS https://git.io/JL6Vw
12:40:35 <LordAro> wait hang on, which one should i be reviewing? they appear to be identical
12:40:53 <TrueBrain> one PR has the other ocmmit, yes
12:41:03 <TrueBrain> if that function is okay
12:43:31 <DorpsGek> [OpenTTD/OpenTTD] LordAro approved pull request #8115: GS methods to let company have exclusive access to the industry https://git.io/JL6wJ
12:43:53 <TrueBrain> will check for that initialize zero part
12:44:36 <LordAro> unless the saveload code is somehow initialising the Industry object without calling the constructor (i wouldn't put it past it...), it's setting those variables already
12:45:18 <TrueBrain> I see no reason why it wouldn't have
12:45:24 <TrueBrain> but let me add an assert, and test it :D
12:47:09 <TrueBrain> right, where is that savegame archive of yours ..
12:47:55 <TrueBrain> owh, title screens only .. I have those too :P
12:53:15 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #7912: Feature: Disallow industry production changes from GS https://git.io/Jej1Z
12:53:41 <TrueBrain> removing code is surprisingly easy :D
13:03:33 <TrueBrain> lol, you still cannot remove a server from the serverlist via the GUI?
13:03:45 <TrueBrain> sometimes I am surprised that some bugs are not fixed over the years :D
13:05:56 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #7912: Feature: Disallow industry production changes from GS https://git.io/JL663
13:07:33 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain approved pull request #7912: Feature: Disallow industry production changes from GS https://git.io/JL66z
13:11:29 *** andythenorth has joined #openttd
13:12:01 <TrueBrain> and how is your hair?
13:15:48 <_dp_> case LT_ARCTIC: TileLoopTreesAlps(tile); break;
13:16:19 <TrueBrain> hmm ... not being able to fast-forward in MP is annoying if you want to test some things :P
13:16:26 <TrueBrain> how often does this link graph scheduler run ...
13:16:56 <_dp_> you can probably safely patch it a bit to enable
13:17:22 <andythenorth> hair? what hair?
13:20:16 <andythenorth> the first thing against the wall
13:20:24 <andythenorth> will be OpenTTD trees
13:21:06 <LordAro> TrueBrain: which one do you want me to merge? :p
13:21:14 <TrueBrain> LordAro: neither :D
13:21:36 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain merged pull request #7912: Feature: Disallow industry production changes from GS https://git.io/Jej1Z
13:22:45 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain dismissed a review for pull request #8115: GS methods to let company have exclusive access to the industry https://git.io/JL6wJ
13:22:46 <TrueBrain> as I want to do it properly ... ^^ :)
13:22:49 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #8115: GS methods to let company have exclusive access to the industry https://git.io/JfZt7
13:23:14 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain approved pull request #8115: GS methods to let company have exclusive access to the industry https://git.io/JL61R
13:23:57 <TrueBrain> now I need to figure out how this linkgraph stuff works in multiplayer :P
13:29:30 <DorpsGek> [OpenTTD/OpenTTD] andythenorth commented on pull request #8403: Feature: Reworked recessions setting to allow durations up to 10 years https://git.io/JL6DT
13:29:52 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain merged pull request #8115: GS methods to let company have exclusive access to the industry https://git.io/JfZt7
13:30:00 <TrueBrain> right, that solves the saveload issue for now :)
13:31:20 * andythenorth wavey hands, some kind of per-region 'economy health' var
13:31:35 <andythenorth> Railroad Tycoon would also dump payment rates in a recession, do we do that currently?
13:31:38 <andythenorth> or just production?
13:32:05 <andythenorth> wiki says we just do cargo production
13:34:14 * andythenorth wonders if cb39 is used by anybody ever?
13:35:20 <andythenorth> I can't find documentation for which vars it would have in scope, except var 18
13:37:43 <TrueBrain> okay, linkgraph is runnng, it just doesn't need to join in multiplayer .. eeuuhh .. how to force that :P
13:50:13 <DorpsGek> [OpenTTD/OpenTTD] ldpl opened pull request #8413: Fix: Don't lower tree density if spreading is not enabled https://git.io/JL6H9
13:50:31 <TrueBrain> STOP MAKING NEW PRS! Ffs :P
13:51:09 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #7081: Change: [Linkgraph] Pause the game when linkgraph jobs lag (#6470) https://git.io/fh2Pg
13:52:58 <_dp_> TrueBrain, ok, ok, I'll make it a draft :p
13:53:13 <andythenorth> 'make trees great again'
13:53:25 <andythenorth> eh it's only a bit more work to move them to newgrf _dp_ :P
13:53:30 <andythenorth> like 2 lines or something, right?
13:53:39 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain approved pull request #7081: Change: [Linkgraph] Pause the game when linkgraph jobs lag (#6470) https://git.io/JL6Q9
13:53:56 <_dp_> andythenorth, well, I have t(h)ree prs planned
13:54:03 <_dp_> andythenorth, and moving to newgrf would be 4th :p
13:54:20 * andythenorth hopes for christmas trees grf
13:54:39 <Samu> I discovered an issue with these new CMake builds: 'Openttd.exe' and 'Openttd.exe (Install)'
13:54:42 <andythenorth> we don't have weather, but we could have autumn
13:54:53 <_dp_> seasons don't rly fit openttd imo
13:54:57 <_dp_> unless they're permanent
13:55:03 <andythenorth> if it's content...
13:55:04 <Samu> the 'Openttd.exe' doesn't create AI/GS API compatibility files
13:55:09 <_dp_> a bit like day/night just slover
13:55:14 <Samu> some AIs will fail to start
13:55:18 <andythenorth> speedruns with hacked grfs don't fit openttd imho :)
13:55:25 <andythenorth> but they're lolz, and there's a group who like it
13:55:28 <Samu> I have to test them with 'Openttd.exe (Install)'
13:55:47 <TrueBrain> _dp_: can you explain to me how your PR fixes the issue?
13:55:56 <andythenorth> seasons definitely don't work for farm production, it's weird to do _that_
13:56:00 <DorpsGek> [OpenTTD/OpenTTD] LordAro commented on pull request #8413: Fix: Don't lower tree density if spreading is not enabled https://git.io/JL67l
13:56:03 <TrueBrain> and you are missing a space before } in else
13:56:14 <_dp_> TrueBrain, it's wrong, ignore it for now
13:56:17 <LordAro> an extra one before the first if as well
13:56:39 <TrueBrain> so you made 2 devs review a PR that is wrong?! OMG! :P
13:56:52 <_dp_> TrueBrain, I made it a draft :p
13:57:24 <_dp_> I realised the mistake right when I pressed the "create" button xD
13:58:41 <_dp_> fwiw it now messes up sparse forests
13:58:52 <Samu> by the way, how would AIs know an industry is exclusive to someone else?
14:01:06 <_dp_> that's gs stuff, ai's aren't expected to deal with that
14:01:15 <_dp_> and gs is supposed to tell player somehow
14:01:23 <FLHerne> That's not really an answer :p
14:02:30 <_dp_> FLHerne, GetExclusiveSupplier should be available for ai's
14:02:54 <_dp_> and GetExclusiveConsumer
14:03:54 <TrueBrain> a patch in a patch in a patch
14:03:59 <TrueBrain> that became really difficult to review :P
14:12:37 <Samu> Company A has exclusive supplier, Company B bought exclusive transport rights in town. Neither company A or B get cargo now :p
14:14:13 <Samu> 0% transported, industry will close
14:16:22 <DorpsGek> [OpenTTD/OpenTTD] JGRennison commented on pull request #7081: Change: [Linkgraph] Pause the game when linkgraph jobs lag (#6470) https://git.io/JL6b7
14:18:02 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain merged pull request #7081: Change: [Linkgraph] Pause the game when linkgraph jobs lag (#6470) https://git.io/fh2Pg
14:18:31 <TrueBrain> that moment you try to rebase a year old PR, to find out that is not easy :D
14:22:11 <_dp_> does 6470 only affect cargodist?
14:23:53 *** frosch123 has joined #openttd
14:24:04 <TrueBrain> I have no clue what the linkgraph does, honestly :)
14:25:33 <andythenorth> is linkgraph the actual cargo flow graph, or is it just the one displayed to player?
14:28:17 <_dp_> also LinkGraphSchedule::SPAWN_JOIN_TICK - 2 looks fishy
14:28:33 <_dp_> in general network commands aren't guaranteed to execute in one tick
14:28:43 <_dp_> not sure if that breaks anything here though
14:28:51 <glx> <TrueBrain> that moment you try to rebase a year old PR, to find out that is not easy :D <-- if it adds files it can be fun with CMake inclusion :)
14:29:16 <_dp_> but still depending on timing of networks commands doesn't seem to be a good idea
14:30:16 <andythenorth> banning a whole network subrange?
14:31:06 <_dp_> I'm more amused by the fact that it was reddit server mod who made that post
14:32:38 <Samu> andythenorth, recessions setting name is "economy"
14:32:59 <Samu> yeah, i don't know why it's named like that
14:33:55 <DorpsGek> [OpenTTD/OpenTTD] ldpl updated pull request #8413: Fix: Don't lower tree density if spreading is not enabled https://git.io/JL6H9
14:34:16 <andythenorth> I mean '?' why do I need to know that? :)
14:34:48 <andythenorth> oh we already have a recessions setting?
14:34:55 <andythenorth> it's not just determined by the economy?
14:35:05 <andythenorth> how does FIRS handle that?
14:35:21 <andythenorth> some of these concepts need deleting
14:35:25 <Samu> it doesn't combine with the economy setting, it is the economy setting!
14:35:48 <andythenorth> what is the economy setting called then?
14:36:05 <Samu> there's a economy type, simply named "type"
14:36:10 <frosch123> "recession" is basically a disaster
14:36:20 <frosch123> your industry does not notice it
14:36:24 <andythenorth> I thought smooth economy prevented recessions :)
14:36:36 <andythenorth> one day I learn OpenTTD
14:36:39 <frosch123> it's just an yeti that eats 50% of all cargo before it arrives at your station
14:36:47 <andythenorth> discord players make me feel like I know nothing :)
14:36:54 <Samu> yes, smooth, frozen and original is part of the economy type setting, named "type"
14:38:10 <_dp_> Samu, it's named "economy.type", just "type" is just a shorthand gs/ai stuff accepts
14:39:14 <Samu> ah, recessions is named "difficulty.economy" then
14:44:49 <frosch123> if you add stuff there, it is not necessary that you edit specs wikis, PR nml and more your self :)
14:45:36 <andythenorth> if nml wiki was in truewiki
14:45:53 <andythenorth> could nml PRs include a docs patch? :P
14:46:59 <FLHerne> Only if the wiki data lived in the nml git repo
14:47:06 <glx> frosch123: about scope and optimisation, I fear adding an extra arg to reduce() will be the only option
14:47:34 <DorpsGek> [OpenTTD/OpenTTD] ldpl commented on pull request #8413: Fix: Don't lower tree density if spreading is not enabled https://git.io/JLiv8
14:47:56 <frosch123> glx: option 2 does not work? adding the scope to the variable in the ast appeared nicer yesterday
14:48:49 <glx> ah maybe, didn't look in this direction
14:49:02 <DorpsGek> [OpenTTD/OpenTTD] ldpl updated pull request #8413: Fix: Don't lower tree density if spreading is not enabled https://git.io/JL6H9
14:49:33 * _dp_ hopes nobody saw that extra space xD
14:51:14 <DorpsGek> [OpenTTD/nml] frosch123 approved pull request #176: Add constants introduced in OpenTTD/OpenTTD#8392. https://git.io/JLifc
14:52:13 <TrueBrain> frosch123: I guess you mean to say here that the 0x47 "ctlflags" of industry was not in the specs yet?
14:52:55 <frosch123> TrueBrain: you solved the riddle :)
14:53:14 <TrueBrain> frosch123: ah .. I expected it was already
14:54:03 <frosch123> pretty sure there are more addition in ottd meanwhile, who forgot to add stuff to specs and nml
14:55:02 <frosch123> for some reason people forget to communicate stuff after PR merge. so noone uses their additions
14:56:09 <TrueBrain> well, I added a line
14:57:59 <TrueBrain> _dp_: also LinkGraphSchedule::SPAWN_JOIN_TICK - 2 looks fishy <- you are right, that "2" is a configuration setting
14:58:12 <TrueBrain> strictly seen, one could run a multiplayer server where ticks are delayed with more than 1
14:58:17 <andythenorth> GS vehicle availability?
14:58:24 <TrueBrain> it was initially meant to tune multiplayer ... we got it right first time round, as it turned out
14:59:16 <FLHerne> frosch123: If someone was using say {UNPRINT} in a language file and didn't want to change it, would just typing \9A\04 in the string instead work?
15:00:11 <_dp_> TrueBrain, it's not just about the value, I don't think relying on any kind of timing of network commands at all is the right approach
15:00:15 <frosch123> FLHerne: pretty sure you have to use the unicode variant
15:00:29 <_dp_> like, there are callbacks if something needs to be ececuted after a certain command
15:00:36 <frosch123> and since the parameter will be invalid unicode in any case, it probably won't work either
15:00:38 <TrueBrain> _dp_: OpenTTD is in lockstep (server vs client), so I don't see a problem there?
15:01:24 <TrueBrain> if you do a DoCommandP on tick N, you are guaranteed it is executed on N + 1 for server + clients
15:01:48 <_dp_> TrueBrain, it may be right for server commands but definitely not client ones
15:01:58 <_dp_> clients are rate limited
15:01:59 <TrueBrain> this is the server executing the command
15:02:07 <TrueBrain> the server prepares a frame full of commands
15:02:19 <_dp_> I know, that's why I'm saying it will probably work
15:02:28 <TrueBrain> isn't it guaranteed to work? :)
15:02:29 <_dp_> just not the good approach imo
15:02:48 <TrueBrain> if you have alternatives, I am happy to hear; but everything we do here is "just in time" concept :)
15:02:57 <frosch123> TrueBrain: client commands are queued on the server, and can be delayed a lot
15:03:04 <_dp_> I didn't study that pr long enough to say it's guaranteed :p
15:03:04 <TrueBrain> client commands, yes :)
15:03:10 <TrueBrain> but this is not a client command ;)
15:03:12 <FLHerne> frosch123: There's a stupid hack in the NFO specs for invalid unicode
15:03:35 <frosch123> FLHerne: yes, but i would expect nml to bail on invalid unicode in langfiles
15:04:26 *** jottyfan has joined #openttd
15:04:31 <FLHerne> I meant literally "\9A\04" in the string
15:04:47 <DorpsGek> [OpenTTD/OpenTTD] SamuXarick left a comment on commit: Feature: Set exclusive access to industry from GS (#8115) https://git.io/JLiTO
15:04:48 <_dp_> that whole +1 tick thing is kinda weird btw
15:04:56 <_dp_> makes a lot of problem for gamescripts and stuff
15:05:01 <frosch123> FLHerne: nml interprets "\" in langfiles? :o
15:05:03 <TrueBrain> _dp_: you mean of the network itself?
15:05:07 <FLHerne> ...do we even have any tests for this?
15:05:19 <_dp_> TrueBrain, it's not network as that happens locally on the server
15:05:29 <TrueBrain> I meant: you meant for networking?
15:05:35 <TrueBrain> that every command is executed a frame later?
15:05:58 <FLHerne> No of course we don't
15:06:00 <_dp_> no, just for gamescript logic, they can't execute more than 1 command/tick
15:06:23 <_dp_> so it takes eternity to update company goals for examply
15:06:33 <TrueBrain> yeah, your sentence left some room for interpretation, hence my question ;)
15:06:46 <TrueBrain> GSes are heavily rate limited yes
15:07:18 <TrueBrain> (well, AIs are too btw)
15:08:05 <_dp_> if I wasn't able to spam commands with a patch I probably would've got rid of that +1 delay altogether by now
15:09:26 <_dp_> but just having a bit of async is a minor annoyance for me right noow
15:10:00 <TrueBrain> if I remember correctly, single player is delayed with +1, as multiplayer is always delayed with +1
15:10:09 <TrueBrain> and to avoid people testing their GS on SP, to find out it crashes and burns on MP
15:10:12 <TrueBrain> stuff is delayed :)
15:10:35 <TrueBrain> ha, I was right too, w00p!
15:10:41 <TrueBrain> 10 year old knowledge ftw :P
15:10:41 <_dp_> TrueBrain, yeah, there is something like that there...
15:11:15 <TrueBrain> for AIs it made sense: AIs should calm down in the rate they build stuff etc
15:11:21 <TrueBrain> it makes for a more .. fair, feel
15:11:25 <TrueBrain> for GS, it is a bit of a different story
15:11:30 <_dp_> though it should probably mention gs testing in comments as those didn't make any sense to me last time I read them
15:11:36 <TrueBrain> but .. GS is just a small modification on AI, because we could :P
15:11:44 <andythenorth> it was a good experiment
15:11:55 <TrueBrain> _dp_: that sentence makes no sense to me :D Sorry, what did you mean?
15:12:31 <_dp_> "and to avoid people testing their GS on SP, to find out it crashes and burns on MP" - this makes sense to me as a reason for delay
15:12:40 <_dp_> whatever is in the comment near that delay doesn't :p
15:12:51 <TrueBrain> `/* Suspend the script player for 1+ ticks, so it simulates multiplayer. This both avoids confusion when a developer launched his script in a multiplayer game, but also gives time for the GUI and human player to interact with the game. */`
15:13:11 <TrueBrain> seems to say the same to me :D
15:13:17 <FLHerne> frosch123: I tried it, the answer is 'yes'
15:13:30 <_dp_> TrueBrain, hm, yeah, not sure what was I looking at then
15:13:43 <_dp_> something about "fairness" for client on serevr
15:13:48 <TrueBrain> _dp_: it is difficult code to understand, I know .. and writing comments is hard when you are neck deep in it :)
15:14:03 <frosch123> FLHerne: i would consider that a bug :)
15:14:15 <Samu> speaking of network tick delays, "More than max_no_competitors could be created in network games" the PR that got derailed :(
15:15:17 <TrueBrain> _dp_: but I think the more we see how people use GS, the more it becomes clear the framework doesn't cut it sufficiently. For example, not having callbacks makes it an event-based system, which feels unfit for the purposes people use GSes
15:16:20 <TrueBrain> sounds like a very clear comment to me too, sorry :P
15:16:29 <FLHerne> frosch123: It doesn't seem to be documented :-/
15:16:51 <Samu> there can be 2 tick delays, I know they can happen
15:17:18 <FLHerne> frosch123: But anyway, in case of "{UNPRINT} is deprecated, consider using ... instead.", what should '...' be?
15:17:34 <FLHerne> Or just "... is deprecated, stop doing stupid things?"
15:18:01 <FLHerne> "is deprecated without a direct replacement." ?
15:18:30 <TrueBrain> _dp_: no clue what more to add to such comment to make it more clear :D But it does require you to know how the whole server/client model works for OpenTTD, that is very true :)
15:18:33 <_dp_> TrueBrain, I may be wrong but I think that also delays gs commands which has nothing to do with fairness
15:19:35 <TrueBrain> not sure if it matters, one way or the other
15:19:43 <TrueBrain> the GS commands will always be delayed till the next frame
15:19:46 <TrueBrain> as otherwise, desyncs
15:19:54 <_dp_> also 1 tick "advantage" in openttd is kinda laughable anyway :p
15:19:55 <TrueBrain> if it is +1 or +2 .. :)
15:20:06 <supermop_Home> andythenorth how far apart should stations be for luxury emus?
15:20:22 <andythenorth> supermop_Home I'd place them in towns :P
15:20:37 <TrueBrain> _dp_: with a frame_freq of 0, you are correct
15:20:46 <TrueBrain> but initially, it was considered that frame_freq could be as high as 20
15:20:51 <TrueBrain> (and we had servers running that)
15:21:00 <_dp_> TrueBrain, I'm pretty sure if it's arranged correctly gs commands can be executed instantly without desyncs
15:21:02 <TrueBrain> that means a server-player has a full second longer to do shit :)
15:21:21 <andythenorth> supermop_Home there's no real specific distance, usually I say '64 tiles' but that's kind of made up
15:21:28 <TrueBrain> _dp_: not in the way OpenTTD is setup. At least, you cannot guarantee that in any way, shape or form :)
15:21:30 <andythenorth> and if you're using cdist on a multi-stop route...
15:21:38 <TrueBrain> the way the network is designed, is to prevent desyncs as much as possible
15:21:47 <andythenorth> the cargo aging thing is a failed hack TBH :)
15:21:49 <TrueBrain> which might not be what you want, but it is what you got :P
15:22:01 <TrueBrain> there are SOME commands that are an exception btw
15:22:07 <_dp_> TrueBrain, iirc frame_freq doesn't even work correctly on non-default setting
15:22:07 <TrueBrain> but they are very carefully tailored
15:22:17 <TrueBrain> 13 years ago, it used to work flawless :P
15:22:19 <_dp_> at least that's the impression I got when I looked at netcode lastt time
15:22:21 <TrueBrain> I cannot speak for the current state :D
15:22:55 <TrueBrain> but again, it was a setting so we could tune stuff .. pretty sure everyone runs it on 0 these days :)
15:22:59 <TrueBrain> the internet changed a lot :P
15:23:14 <andythenorth> supermop_Home failing to make money with them?
15:23:19 <andythenorth> they're quite brittle
15:23:34 <TrueBrain> for the first few years some people ran it on "1" instead of "0", as with 300+ms latency, it was becoming easy to hit the lag-protection
15:23:41 <supermop_Home> will an equivalent capacity of medium wagons load faster than long wagons?
15:24:08 <andythenorth> nope, all wagon length stats are normalised as far as possible
15:24:16 <andythenorth> there are zero economic choices to make there
15:24:37 <supermop_Home> andy im using a griffon with 4-6 medium wagons
15:24:45 <supermop_Home> but it looks kinda dumb
15:25:12 <andythenorth> the sprite, or the resulting train?
15:25:24 <TrueBrain> _dp_: this is btw one of the drawbacks of what we nowedays call "lockstep" server/client, or "fully deterministic" :) It needs a delay to ensure server and client are doing EXACTLY the same :P
15:25:35 *** jottyfan has quit IRC (Quit: jottyfan)
15:26:06 <supermop_Home> the short little train.. like i feel like i should be using something else
15:26:18 <TrueBrain> I doubt there is an alternative for OpenTTD that would work as robust shrug
15:26:25 <andythenorth> supermop_Home pax or freight?
15:26:26 <supermop_Home> around 40-54 tiles manhattan between towns
15:26:38 <_dp_> TrueBrain, I kinda lost you on "the way the network is designed, is to prevent desyncs as much as possible"
15:26:47 <_dp_> it sure doesn't look that way :p
15:27:13 <TrueBrain> did you ever notice have incredible stable our network games are? :)
15:27:17 <andythenorth> supermop_Home you can probably use anything except high-speed and TGV on that
15:27:18 <supermop_Home> HST would be overkill
15:27:29 <TrueBrain> the only desyncs happen because people are using random over interactiverandom, or memory that is not initialized, basically :p
15:27:37 <TrueBrain> and for that I just say: NewGRF :P
15:27:56 <andythenorth> supermop_Home I have a 62 tile route with a mix of luxury emus, pax engines, all making money
15:28:00 <TrueBrain> but games do not desync because a client is lagging behind, etc
15:28:03 <andythenorth> 3 stations, one dead in the middle
15:28:04 <frosch123> FLHerne: there is no valid usecase for UNPRINT. consider it non-existing :)
15:28:07 <_dp_> TrueBrain, yeah, but that's just what command queue gives you
15:28:37 <TrueBrain> _dp_: a lock-stepped command queue, yes :) With that we mean: every N ms the server tells all clients: make a step, wait for the next
15:29:01 <_dp_> also I recently found out that some ppl enable pause on join for stability
15:29:06 <frosch123> FLHerne: nml and/or eints do explicitly not support some control chars, to not expose things that were considered bad features
15:29:31 <TrueBrain> _dp_: pause-on-join should only prevent the command-queue from increasing in size while a client is joining
15:29:43 <TrueBrain> as in, there are games where running commands is about as fast as the game runs
15:29:54 <TrueBrain> but those servers have other problems :P
15:30:11 <milek7> well, not necessarily. delay can be avoided if there's ability to rewind game state :P
15:30:16 <TrueBrain> pause-on-join is not needed if you make sure your server is slower than all clients btw :D
15:30:31 <TrueBrain> milek7: yes, when I sit on an unicorn, I love how the world looks to me :)
15:30:51 <TrueBrain> in OpenTTD, "game state" is rather .. fluent :P
15:30:51 <andythenorth> unicorns for christmas pls
15:31:19 <TrueBrain> and so many people came (and went), swearing to fix that :P
15:31:57 <_dp_> TrueBrain, I know that, but if client is lagging on simulation it's hopeless anyway, pause or join or not
15:32:07 <TrueBrain> _dp_: I fully agree :)
15:32:11 <_dp_> also there was a setting to increase catchup time iirc
15:32:17 <TrueBrain> I believe openttdcoop was to blame for that functionality btw :P
15:32:33 <TrueBrain> yeah ... but if your client cannot keep up, increasing that is totally pointless too :D
15:32:38 <milek7> yeah rewinding ottd state is hard to do, but I commented generally about "lockstep" server/client
15:32:38 <_dp_> TrueBrain, oh, that explains everything xDDD
15:33:40 <TrueBrain> many people couldn't join openttdcoop games after a while :P
15:34:00 <TrueBrain> and pause-on-join does help with slow connections
15:34:06 <TrueBrain> like .. 15 years a go that was a real thing
15:34:06 <_dp_> milek7, rewind what? as long as command order is maintened it doesn't matter if it's delayed or not
15:34:42 <TrueBrain> these days if you need pause-on-join, I would dare to claim you have other problems already :D
15:34:49 <TrueBrain> if you can watch a youtube movie, that shouldn't be any issue :D
15:35:00 <TrueBrain> well, for 4k by 4k servers maybe
15:35:07 <TrueBrain> but then again ... you have other issues :P
15:35:48 <_dp_> btw, nowdays map loading time is mostly limited by server archiving speed
15:35:59 <TrueBrain> _dp_: I think milek7 refers to a way of lockstep, which is an async-lockstep, where you run commands as soon as you receive them, and roll back if it turns out there was one out-of-order
15:36:06 <TrueBrain> it is used in RPG games for example
15:36:14 <TrueBrain> not all, but just some, key frames cause a rollback
15:36:16 <milek7> _dp_: I mean that you can design game with lockstep multiplayer without effective lag (apply late arriving commands in the past state, and resimulate everything after that)
15:36:56 <_dp_> milek7, ah, no, I just meant servers own commands, clients will be delayed as usual ofc
15:38:17 <TrueBrain> for action games it works really well, async-lockstep, as a lot of information is irrelevant in terms of time
15:38:22 <TrueBrain> or the health of your oponent
15:38:24 <DorpsGek> [OpenTTD/OpenTTD] SamuXarick commented on pull request #8403: Feature: Reworked recessions setting to allow durations up to 10 years https://git.io/JLiOt
15:38:29 <TrueBrain> so rollbacks are very rare
15:38:41 <TrueBrain> (information just resets when-ever you get the next update)
15:38:48 <TrueBrain> OpenTTD is not that kind of game :)
15:39:02 <TrueBrain> every change has a HUGE influence on the game-state :P
15:39:02 <frosch123> factorio has some kind of latency hiding
15:39:34 <frosch123> when you build stuff, the drawing stuff pretends it's there immediately, but it only starts existing/working when the server confirmed it
15:39:35 <TrueBrain> I hope they didn't use the OpenTTD approach for a lot of things .. game-state for sure is one of them :D
15:39:51 <TrueBrain> lot of games use that trick, yes :) Minecraft is known for that :D
15:39:58 <TrueBrain> (you see a block, and poof, it is gone)
15:40:09 <frosch123> oh, they tried something else first. it failed, then they went the ottd way
15:40:10 <_dp_> TrueBrain, yeah, stuff like that doesn't even cross my mind when it comes to openttd :p
15:40:37 <TrueBrain> the network we had in the very first version was funny
15:40:43 <TrueBrain> depending on your latency, you would desync instantly
15:41:01 <TrueBrain> when I rewrote the network to TCP, lot of people were telling me I was doing it wrong, games should be UDP, and blablablablablabla
15:41:05 <TrueBrain> this was 2005 or what-ever
15:41:18 <TrueBrain> fucking hilarious, looking back :)
15:42:02 <_dp_> udp? in openttd?? suuure...
15:42:23 <TrueBrain> "all games do UDP, so OpenTTD should too", I remember people saying :P
15:42:27 <andythenorth> game state should be XML :P
15:42:32 <andythenorth> oh wait, that was newgrf
15:42:41 <TrueBrain> you really REALLY want to test out my @kick macros, dont you andythenorth ?
15:42:50 <andythenorth> everything should be tested occasionally
15:43:04 <andythenorth> I am going out in 20 mins, so kick me then, and it will look like I rage quit
15:44:00 <_dp_> anyway, history is fun but why did we even start on that topic? xD
15:44:15 <TrueBrain> _dp_: owh, you said you didn't understand a comment in the code :D
15:44:42 <frosch123> are PR templates a good idea, or are they just inconvenient?
15:44:48 <TrueBrain> frosch123: a very good idea
15:44:53 <TrueBrain> this is just annoying, honestly :)
15:44:54 <andythenorth> just keep them simples
15:45:18 <_dp_> meanwhile I still haven't figured if that linkgraph pr affects non-cargodist games
15:45:20 <andythenorth> 'why', 'screenshot if relevant' ' how to test', 'what I've tested'
15:45:41 <frosch123> i thought i could make a single template. but there are so many things for features-only, that make no sense for bugs. but adding multiple templates to choose from makes it harder to pick the right one
15:45:51 <JGR> @dp if there are no link graph jobs running, it will have no effect
15:45:59 <andythenorth> "The [product I make name] feature that asks ‘why is this idea important’ is great because it makes the respondent think twice about what they’re saying and then argue the case for their idea."
15:46:06 <TrueBrain> frosch123: "bugs", "features", sounds sane to me .. but l ike 2 or 3
15:46:23 <_dp_> JGR, yeah, but that doesn't quite answer it ...
15:46:29 <andythenorth> frosch123 "Heading [optional - delete if not needed]"
15:46:33 <andythenorth> get people to deliberate a bit
15:46:34 <TrueBrain> what does link graph do, is the question I guess :D
15:46:41 <_dp_> like, I know some cargodist/linkgraph/something is still running even if it's off
15:46:46 <DorpsGek> [OpenTTD/OpenTTD] SamuXarick commented on pull request #8404: Feature: Snow line height can be autodetermined upon world generation https://git.io/JLi3h
15:46:49 <_dp_> I justt not sure what exactly
15:47:03 <frosch123> TrueBrain: main toolbar, map menu
15:47:33 <andythenorth> is it only the visual display?
15:48:02 <frosch123> no, it also routes stuff, but the display makes it obvious why it is called "graph"
15:48:33 <frosch123> is explaining things with pictures no longer cool?
15:48:35 <TrueBrain> so without cargodist, it also has plenty of stuff to do? :)
15:48:54 <andythenorth> it should be doing nothing without cdist AIUI
15:48:55 <_dp_> at the very least it has plenty of ram to eat :p
15:50:04 <andythenorth> oh, if I turn off cdist, I still get a cargo map?
15:51:09 <JGR> Edges are still recorded even if cargodist does not do anything with them
15:51:10 <Eddi|zuHause> that's why it's not called "off"
15:51:47 <andythenorth> every day is a school day
15:53:02 <_dp_> can it all be just ripped off on the server safely I wonder...
15:53:32 <_dp_> clients can linkgraph themselves if they need it :p
15:53:39 <JGR> Why bother, the performance impact is negligible with cargodist set to manual
15:54:08 <_dp_> JGR ram consumption isn't entirelyl negligible
15:54:10 <TrueBrain> _dp_: if you go down that route, many more places you could just disable the code for servers :P
15:54:45 <_dp_> though I guess 64k stations isn't exactly a common usecase...
15:55:03 <TrueBrain> what gave that away? :D
15:55:05 <Eddi|zuHause> i was about to ask: how many stations are you getting?
15:56:20 <_dp_> anyway, I was just wondering, if it was a real problem I would've probably done it already :P
15:57:07 <FLHerne> There are definitely games with thousands of station
15:57:19 <JGR> @dp It's certainly possible to significantly reduce the RAM use of the linkgraph, but it's still not exactly near the top of memory wasters
15:59:09 <Eddi|zuHause> FLHerne: surely in a game with thousands of stations, the actual cargo packets are orders of magnitude more
15:59:29 <_dp_> with openttd release cycle a lot of my thinking goes the way "is this something I may be concerned about the next year" :p
15:59:50 <FLHerne> I have a few cities with about 150 bus stops each, the cargo:station ratio is probably fairly low
15:59:54 <JGR> When you have thousands of stations, the N^2 parts of the linkgraph storage starts becoming significant
16:01:32 <Eddi|zuHause> why is there N^2 storage? wouldn't a sparse lookup table be better?
16:01:35 <JGR> In trunk you end up with loads of invalidated FlowStats which use a lot of memory and make the linkn grpah overlay inaccurate
16:03:41 *** Flygon has quit IRC (Quit: A toaster's basically a soldering iron designed to toast bread)
16:05:04 <TrueBrain> efficiency of 300% ... I think this PR still has a bug :D
16:06:01 <TrueBrain> it also gets a big wonky if it takes more than a year to travel from station to station :D hihi
16:16:09 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #7353: Feature: Measure vehicle capacity utilisation efficiency https://git.io/fhho4
16:16:12 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #7353: Feature: Measure vehicle capacity utilisation efficiency https://git.io/JLiWZ
16:16:32 <TrueBrain> it is a really nice feature :) Just needs a bit more work :) But at least rebased now! I think .... it was difficult :P
16:16:47 * andythenorth wonders if github offers a way to discover nml repositories
16:30:42 <milek7> TrueBrain: sorry if I'm asking too often, but is there something required from me about keyauth proposal? or I should just wait for feedback?
16:40:23 <TrueBrain> I wanted to pick that up today, but got distracted heavily :D
16:41:02 <TrueBrain> I want to make a Discussion thread about it to collect some info from others, so we can pick a way to go forward
16:41:15 <TrueBrain> Hopefully tonight, otherwise tomorrow
16:42:02 <TrueBrain> And we need to find wording that the average Joe is going to understand .. ugh :p
16:51:22 *** ZirconiumY has joined #openttd
16:52:22 *** ZirconiumX has quit IRC (Ping timeout: 480 seconds)
16:52:38 <andythenorth> I should...test a headset or something
16:52:50 <andythenorth> find a room without wives, children
16:53:00 <andythenorth> hide anything personal
16:53:40 <TrueBrain> You are still welcome to join LordAro
16:53:57 <TrueBrain> But it does require testing beforehand :)
16:54:36 <LordAro> mm, especially because i'm now on my laptop which i've never used the webcam with :)
16:55:17 *** Artea has quit IRC (Ping timeout: 480 seconds)
16:55:23 *** Westie has quit IRC (Ping timeout: 480 seconds)
16:55:39 <TrueBrain> So get yourself on Discord, and we can test tonight :)
17:05:37 <Xaroth> It's a tourist attraction
17:05:42 <Xaroth> a busride to the top of the mountain
17:06:04 <Xaroth> or a kickass sleighride downwards
17:22:04 <DorpsGek> [OpenTTD/OpenTTD] SamuXarick updated pull request #7822: Fix #7670: Cache the origin tile to prevent recurring calls to the road pathfinder when a vehicle is blocked by another https://git.io/JegbJ
17:22:49 <andythenorth> releases are so boring :P
17:22:55 * andythenorth hasn't automated any release stuff
17:51:29 *** Wormnest has joined #openttd
17:53:51 <FLHerne> andythenorth: Is that true?
17:54:02 <andythenorth> that it runs amok?
17:54:29 <FLHerne> Can't you block creation based on industry_count, industry_distance etc?
17:54:46 <andythenorth> I can block it trivially using probability
17:55:30 <FLHerne> So why is "The rate of industry spawning is nothing that newgrf can influence...except to prevent it completely." true?
17:55:38 <_dp_> pretty sure you can change rates one way or the other
17:55:42 <andythenorth> there's nothing I can do to make more spawn
17:55:47 <_dp_> but may not be the most convenient way
17:55:56 <andythenorth> or at least, last time it was investigated, it was thought there was nothing
17:56:10 <FLHerne> Sure, but the complaint is about too many
17:56:58 <FLHerne> Default OTTD behaviour is 'amok' :p
17:57:03 <FLHerne> Particularly on large maps
17:58:16 <FLHerne> Still, the answer to this particular complaint looks like "set to 'funding only'"
17:58:31 <andythenorth> yes, but then they get a map with no industries :)
17:58:46 <andythenorth> which is kinda odd
17:59:56 <FLHerne> You can change it after map creation
18:00:56 <andythenorth> I have been changing it using the FIRS parameter during gameplay
18:01:12 <andythenorth> but that's equally limiting, as it prevents funding
18:01:30 <andythenorth> it's like every version of this is just FAIL :)
18:01:30 <FLHerne> andythenorth: Same setting
18:02:29 <FLHerne> Or is that a JGR thing?
18:02:55 <andythenorth> 'funding' won't be found as a search string
18:03:44 <FLHerne> Maybe I'm not hallucinating today
18:14:08 <DorpsGek> - Update: Translations from eints (by translators)
18:21:37 <frosch123> hmm, looks like github has not implemented a PR template chooser
18:21:53 <frosch123> you can only have one PR template
18:25:56 <TrueBrain> frosch123: that is ... lame
18:27:54 <TrueBrain> I think PR creators cannot set labels
18:28:02 <TrueBrain> so not sure about that last part
18:28:17 <TrueBrain> and it is common to put this in <!-- -->, so if someone posts it, it is not rendered
18:28:21 <frosch123> yes, some of them are for the reviewers
18:28:26 <TrueBrain> (not the ##, but the rest)
18:28:43 <TrueBrain> ah, yeah, okay, that is fine :)
18:29:01 <TrueBrain> "Checklist for reviewer" maybe as title instead?
18:29:05 <frosch123> i'll add the <!-- --> when the rest is fine
18:29:10 <TrueBrain> if I would make a PR, I would remove that part not understanding it :D
18:29:36 <TrueBrain> Limitations: add something like "do you know of a bug or a corner case that doesn't work?" ?
18:30:21 <TrueBrain> motivation, feature: "Which group of players you think would enjoy this feature?" or something?
18:30:42 <TrueBrain> but yeah, looks fine!
18:38:08 <frosch123> andythenorth: noone reads that
18:41:03 <Eddi|zuHause> if you link it, they will come :p
18:41:54 <Samu> glx, how do i test regression now after the cmake?
18:42:41 <Eddi|zuHause> possibly also link to codestyle and commit message style?
18:43:01 <frosch123> github already does that
18:43:09 <frosch123> does not mean people read it
18:43:17 <frosch123> but it is automatically checked, so no problem
18:44:01 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #8413: Fix: Don't lower tree density if spreading is not enabled https://git.io/JLiPD
18:44:26 <Eddi|zuHause> i haven't made a PR in a while, but i never saw a prominent place where commit message style was explained. and had to bury deep in the compiler output to find out what exactly was wrong
18:44:32 <TrueBrain> what is cool, if you change contribution.md, it tells you it has changed since your last PR :)
18:44:53 <glx> Samu: there should be a "Test" menu with "Run CTest" in
18:44:54 <TrueBrain> GitHub interface tries really hard to make people read it :D It is nice :)
18:45:33 <glx> Samu: other option is to switch solution explorer view to show targets
18:45:56 <glx> and build "regression" target
18:47:00 <Samu> i don't have anything listed in the solution explorer when show targets is selected
18:48:14 <glx> double click on cmake target view
18:49:01 <DorpsGek> [OpenTTD/OpenTTD] ldpl commented on pull request #8413: Fix: Don't lower tree density if spreading is not enabled https://git.io/JLiXn
18:49:51 *** jottyfan has joined #openttd
18:51:21 *** Samu has quit IRC (Quit: Leaving)
18:52:36 *** JGR has quit IRC (Quit: Leaving)
18:53:29 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #8403: Feature: Reworked recessions setting to allow durations up to 10 years https://git.io/JLiX9
18:53:32 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain closed pull request #8403: Feature: Reworked recessions setting to allow durations up to 10 years https://git.io/JLask
18:54:47 <TrueBrain> _dp_: cheers, good explanation :D
18:54:54 <TrueBrain> let me build a preview and test it out :)
19:00:30 <andythenorth> discuss in stream or not?
19:01:59 <frosch123> i am preparing a list of all vehicle/group list related things. we have more of those
19:02:25 <frosch123> i don't think they should be discussed in isolation
19:04:08 <TrueBrain> andythenorth: did you manage to prepare a nice map for that one ticket I asked by any chance? :D
19:07:39 <andythenorth> no, but I can look
19:07:53 <andythenorth> I have finished a Horse release, to-do list is shorter :P
19:08:29 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain requested changes for pull request #8401: Change: Towns don't build dead-end road bridges https://git.io/JLiDY
19:08:43 <TrueBrain> andythenorth: yes! And I want some nice looking graphics :D
19:08:57 <TrueBrain> (with files on BaNaNaS plz, so I can download them easily :D)
19:09:00 <andythenorth> rainbow colour schemes :P
19:09:57 <TrueBrain> I like how #8401 went slightly overboard with comments :) That is a nice problem to have :)
19:10:33 <andythenorth> I used to comment every line when I was learning to code
19:10:55 <andythenorth> now I only comment for apologies to my future self
19:13:42 <TrueBrain> my truck is 30% efficient ...booooo
19:14:05 <TrueBrain> guess most vehicles are ~50% efficient :P
19:14:21 <Timberwolf> I tend to have a big thing about "why, not what" with dev teams.
19:14:58 <TrueBrain> there is a bit of balance between those comments :)
19:15:00 <andythenorth> I used to write comments like "this adds two numbers together" :P
19:15:01 <michi_cc> TrueBrain: Do you have time for a stream test, or are you still blowing leafs?
19:15:08 <TrueBrain> michi_cc: nope, I have time!
19:15:18 <andythenorth> I don't understand even the ticket
19:15:33 <Timberwolf> TrueBrain: yeah, it's more of a subtle balance.
19:16:09 <Timberwolf> "does this tell me something I can't see from the code in front of me?" is a good question imo.
19:16:10 <frosch123> Timberwolf: "code with comments" in the intermediate state between "bad code without comments" and "code that is so good it does not need comments" :)
19:16:10 * andythenorth steps away from issues
19:16:33 *** blathijs has quit IRC (Ping timeout: 480 seconds)
19:16:39 <Timberwolf> And also, "why doesn't tge code tell me that?"
19:20:58 <Timberwolf> I had a coworker years ago whose comments mainly explained his state of mind.
19:22:05 <Timberwolf> They were all fairly low value "explaining the method name again" type ones, but with more swearing and ranting the angrier he'd been.
19:22:44 *** blathijs has joined #openttd
19:23:10 <andythenorth> I think most comments can be distilled down to 3 things
19:23:32 <andythenorth> 1. a url to Stack Overflow for Why This Doesn't Work And What I Had To Do About It
19:30:09 <TrueBrain> LordAro: so ... tomorrow .. in or out? :)
19:31:09 <LordAro> TrueBrain: i think i should be able to do tomorrow
19:31:16 <TrueBrain> can we test today? :)
19:31:22 <LordAro> need to work out how to make my microphone work though
19:31:25 <LordAro> alsa isn't playing ball
19:31:31 <michi_cc> TrueBrain: Did you try to convince jgr, btw?
19:31:41 <TrueBrain> I have not asked, no
19:33:29 <TrueBrain> LordAro: well, hopefully you get it to work today :D Would be nice to test it :P
19:33:39 <TrueBrain> I don't really feel like testing it 1 hour before stream, so any time before that would be good :)
19:34:09 <TrueBrain> it is hard to configure OBS :(
19:35:19 *** gelignite has joined #openttd
19:53:49 *** andythenorth has quit IRC (Quit: andythenorth)
20:11:40 <DorpsGek> [OpenTTD/OpenTTD] ldpl opened pull request #8415: Feature: Add an option to disable tree growth completely https://git.io/JLi5i
20:11:53 *** jottyfan has joined #openttd
20:17:25 *** jottyfan has quit IRC (Quit: jottyfan)
20:20:26 <_dp_> next pr will be "now that I've fixed everything lets remake trees from scratch and try not to break"
20:20:30 <TrueBrain> LordAro: and as you find out on Discord, very painfully, is that nobody knows who you are :P I had hilarious conversations with people who were telling me things about openttd.org :P
20:20:34 <_dp_> that is, if I ever get to doing that :p
20:22:18 <frosch123> TrueBrain: so that's why you do not want that purple colour?
20:22:29 <frosch123> you want honest feedback :)
20:22:38 <TrueBrain> haha, yes, that is the excuse that works :)
20:22:57 <TrueBrain> honestly, I find it hilarious .. users says BaNaNaS doesn't work, I tell him what he is doing wrong, and he goes off on me that he knows better how it works
20:23:01 <TrueBrain> those things, just make my day :)
20:25:43 <_dp_> haha, similarly there is no end to people trying to tell me how towns grow xD
20:34:08 <_dp_> though I kinda know devs opinion on this already it would be interesting to hear viewers/players
20:34:21 <_dp_> if you ever manage to get their feedback somehow...
20:34:29 <TrueBrain> that sure is a list :P
20:34:41 <TrueBrain> maybe for next time, if there is a next time :D
20:34:45 <_dp_> that's kind of my TODO-OR-NOT-TODO list :p
20:35:00 <TrueBrain> I have no clue how many people we can expect, and it they ever want a second one :P
20:40:42 <michi_cc> The list of people that stream for 1-10 viewers in Twitch is looong.
20:40:53 <michi_cc> As long as streaming is fun...
20:43:27 <_dp_> out of that list I personally am most interested to see some popup style and usage for new features as there are currently 2 trends in patchpacks
20:44:01 <TrueBrain> scriptable UIs next you say?
20:44:41 <_dp_> that's too big of a pie :p
20:44:51 <_dp_> I'm more realistic so just ui for scripts ;)
20:45:37 <glx> I often can't watch streams because twitch doesn't enable transcoding
20:45:47 <_dp_> (I know I missed some words somewhere but it's too late to correct)
20:47:47 <_dp_> speaking of ui for scripts, something needs to be done about blood company rating table
20:48:06 <_dp_> did nielsm pr his changes on that?
20:48:16 <michi_cc> glx: Transcoding is only guaranteed for partnered streamers and often also for affiliates. "Nobodies" often don't get it for their stream.
20:48:41 <nielsm> I didn't finish design for the GS-controlled company ratings
20:48:47 <glx> but it highly depends on the start hour too
20:49:13 <_dp_> nielsm, you have a lot of cool unfinished things I noticed ;)
20:49:20 <_dp_> well, so do I, I guess xD
20:51:15 <TrueBrain> he made drafts out of them :P
20:52:33 <TrueBrain> michi_cc: so we will be streaming 1080p to everyone? Well, that will be interesting :D
20:53:22 <michi_cc> If you don't have that much moving content, you don't need a high bit-rate though.
20:53:56 <michi_cc> I've seen non-game streams on Twitch where 1080p with 3000kpbs or so looks totally fine.
20:56:21 <nielsm> btw if you need to catch me for something specific please PM me, I might not notice channel messages until much later
20:56:26 <TrueBrain> I have min 4500 max 6000, guess I can tune that a bit
20:56:36 *** Tirili has quit IRC (Quit: Leaving)
20:57:19 <TrueBrain> trying to recreate a happy accident I had when making an intro .... but I am failing to do so :D
20:57:29 <TrueBrain> annoying, how things can just happen, and when you want it to happen, you fail :P
20:59:04 <TrueBrain> ...... keep that attitude for tomorrow would you? :P
21:04:36 <_dp_> hm... is there any particular reason why cargo ages when vehicle is loading/unloading?
21:04:59 <_dp_> I mean... it doesn't age on station
21:05:45 <_dp_> and it would make some things better if it didn't
21:08:51 <frosch123> i like that it penalises too long trains that load for 2 years
21:09:03 <frosch123> i wouldn't mind if cargo also ages while in stations
21:10:26 <_dp_> without station aging long trains can do some timetabling/delaying schenanigans to avoid the issue
21:15:18 <_dp_> it also kinda contradicts station rating calculation as that rewards you for having a train waiting at all times
21:15:25 <TrueBrain> right .. that is NOT helpful GitHub ...
21:15:33 <TrueBrain> GitHub Actions has encountered an internal error when running your job.
21:15:44 <TrueBrain> how about a rerun .. does that fix anything :D
21:16:29 <TrueBrain> at least info@ received an email about the failure :)
21:17:31 <frosch123> _dp_: but isn't that a good thing? it means there is some optimum in the middle which you can aim for. instead of some extreme
21:19:26 <glx> TrueBrain: seems MacOS runner has starting issues
21:19:35 <_dp_> frosch123, I'm not sure I like that kind of micro
21:19:40 <TrueBrain> I was hoping they would fix that last week
21:19:45 <_dp_> frosch123, managing train delays is a pita
21:20:48 <frosch123> pff, there always has to be room for better players
21:21:00 <frosch123> don't adjust the game to what you are good at
21:21:06 <_dp_> frosch123, but, more importantly, it penalizes servicing industries that produce less cargo
21:21:16 <_dp_> not only you get less but also more affected by ratings/aging
21:21:54 <frosch123> if you go for the same train length, yes
21:22:07 <frosch123> you have to use shorter trains for smaller industries
21:22:19 <_dp_> if you go for shorter trains you lose more on running costs
21:22:23 <frosch123> and imo a quite good one
21:23:40 <_dp_> frosch123, I may not be the only one who doesn't find it fun :p
21:23:50 <frosch123> ok, so you want a flat map, equally spaced industries with constant production, ... did you consider playing excel instead? :p
21:23:53 <_dp_> though most players don't even know how cargo ages...
21:24:58 <_dp_> also it's kind of irrelevant to how I play tbh
21:27:00 <TrueBrain> glx: can't find reports of people complaining about it .. hope it isn't just us :D
21:27:58 <TrueBrain> owh, we are running on macos-11.0
21:28:05 <TrueBrain> not macos-latest (which is 10.15)
21:28:11 <TrueBrain> there might be less runners or something for 11.0?
21:28:30 <TrueBrain> "The macOS 11.0 virtual environment is currently provided as a preview only."
21:30:25 <TrueBrain> CI is running macos-latest
21:30:29 <TrueBrain> which is much more responsive
21:30:36 <TrueBrain> but I guess we wanted the 11.0 for ARM?
21:33:06 <TrueBrain> anyway, nightlies are not being produced because of this .. guess we will try again tomorrow :D
21:38:54 *** andythenorth has joined #openttd
21:39:51 <frosch123> let's see what essays people will write
21:47:30 <andythenorth> let's have a beer!
21:50:05 <FLHerne> /me has a big mug of cocoa and some porridge :p
21:50:42 <FLHerne> I did drink some whisky with chili in it earlier for some reason
21:54:56 *** Samu has quit IRC (Quit: Leaving)
21:56:26 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
22:06:24 <Timberwolf> I got back into brewing my own during one of the various lockdown phases.
22:07:09 <Timberwolf> Currently have a cherry chocolate imperial stout, which is very nice (except the time I decided to have an entire pint of it during our work "Friday end of week drinks" call)
22:07:44 <Timberwolf> Well, it was still nice except one of our developers was very keen to get some infrastructure into Terraform immediately after the social call :)
22:31:19 <Timberwolf> It was on the list at one point!
22:31:31 <andythenorth> your list is longer than mine yet? :)
22:31:39 <andythenorth> do an industry set, that raises the game
22:31:48 <Timberwolf> I have most of the model, it was going to be an alternative livery for whatever the diesel-only version of it is.
22:32:06 <andythenorth> mine's more fake than the 88
22:35:08 <andythenorth> there's a picture somewhere of the big eurodual on test next to an 88
22:38:11 <andythenorth> google image search is very bad for Horse completedness
22:38:16 *** sla_ro|master has quit IRC ()
22:38:16 <andythenorth> I keep finding new things to draw
22:38:47 <andythenorth> how's the stations?
22:39:09 <Timberwolf> I should add some more to those.
22:39:35 <andythenorth> full baseset replacement or go home :P
22:39:44 <andythenorth> ogfx has served its time :P
22:39:56 <andythenorth> at least, the building sprites
22:40:03 <andythenorth> ogf landscape is good, and the vehicles are ok
22:40:14 <Timberwolf> I've been quite tempted to make a town set at some point.
22:40:47 <andythenorth> towns, planes, I am avoiding
22:41:05 <andythenorth> the current state of grf is remarkably good though
22:41:07 <Timberwolf> My problem with planes is I don't really like them (in game)
22:41:10 <andythenorth> so many good sets appearing
22:41:23 <andythenorth> planes, AV 9.8, done
22:41:27 <Timberwolf> So I'd do something weird and idiosyncratic to "fix the problems I see"
22:41:37 <andythenorth> they're just for FIRS supplies :P
22:45:10 * andythenorth wonders what voxel ships look like
22:45:15 <andythenorth> ships are soooo boring to draw
22:48:21 <DorpsGek> [OpenTTD/OpenTTD] 2TallTyler commented on pull request #8401: Change: Towns don't build dead-end road bridges https://git.io/JLPTS
22:48:33 <DorpsGek> [OpenTTD/OpenTTD] 2TallTyler commented on pull request #8401: Change: Towns don't build dead-end road bridges https://git.io/JLPTQ
22:55:13 <DorpsGek> [OpenTTD/OpenTTD] 2TallTyler updated pull request #8401: Change: Towns don't build dead-end road bridges https://git.io/JLzst
22:56:30 *** jottyfan has joined #openttd
22:56:37 <DorpsGek> [OpenTTD/OpenTTD] 2TallTyler commented on pull request #8401: Change: Towns don't build dead-end road bridges https://git.io/JLPkV
23:20:29 <DorpsGek> [OpenTTD/OpenTTD] JGRennison opened pull request #8416: Change: [Linkgraph] Allow job threads to be aborted early when clearing schedule https://git.io/JLPtt
23:21:59 <andythenorth> nice use of template :D
23:27:00 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
23:27:12 *** WormnestAndroid has joined #openttd
23:37:24 <frosch123> i made a compilation of all the group list prs
23:37:33 <frosch123> those merged last week and those still open
23:37:52 <frosch123> some of them proof that there are also singleplayer people :)
23:38:02 <frosch123> (read: broken in multiplayer)
23:38:46 <frosch123> but still a welcome distraction from the multiplayer fanboys
23:38:57 <DorpsGek> [OpenTTD/OpenTTD] michicc commented on pull request #8416: Change: [Linkgraph] Allow job threads to be aborted early when clearing schedule https://git.io/JLPma
23:39:13 <TrueBrain> We will see tomorrow live what you made out of this :D
23:40:11 <frosch123> still unsure how to incorporate the troll pr
23:43:12 <TrueBrain> We will just see how it goes :p so many ways to troll with it
23:43:55 <TrueBrain> If it can crash with trains I have some ideas :p
23:44:43 <frosch123> i'll add that tomorrow
23:44:44 <DorpsGek> [OpenTTD/OpenTTD] JGRennison commented on pull request #8416: Change: [Linkgraph] Allow job threads to be aborted early when clearing schedule https://git.io/JLPYc
23:45:12 <TrueBrain> "And now to our sponsor ... do you also hate that ships are useless? They just float a bit, but they cannot even crash trains! Wait no longer .. hovercrafts are here!"
23:45:36 <TrueBrain> Followed by boats ramming into trains
23:46:19 <frosch123> i am still amazed how dead the forums are
23:46:25 <TrueBrain> "What is the shortest path from A to B? Wrong! Hovercrafts "
23:47:31 <TrueBrain> Reddit is more active than the forums, yes
23:47:49 <frosch123> not sure about that
23:47:57 <frosch123> reddit is a single thread
23:48:07 <frosch123> while forum is scattered over 200 subforums
23:48:11 <TrueBrain> Btw, did any of the merged PRs for groups break multiplayer?
23:48:25 <frosch123> they break the gui in some places
23:48:39 <frosch123> but didn't spot multiplayer issues
23:48:52 <frosch123> i'll mention the issues on the stream, maybe someone fixes them :p
23:49:57 <frosch123> my favorite pr is the one that goes: this adds feature X. me: hmm, didn't we already have that? me: let's test, oh, it's still there, but kind of broken, so the PR adds an alternative solution instead of fixing the original...
23:50:45 <TrueBrain> Who needs hovercrafts
23:50:50 <frosch123> glx: nah, just missing widget sizing, so texts are truncated and unreadable
23:51:26 <TrueBrain> Anyway, time for some zzz
23:52:11 *** frosch123 has quit IRC (Quit: be yourself, except: if you have the opportunity to be a unicorn, then be a unicorn)
23:59:22 <andythenorth> _dp_ that is gold
23:59:33 <andythenorth> also FFS what time is ti?
23:59:38 <andythenorth> why do I stay up so late?
23:59:43 <andythenorth> wife wakes up at 5am, this hurts
continue to next day ⏵
