Config
Log for #openttd on 8th June 2021:
Times are UTC Toggle Colours
00:11:55  *** tokai has joined #openttd
00:11:55  *** ChanServ sets mode: +v tokai
00:13:24  *** HerzogDeXtEr has quit IRC
00:18:46  *** tokai|noir has quit IRC
00:20:44  *** debdog has quit IRC
00:24:01  *** debdog has joined #openttd
00:51:36  *** debdog has quit IRC
00:56:13  *** peter1138 has quit IRC
00:59:29  *** peter1138 has joined #openttd
00:59:29  *** ChanServ sets mode: +o peter1138
01:04:42  *** debdog has joined #openttd
02:07:18  *** Wormnest has quit IRC
02:11:02  *** norri has quit IRC
02:58:22  *** D-HUND has joined #openttd
03:01:44  *** debdog has quit IRC
03:10:39  *** glx has quit IRC
04:14:26  *** tokai|noir has joined #openttd
04:14:26  *** ChanServ sets mode: +v tokai|noir
04:21:13  *** tokai has quit IRC
04:35:30  *** Flygon has joined #openttd
04:50:28  *** virtualrandomnumber has joined #openttd
04:50:44  *** virtualrandomnumber has quit IRC
05:18:27  *** andythenorth has joined #openttd
05:22:21  *** tokai has joined #openttd
05:22:21  *** ChanServ sets mode: +v tokai
05:29:15  *** tokai|noir has quit IRC
05:30:06  *** andythenorth has quit IRC
05:49:55  *** sla_ro|master has joined #openttd
06:02:37  *** urdh has quit IRC
06:03:24  *** urdh has joined #openttd
06:50:03  *** andythenorth has joined #openttd
06:59:20  <peter1138> Yes/no
07:00:45  <andythenorth> probably
07:00:52  <andythenorth> another day, another covid test for my kids
07:00:53  <andythenorth> hurrah
07:17:08  *** HerzogDeXtEr has joined #openttd
07:24:54  <TrueBrain> it doesn't stop ! :P
07:27:41  <andythenorth> such admin
07:28:15  *** tokai|noir has joined #openttd
07:28:15  *** ChanServ sets mode: +v tokai|noir
07:35:10  *** tokai has quit IRC
07:36:27  *** andythenorth has quit IRC
07:37:09  *** tokai has joined #openttd
07:37:09  *** ChanServ sets mode: +v tokai
07:39:18  <TrueBrain> so .. how am I going to solve this freaking VS2017 issue without making the code really weird
07:39:19  <TrueBrain> hmm
07:42:15  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #9339: Codechange: add and use the ability to store (lists of) structs in savegames https://git.io/JZLZC
07:44:04  *** tokai|noir has quit IRC
07:45:38  <TrueBrain> I really think we should give tokai an award for most unstable connection in existence :)
07:55:40  *** EER has joined #openttd
08:01:49  <peter1138> SL_*_INCLUDE did look weird
08:02:11  <peter1138> inline static const? hmm
08:03:54  <TrueBrain> I did not know they existed :P
08:03:58  <TrueBrain> but they do
08:04:08  <TrueBrain> just MSVC is being a little child and doesn't want to play ball
08:04:41  <peter1138> Cannot access private member
08:04:52  <peter1138> Oi, you, leave my members alone
08:05:11  <TrueBrain> the fact it needs to access private members is already ... euhmz .. yeah
08:05:12  <TrueBrain> but okay
08:05:45  <peter1138> friend?
08:06:43  <TrueBrain> yeah .. so on all compilers it works to make the class a friend
08:06:48  <TrueBrain> where the inline static const is defined in
08:06:53  <TrueBrain> but .. MSVC2017 says: NO
08:07:22  <TrueBrain> trying to reproduce it on godbolt, but I don't know how to get MSVC2017 on there in C++17 mode :P
08:07:48  <TrueBrain> c++latest doesn't seem to be it
08:09:38  <TrueBrain> MSVC version numbers are just weird
08:11:08  <TrueBrain> okay, so it works with MSVC CL 19.14
08:11:10  <TrueBrain> it breaks with 19.16
08:11:19  <TrueBrain> fixed with 19.20
08:11:22  <TrueBrain> .... lol?
08:14:10  *** roadt has joined #openttd
08:20:21  <TrueBrain> https://developercommunity.visualstudio.com/t/friend-declaration-not-respected-with-us/257765?space=8&q=friend <- think this is the bug
08:21:37  <EER> so solved in vc2019 and not 2017? It looks like the C++ part of microsoft development is a lot less open than the .net efforts :)
08:23:00  *** tokai|noir has joined #openttd
08:23:00  *** ChanServ sets mode: +v tokai|noir
08:27:18  <TrueBrain> https://godbolt.org/z/f5dro8vhz
08:29:53  *** tokai has quit IRC
08:30:54  <TrueBrain> so just 2 versions they released are broken .. ugh .. so how to work around this .. hmm
08:31:55  <TrueBrain> non-inline variant works
08:31:58  <TrueBrain> but that is so uggglllyyyy
08:32:39  <EER> #ifdef MSVC_1916? speaking of ugly code :P
08:34:18  <EER> Based on this:
08:34:19  <EER> https://dev.to/yumetodo/list-of-mscver-and-mscfullver-8nd
08:34:41  <EER> The 1920 was released with VS2019, so they apparently never bothered to fix 2017
08:34:47  <TrueBrain> sadly, that doesn't really help, as I need to move a block of code :(
08:34:56  <TrueBrain> and no, that is how MSVC works .. VS2017 stopped at 1916
08:35:12  <TrueBrain> I believe only security-related stuff is backported
08:35:31  <EER> very attentive of them to break it in the final release of 2017
08:35:42  <TrueBrain> many more bugs :)
08:35:43  <Rubidium> keep believing?
08:35:44  <TrueBrain> this is just 1 ;)
08:35:56  <TrueBrain> they just want you to upgrade to their latest
08:36:01  <TrueBrain> which ... I cannot really blame them for
08:36:09  <EER> is there a reason not to for openttd?
08:36:18  <TrueBrain> people still compile OpenTTD for WinXP
08:36:22  <TrueBrain> nuff said on that front :P
08:36:36  <EER> with vs2017?
08:36:45  <TrueBrain> I do not know how, and I don't care how :P
08:37:01  <TrueBrain> but as soon as we drop any target, there will be some QQ
08:37:08  <TrueBrain> either an OS or a compiler
08:37:14  <TrueBrain> "they demand a good reason for ...."
08:37:37  <EER> I guess "I wanted to inline something" doesn't count ;)
08:37:46  <TrueBrain> it might or might not
08:37:53  <TrueBrain> just not in any mood to deal with those type of people
08:38:00  *** andythenorth has joined #openttd
08:38:13  <EER> At work we still have a vm with VS6 to compile our old shit, every couple of years :')
08:38:28  <TrueBrain> in 90% of the cases it is a user that contributed NOTHING to the project, but do complain they can no longer compile it theirselves on their ancient compiler :P
08:39:26  <TrueBrain> owh, joy, the "non-inline" variant doesn't work, as the length of the array is unknown because of that
08:39:26  <Rubidium> I'd say, update the compiler of the build farm to something new enough and then commit it ;)
08:39:38  <TrueBrain> Rubidium: the CI runs both VS2019 and VS2017
08:39:45  <TrueBrain> so it is more: drop a compiler from the CI :P
08:39:48  <EER> If I put myself in a classicist shoes like that, I would be using VS2003, maaaaybe 2005. 2017/9 isn't that big of a change imho
08:39:52  <andythenorth> oof is TrueBrain complaining about mac users again? :P
08:40:06  <Rubidium> TrueBrain: yeah, because obviously it doesn't support C++17 ;)
08:40:30  <tokai|noir> TrueBrain: I can forward it to my "ISP" :)
08:40:41  <TrueBrain> tokai|noir: :D
08:40:44  <TrueBrain> it really is insane
08:40:52  <TrueBrain> that you can do anything with your connection
08:41:03  <TrueBrain> for months it was "stable", but lately it just went bananas again
08:41:42  <TrueBrain> not complaining btw; just made me amazed how unstable a connection can be :)
08:42:06  *** m1cr0man has quit IRC
08:42:35  *** m1cr0man has joined #openttd
08:43:53  *** tneo has quit IRC
08:44:12  <tokai|noir> TrueBrain: if the missing stability was the only problem. The connection's poor bandwidth makes people flabbergasted all the time, when I tell them. It's straight out of the early 2000s. :D
08:44:49  *** andythenorth has left #openttd
08:44:55  <TrueBrain> StarLink? :D
08:45:00  <TrueBrain> as in: time for StarLink?
08:45:45  *** tneo has joined #openttd
08:47:36  <TrueBrain> okay, lets try this ..
08:47:40  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #9339: Codechange: add and use the ability to store (lists of) structs in savegames https://git.io/JZLZC
08:47:53  <TrueBrain> best part is that you can leave a rant about it :D
08:50:00  <TrueBrain> I could now if-def it
08:50:07  <TrueBrain> but I am not :P
08:50:19  <TrueBrain> or am I .. hmmm :P
08:50:21  <EER> Think about the children!
08:50:32  <TrueBrain> it makes it more explicit that it is only MSVC .. hmmm
08:51:38  <DorpsGek> [OpenTTD/OpenTTD] bootandy commented on discussion #9299: Cities build irrational bridges: right on the shore, along the lake, next to each other https://git.io/JGLMK
08:52:20  <tokai|noir> TrueBrain: Probably would an improvement. Just so darn expensive. :)
08:52:50  <TrueBrain> haha, yeah
08:59:26  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #9339: Codechange: add and use the ability to store (lists of) structs in savegames https://git.io/JZLZC
08:59:59  <TrueBrain> lets see if MSVC appreciates this solution ..
09:03:07  <TrueBrain> noooooooo
09:08:58  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #9339: Codechange: add and use the ability to store (lists of) structs in savegames https://git.io/JZLZC
09:09:40  <Timberwolf> Everyone loves MSVC. Back in the days of ~2005, a friend sent me some code... which VC turned into a console app which did nothing.
09:09:59  <Timberwolf> Turned out that if a file had unix-style line breaks, MSVC was treating it as an empty file.
09:10:12  <LordAro> classic
09:10:18  <Timberwolf> For... C, that notoriously whitespace-sensitive language.
09:10:55  <TrueBrain> that sounds like MSVC :)
09:11:15  <TrueBrain> really happy they developed VSCode, as it is just better in so many ways
09:11:27  <TrueBrain> as I liked the MSVC editor for the most part
09:11:35  <TrueBrain> just everything after the editor was .. euh ........ shit?
09:12:51  <TrueBrain> so reading a bit of bug-reports, the timeline seems to be this: in older versions of CL, "inline static const" ignored "friend" and could access any private variable
09:12:59  <TrueBrain> they fixed it, and now it couldn't access ANY private variable
09:13:06  <TrueBrain> they fixed that and it now works as it should
09:13:14  <TrueBrain> so ... LEFT, no RIGHT, no, WTFFFFFFF
09:19:16  <TrueBrain> all green \o/
09:29:09  *** sla_ro|master has quit IRC
10:13:55  *** andythenorth has joined #openttd
10:14:09  <andythenorth> someone broke the internet
10:14:17  <LordAro> indeed
10:14:23  <LordAro> fastly is current suspect
10:20:15  <TrueBrain> People still use that?
10:20:58  <LordAro> quite a lot of people, apparently
10:21:23  <LordAro> stackoverflow,reddit,python.org...
10:22:38  <TrueBrain> Surprising
10:24:15  *** WormnestAndroid has quit IRC
10:24:28  *** WormnestAndroid has joined #openttd
10:24:51  <TrueBrain> I really thought it died out :D oops, as clearly I am wrong :p
10:26:10  *** skrzyp has quit IRC
10:41:04  <FLHerne> I don't understand why these big sites trust CDNs with their entire business this way
10:41:49  <FLHerne> Looks like Paypal, Cloudflare had short outages but then failed-over to another provider
10:42:06  <FLHerne> Everyone else is just dead until further notice
10:42:48  <andythenorth> it's parallel to why do they trust security to services like solar winds (and similar)
10:43:11  <andythenorth> giving privileged system access and trust to a remote service
10:43:21  <andythenorth> which is a high value attack target
10:43:46  <andythenorth> "everything is fine until it isn't"
10:53:43  *** Strom has quit IRC
10:57:35  <TrueBrain> On the other hand.. 1 disruption in how many months?
10:58:21  <TrueBrain> Availability is expensive for most sites, so .. yeah
10:59:12  <andythenorth> it's funny that we (me at work) don't use the service, it's 4th, 5th or 6th party to us
10:59:19  <andythenorth> but we still have degraded services and ops
11:01:25  *** Strom has joined #openttd
11:01:54  <TrueBrain> as long as these things remain rare ... pretty sure you surive :D
11:02:00  <TrueBrain> not worth the expensive to fix it :P
11:04:40  <andythenorth> it's totally unrealistic to have a failover for e.g. vimeo, or eventbrite
11:05:10  <andythenorth> we already isolate from pypi, but it looks like it wasn't complete
11:05:16  <andythenorth> so that smacked us
11:06:45  <andythenorth> having a failover for Stack Overflow, that's a lol idea :)
11:06:54  <andythenorth> for when we need to copy-paste some code :P
11:10:46  <Rubidium> a "lol idea"? How is this issue going to be solved if the engineers of fastly can't find the solution on stackoverflow?
11:11:00  <andythenorth> let's clone SO
11:12:21  <TrueBrain> they solved it within an hour, it seems
11:12:25  <TrueBrain> which is pretty nice
11:14:56  <Timberwolf> failover for StackOverflow = find someone with an attitude of "it compiles, therefore it works", frequently draws inexplicable and bizarre conclusions as to why something does or doesn't work, and is adamant their view of a problem is right and everyone else is wrong. Ask them.
11:15:05  <FLHerne> andythenorth: SO database is publically available under CC license
11:15:12  <FLHerne> so you could absolutely do that
11:15:15  <andythenorth> oh
11:15:18  <andythenorth> I was only trolling :(
11:15:25  <andythenorth> you deflated me
11:16:36  <andythenorth> I wonder who status.io outsource their status page to? https://status.status.io/
11:21:04  <peter1138> Yeah, how am I meant to get any work done if I can't copy & paste from StackOverflow?
11:21:16  <andythenorth> pls send me the codes
11:21:42  <peter1138> Undo knobs
11:25:30  *** tokai has joined #openttd
11:25:30  *** ChanServ sets mode: +v tokai
11:32:16  *** tokai|noir has quit IRC
11:33:58  <Timberwolf> peter1138: Copy and paste it from personal tech blogs!
11:34:35  *** tokai|noir has joined #openttd
11:34:35  *** ChanServ sets mode: +v tokai|noir
11:35:13  <Timberwolf> It takes a little longer as you have to wade through 20 paragraphs of, "I remember as a little boy when my grandma would set up a Kubernetes cluster. The gentle sounds of trees flexing in the autumnal breeze, melding with the surprisingly extensive list of curse words granny had assembled over a lifetime in SRE."
11:35:30  <LordAro> hahaha
11:36:03  *** Funke has joined #openttd
11:37:38  <peter1138> Got to get those SEO results up...
11:39:49  *** Strom has quit IRC
11:41:21  *** tokai has quit IRC
11:41:57  *** gelignite has joined #openttd
11:42:34  *** tokai has joined #openttd
11:42:34  *** ChanServ sets mode: +v tokai
11:49:25  *** tokai|noir has quit IRC
12:14:29  *** glx has joined #openttd
12:14:29  *** ChanServ sets mode: +v glx
12:17:02  *** Samu has joined #openttd
12:26:32  <FLHerne> https://keepthescore.co/blog/posts/deleting_the_production_database/
12:26:58  <FLHerne> "What have we learned? Why won’t this happen again?"  > "We’ve learned that having a function that deletes your database is too dangerous to have lying around"
12:30:51  *** tokai|noir has joined #openttd
12:30:51  *** ChanServ sets mode: +v tokai|noir
12:32:38  <andythenorth> I often reach for this https://www.youtube.com/watch?v=SlA9hmrC8DU
12:37:50  *** tokai has quit IRC
12:40:31  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #9339: Codechange: add and use the ability to store (lists of) structs in savegames https://git.io/JZLZC
12:40:37  <TrueBrain> right, mostly figured out what stations were trying to do .. boy, that was difficult code ..
12:40:42  <TrueBrain> still the old stations are .. weird
12:41:30  <TrueBrain> and the switch from STNS to STNN didn't have a savegame bump, which makes for annoying code
12:42:21  <peter1138> Might run out of versions!
12:43:11  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #9339: Codechange: add and use the ability to store (lists of) structs in savegames https://git.io/JZLZC
12:43:19  <TrueBrain> yeah ...
12:43:22  <TrueBrain> and biting us in the ass now
12:43:23  <TrueBrain> great!
12:43:23  <TrueBrain> :P
12:43:37  <glx> of course https://github.com/OpenTTD/OpenTTD/compare/master...glx22:saveload doesn't work, need to redo chunks registration
12:47:21  <TrueBrain> what has the chunk registration have to do with it?
12:56:48  *** Kitrana1 has quit IRC
13:06:05  <glx> the main issue is static const ChunkHandler map_chunk_handlers[]
13:07:36  <glx> causes slicing, and if I remove default copy constructor it doesn't compile (expected)
13:13:55  <TrueBrain> I do wonder if those "save" booleans are really needed, tbh
13:14:15  <glx> some chunks are load only
13:14:41  <TrueBrain> yeah, I too have struct-parts that are load only, but I found other ways to deal with that
13:14:47  <glx> and there's an early return in not saving
13:15:32  <TrueBrain> as it looks to me that if saveload.[cpp|h] defines RiffChunkHandler, ArrayChunkHandler, etc
13:15:42  <TrueBrain> and the default ChunkHandler has a noop Save()
13:15:45  <TrueBrain> you get similar behaviour
13:15:55  <TrueBrain> but I honestly just skimmed it over when I wanted to start with it
13:17:06  <TrueBrain> Basically, I think it won't be a bad thing to have a LoadOnlyChunkHandler, of sorts :)
13:17:18  <TrueBrain> as now it is a bit .. implied that it was intentional :P
13:17:53  <glx> I tried something like that, with RiffChunkHandler being subclass of ChunkHandler, depending on declared chunk type, but the issue is on loading some now CH_ARRAY were CH_RIFF
13:18:15  <TrueBrain> you have to depend on what is in the code, yes :)
13:18:22  <TrueBrain> the type in the savegame is there to be pretty :P
13:18:42  <TrueBrain> it does some bookkeeping
13:18:46  <TrueBrain> but that is about it :D
13:19:03  <glx> saveload code is a pain ;)
13:19:16  <TrueBrain> no kidding :)
13:19:21  <TrueBrain> years of old code on top of old code
13:19:23  <TrueBrain> hack on hack on hack
13:23:37  *** andythenorth has left #openttd
13:24:34  <TrueBrain> the worst thing I have, that weird hacks stored on disk we can never remove
13:24:37  <TrueBrain> as we need to support loading it :P
13:32:31  *** nielsm has joined #openttd
13:47:10  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #9339: Codechange: add and use the ability to store (lists of) structs in savegames https://git.io/JZLZC
13:47:28  <TrueBrain> wasted 3 hours on converting station-chunks, only to find out 1) there is a bug somewhere, 2) it is completely not reviewable :P
13:47:33  <TrueBrain> so .. I need another approach for that :D
13:52:58  *** gelignite has quit IRC
13:53:09  *** tokai has joined #openttd
13:53:09  *** ChanServ sets mode: +v tokai
13:54:55  *** sla_ro|master has joined #openttd
13:57:38  *** gelignite has joined #openttd
13:59:00  <TrueBrain> I have this pretty big collection of savegames from different revisions .. but a huge hole between 218 and 287 :P
13:59:09  <TrueBrain> guess I should scape the GitHub issue tracker ..
14:00:08  *** tokai|noir has quit IRC
14:01:00  <LordAro> or flyspray
14:01:11  <TrueBrain> do not think they help with that hole
14:01:20  <LordAro> 218 too new?
14:01:22  * LordAro forgets
14:01:43  <TrueBrain> 218 is 1.10
14:01:47  <TrueBrain> owh
14:01:54  <TrueBrain> I am being a donkey
14:02:01  <TrueBrain> 220 .. 286 isn't used by us
14:02:02  <TrueBrain> LOL
14:02:03  <TrueBrain> oops :D
14:02:05  <LordAro> is that the block that you- :D
14:14:49  <TrueBrain> 1.11 cannot open all coop-public-server games :P
14:15:03  <LordAro> :o
14:15:11  <TrueBrain> especially really old ones
14:15:27  <TrueBrain> did not really investigate, other than they failing to load
14:15:36  <LordAro> file a bug!
14:16:52  <TrueBrain> first I want a way to know if my code is a regression :P
14:29:21  <TrueBrain> right, bunch of savegames that load with 1.11.2 don't load with my branch .. so this is useful at least :D
14:29:53  <TrueBrain> it is interesting that many savegames load files without having the right NewGRFs .. and some crash :P
14:30:21  <TrueBrain> files = fine
14:30:25  <TrueBrain> weird typo
14:32:28  <TrueBrain> hmm, they also fail on master
14:32:29  <TrueBrain> oh-oh
14:33:29  *** tokai|noir has joined #openttd
14:33:29  *** ChanServ sets mode: +v tokai|noir
14:36:08  <TrueBrain> hmm .. a  month ago master also fails on some ..
14:36:10  <TrueBrain> curious
14:40:23  *** tokai has quit IRC
14:41:40  <TrueBrain> newgrf_engine.cpp:1145: void GetRotorOverrideSprite(EngineID, const Aircraft*, bool, EngineImageType, VehicleSpriteSeq*): Assertion `!(e->u.air.subtype & AIR_CTOL)' failed
14:41:46  <TrueBrain> most likely because I don't have NewGRFs loaded atm
14:41:53  <TrueBrain> owh, 1.11.2 doesn't have asserts on
14:41:54  <TrueBrain> duuuhhhh
14:41:55  <TrueBrain> lol
14:41:58  <TrueBrain> what a waste of time :P
14:42:01  *** Kitrana has joined #openttd
14:42:42  <TrueBrain> testing with NewGRFs is ssssoooo sslslllllooowwwww
14:43:55  *** Kitrana1 has joined #openttd
14:50:05  *** Kitrana has quit IRC
14:58:28  <TrueBrain> in the set of savegames I have atm, which is mostly coop and some others, I need 270 NewGRFs I didn't had yet ..
14:58:29  <TrueBrain> holy crap
15:05:22  <Rubidium> can't your savegame reader read the savegame and then only place the right ones in the right folder ;)
15:05:34  <TrueBrain> that is what I am constructing
15:05:42  <TrueBrain> but it is idiotic that it is needed
15:05:48  <TrueBrain> we really need to cache NewGRF results :)
15:06:03  <TrueBrain> frosch123: yes, I know ... :P
15:07:06  *** Kitrana has joined #openttd
15:07:58  *** Kitrana1 has quit IRC
15:08:33  <TrueBrain> "fun" fact, in order to do that, I need to convert all grfs into their md5-name :P
15:31:39  <TrueBrain> this is also odd .. I have a tar with a grf in there, of which "grfid -m" tells me it has a certain hash
15:31:43  <TrueBrain> ingame it tells me: NAH! You don't have it!
15:32:17  <TrueBrain> but, you know what, I did find a GRF with the same ID, I will load that instead
15:32:20  <TrueBrain> after which it loads the right one
15:33:12  *** Wormnest has joined #openttd
15:39:13  <TrueBrain> the error line stating which file is not found, shows the replaced md5sum
15:39:16  <TrueBrain> not the one missing
15:39:17  <TrueBrain> lol?
15:46:18  <TrueBrain> this has confused me several times in crash reports .. but now I get why :P
15:51:46  <TrueBrain> dbg: [sl] Found waypoint tile 233444 with invalid position
15:51:46  <TrueBrain> lol
15:52:58  *** EER_ has joined #openttd
15:58:19  *** EER has quit IRC
16:00:02  *** Progman has joined #openttd
16:08:24  <FLHerne> TrueBrain: Is the idea with this 'savegame reader' to create a library that can be used by both OTTD and third-party tools?
16:19:43  <TrueBrain> To proof my header PR works
16:19:55  <TrueBrain> I have a lot of short-term goals :D
16:32:57  *** virtualrandomnumber has joined #openttd
16:36:20  *** virtualrandomnumber has quit IRC
16:49:26  <Timberwolf> TrueBrain: collection of old save games: http://timberwolf.woaf.net/stuff/openttd/saves.zip
16:50:01  <Timberwolf> Some have a very old version of UKRS2 with different vehicle lengths, and one has a reference to the original Project 1000 file.
16:50:17  <Timberwolf> Also a few of them have a newgrf which messes with the UKRS properties.
16:51:36  <TrueBrain> tnx :D
16:51:48  <TrueBrain> downloaded :)
16:52:11  <LordAro> i've got a boatload of saves on my linux partition somewhere
16:52:18  <LordAro> mostly from the 0.4 era
16:52:21  <LordAro> or even TTD saves
16:52:35  <Timberwolf> There are two TTDPatch ones in there from ~2002.
16:52:36  <TrueBrain> 0.4 and older are welcome
16:52:54  <DorpsGek> [OpenTTD/OpenTTD] rubidium42 commented on pull request #9338: Several code refactors of the SaveLoad code https://git.io/JZOSd
16:53:55  <TrueBrain> I really want GitHub to tell when stuff is moved ..
16:55:26  <TrueBrain> I love my local git
16:55:34  <TrueBrain> struggling with station save/load code
16:55:39  <TrueBrain> well, git fetch later, and I can see it :D
16:56:08  <LordAro> what did you have your git diff settings as?
16:56:30  <TrueBrain> [diff]
16:56:30  <TrueBrain>         colorMoved = zebra
16:56:37  <TrueBrain> the most epic addition to your local git evah
16:58:44  <LordAro> :+1:
16:59:00  *** andythenorth has joined #openttd
16:59:07  * LordAro looks sadly at svn
16:59:26  <andythenorth> is there another way to look at it? :P
16:59:28  <andythenorth> oof silly andythenorth
16:59:35  <TrueBrain> BURN
16:59:43  <Rubidium> svn is good in certain circumstances
17:00:01  <Rubidium> (or at least good enough)
17:00:48  <LordAro> TrueBrain: apparently there's also 'dimmed-zebra'
17:01:23  <TrueBrain> I am good with zebra :D
17:01:31  <TrueBrain> just ... GitHub .. get it implemented ... QUICK
17:01:47  <Rubidium> think about some programming language that uses binary blobs that aren't really diffable, but are quite large. Then not having the whole history of all files might be quite beneficial if it is a long running repository with many changes. Having said that, for anything that is (easily) diffable there are definitely better VCSes
17:03:09  *** frosch123 has joined #openttd
17:03:30  <TrueBrain> Timberwolf: those two SV1s still load fine :D That is cool :)
17:04:05  <TrueBrain> missing several GRFs, but they load fine, so what-ever :)
17:06:03  <Timberwolf> Oh yes, I did think that would have been from the GRF era, the three-unit Dash was a bit sus.
17:07:43  <TrueBrain> right, some ottdc public games still crash
17:07:48  <TrueBrain> most likely because I cannot find the right grfs
17:07:53  <V453000> g_g
17:07:55  <TrueBrain> someone happen to have grfpack 7.0 on their disk? :D
17:08:06  <V453000> that must be some old sheit
17:09:03  <V453000> I don't think I ever had 7.0 in my time
17:09:05  <V453000> I'd say don't bother
17:09:13  <V453000> some past is just too far
17:09:28  <TrueBrain> are you now telling me not to bother with something I want to do? :D
17:10:16  <TrueBrain> hmm .. OpenGFX NewGUI 0.2 .. what are the odds of me finding that
17:10:23  <V453000> alright :)
17:10:36  <V453000> enjoy :P
17:11:35  <TrueBrain> it is too bad the old grfpacks are not in bundles.ottdc anymore
17:11:36  <TrueBrain> meh
17:11:49  <V453000> I've had a look too, yeah
17:11:54  <V453000> only version 8 seems to be there
17:14:15  <TrueBrain> SVN also is no longer active :P
17:14:20  <TrueBrain> or at least, the WebDAV
17:16:18  <TrueBrain> and mercurial needs authentication, which I do not have :P
17:16:53  <V453000> everything ded
17:17:04  <V453000> world burning
17:17:30  <TrueBrain> you should give andythenorth a hand
17:17:36  <V453000> or a head
17:17:46  <TrueBrain> I would think: I am trying to dig up something that is 13 years old .. not realistic to still expect it to be available :)
17:18:00  <V453000> yeah, it's all been so long :)
17:18:17  <frosch123> TrueBrain: yes, i have all of them
17:18:45  <frosch123> but at some point i ran that tool that deletes duplicate files
17:18:53  <frosch123> to save on newgrf scanning time
17:19:01  <TrueBrain> that is more than fine
17:19:05  <TrueBrain> I need the unique files anyway :)
17:19:10  <frosch123> so while i have all the files, not all of them are in the pack 7.0 folder :)
17:25:22  <TrueBrain> right, got a copy of the repo :D
17:26:43  <glx> seems I have ottc-7.1 .2 and .3
17:27:18  <frosch123> 200MB tar.xz :p
17:27:29  <frosch123> where do i put this shit?
17:27:50  <Rubidium> shithub?
17:27:51  <TrueBrain> sorry, already got a copy of all grfpacks :D
17:27:54  <TrueBrain> but tnx anyway :)
17:28:17  <TrueBrain> 290 grfs, total of 93MB
17:28:50  <glx> haha and I even have a duplicate 7.3
17:28:51  <frosch123> yeah, i did not just tar the cooppacks
17:28:56  <frosch123> but all the other random stuff
17:28:58  <TrueBrain> :D
17:29:01  <frosch123> probably 3 snapshots of yeti
17:29:05  <TrueBrain> well .. zip it and put it on a GitHub issue? :P
17:29:29  <TrueBrain> or what I do: create a gist with a file named ending on .md
17:29:31  <Rubidium> doesn't IRC have file sending capabilities?
17:29:35  <TrueBrain> and after that you can drag there too :P
17:29:39  <TrueBrain> my client doesn't have that Rubidium :D
17:30:02  <Rubidium> sneakernet then
17:30:24  <TrueBrain> or: https://filetransfer.kpn.com/welcome/intro
17:31:05  <DorpsGek> [OpenTTD/team] frosch123 commented on issue #226: [es_ES] Translator access request https://git.io/JZvUR
17:31:26  <glx> oh I even have a fuzzle.org folder
17:32:48  <TrueBrain> in the higher-numbers I now have most savegame versions .. in the lower numbers I am lacking
17:32:55  <TrueBrain> 5 .. 17, for example
17:33:09  <TrueBrain> maybe I should make a website for this .. :P
17:33:45  <TrueBrain> Timberwolf: can I publish the savegames you send, or rather not?
17:34:09  <frosch123> didn't our lord start something like that?
17:34:13  <TrueBrain> he did
17:34:16  <TrueBrain> I extended it a bit already
17:34:21  <TrueBrain> and now .. it is becoming large :P
17:34:33  <frosch123> exceeded gh limit? :p
17:34:42  <TrueBrain> for sure
17:34:46  <TrueBrain> 517 NewGRFs :D
17:35:35  <frosch123> so we need a private repo with lfs and all the non-free grfs and savegames? :p
17:35:40  <glx> only 517 ?
17:36:33  <TrueBrain> glx: -required- for the set of savegames I have :P
17:36:37  <glx> last_newgrf_count = 1314 <-- but maybe it includes duplicates in the total
17:46:09  <TrueBrain> okay, 1 of the public games I can load, but time doesn't happen .. it seems to get stuck on something :P
18:01:25  <TrueBrain> holy crap, "-dgrf=1" is noisy
18:02:17  <TrueBrain> dbg: [grf] The file GM-TTO.CAT specified in baseset/orig_tto.obm is missing
18:02:19  <TrueBrain> so many times
18:02:20  <TrueBrain> lol
18:05:28  <TrueBrain> over 150 lines of code broadcast on grf=1
18:07:12  <TrueBrain> we might want to split the facility into "logs for NewGRF developers" and "logs for OpenTTD developers" :)
18:13:13  <frosch123> the former does not exist
18:13:24  <frosch123> no newgrf developer understands the output from ottd
18:13:55  <TrueBrain> okay, lets make it: we might want to split the facility into "logs for NewGRF OpenTTD developers" and "logs for sane OpenTTD developers" :)
18:13:56  <frosch123> and for the latter, usually it's easier to set a breakpoint :p
18:13:56  <TrueBrain> :P
18:14:12  <frosch123> so, not sure whether anyone notices, if you delete them all :p
18:14:14  <TrueBrain> currently I cannot see when it loads replacements GRFs between the shitstorm of other messages :)
18:14:23  <TrueBrain> haha, well, that works too :P
18:18:04  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain opened pull request #9340: Fix: compatible NewGRFs in crash-log reported wrong md5 https://git.io/JZ3v7
18:20:00  *** gelignite has quit IRC
18:22:11  <TrueBrain> frosch123: I found a LZO savegame
18:22:19  <TrueBrain> I cannot load it in Python :P
18:23:05  *** Flygon has quit IRC
18:23:51  <frosch123> haha :)
18:25:45  <DorpsGek> [OpenTTD/OpenTTD] frosch123 approved pull request #9340: Fix: compatible NewGRFs in crash-log reported wrong md5 https://git.io/JZ3Js
18:26:32  <TrueBrain> did you appreciate my motivation frosch123 ? :D
18:26:37  <frosch123> TrueBrain: try to upload it to bananas, and file a bug report, if it doesn't work :p
18:27:14  <frosch123> your what?
18:27:19  <TrueBrain> #9340!
18:27:26  <TrueBrain> didn't you read my motivation? :(
18:27:55  <frosch123> ah that, already forgot it again :)
18:28:04  <TrueBrain> tssk
18:28:20  <frosch123> something something waisted a lot of time something something :)
18:28:26  <TrueBrain> :D
18:28:28  <frosch123> -i
18:30:48  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain merged pull request #9340: Fix: compatible NewGRFs in crash-log reported wrong md5 https://git.io/JZ3v7
18:31:40  *** norri has joined #openttd
18:34:20  <norri> I was looking for some way of letting trains control the curve speed limit after this commit made it (more or less) impossible https://github.com/OpenTTD/OpenTTD/commit/0f91cb04791b329c5e457588a39d1e7a43e43136
18:34:50  <norri> I gave it a shot and came up with this: https://github.com/vituscze/OpenTTD/commit/6e22762af9eb68f7c6254560cfa01ed728100516
18:35:11  <norri> bofore I open a PR, I'd like to get some feedback on whether this is actually useful or even the right way of doing a new NewGRF feature
18:36:27  <V453000> hello :) It's incredibly useful IMO, it allows doing stuff that isn't possible anymore, and doing int in a much more sensible, and even more powerful way. Big bonus points for allowing negative numbers, too high curve speeds has been something that NUTS was quite limited by, I'd for sure use this.
18:36:39  <V453000> frosch123: what do you think? :)
18:36:53  <frosch123> why did you login twice to irc?
18:36:57  <V453000> LOL
18:36:57  <V453000> XD
18:37:24  <V453000> you know way too well how much I can't code :)
18:37:33  <LordAro> TrueBrain: https://lordaro.co.uk/~/lordaro/lordarosave.tar.gz lots of duplicates, but i figure you're better placed to work out what's interesting :p
18:37:40  <LordAro> ~90MB
18:37:58  <TrueBrain> cheers! Can I publish those?
18:38:08  <norri> we're different people, I swear
18:38:13  <TrueBrain> "othersave/stable save/arg!!!!!!!!!!!.sav" :)
18:38:29  <LordAro> TrueBrain: probably
18:38:38  <LordAro> also, i have almost no memory of what any of them are
18:38:55  <TrueBrain> will check it out, tnx
18:39:02  <TrueBrain> now I need to find a way to unpause the game on start ..
18:39:18  *** iSoSyS has joined #openttd
18:41:34  <LordAro> if (paused) unpause();
18:41:47  *** iSoSyS has quit IRC
18:41:49  <TrueBrain> I rather not modify the source for this :P
18:41:54  <TrueBrain> I am trying console scripting :D
18:42:47  <LordAro> :o
18:43:15  <TrueBrain> unpause only works in network games
18:43:16  <TrueBrain> REALLY?!
18:43:17  <TrueBrain> wth
18:43:29  <TrueBrain> why? :P
18:45:02  <TrueBrain> after loading I run each game for 256 ticks, to ensure it was not just the save that loaded in fine, but also that nothing is horribly wrong :)
18:45:13  <TrueBrain> but .. ofc that means pause needs to be off :P
18:47:03  <frosch123> norri: diff looks fine, but how did you come up with this weird unit? normalized to 19?
18:47:20  <V453000> it allows negative bonus of up to -95%
18:47:25  <V453000> 5% per step * 19
18:47:34  <frosch123> usually we do fractional numbers with a fixed number of fractional bits
18:47:46  <norri> well, I thought the tilt's 20% was too coarse and I wanted some smaller unit
18:47:49  <norri> so I went with 5%
18:47:53  <frosch123> so property would be signed 16 bit with 8 fractional part or so
18:47:58  <norri> originally it was 0% to 5*255%
18:48:06  <norri> but v453000 asked me to include negative mods
18:48:21  <norri> but yeah, I can change it
18:48:39  <frosch123> callback would return 15 bit integer, which means up to 6400%
18:50:10  <frosch123> the result in GetCurveSpeedLimit may need some clamping though
18:50:34  <frosch123> i think there is a maximum speed that trains can do
18:51:47  <norri> alright, I'll look into it
18:52:44  <V453000> I'll just say, can we pleeeeease keep the negative numbers? It would be SO useful for me, I don't know if you're considering to remove it, but that's a really significant feature
18:53:09  * andythenorth has ideas for it
18:53:19  <andythenorth> logging truck with permanently locked diffs
18:53:23  <andythenorth> slower in curves :P
18:53:28  <norri> yeah, I plan to
18:53:45  <andythenorth> V453000 rocket trains?
18:53:53  <andythenorth> fast in a straight line, can't turn
18:53:54  <andythenorth> rocket slug
18:54:00  <frosch123> V453000: negative numbers are fine, but also make them negative in the nfo :) 0 should be 0, -1 should be -1
18:54:15  <V453000> Yeah, sure, default should be 0 with 0 bonus
18:55:53  <V453000> For NUTS, I'd set some negative value on each fast train. Basically, now it works is that fast class is always the poorly acelerating one. Which basically means that eventually once you get a lot of traffic, it stops being viable. It would be a lot more interesting if the fast class could really strictly require some curve lengths, while getting a lot more power. As building longer curves gets more difficult as you expand your network, this wou
18:56:44  <andythenorth> gr8 thread https://steamcommunity.com/app/1536610/discussions/0/3113656428095596209/
18:58:41  <LordAro> proper popcorn fodder
18:58:51  <TrueBrain> pff, I already linked that a few days ago :P
19:01:04  <andythenorth> I was inattentive
19:01:10  <andythenorth> please accept my apologies
19:01:14  <andythenorth> profoundly
19:01:17  * andythenorth /sarcasm
19:01:22  <TrueBrain> wait, what?
19:01:25  <TrueBrain> :'(
19:02:52  * andythenorth goes it copy lomo?
19:02:58  <andythenorth> lomo FIRS
19:03:29  <peter1138> Is it tea time?
19:03:38  <LordAro> yes
19:03:46  <peter1138> Alright.
19:03:48  <TrueBrain> andythenorth: is being a meany, so yes
19:03:55  <peter1138> It's hard work cycling that slow.
19:04:04  <LordAro> oof
19:04:04  <peter1138> Test of balance :p
19:04:38  <andythenorth> TrueBrain I will go play Tanks
19:04:41  <TrueBrain> no
19:04:44  <andythenorth> oh
19:04:44  <TrueBrain> go play with V453000
19:04:45  <TrueBrain> :)
19:06:28  <V453000> V is reverting 15 railtypes in nuts in hopes the curve speed thingy gets merged eventually :P
19:06:52  <TrueBrain> no pressure :P
19:08:16  <TrueBrain> frosch123: soon I need a way for OpenTTD to convert a savegame from an old format to a new :P
19:08:34  <norri> frosh123: is this what you meant? https://github.com/vituscze/OpenTTD/commit/86fe82e68c3f02058f97110cebf53ce8aa663ecc
19:09:31  <norri> not sure if the value should be clamped between 0 and absolute_max_speed or between 2 and absolute_max_speed
19:09:44  <norri> 0 km-ish/h speed limit doesn't make much sense though
19:10:49  <TrueBrain> FREEZE! I'm ma baker, put yours hands in the air and give me all your money
19:11:57  <TrueBrain> I now have 445 savegames which require 567 NewGRFs to work .. around 20 games have no NewGRF .. "worst" game has 57 ..
19:13:07  <andythenorth> what's the limit?
19:13:14  <TrueBrain> 62? Always forget
19:13:16  <andythenorth> apparently the limit is far too low
19:13:16  <TrueBrain> soon 255 :P
19:13:20  <andythenorth> you wat now
19:13:23  <TrueBrain> those 445 games span 115 different savegame version
19:13:23  <andythenorth> outrage
19:13:38  <TrueBrain> and 29 different map-sizes
19:13:39  <andythenorth> can we bring back musa?
19:13:47  <TrueBrain> I can sing for you?
19:13:49  * andythenorth considering 1 grf per vehicle
19:13:52  <TrueBrain> also a bit of mussssic
19:13:58  <andythenorth> but uploading through web....brrrr
19:14:06  <TrueBrain> I consider a pay-after-your-10th-upload subscription on BaNaNaS
19:14:17  <andythenorth> charge the creators?
19:14:18  <andythenorth> interesting model
19:14:23  <andythenorth> how much is your SDK?
19:14:33  <andythenorth> and is an engineering support package available?
19:14:46  <TrueBrain> for 150 euro an hour, you can get support
19:14:49  <TrueBrain> which is really cheap, considering
19:15:16  <andythenorth> is there an SLA / some kind of warranty?
19:15:32  <TrueBrain> pay-as-you-go
19:15:35  <TrueBrain> and no
19:15:38  <andythenorth> can I buy my spec done, at a fixed fee?
19:15:51  <andythenorth> lol fixed spec, fixed fee, fixed deadline, I actually used to sell that
19:15:51  <TrueBrain> fixed fee is possible, but you won't like the number
19:15:52  <andythenorth> lol
19:16:32  <TrueBrain> it takes 13 minutes to analyze those 445 savegames
19:16:42  <TrueBrain> where it simulates 256 ticks (so all tiles cycled at least once)
19:16:47  <norri> frosch123: whoops, typo'd your name; commit is above
19:17:01  <TrueBrain> 8 out of those fail loading with current master
19:17:11  <TrueBrain> most likely as I don't have the right NewGRFs :)
19:19:56  <LordAro> TrueBrain: how many of mine are useful?
19:20:06  <TrueBrain> haven't looked at those, sorry :)
19:20:10  <LordAro> :<
19:22:49  <DorpsGek> [OpenTTD/OpenTTD] DorpsGek pushed 1 commits to master https://git.io/JZ3Oz
19:22:50  <DorpsGek>   - Update: Translations from eints (by translators)
19:24:05  <Rubidium> TrueBrain: probably the 57 is the actual limit, the rest are the base GRFs
19:27:23  <TrueBrain> LordAro: some interesting savegames in there, version-wise
19:27:33  <frosch123> norri: i even highlight on that spelling :p
19:29:11  *** gelignite has joined #openttd
19:31:35  <TrueBrain> LordAro: keeping 16 of those savegames :)
19:31:54  <LordAro> :)
19:35:43  *** jottyfan has joined #openttd
19:36:28  <frosch123> norri: https://github.com/frosch123/OpenTTD/commit/7eb09bbef64f6b087e1ecd343871ca7c40ee0802 <- i think you need something like that
19:40:56  <norri> oh, does the callback only return 15 bits?
19:44:55  <norri> frosch123: alright, should have paid more attention when reading the spec... should I add that to my commit or do you want to do it separately?
19:45:08  <norri> well, my PR
19:45:12  <norri> probably two commits at that point
19:46:43  <frosch123> add it to your PR please
19:46:59  <norri> alright, anything else?
19:47:05  <frosch123> maybe the "if" in GetEngineProperty can be cascaded differently
19:47:16  <frosch123> norri: get V to test it :)
19:47:51  <V453000> V is preparing NUTS to work with it already, I'm just finishing the railtype revert, I'll add the values in soon
19:48:30  <norri> is there something I need to do to support the 0x36 callback in NML?
19:48:42  <peter1138> What is that commit fixing?
19:48:47  <peter1138> (Or is it adding...)
19:49:01  <frosch123> peter1138: adding
19:49:34  <glx> <norri> is there something I need to do to support the 0x36 callback in NML? <-- you can look how other variables are supported
19:49:56  <glx> I think there's some dict somewhere
19:49:58  <frosch123> action3_callbacks.py:60 etc
19:52:25  <TrueBrain> guess nobody has ottdc grfpack 6beta? :P
19:52:31  <TrueBrain> missing 2 files .. :D
19:52:47  <norri> alright, found it... do I need the purchase stuff? not sure what that's for
19:52:59  <glx> I don't have this pack, but maybe I have the files
19:53:20  <TrueBrain> if you only have newer packs, that is doubtful glx :)
19:53:29  <norri> https://github.com/OpenTTD/nml/blob/09de5232a501df0f5fe4a934b90a06ba708b1043/nml/actions/action3_callbacks.py#L68
19:53:31  <norri> this thing
19:53:39  <glx> oh I have many individual grfs too
19:53:42  <Rubidium> TrueBrain: those weren't in the repository I gave you?
19:53:47  <TrueBrain> Rubidium: nope, starts at 6
19:53:50  <TrueBrain> 6beta is not in there
19:54:12  <TrueBrain> euh .. now I have to remember how I start a PM to you glx :P
19:54:16  <TrueBrain> I can never remember that with this client
19:54:50  <TrueBrain> found it!
19:55:09  <frosch123> norri: no, no curve speed in the purchase list :)
19:55:11  <LordAro> TrueBrain: was there anything interesting about my saves except for their versions?
19:55:20  <norri> alright cool
19:55:26  <TrueBrain> I haven't looked at them visually, if that is what you mean :)
19:55:27  <norri> I'll be back!
19:55:30  <TrueBrain> why you ask?
19:57:01  <LordAro> just curious what was interesting about the 16 saves :p
19:57:19  <TrueBrain> savegame version :)
19:58:10  <LordAro> suitably old?
19:58:14  <TrueBrain> yes
19:58:15  <TrueBrain> very :)
19:59:19  <peter1138> Like me.
20:02:31  <TrueBrain> LordAro: after the titlegame, yours is now the oldest I have :)
20:04:28  <LordAro> :D
20:08:53  <Timberwolf> I feel the urge to go spelunking. I found some that are from an Asus Eee PC, but those will still be ~2010 or so.
20:12:53  <Timberwolf> Aw yeah, I have a disk full of "Date modified: 2007"!
20:14:39  <TrueBrain> do not change bash files while it is executing
20:14:42  <TrueBrain> that is not going as you expect
20:15:16  <TrueBrain> hmm .. I should also still have my OpenTTD folders on archive somewhere .. see if I can find some old savegames .. well, not tonight :)
20:17:41  <TrueBrain> peter1138: do you happen to have older versions of ukwaypointsw.grf? :)
20:17:46  <TrueBrain> the grf is credited to you ... ;)
20:21:20  <LordAro> TrueBrain: which one is oldest?
20:21:49  <TrueBrain> LordAro: sorry? which one what?
20:21:53  *** nielsm has quit IRC
20:22:20  <peter1138> Gods
20:22:26  <peter1138> Did I make that?
20:22:31  <LordAro> TrueBrain: which of my saves
20:22:35  <TrueBrain> the websites claims so peter1138 :)
20:22:46  <TrueBrain> LordAro:   - Magnetic_Corp.,_30th_Dec_2090_CLI.sav
20:22:46  <peter1138> Only the coding of course, not actual drawing.
20:23:02  <TrueBrain> gave me a hope you might have a copy of the md5 I am looking for :D
20:23:13  <LordAro> interesting
20:23:22  <LordAro> that doesn't seem like one of mine
20:23:37  <V453000> get this: I actually started a branch in my hobby project
20:23:40  <V453000> that's a first
20:23:57  <TrueBrain> LordAro: awh :(
20:24:19  <LordAro> (as in, downloaded from somewhere else)
20:24:23  <peter1138> I guess I could search for it...
20:24:38  <LordAro> i imagine orudge could help you download all the .sav attachments from tt-forums :p
20:24:46  <TrueBrain> peter1138: only if you want to. It doesn't have to :)
20:25:29  <TrueBrain> peter1138: in case you are, I am looking for the version of which the md5 starts with "f0bf75" :)
20:26:24  <peter1138> Damn, the GRFID is my traditional PN..
20:26:37  <peter1138> PN 00 01. Worry.
20:26:57  <Timberwolf> OK, this is interesting - I have a bunch of saves from at latest 2007, which clearly had newgrfs installed but don't find the GRFIDs when loading.
20:27:35  <peter1138> TrueBrain, I won't have any other versions though.
20:27:45  <TrueBrain> in general, anyone who happen to have any of these on their machine, I would love a copy: https://gist.github.com/TrueBrain/624866655e843c3b84b5a614dbc3cabb :)
20:27:45  <peter1138> If I have any,.
20:27:48  <DorpsGek> [OpenTTD/nml] frosch123 opened issue #221: Support industry cargo subtype display without waiting amount https://git.io/JZ38r
20:27:49  <TrueBrain> peter1138: too bad :)
20:28:02  <glx> maybe the 2 firs can be rebuilt
20:28:19  <TrueBrain> all those GRFs are REALLY old, so don't bother if you have nothing from before ... 2012 or so?
20:28:56  <peter1138> Just found a load of files from 2005 though.
20:29:20  <TrueBrain> but, just to point out the insanity here: I can load games from years ago just fine, all the way up till master
20:29:30  <peter1138> 996a2ea20be9caec51197d4e20dd63c1  pb_viaduct.grf
20:29:33  <TrueBrain> we really are still backwards compatible for most part :P
20:29:45  <TrueBrain> (like 4 savegames crash .. need to investigate)
20:30:04  <TrueBrain> peter1138: on the list!
20:30:16  <peter1138> I've got newwater_269.grf, so even older...
20:30:25  <TrueBrain> so zip it, and drag it as comment on the gist ;)
20:30:39  <TrueBrain> (you don't have to submit the comment; you can also PM me the link
20:32:13  <peter1138> OpenTTD r3234M (24.11.05)
20:32:15  <peter1138> Nice.
20:33:37  *** jottyfan has quit IRC
20:34:52  <peter1138> gist does not support zip :(
20:35:16  <andythenorth> goes it refactor FIRS more?
20:35:18  <andythenorth> probably
20:35:28  <andythenorth> got up at 05.30 so this will go well
20:35:34  <TrueBrain> peter1138: try again :)
20:35:46  <TrueBrain> I think I have to make the file a markdown before it works
20:36:30  <andythenorth> did we document nml procedures with parameters?
20:36:34  * andythenorth forgot how to do them :P
20:37:17  <glx> if you never wrote the doc there's no doc :)
20:38:10  <andythenorth> I never wrote the doc
20:38:22  <peter1138> custombridgeheads17.diff.gz
20:38:23  <peter1138> Nice.
20:38:28  <andythenorth> one year old :( https://github.com/OpenTTD/nml/issues/105
20:38:31  <andythenorth> bad andythenorth probably
20:38:39  <peter1138> diagstations3.diff.gz
20:38:48  <peter1138> Well I still have this apparently...
20:38:53  <glx> but the format is easy, instead of 'switch(..., name, exrpression)' it's 'switch(..., name, p1, [...], expression)
20:38:57  <frosch123> andythenorth: do you use the input/output cargo information in cargo_subtype_display?
20:39:11  <frosch123> there is a disagreement between ottd implementation and newgrf spec
20:39:13  <andythenorth> not sure
20:39:46  <andythenorth> https://github.com/andythenorth/firs/blob/master/src/templates/extra_text_secondary.pynml#L40
20:39:52  <LordAro> peter1138: try applying them :p
20:40:36  <andythenorth> frosch123 do you want some compiled nml?
20:40:44  <frosch123> andythenorth: getbits(extra_callback_info2, 16, 23) <- that 23 should be a 8 :p
20:41:00  <andythenorth> wonder why it's a 23
20:41:10  <frosch123> 23 = 16 + 8 - 7
20:41:18  <frosch123> 23 = 16 + 8 - 1
20:41:23  <frosch123> close :p
20:41:45  * andythenorth wonders what breaks :)
20:41:51  <andythenorth> I never defined anything with 8 cargos yet
20:42:25  <glx> ah yes it's not (begin, end), it's (begin, count) ;)
20:42:57  <andythenorth> oof
20:43:08  <andythenorth> I should fix it :)
20:43:21  <andythenorth> but I just ran out of brain
20:43:25  <andythenorth> 05.30 seems a long time ago
20:43:44  <peter1138> Jesus. I've found my old newgrf bridges patch...
20:43:54  <TrueBrain> sorry for this trip down memory lane :P
20:44:12  <andythenorth> bridges over stations?
20:44:30  * andythenorth thinks peter1138 might be JGR in disguise
20:45:10  <peter1138> Includes an early version of visible bounding-boxes...
20:45:32  <peter1138> image = b[12 + (ti->map5&0x01)]
20:45:42  <peter1138> Back when map access looked like that.
20:45:52  <TrueBrain> pretty :)
20:46:13  <glx> oh from before tfc_newmap
20:46:58  <TrueBrain> owh, right, tomorrow is a big day for OpenTTD ..
20:47:03  <glx> gog ?
20:47:03  <TrueBrain> I should prepare a news-post ..
20:47:16  <LordAro> :o
20:47:41  <TrueBrain> guess I do that tomorrow :P
20:48:24  <TrueBrain> https://gist.github.com/TrueBrain/624866655e843c3b84b5a614dbc3cabb <- down to 7 missing :)
20:49:53  <norri> frosch123: alright, gave modified openttd and nml to V453000 for testing, we'll get back on that's done
20:50:02  <norri> thank you for the help
20:53:04  <TrueBrain> funny ... after adding the grf I just got, some games started to crash :D
20:53:11  <TrueBrain> by having the right grf, the wrong thing happens :P
20:53:24  <peter1138> Nice.
20:54:08  *** Samu has quit IRC
20:54:13  <TrueBrain> hmm .. one has an invalid chunk .. lets try 1.11.2 to be sure (I was trying master)
20:54:56  <frosch123> maybe it's from miniin or chillpp :p
20:55:27  <V453000> <3 frosch
20:55:32  <glx> oh miniin was fun
20:55:35  <TrueBrain> it is one of the public games from ottdc .. I think not :D
20:55:50  <glx> and a pain to sync
20:57:10  <frosch123> TrueBrain: which one? sometimes they wrote down which version they played
20:57:36  <V453000> iirc all of the archived games have a version noted
20:57:36  <TrueBrain> PSG39
20:57:49  <TrueBrain> okay, also fails on 1.11.2, good
20:57:51  <V453000> r9769
20:58:32  <glx> oh a nightly then, can have bugs
20:59:02  <TrueBrain> but I can't remember it failed before I added a certain NewGRF .. let me check
20:59:49  <TrueBrain> no, it seems to always fail
21:00:05  <frosch123> TrueBrain: at least there is special code for that revision range. there was something broken back then
21:00:45  <frosch123> https://github.com/OpenTTD/OpenTTD/blob/master/src/saveload/afterload.cpp#L1607 <- i mean, when do you do "before 57" else if "before 59"
21:00:50  <frosch123> only when 58 was broken, right?
21:01:37  <TrueBrain> comment is funny
21:01:50  <TrueBrain> okay, here I have a game I can load without newgrfs, but I cannot load with the newgrfs I have :D
21:01:54  <TrueBrain> that is priceless
21:02:36  <TrueBrain> PSG02 also doesn't load, but with this: OpenTTD/src/core/pool_type.hpp: index < this->first_unused
21:03:09  <TrueBrain> I need to download older OpenTTDs to see if it is a recent issue for these kind of games :)
21:05:10  *** EER_ has quit IRC
21:06:19  <glx> trace helps a lot with this assert
21:07:54  <TrueBrain> I doubt that :P I think that requires a whole different set of debugging tbh :D
21:08:39  <glx> at least the trace tells which pool :)
21:09:31  <TrueBrain>  [10] releases/1.11.2-assert/openttd(_Z13GetObjectTypej+0x30) [0x5635f77cdf20]
21:09:32  <TrueBrain> :D
21:10:36  <glx> ah in MSVC I usually get the pool type
21:11:05  <TrueBrain> #3  0x0000555555833ee9 in Pool<Vehicle, unsigned int, 512ul, 1044480ul, (PoolType)1, false, true>::Get (
21:11:05  <TrueBrain>     index=<optimized out>, this=<optimized out>, this=<optimized out>, this=<optimized out>, index=<optimized out>)
21:11:07  <TrueBrain> type 1 :P
21:11:19  <glx> vehicle
21:11:21  <TrueBrain> but again, this is just how it presents the problem .. I think the real problem is a few layers deeper :)
21:11:57  <TrueBrain> and possibly it is a regression, as this is afterload stuff
21:12:08  <TrueBrain> but .. something for another day :)
21:13:16  <TrueBrain> now I can first continue my SaveLoad changes, without being scared I completely break backwards compatibility :)
21:19:43  *** Funke has quit IRC
21:21:35  *** Strom has joined #openttd
21:22:17  <TrueBrain> okay, so now I am sure: there is no regression with 500 savegames from 1.11.2 to my PR :) Nice .. now it is time to break stations again :P
21:39:08  *** andythenorth has quit IRC
21:57:45  *** sla_ro|master has quit IRC
22:04:20  *** gelignite has quit IRC
22:06:10  *** frosch123 has quit IRC
22:34:55  *** norri has quit IRC
22:37:46  *** HerzogDeXtEr has quit IRC
22:45:24  *** Progman has quit IRC
22:56:58  *** Compu has joined #openttd
23:31:54  *** tokai has joined #openttd
23:31:54  *** ChanServ sets mode: +v tokai
23:38:48  *** tokai|noir has quit IRC

Powered by YARRSTE version: svn-trunk