Config
Log for #openttd on 3rd November 2019:
Times are UTC Toggle Colours
01:00:07  <DorpsGek_III> [OpenTTD/OpenTTD] ldpl commented on issue #7814: More track for the same cost https://git.io/JezxN
01:42:45  *** HerzogDeXtEr has quit IRC
01:44:11  *** HerzogDeXtEr has joined #openttd
01:46:51  *** HerzogDeXtEr has joined #openttd
01:49:45  *** HerzogDeXtEr has quit IRC
01:50:59  *** HerzogDeXtEr has joined #openttd
01:53:34  *** HerzogDeXtEr has quit IRC
02:19:07  <DorpsGek_III> [OpenTTD/OpenTTD] glx22 commented on pull request #7809: Fix #7631: replace produced&transported with cargo sprites in industry directory https://git.io/Jeg97
03:05:57  *** Wormnest_ has joined #openttd
03:43:23  *** Wormnest_ has quit IRC
03:52:08  *** D-HUND has joined #openttd
03:55:27  *** debdog has quit IRC
04:02:27  *** D-HUND is now known as debdog
04:16:49  *** Wormnest_ has joined #openttd
04:38:05  *** glx has quit IRC
05:46:26  <DorpsGek_III> [OpenTTD/OpenTTD] Wirdal commented on issue #7592: Road vehicles don't balance between multiple loading bays https://git.io/fjlL0
05:47:36  *** Wormnest_ has quit IRC
05:54:01  *** sla_ro|master has joined #openttd
07:36:26  *** Samu has joined #openttd
07:41:12  *** andythenorth has joined #openttd
07:42:40  <andythenorth> o/
07:51:38  <Samu> hi
07:54:35  <andythenorth> oof, caboose as lead engine of consist
07:54:42  <andythenorth> I tried to add support for that, not wise
08:16:12  <Samu> thinking about the solution for #7670
08:16:53  <Samu> cache could be part of it
08:17:15  <Samu> just not sure how
08:18:04  <Samu> the first time the pathfinder is called, the cache is empty, so it pathfinds
08:19:27  <Samu> something is cached. Then the vehicle finds itself blocked the current tick and doesn't advance. However, on the next tick, something is cached already
08:19:36  <andythenorth> @seen pikka
08:19:36  <DorpsGek> andythenorth: pikka was last seen in #openttd 5 days, 20 hours, 7 minutes, and 57 seconds ago: <Pikka> oops
08:19:54  <Samu> but it doesn't pick up the cache
08:20:33  <Samu> maybe the cache isn't prepared to handle the same tile
08:21:41  <Samu> it expects the vehicle to have advanced
08:21:50  <andythenorth> hmm
08:22:24  <andythenorth> buy menu is ordered by engines (anything with purchase power != 0), then wagons (anything with purchase power = 0)
08:22:45  * andythenorth has a pony wish 
08:23:06  <andythenorth> engines, powered wagons, wagons
08:23:34  <andythenorth> (powered wagons with cb 36, not the other way)
08:32:28  *** Smedles has quit IRC
08:33:47  *** Smedles has joined #openttd
08:34:01  *** Maarten has quit IRC
08:37:07  *** Progman has joined #openttd
08:53:58  <Samu> roadveh cache is complex, it caches tiles where there's intersections, not just any tile
09:43:33  *** HerzogDeXtEr has joined #openttd
09:58:22  *** frosch123 has joined #openttd
10:02:36  <DorpsGek_III> [OpenTTD/OpenTTD] andythenorth commented on pull request #7809: Fix #7631: replace produced&transported with cargo sprites in industry directory https://git.io/JegFq
10:21:24  *** andythenorth is now known as Guest6957
10:21:25  *** andythenorth_ has joined #openttd
10:21:25  *** andythenorth_ is now known as andythenorth
10:23:07  *** Guest6957 has quit IRC
10:47:20  <Samu> I have a fix for 7670, but it's not fully tested, I'm unsure if I make a pr of it
10:47:55  <Samu> actually, the fix involves caching
10:48:11  <Samu> which is a YAPF feature, so it doesn't fix for NPF
11:06:50  <DorpsGek_III> [OpenTTD/OpenTTD] SamuXarick opened 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
11:07:32  <Samu> who can test?
11:07:53  *** gelignite has joined #openttd
11:09:35  * andythenorth making trains
11:09:36  <andythenorth> https://www.tt-forums.net/viewtopic.php?f=26&t=68616&p=1226514#p1226514
11:41:59  <Samu> well I'm not on the celeron now, which was where I first found the problem
11:42:23  <Samu> that initial savegame now averages 5 ms for road vehicles
11:42:38  <Samu> down from 65 ms
11:43:48  <Samu> on the celeron I reported 135-160 ms
11:46:25  <Samu> @calc 160/65 * 5
11:46:25  <DorpsGek> Samu: 12.3076923077
11:46:47  <Samu> estimated 12 ms for the celeron
11:46:54  <Samu> down from 160
12:06:04  *** Wolf01 has joined #openttd
12:12:17  *** tokai|noir has joined #openttd
12:12:17  *** ChanServ sets mode: +v tokai|noir
12:18:09  * andythenorth wonders if substituting palette animation for frame animation would be faster :P
12:18:47  <andythenorth> it sucks FPS from 32 to 21, in an otherwise bare map
12:19:13  *** tokai has quit IRC
12:19:59  <Eddi|zuHause> i'm not convinced
12:20:54  <Eddi|zuHause> because suddenly you added code paths for checking whether sprites changed, in places that did not need changing before (like ground sprites)
12:22:58  * andythenorth wonders where the palette cycle is applied
12:23:09  <andythenorth> is there one final composed bitmap for the screen?
12:24:34  <Eddi|zuHause> yes
12:24:52  <frosch123> there is a second bitmap keeping track of animated pixels
12:25:14  <frosch123> which is constantly blitted over the screen (mostly transparent)
12:25:51  <Eddi|zuHause> on which corner of youtube did i end up now? https://www.youtube.com/watch?v=FHMO6VOBsgM
12:26:21  <andythenorth> your YT is not like mine
12:28:02  <Eddi|zuHause> this is only like 4 steps away from a channel i recently subscribed to
12:28:59  <Eddi|zuHause> and this channel has songs called "Lala", "Lalala" and "Lalalala"
12:29:48  <Eddi|zuHause> (i would assume there's also a "La", but i haven't seen that one yet)
12:56:52  *** glx has joined #openttd
12:56:52  *** ChanServ sets mode: +v glx
13:05:18  *** Flygon has quit IRC
13:36:40  <DorpsGek_III> [OpenTTD/OpenTTD] glx22 updated pull request #7809: Fix #7631: replace produced&transported with cargo sprites in industry directory https://git.io/JeuaI
13:57:15  *** andythenorth has quit IRC
14:01:28  *** supermop_Home_ has quit IRC
14:07:14  *** supermop_Home has joined #openttd
15:18:36  *** supermop_Home has quit IRC
15:22:17  <DorpsGek_III> [OpenTTD/OpenTTD] Eddi-z updated pull request #7000: Some NewGRF variables concerning railtypes https://git.io/fhI7h
15:23:40  <Eddi|zuHause> should i have compiled before pushing?
15:43:46  <Eddi|zuHause> "if (param == 0x88 && (condtype < 0x0B || condtype > 0x0E)) {" <-- looks like this line should have been touched for NRT, but wasn't
15:44:35  <Eddi|zuHause> (not that the situation it wants to catch would appear in any real use case anyway)
15:47:14  <Eddi|zuHause> how do git people edit to branches at the same time?
15:49:56  *** Wormnest_ has joined #openttd
15:51:30  *** gelignite has quit IRC
15:51:58  <LordAro> Eddi|zuHause: define "at the same time"
15:52:14  <Eddi|zuHause> LordAro: like, having two working copies
15:52:29  <glx> dual clone maybe
15:52:34  <LordAro> either by having two working copies, or just by switching between them :p
15:52:42  <Eddi|zuHause> LordAro: edit a file in master, and not lose your unfinished/uncommited work in a branch
15:52:52  <glx> git stash
15:53:07  <LordAro> temporary commit in the branch, or ^
15:53:11  <Eddi|zuHause> glx: that would still modify the file on disk, and confuse the open editor
15:53:29  <Eddi|zuHause> so, second clone it is
15:53:42  <LordAro> sounds like you need a better editor :p
15:53:55  <glx> let's try to rebase cmake branch
15:54:02  <LordAro> \o/
15:55:39  <Eddi|zuHause> LordAro: the editor would probably be fine, but i would probably lose any unsaved changes
15:56:08  <LordAro> keeping unsaved changes in the editor is always dangerous
15:58:49  <DorpsGek_III> [OpenTTD/OpenTTD] Eddi-z opened pull request #7823: Fix c02ef3e456: Special case of Var 0x88 was not handled for NRT Acti… https://git.io/Je2vL
15:59:53  <Eddi|zuHause> was probably silly to do this, as it will now conflict with #7000, which would silently have fixed this in the next iteration :p
16:03:22  <glx> so nice, conflicts in each cmake commit
16:03:30  <LordAro> haha
16:03:43  <glx> most of them are in deleted files ;)
16:04:23  <glx> ah finally the README.md conflicts
16:05:16  <LordAro> relatedly, i do need to backport something similar to https://github.com/JGRennison/OpenTTD-patches/commit/f479559a4e61aefb2bb5973d60dbf7cd4dc0c166
16:05:37  <LordAro> MinGW doesn't configure successfully without SDL2 at all - i have to specify --without-sdl
16:05:57  <LordAro> glx: luckily we haven't merged andy's rewrite yet then :p
16:06:01  <glx> I think I already handle that in cmake
16:06:17  <LordAro> there are a few things handled "already" in cmake
16:06:21  <LordAro> unfortunately...
16:06:40  <glx> hmm I have not tested mingw without SDL2 in master
16:07:28  *** andythenorth has joined #openttd
16:08:24  <glx> oh I need to be careful and apply the cmake readme changes into compiling
16:08:59  <andythenorth> ouch sorry :)
16:09:03  <andythenorth> I changed stuff
16:09:21  <glx> that's the fun with rebasing :)
16:10:20  <andythenorth> stash or commit in the branch
16:10:45  <andythenorth> one trick is using --amend and also not pushing to any CI that will get angry about broken
16:15:30  <Eddi|zuHause> action7 is... weird
16:18:12  <Eddi|zuHause> also, newgrf-specs is missing all NRT stuff
16:18:31  <andythenorth> yes
16:19:01  <andythenorth> well the stubs are there :P https://newgrf-specs.tt-wiki.net/index.php?title=Action0/Roadtypes&action=edit&redlink=1
16:19:27  <andythenorth> https://github.com/OpenTTD/nml/issues/46
16:20:07  <Eddi|zuHause> i mean, like the new action7 conditions for definedness
16:26:47  <Eddi|zuHause> andythenorth: in NRT action7, what's the difference between condition 0x0F and 0x11, or 0x10/0x12 respectively?
16:26:56  <Eddi|zuHause> andythenorth: they're the exact same code
16:27:16  <Eddi|zuHause> andythenorth: should one of them be Tram?
16:27:53  <andythenorth> not sure, but I'm going to find the PR
16:29:10  <andythenorth> Eddi|zuHause: which src file?
16:29:18  <Eddi|zuHause> newgrf.cpp
16:29:24  <Eddi|zuHause> line 6663-ish
16:30:57  <andythenorth> hmm, seems 4 would be 2 too many
16:31:14  <andythenorth> oh wait, there's valid and not valid checks
16:31:29  <Eddi|zuHause> yes, that's 2...
16:31:34  <Eddi|zuHause> but what are the other 2 for?
16:33:07  <andythenorth> either (a) it's vestigal, maybe it was intended to be TRAMTYPE in the extra 2 or (b) the second two are tramtypes due to some magic or (c) I don't know
16:33:36  <andythenorth> I'm reading more of that file
16:34:05  <Eddi|zuHause> andythenorth: i don't see that being helpful. would be more interesting to track the history of that change
16:34:16  <Eddi|zuHause> from before squashing everything
16:34:26  <andythenorth> well we threw away the history :P
16:34:29  <andythenorth> due to git mistakes
16:34:40  <Eddi|zuHause> that is... problematic
16:35:46  <Samu> Is this good english for a comment? "Randomize all settings the Script requested to be randomized, but when
16:35:46  <Samu>  'all' is set to false, then all but "start_date" settings are randomized." https://github.com/OpenTTD/OpenTTD/pull/7486#discussion_r329305287
16:36:48  * andythenorth exploring
16:37:03  <andythenorth> blame is crashing in github
16:37:36  <DorpsGek_III> [OpenTTD/OpenTTD] Eddi-z updated pull request #7823: Fix c02ef3e456: Special case of Var 0x88 was not handled for NRT Acti… https://git.io/Je2vL
16:38:13  <DorpsGek_III> [OpenTTD/OpenTTD] Eddi-z commented on pull request #7823: Fix c02ef3e456: Special case of Var 0x88 was not handled for NRT Acti… https://git.io/Je2fu
16:38:42  <andythenorth> Eddi|zuHause: https://github.com/andythenorth/NotRoadTypes/blob/road-and-tram-types/src/newgrf.cpp#L6406
16:38:54  <DorpsGek_III> [OpenTTD/OpenTTD] Eddi-z updated pull request #7823: Fix c02ef3e456: Special case of Var 0x88 was not handled for NRT Acti… https://git.io/Je2vL
16:39:14  <andythenorth> I think it's vestigal, from ROADTYPE_TRAM being split from ROADTYPE_ROAD, which I think was changed
16:40:48  <Eddi|zuHause> so, remove the two superfluous ones?
16:41:44  <frosch123> noone knows nrt
16:43:47  <andythenorth> it's a meta game :)
16:44:40  <DorpsGek_III> [OpenTTD/OpenTTD] Eddi-z updated pull request #7823: Fix c02ef3e456: Special case of Var 0x88 was not handled for NRT Acti… https://git.io/Je2vL
16:44:47  <Eddi|zuHause> i'ma reuse that PR
16:45:10  <DorpsGek_III> [OpenTTD/OpenTTD] michicc approved pull request #7809: Fix #7631:  16 out cargo support for industry directory https://git.io/Je2fx
16:45:10  <Eddi|zuHause> should probably squash or reorder those two commits now
16:45:39  <frosch123> anyway, i think peters idea was to separate road/tram on the spec side, and then not bother/check in the implementation
16:45:49  <andythenorth> ooh 7631 is fixed :)
16:45:50  <frosch123> but no idea, all myths
16:46:17  <LordAro> one would hope that between peter1138, andythenorth & Wolf01, some sort of spec could be produced
16:46:38  <andythenorth> I am kinda writing nml examples slowly
16:46:43  <andythenorth> as test cases
16:46:54  <andythenorth> that kinda...assumes I did the nml patch correctly :P
16:47:56  <frosch123> it's the hyrum's spec
16:48:47  <DorpsGek_III> [OpenTTD/OpenTTD] Eddi-z updated pull request #7000: Some NewGRF variables concerning railtypes https://git.io/fhI7h
16:51:11  <andythenorth> I think of it as an adventure
16:51:21  <andythenorth> and a lesson :P
16:51:49  <andythenorth> shall we do a new station spec next? o_O
16:52:33  <frosch123> sounds political. make a station spec to distract from no-nrt spec?
16:52:44  <DorpsGek_III> [OpenTTD/OpenTTD] Eddi-z commented on pull request #7000: Some NewGRF variables concerning railtypes https://git.io/Je2Jk
16:53:45  <DorpsGek_III> [OpenTTD/OpenTTD] glx22 updated pull request #7270: Introduce CMake (and removing all other project-related code) https://git.io/fhbqc
16:53:56  <andythenorth> I have been relying on this one :) https://wiki.openttd.org/Frosch/NotRoadTypes
16:54:01  <glx> works locally, let's see what the CI thinks
16:54:22  <Eddi|zuHause> frosch123> anyway, i think peters idea was to separate road/tram on the spec side, and then not bother/check in the implementation <-- ok, this kinda makes sense, in case we want to completely reimplement it later (as if that would ever happen)
16:54:37  <frosch123> Eddi|zuHause: noone knows :)
16:56:41  <frosch123> if there are no checks, hyrum's law will result in grfs using the spec incorrectly
16:56:47  <frosch123> though since there is no spec, all bets are off
16:58:16  <Eddi|zuHause> "error: cannot initialize a variable of type 'RailType' with an rvalue of type 'uint8' (aka 'unsigned char')" <-- i must have failed with some assumption here that GetReverseRailtypeTranslation would return a RailType...
16:59:07  <DorpsGek_III> [OpenTTD/OpenTTD] Eddi-z updated pull request #7000: Some NewGRF variables concerning railtypes https://git.io/fhI7h
17:00:02  <Eddi|zuHause> this whole "switch (condtype)" could use some commenting
17:01:45  <Eddi|zuHause> also note that this whole code is completely untested
17:06:48  *** gelignite has joined #openttd
17:10:46  <DorpsGek_III> [OpenTTD/OpenTTD] SamuXarick updated pull request #7486: Fix: AI/GS settings with the flag SCRIPTCONFIG_RANDOM could be altered after loading from a savegame. https://git.io/fjtsp
17:12:21  <Eddi|zuHause> "result = !!(param_val & (1 << cond_val));" <-- this looks really odd :/
17:12:27  <andythenorth> oof I found a bug in Horse :(
17:12:37  <andythenorth> cases that weren't being exposed, and now are
17:12:40  <Eddi|zuHause> Is it a Horsefly?
17:12:43  <andythenorth> yes
17:12:52  <andythenorth> Horse is such a nest of spaghetti
17:13:12  <andythenorth> the nml compile part is fine, and the graphics renderer is fine
17:13:18  <andythenorth> but the bits that connect those two are horrible
17:13:56  <Eddi|zuHause> fun fact: the german translation of "Horsefly" is "Bremse" [Brake]
17:14:05  <andythenorth> hmm
17:16:56  <andythenorth> should we use these as screenshots? o_O https://www.tt-forums.net/viewtopic.php?t=73800#p1226449
17:17:24  <Eddi|zuHause> andythenorth: not without asking
17:17:32  <andythenorth> obvs :)
17:17:47  <Eddi|zuHause> and only the first one
17:17:48  <andythenorth> I rephrase, should I ask if we can use them as screenshots
17:17:49  <frosch123> screenshots without infrastructure?
17:18:50  <andythenorth> has a castle
17:18:56  <andythenorth> always a winner
17:20:37  <frosch123> ah, the famous palace in altdorf
17:22:41  <Eddi|zuHause> andythenorth: so, this is now the place where we would develop test cases for #7000
17:23:28  <andythenorth> :)
17:24:00  <Eddi|zuHause> it's probably "done", except for removing the hacky var3F stuff
17:24:19  <Eddi|zuHause> and potentially adding NRT versions of the other stuff
17:32:37  <frosch123> Eddi|zuHause: why do you use GetReverseRailTypeTranslation?
17:32:49  <andythenorth> eh can we do fog? https://railpictures.net/photo/714362/
17:33:08  <frosch123> that will only return something if the grf knows the railtype, which defeats the entire point?
17:33:51  <Eddi|zuHause> frosch123: how would a grf ask about a railtype it doesn't know?
17:35:00  <frosch123> wasn't the idea that a railtype grf could define E1, E2, E3, E4, and the train grf can still figure out whether ELRL or 3RDR is available?
17:35:51  <Eddi|zuHause> frosch123: the idea is to throw two RTT entries at it (which might have been subject to Action6-style fallback replacement earlier in the GRF, if you're thinking in NML terms), and test whether those are actually the same thing
17:36:25  <frosch123> yes, but your patch doesn't do that
17:37:00  <frosch123> or rather, it only does that if the grf also knows the railtype the other two are mapped to
17:38:10  <Eddi|zuHause> hm...
17:38:24  <Eddi|zuHause> that might be a corner case i have missed
17:44:59  <frosch123> i think you want to compare the results of GetRailTypeByLabel(_cur.grffile->railtype_list[param1/2])
17:48:04  <frosch123> looks like railtype_map is only relevant to the railtype grf, so the general case should use railtype_list
17:48:24  <frosch123> though all other a7 checks use the label directly
17:48:47  <frosch123> not sure whether there is a good reason for that, or whether it predates the cargo translation table :p
17:50:39  <Eddi|zuHause> ok, let's try that...
17:51:06  <DorpsGek_III> [OpenTTD/OpenTTD] Eddi-z updated pull request #7000: Some NewGRF variables concerning railtypes https://git.io/fhI7h
17:52:34  <DorpsGek_III> [OpenTTD/OpenTTD] Eddi-z updated pull request #7000: Some NewGRF variables concerning railtypes https://git.io/fhI7h
20:03:01  *** Wormnest_ has quit IRC
20:09:36  *** cHawk- has joined #openttd
20:13:41  *** cHawk has quit IRC
20:15:54  <DorpsGek_III> [OpenTTD/OpenTTD] SamuXarick opened pull request #7824: Fix #7697: Add a Missing AddProducedCargo proc for HQ https://git.io/Je2I2
20:25:48  *** sla_ro|master has quit IRC
20:36:59  *** sla_ro|master has joined #openttd
20:51:35  *** gelignite has quit IRC
21:04:41  <DorpsGek_III> [OpenTTD/OpenTTD] nikolas commented on issue #7810: SDL2: IME input problem for non-english language on Linux https://git.io/JeuoP
21:11:04  *** sla_ro|master has quit IRC
21:18:04  <_dp_> experimenting with town zones highlight: https://i.imgur.com/nYyKbGN.png
21:20:03  <Eddi|zuHause> we should probably rescale the radius of some of these
21:22:08  <_dp_> plz no :p
21:22:15  <andythenorth> quite interesting
21:22:15  <_dp_> not without an option at least
21:22:29  <_dp_> why rescale though?
21:22:41  <andythenorth> grid is all wrong though eh :)
21:23:11  <_dp_> andythenorth, what grid?
21:23:31  <andythenorth> https://github.com/OpenTTD/OpenTTD/pull/7745
21:23:55  <andythenorth> I tried some interesting adjustments also locally :P  https://github.com/OpenTTD/OpenTTD/pull/7745#issuecomment-549037785
21:27:10  <Eddi|zuHause> _dp_: the innermost town zone tends to get too big too quickly, and the other ones have this very narrow band instead of forming proper areas
21:28:16  <Eddi|zuHause> besides, i'm not too fond of the concentric circles anyway
21:29:34  <Eddi|zuHause> _dp_: what would the option do and which target audience wants to meddle with that option anyway, besides the "everything must stay as it was in 1996" crowd?
21:31:36  <frosch123> yay, finally some tile highlighting that doesn't make the whole viewport busy as in dwarf fortress
21:32:52  <andythenorth> ooh we could do ASCII overlays? o_O
21:33:10  <andythenorth> Eddi|zuHause: I wondered about grid varying by town zone :P
21:33:15  <andythenorth> but then...town zone moves
21:33:42  <Eddi|zuHause> andythenorth: that's horrible, because grids don't care about existing roads
21:37:51  <DorpsGek_III> [OpenTTD/OpenTTD] LordAro commented on pull request #7817: Feature: minimap command https://git.io/Je2t3
21:38:25  <DorpsGek_III> [OpenTTD/OpenTTD] LordAro approved pull request #7823: Fix c02ef3e456: Special case of Var 0x88 was not handled for NRT Acti… https://git.io/Je2ts
21:39:36  <DorpsGek_III> [OpenTTD/OpenTTD] LordAro approved pull request #7824: Fix #7697: Add a Missing AddProducedCargo proc for HQ https://git.io/Je2tZ
21:40:26  *** supermop_Home has joined #openttd
21:40:29  <supermop_Home> yo
21:42:22  <_dp_> Eddi|zuHause, citybuilders are designed to work with a current town mechacnics
21:42:45  <_dp_> Eddi|zuHause, any abrupt change can throw off gameplay there and void years worth of scores history
21:42:58  <Eddi|zuHause> _dp_: how is that an argument?
21:45:08  <DorpsGek_III> [OpenTTD/OpenTTD] LordAro requested changes for pull request #7809: Fix #7631:  16 out cargo support for industry directory https://git.io/Je2tl
21:46:03  <DorpsGek_III> [OpenTTD/OpenTTD] LordAro approved pull request #7808: Removal of SVN ID line as it is not needed. https://git.io/Je2t0
21:48:18  <DorpsGek_III> [OpenTTD/OpenTTD] LordAro commented on pull request #7805: Fix #7646: incomplete cleanup for non-threaded world generation failure https://git.io/Je2tu
21:48:51  *** Wormnest_ has joined #openttd
21:49:17  <DorpsGek_III> [OpenTTD/OpenTTD] LordAro commented on pull request #7782: Codechange: Improve NetworkAddress usage https://git.io/Je2tg
21:49:17  <DorpsGek_III> [OpenTTD/OpenTTD] LordAro closed pull request #7782: Codechange: Improve NetworkAddress usage https://git.io/JeBKX
21:53:05  <DorpsGek_III> [OpenTTD/OpenTTD] frosch123 requested changes for pull request #7823: Fix c02ef3e456: Special case of Var 0x88 was not handled for NRT Acti… https://git.io/Je2tr
21:53:36  <frosch123> at some point someone needs to define what nrt should do
21:53:47  <frosch123> randomly changing stuff makes it even worse as it already is
21:54:00  <LordAro> frosch123: this is why i don't always merge stuff automatically, thanks :)
21:54:06  <DorpsGek_III> [OpenTTD/OpenTTD] Eddi-z commented on pull request #7823: Fix c02ef3e456: Special case of Var 0x88 was not handled for NRT Acti… https://git.io/Je2t6
21:54:54  <Eddi|zuHause> frosch123: we can leave out the second change for now, and probably forget about it. but the first change is definitely valid
21:55:08  <DorpsGek_III> [OpenTTD/OpenTTD] LordAro requested changes for pull request #7000: Some NewGRF variables concerning railtypes https://git.io/Je2tX
21:55:31  <frosch123> yes, though i wonder whether the if cases can be reordered somehow
21:57:52  <frosch123> but i see no clever way without changing the behaviour
21:58:57  <frosch123> also, how do i stop gh from always wanting secondary authentification?
21:59:06  <frosch123> do i need to disable some ff privacy stuff?
21:59:25  <DorpsGek_III> [OpenTTD/OpenTTD] Eddi-z commented on pull request #7000: Some NewGRF variables concerning railtypes https://git.io/Je2tH
22:00:02  <LordAro> frosch123: 2FA?
22:00:08  <LordAro> or something else?
22:00:28  <frosch123> i do not use 2fa
22:00:41  <Eddi|zuHause> i've had to insert some random email code every now and then
22:00:51  <DorpsGek_III> [OpenTTD/OpenTTD] LordAro commented on pull request #7000: Some NewGRF variables concerning railtypes https://git.io/Je2td
22:00:54  <frosch123> but everytime i login, gh sends a email with a verification code
22:01:06  <frosch123> so i always have to wait 10 seconds until the email arrives or so
22:01:08  <LordAro> sounds like you've ticked a box somewhere
22:01:29  <DorpsGek_III> [OpenTTD/OpenTTD] Eddi-z commented on pull request #7000: Some NewGRF variables concerning railtypes https://git.io/Je2tb
22:02:02  <Eddi|zuHause> LordAro: i definitely didn't tick any box, it just sort of happened at some point
22:02:37  <LordAro> perhaps a default has changed somewhere
22:04:09  <DorpsGek_III> [OpenTTD/OpenTTD] Eddi-z updated pull request #7000: Some NewGRF variables concerning railtypes https://git.io/fhI7h
22:05:57  <DorpsGek_III> [OpenTTD/OpenTTD] Eddi-z updated pull request #7000: Some NewGRF variables concerning railtypes https://git.io/fhI7h
22:06:53  <DorpsGek_III> [OpenTTD/OpenTTD] Eddi-z commented on pull request #7000: Some NewGRF variables concerning railtypes https://git.io/Je2qe
22:07:17  <frosch123> oh, gh has "saved replies"
22:07:28  <frosch123> useful for rejecting feature requests :)
22:08:46  <andythenorth> my intent for NRT is to grind through the docs sometime
22:08:52  <Eddi|zuHause> i really wish it would be easier to see the actual CI failure, instead of digging 3 levels deep
22:09:02  <supermop_Home> years of scores history?
22:09:16  <LordAro> Eddi|zuHause: GH Actions Soonᵀᴹ
22:09:43  <andythenorth> supermop_Home: moar Horse https://www.tt-forums.net/viewtopic.php?f=26&t=68616&start=500#p1226514
22:09:47  <frosch123> how popular do you need to be to make gh's "available for hire" flag useful?
22:10:02  <andythenorth> dunno :)
22:10:02  <LordAro> that thought did cross my mind
22:10:15  <andythenorth> don't most engineers have far more job offers than they want?
22:10:51  <frosch123> depends how picky you are
22:11:11  * peter1138 is copying files from his server, finally.
22:11:33  <andythenorth> I see these weird SO or hackernews posts, with engineers who say things like 'my job sucks, but I can't get another'
22:11:37  <andythenorth> which makes no sense :P
22:12:14  <frosch123> if you want to apply worldwide, you can use the job offers on reddit/cpp
22:12:51  <frosch123> though i am not sure the people posting there are aware of that
22:13:25  <andythenorth> also in the link above, the 10hp looks weird :P
22:13:26  <andythenorth> eh
22:13:33  * andythenorth changes subject
22:13:33  <supermop_Home> like the dvt
22:13:48  <supermop_Home> happy to have it
22:14:03  <frosch123> andythenorth: line spacing too small?
22:16:26  <supermop_Home> can you make a apt-ish train with dvts on both ends and engine in the center?
22:17:12  <andythenorth> yes
22:17:16  <supermop_Home> neato
22:17:28  *** Progman has quit IRC
22:17:41  <andythenorth> frosch123: more that it's weird having to set power :)
22:17:48  <andythenorth> but that's the way we get an engine, so eh
22:18:01  <andythenorth> snowploughs will also get power :P
22:18:17  <andythenorth> I did try some tricks with a hidden railtype and cb36, but it didn't work
22:19:50  <frosch123> oi, i wonder whether i had grasped that if i were more awake :)
22:20:08  <supermop_Home> cabbage in NA horse?
22:20:44  <andythenorth> supermop_Home: yes, this is prep for that
22:20:59  <andythenorth> and snowploughs :P
22:21:05  <andythenorth> snowploughs are so pointless :)
22:22:36  <andythenorth> https://www.tauntontrains.co.uk/oldsite/images/2016-MAR/57306-66431-14-03-16-7Z75-0920-FWATER-FWATER-VIAHIGHBRIDGE-40STEP-NW.jpg
22:24:08  <andythenorth> oh there are snowblowers too :P
22:25:06  <supermop_Home> you have them in the uk?
22:25:09  <supermop_Home> why?
22:26:51  <andythenorth> https://www.rmweb.co.uk/community/index.php?/topic/131774-network-southeast-snow-train/
22:28:35  <supermop_Home> ah yes. the south east of england
22:28:48  <supermop_Home> famous for snowbound railways
22:30:17  <Eddi|zuHause> the rarer snow is, the more catastrophic it tends to be when it actually happens
22:31:05  *** Wolf01 has quit IRC
22:33:31  <supermop_Home> i need this: https://acaia.co/products/pyxis-ultralight-weighing-scale
22:34:11  <supermop_Home> Eddi|zuHause: Atlanta is famously paralyzed everytime they get flurries that barely even stick to the ground...
22:34:42  <milek7> what does it have to do with.. coffee?
22:35:11  <Eddi|zuHause> supermop_Home: when i was young, they said that about rome. nowadays this same problem occurs to our cities
22:36:24  <Eddi|zuHause> supermop_Home: and i live more northern than most canadians.
22:37:16  <supermop_Home> the coffee scale i need regardless of snowfall
22:37:23  <Eddi|zuHause> and i believe it's also more northern than london
22:37:36  <supermop_Home> why the universe has conspired to make it so expensive i don't know
22:40:25  <Eddi|zuHause> ok, London is pretty much the same latitude
22:40:44  <Eddi|zuHause> supermop_Home: it's called capitalism
22:41:58  <Eddi|zuHause> supermop_Home: it's a thing that was invented outside the USA, and like everything else, the USA imported it and brought it to such extremes that it's become unhealthy.
22:46:17  <andythenorth> capitalism is somewhat Dutch
22:47:04  <andythenorth> the shareholder part was somewhat co-operative, it was to support purchasing of windmills by communities, to drain land that had too much peat extracted and was below sea level
22:47:33  <andythenorth> transactions via paper and accounting was also somewhat dutch, but also hanseatic league
22:47:47  <andythenorth> at least, according to a couple of books I read :P
22:48:48  <supermop_Home> damn hanseatic league screws me over once again
22:50:06  <andythenorth> ouch, once again, a whole day to write 14 lines of code
22:50:12  <andythenorth> but they are the correct lines :P
22:50:42  <andythenorth> and they are fewer than the old non-working lines
22:58:40  <peter1138> That's nothing.
22:58:45  <peter1138> I had a whole day to eat food.
23:07:10  <andythenorth> did you eat any?
23:07:58  <frosch123> i have heard that programmers turn coffee into code
23:08:12  <frosch123> but if peter started eating code lines ...
23:08:30  <andythenorth> I turn coffee into mild anxiety and a need for ant-acid tablets
23:08:49  <andythenorth> but in the morning coffee turns me into 'adequate'
23:12:05  <andythenorth> supermop_Home: :P https://dev.openttdcoop.org/attachments/download/9527/DVTs.png
23:12:09  <andythenorth> not quite right, but eh
23:12:26  <supermop_Home> hehe
23:13:08  <supermop_Home> probably not worth the callback hell to blank out cabs when in middle of consist, if consist has DVTs
23:13:29  <andythenorth> can now do this http://www.rmweb.co.uk/community/uploads/monthly_12_2017/post-7258-0-38103200-1513989145_thumb.jpg
23:13:50  <andythenorth> using livery thing
23:15:09  <andythenorth> did we do a 3rd company colour yet? :P
23:15:48  <Eddi|zuHause> what for?
23:16:12  <Eddi|zuHause> that image certainly doesn't need it
23:16:36  <supermop_Home> 3rd cc is 'slightly faded version of 1st cc'
23:16:40  <peter1138> I did 32 company colours.
23:17:30  <supermop_Home> i was wondering.... about rendered sprites in 8bpp and 'palletted' but not 32bpp
23:17:36  <Eddi|zuHause> supermop_Home: you can do that with random bits, doesn't need special cc
23:18:08  <andythenorth> the dark grey around the windows in the image...could be 3CC :P
23:18:45  <supermop_Home> like if you could render sprites, but clamp the output to something like the TT palette without it looking bad
23:19:24  <Eddi|zuHause> supermop_Home: there are conversion scripts into 8bpp?
23:20:47  <supermop_Home> one idea i had was that you would let saturation etc vary over the whole range but limit the hues to those in the palette
23:21:50  <supermop_Home> assuming that those colors are the only paints available in TT land, but of course shadows could make them any amount of darker or lighter
23:22:12  <supermop_Home> (maybe only luminance varies and saturation is also limited)
23:23:20  <supermop_Home> basically instead of locomotion where models were rendered them forced to the palette, which made them look dead,
23:23:48  <supermop_Home> you'd have full 32bpp, but only cheerful 8bpp style colors
23:24:12  <Eddi|zuHause> i have no clue what you're saying
23:24:18  <Eddi|zuHause> you either have 32bpp or not.
23:24:31  <Eddi|zuHause> you have no choice in the selection of colours in 8bpp mode
23:24:57  <supermop_Home> if you have 32bpp you don't have to use all those millions of colors
23:25:29  <Eddi|zuHause> yes. but that is your artistic choice and the game has nothing to do with that
23:26:11  <supermop_Home> you could use just a few thousand that are more or less thematic to the original palette
23:26:52  <supermop_Home> this wouldn't be something for the game to do, it would be part of the work flow in modelling / rendering / photoshopping
23:26:59  *** frosch123 has quit IRC
23:27:46  <supermop_Home> its more of an idea for a set of 'best practices' or methods to achieve a 32bpp look that is not out of place in the game, as many rendered sets are
23:28:25  <Eddi|zuHause> i think you're talking to the wrong audience for that
23:28:47  * andythenorth bed
23:28:48  *** andythenorth has left #openttd
23:28:50  <DorpsGek_III> [OpenTTD/OpenTTD] JGRennison commented on issue #7810: SDL2: IME input problem for non-english language on Linux https://git.io/JeuoP
23:30:07  <supermop_Home> maybe
23:30:47  <supermop_Home> been doing a fair amount of rendering for work lately so it's on my mind
23:32:17  <supermop_Home> is it better to have a house set with 1000 houses, that you have to draw etc, or 250 houses with 4 permutations each
23:33:13  <supermop_Home> would the bigger grf size or the code to randomly combine bits in game be worse
23:33:16  *** Samu has quit IRC
23:41:38  <Eddi|zuHause> better using what metric?
23:45:22  <supermop_Home> game performance
23:46:27  <Eddi|zuHause> i doubt it matters either way
23:47:09  <Eddi|zuHause> if you're not hitting sprite cache limits, or doing crazy animation, you probably don't need to worry
23:51:36  <supermop_Home> it seems like using some sort of combinatorial magic would be a lot faster for the author/artist than drawing 1000s of sprites even if you are reusing a lot elements
23:53:38  <supermop_Home> but for someone like me it might be easier to just slog through in photoshop
23:53:51  <Eddi|zuHause> you can compose sprites from sub-sprites
23:54:38  <Eddi|zuHause> say, you construct a multi-level building from different variations for each level
23:56:42  <Eddi|zuHause> like, you're designing houses for an early industrial european town center, you could make the ground level from variations of shops, varying the number of upper levels, and also varying the roof
23:59:22  <DorpsGek_III> [OpenTTD/OpenTTD] James103 commented on issue #7816: SDL2: in console, typing _ or + zooms in and out the map https://git.io/JegWa

Powered by YARRSTE version: svn-trunk