IRC logs for #openttd on OFTC at 2019-05-10
⏴ go to previous day
03:37:52 <DorpsGek_II> [OpenTTD/OpenTTD] stale[bot] commented on pull request #7339: Fix #7311: External configuration file does not change directories https://git.io/fjCWe
05:13:38 *** Flygon_ has joined #openttd
05:22:03 <DorpsGek_II> [OpenTTD/OpenTTD] glx22 opened pull request #7578: Fix: ignore C++ standard library headers in depend https://git.io/fjClf
07:20:19 *** Supercheese has joined #openttd
07:44:30 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN opened pull request #7579: Fix #7576: Crash when destroying rail/tram crossing. https://git.io/fjC8O
08:16:44 *** Supercheese has joined #openttd
08:38:55 <Eddi|zuHause> hm, youtube decided to finally remove the unsubscribed videos from my feed... took only like 2 days to trickle down some databases?
08:55:09 <DorpsGek_II> [OpenTTD/OpenTTD] LordAro approved pull request #7578: Fix: ignore C++ standard library headers in depend https://git.io/fjC8x
09:38:49 *** kiwitree has joined #openttd
09:52:15 *** andythenorth has joined #openttd
10:31:45 <andythenorth> when is lunch peter1138 ?
11:19:36 <Eddi|zuHause> great visual studio... i have to replace a bunch of strings of the form "%QX1234.5" with "%Q*" and it won't let me use regex...
11:34:36 *** andythenorth has joined #openttd
11:54:28 <LordAro> Eddi|zuHause: this is where you exit VS and use a proper editor
13:02:02 *** alekseiytalanov has joined #openttd
13:02:57 *** alekseiytalanov has left #openttd
13:35:53 *** supermop_work has joined #openttd
14:22:10 <DorpsGek_II> [OpenTTD/OpenTTD] glx22 merged pull request #7578: Fix: ignore C++ standard library headers in depend https://git.io/fjClf
14:37:51 <DorpsGek_II> [OpenTTD/OpenTTD] stale[bot] commented on pull request #7490: Add: Press ctrl to build diagonal rivers in scenario editor. https://git.io/fjCgR
14:41:25 *** kiwitree has joined #openttd
14:55:41 <DorpsGek_II> [OpenTTD/OpenTTD] btzy opened issue #7580: Feature Request: When loading a saved game, the file picker should remember the last used directory https://git.io/fjCgd
15:05:24 *** sla_ro|master has joined #openttd
15:28:57 <andythenorth> I forgot to add an emoji
15:29:02 <andythenorth> sounds like station ratings
16:16:32 <nielsm> ah, new posts in the macos slowness thread
16:16:45 <nielsm> conclusion is (still), it's not a cause we are measuring
16:17:18 <nielsm> this will probably need some TIC/TOC measurements all around the quartz driver
16:28:08 *** Wormnest has joined #openttd
17:16:33 *** Progman has joined #openttd
17:25:34 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN commented on pull request #7573: Fix #7561: Remove assumption between power and cost https://git.io/fjCrd
17:50:29 *** FoxsideMiners has joined #openttd
18:16:21 *** HerzogDeXtEr has joined #openttd
19:22:34 *** gelignite has joined #openttd
19:25:39 <DorpsGek_II> [OpenTTD/OpenTTD] stormcone opened pull request #7581: Fix #7224: Drag and drop vehicle group creation does not work correctly. https://git.io/fjC6Z
19:36:05 <DorpsGek_II> [OpenTTD/OpenTTD] nielsmh approved pull request #7581: Fix #7224: Drag and drop vehicle group creation does not work correctly. https://git.io/fjC62
19:45:37 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN commented on pull request #7581: Fix #7224: Drag and drop vehicle group creation does not work correctly. https://git.io/fjC61
19:49:55 <DorpsGek_II> [OpenTTD/OpenTTD] nielsmh merged pull request #7581: Fix #7224: Drag and drop vehicle group creation does not work correctly. https://git.io/fjC6Z
19:59:02 <V453000> I did some testing, and holy shit
19:59:24 <V453000> new nuts maglev, 25fps, new nuts rail strong class 33fps, old nuts rail strong class 45fps
19:59:52 <V453000> that's 45fps -> 33fps probably mostly because of the shitload of graphics switches
20:01:00 <V453000> I believed it would not be so bad tbh, the graphics master switches are big, but in this case they end with 4th entry at most, I guess they should not go through all the possible outcomes then?
20:03:57 <V453000> there are two giant switches like switch_rail_freight_graphics_front , but vast majority of those outcomes are quite rare. most of the time it should just end up at something like 0..8 (that's 4 engines per train) ... in the save I have 4 units ... is my understanding correct that if a switch finds the correct value soon, it does not go through the rest? I remember frosch telling me something like that at some point
20:29:55 <DorpsGek_II> [OpenTTD/OpenTTD] stormcone opened pull request #7582: Use shitf + click to show vehicles' group. https://git.io/fjCid
20:31:02 <peter1138> V453000, I have no idea what all those long lists of switches are meant to achieve, but I'm guessing it could probably be simplified.
20:55:25 *** frosch123 has joined #openttd
20:56:32 <V453000> peter1138: 1. check if the last unit in the train is an engine, 2. count how many engine veh_ids are in the train 3. draw that amount of engines at the front of the train, or if 1. was an engine, draw them from the front and end in alternating fashion
20:57:38 <V453000> plus if the engine draws as a wagon, it checks if the vehicle is stopped, depotted, and if it's not at the ends of the train, and if it is an engine_id on self, then it draws an overlay through layers
20:58:04 <V453000> I don't know how to simplify it really and keep functionality
20:58:14 <peter1138> Why do you have engines drawing as wagons?
20:58:24 <V453000> the only idea I can think about is massive gutting, NUTS 2.0, simplify drawing to the maximum
20:58:31 <peter1138> I would have... an engine that has an engine sprite, and a wagon that has a wagon sprite.
20:58:46 <frosch123> V453000: do you really need count_veh_id(..) + ... + count_veh_id(..) ?
20:59:30 <V453000> because all engines are consistently purchased as 1-tile units - so 0.5 tile engines have 0.5 tile wagon ... so in order to get engines look like they are all in the front, it must draw them as wagons, or engines depending on situation :)
20:59:40 <V453000> frosch123: I guess not that part, was just code simplification
20:59:48 <V453000> I can have this giant switch for each vehicle separately
21:00:10 <V453000> because that's the length of dual headed and some long articulated steamers
21:00:19 <andythenorth> so instant replace
21:00:23 <andythenorth> no length change on replace
21:00:36 <V453000> because if any vehicles are shorter or longer, it means you can get lossy autoreplace and can't go back
21:00:38 <andythenorth> iz for coop games?
21:00:40 <peter1138> Seems odd to do all this stuff when the game already knows how to draw engines and wagons.
21:00:56 <andythenorth> it's a solution to lack of consist management
21:01:21 <andythenorth> FWIW there is 'template based train replacement' patch, but I didn't understand it
21:01:22 <V453000> one of the most interesting aspects of NUTS is that you get many different engine classes and you need to figure out which one to use - and the only real way to do that is to autoreplace and see how it performs
21:01:46 <V453000> which is why engines, and wagons need to always be purchased in consistent lengths
21:01:59 <V453000> 0.5 tiles for wagons is not a big deal, just means using doubled 4/8 wagons
21:02:30 <V453000> for engines I'd love to do 0.5, but I don't know how to solve it properly with some of the engines which have 16/8 graphics
21:02:48 <V453000> I guess they could just start drawing properly once you have a train longer than 0.5
21:03:45 <andythenorth> what's ultimate TL?
21:04:02 <V453000> also for example the ctrl+click for reversed vehicles is a really secret functionality, so dual heading is really, really nice to have
21:04:19 <andythenorth> so do all trains fixed consist TL5
21:04:52 <andythenorth> most important question is always 'where are the lolz?'
21:05:00 <andythenorth> I try to teach people this in WOT Blitz
21:05:11 <andythenorth> but no They Are Very Serious
21:05:46 <peter1138> Or just get someone to implement the opposite of wagon remvoal.
21:05:54 <peter1138> And remove all the pointlessness.
21:06:10 <V453000> what I'm thinking right now in a context of NUTS2 would be: 1. throw away all this and have 0.5 tile engines. Dual heading would be automatic and ctrl+clicking reverses the rear vehicle to be back to front-facing. 2. provide wagons which are for 1 cargo only. I miss that honestly, and I could then have the simplest wagon drawing possible.
21:06:25 <V453000> "get someone to do X" :D
21:06:55 <V453000> it's not pointless peter1138, the reasons might just feel obscure to you, and admittedly they are, but they all have reasoning
21:07:16 <V453000> being able to autoreplace from anywhere to anywhere is a seriously important concept for nuts
21:07:20 <peter1138> I also don't see why you need massive blocks concerning length.
21:07:35 <peter1138> Don't you only care about the half-tile fake-wagon-engine after the engine?
21:08:02 <V453000> because then the "real engine" needs to draw like a wagon otherwise in a 2-engine train there would draw 3 engines
21:09:19 <andythenorth> peter1138: I had an idea $sometime about autoreplace
21:09:31 <andythenorth> add a float box next to the vehicle you're replacing
21:09:43 <andythenorth> so Red Hopper = 1.5x Blue Hopper
21:10:00 <andythenorth> or Thomas = 0.8x Percy
21:11:21 *** tokai|noir has joined #openttd
21:11:21 *** ChanServ sets mode: +v tokai|noir
21:12:56 <peter1138> You're solving one problem by creating a horrible mess elsewhere. Oh well. It's your CPU :p
21:14:58 <V453000> As I said, I'll consider having 0.5 tile engines instead
21:15:06 <Markk> Guten abend, everyone! I've got a small question, if I run a train 600 tiles in a diagnoal line, will it make the same money as if it was running in a straight (or paralell to the map) line? With the cargo, speed, length and so on being the same.
21:15:13 <V453000> which will complicate it a little bit for some cases, but nothing this crazy
21:15:56 <nielsm> Markk: it will probably make more money as it makes the delivery faster
21:16:11 <nielsm> but the "distance delivered" is the same
21:16:22 <glx> that's the kind of solution I see everywhere regarding this GCC8 warning
21:16:38 <Markk> I'll need to make a controlled experiment
21:17:09 <glx> well the extra cast, not the way I apply it
21:18:09 <andythenorth> I don't know how to fix FIRS :|
21:18:53 <nielsm> andythenorth: do you know if FIRS uses a custom station rating calculation for its cargos?
21:19:32 <andythenorth> hmm where are FIRS docs now?
21:19:36 <andythenorth> oh they aren't :(
21:19:39 <nielsm> since one of my suspicions would be an overflow bug with vehicle age
21:20:18 <andythenorth> station rating removed in FIRS 2.0.0 and newere
21:21:17 <andythenorth> unrelated, what should I do about FIRS?
21:21:22 <andythenorth> move it back to devzone?
21:22:02 <andythenorth> hmm but bundles is broken
21:22:08 <andythenorth> so that won't helkp
21:23:41 <andythenorth> peter1138: lemon chicken, rice, chips....carolies? :o
21:24:39 <Samu> i was having so much fun with the build
21:27:06 <Samu> I can't contain myself, arg... i'm so annoyed
21:28:16 <Samu> then they wonder why there's no melee or self cast builds
21:29:00 <nielsm> grr why are savegames always loaded with a million newgrfs
21:30:17 <nielsm> and seeing many of the grfs missing have a "w" at the end they're probably really old
21:34:30 <V453000> BTW if a train set defines a wagon which can carry only a single cargo type, and that cargo is not available in the current map, is the wagon purchaseable?
21:35:13 <michi_cc> glx: What happens if you try e.g. "typedef BOOL(FAR WINAPI * PFNGETFONTRESOURCEINFO)(LPCTSTR, LPDWORD, LPVOID, DWORD);"
21:37:05 <V453000> well the alternative solution is to give all engines a 1-tile graphic
21:37:21 <V453000> no articulated wagons
21:37:56 <V453000> which would kind of look wtf with short trains, but dual headed variants would work ok
21:39:30 <andythenorth> V453000: not purchaseable if cargo not available
21:39:44 <V453000> good, iz what I thought & wanted to hear :)
21:39:54 <V453000> honestly there is some magic to buying a wagon dedicated to 1 cargo
21:40:03 <V453000> I can't really describe it, is it just nostalgia?
21:40:27 <V453000> so I'll probably add unique wagons per every single cargo, plus universal wagons
21:40:37 <V453000> questionnable whether I want to do unique wagons for FIRS cargoes
21:40:52 <V453000> plus in all the wagon generations I have
21:43:13 <glx> michi_cc: D:/developpement/GitHub/glx22/OpenTTD/src/fontcache.cpp: In function 'void LoadWin32Font(FontSize)':
21:43:13 <glx> D:/developpement/GitHub/glx22/OpenTTD/src/fontcache.cpp:985:147: warning: cast between incompatible function types from 'FARPROC' {aka 'long long int (*)()'} to 'PFNGETFONTRESOURCEINFO' {aka 'int (*)(const wchar_t*, long unsigned int*, void*, long unsigned int)'} [-Wcast-function-type]
21:43:13 <glx> static PFNGETFONTRESOURCEINFO GetFontResourceInfo = (PFNGETFONTRESOURCEINFO)GetProcAddress(GetModuleHandle(_T("Gdi32")), "GetFontResourceInfoW");
21:43:53 <glx> no change, it's really the conversion from FARPROC to something else
21:44:10 <peter1138> andythenorth, rice AND chips? Why two carbs? No veg/greens?
21:44:32 <glx> and casting to void(*)() before the actual cast removes the warning
21:45:40 <glx> the warning is correct, but annoying in our case :)
21:45:49 <glx> because we know what we are doing
21:46:49 <glx> and many projects seems to be annoyed as well ;)
21:50:43 <andythenorth> peter1138: lettuce and tomatoes
21:56:58 <nielsm> okay I can load and run that savegame despite the missing newgrfs
21:57:35 <nielsm> and now set up a tracepoint that prints the old/new rating for that cargo at that station every time it's recalculated
21:58:21 <DorpsGek_II> [OpenTTD/OpenTTD] dorobouNeko commented on pull request #7573: Fix #7561: Remove assumption between power and cost https://git.io/fjCXU
22:05:10 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN commented on pull request #7398: Fix #7371: Avoid dependency on foundations of town tile during saveload https://git.io/fjCXq
22:05:52 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN opened pull request #7583: Fix #7577: Check if linkgraph station index is valid before dereferencing. https://git.io/fjCXm
22:06:38 *** SimYouLater has joined #openttd
22:06:59 <DorpsGek_II> [OpenTTD/OpenTTD] michicc approved pull request #7579: Fix #7576: Crash when destroying rail/tram crossing. https://git.io/fjCXY
22:07:02 <SimYouLater> I apologize for getting upset over something that was not the fault of this IRC.
22:07:41 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN merged pull request #7579: Fix #7576: Crash when destroying rail/tram crossing. https://git.io/fjC8O
22:10:20 <SimYouLater> Can you please unban the IP address of my laptop? I do all my GRF development there and I promise not to make a nuisance of myself from now on.
22:12:49 <Eddi|zuHause> you know that by the standard IRC etiquette, circumventing a ban to beg for getting unbanned just results in another ban, right?
22:13:21 <SimYouLater> I had nmo way to contact to ask how long the ban was for.
22:13:32 <SimYouLater> I will leave if you want me to.
22:14:44 <SimYouLater> Also, I didn't get banned until I refused to apologize, which feels kind of dumb but I guess that's your choice.
22:15:44 <peter1138> Damned Nazi's in Doom are waaaay too fast.
22:16:09 <peter1138> SimYouLater, do you have anger management issues perhaps?
22:16:44 <SimYouLater> I stayed calm when I was told to calm down before returning.
22:17:07 <Eddi|zuHause> if you don't understand why you were banned, maybe it's better for everybody if you just stayed away
22:17:24 <SimYouLater> I can't. All the GRF development assistance is here.
22:17:37 <SimYouLater> Why was I banned?
22:18:13 <SimYouLater> All I know is that I was asked to apologize and I calmly refused before leaving.
22:18:26 <SimYouLater> When I tried to come back, I was banned.
22:18:30 <peter1138> Who cares, it was weeks ago.
22:18:41 <SimYouLater> Then why am I still banned?
22:18:53 <peter1138> Why do you even need to stay calm? What is there to be angry about all the time?
22:19:03 <SimYouLater> I was angry once.
22:19:14 <SimYouLater> I swear, it's not going to be an issue.
22:19:24 <andythenorth> peter1138 the speed of nmlc, brexit, the number of carolies in a mars bar/
22:19:34 <andythenorth> actually mostly lolz
22:19:42 <nielsm> well andythenorth, I think I found the explanation for that station rating question
22:20:13 <nielsm> it's because transferred cargo waiting at another station than its originating station counts against the rating of the originating station
22:20:27 <peter1138> I just ate almonds, that's a lot of carolies.
22:20:39 <nielsm> cargo picked up at station A, unloaded at station B, and waiting for further pickup at B, will count against the rating at A
22:20:50 <nielsm> (but not against the rating at B)
22:20:57 <andythenorth> even with cargodist?
22:21:06 <nielsm> this is manual distribution
22:21:09 <DorpsGek_II> [OpenTTD/OpenTTD] michicc approved pull request #7583: Fix #7577: Check if linkgraph station index is valid before dereferencing. https://git.io/fjCXl
22:21:12 <andythenorth> it's too prevent the dump-at-B exploit yes
22:21:16 <nielsm> I think automatic distribution does the same
22:21:31 <andythenorth> I totally forget about it when I play
22:21:41 <nielsm> I didn't know that was even a mechanic at all
22:21:47 <andythenorth> probably explains why I have terrible ratings sometimes on ship transfer routes
22:22:07 <andythenorth> are exploits worth preventing? :P
22:22:08 <DorpsGek_II> [OpenTTD/OpenTTD] michicc opened pull request #7584: Codechange: [AzurePipelines] Speed up non-published CI-builds by doing Debug Windows builds. https://git.io/fjCX8
22:22:15 <SimYouLater> So am I permabanned or is there a point in time I have to wait for?
22:22:29 *** peter1138 sets mode: -b *!~oftc-webi@d23-16-41-35.bchsia.telus.net
22:22:34 <peter1138> FFS, nobody gives a shit.
22:22:56 <peter1138> We get on with the stuff we do, we're not here to faff about with your angermanagementdramallama.
22:23:21 <nielsm> so yea that truck transfer station is his explanation
22:23:31 <nielsm> it somehow suddenly gets much more than it can handle
22:23:43 <SimYouLater> I'm sorry. All I was asking was how to regain your trust. I'll go do something else for a while.
22:30:19 <DorpsGek_II> [OpenTTD/OpenTTD] michicc commented on pull request #7584: Codechange: [AzurePipelines] Speed up non-published CI-builds by doing Debug Windows builds. https://git.io/fjCXz
22:35:40 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN approved pull request #7584: Codechange: [AzurePipelines] Speed up non-published CI-builds by doing Debug Windows builds. https://git.io/fjCXr
22:35:56 <Markk> Hmm, strange. I built a quick test track, both 300 tiles long, one is a long straight, paralell to the map edge, and the second one is the same lengthm but with the tracks being diagnoal instead.
22:36:06 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN merged pull request #7583: Fix #7577: Check if linkgraph station index is valid before dereferencing. https://git.io/fjCXm
22:36:38 <DorpsGek_II> [OpenTTD/OpenTTD] glx22 approved pull request #7584: Codechange: [AzurePipelines] Speed up non-published CI-builds by doing Debug Windows builds. https://git.io/fjCXK
22:36:48 <Markk> The straight line was the one arriving first at the destination, while the diagnoal still is 50 tile behind.
22:36:59 <DorpsGek_II> [OpenTTD/OpenTTD] michicc merged pull request #7584: Codechange: [AzurePipelines] Speed up non-published CI-builds by doing Debug Windows builds. https://git.io/fjCX8
22:37:34 <Markk> I really thought it would be the other way around!
22:43:00 <Markk> Hmm, even stranger. The diagonal train, that arrived second, earned about £278k and the straight line train earned abut £154k.
22:44:04 <Markk> Even though I don't understand why, I still have my answer.
22:57:13 *** Rubidium has joined #openttd
22:57:13 *** ChanServ sets mode: +o Rubidium
22:58:45 <FLHerne> peter1138: That seems a bit harsh, I've definitely said stupid things and then regretted them a bit later
22:59:02 <andythenorth> I've said stupid things and not regretted them even :P
22:59:11 <FLHerne> (that said, I didn't see what they were originally banned for...)
23:15:24 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN dismissed a review for pull request #7573: Fix #7561: Remove assumption between power and cost https://git.io/fjcJy
23:33:16 <DorpsGek_II> [OpenTTD/OpenTTD] glx22 commented on pull request #7398: Fix #7371: Avoid dependency on foundations of town tile during saveload https://git.io/fjC1X
23:35:16 <DorpsGek_II> [OpenTTD/OpenTTD] glx22 commented on pull request #7516: Memory measurement and limits for Squirrel https://git.io/fjC1D
continue to next day ⏵
