Times are UTC Toggle Colours
00:11:07 *** _dp_ has quit IRC 00:12:01 *** Kitrana has joined #openttd 00:15:38 *** Kitrana1 has joined #openttd 00:19:48 *** WormnestAndroid has quit IRC 00:20:02 *** Kitrana has quit IRC 00:22:51 *** WormnestAndroid has joined #openttd 00:51:28 *** WormnestAndroid has quit IRC 00:53:22 *** WormnestAndroid has joined #openttd 01:16:27 *** yeoldetoast has joined #openttd 01:17:53 *** WormnestAndroid has quit IRC 01:22:25 *** WormnestAndroid has joined #openttd 01:30:28 *** WormnestAndroid has quit IRC 01:30:55 *** WormnestAndroid has joined #openttd 02:11:02 *** WormnestAndroid has quit IRC 02:18:25 *** WormnestAndroid has joined #openttd 02:46:56 *** debdog has joined #openttd 02:49:37 *** sim-al2 has joined #openttd 02:50:16 *** D-HUND has quit IRC 03:06:12 *** glx has quit IRC 03:31:11 *** roadt_ has joined #openttd 03:33:48 *** sim-al2 has quit IRC 03:38:13 *** roadt has quit IRC 04:58:28 *** yeoldetoast has quit IRC 04:59:28 *** yeoldetoast has joined #openttd 05:03:31 *** Flygon has joined #openttd 05:19:13 *** snail_UES_ has quit IRC 05:36:27 *** nielsm has joined #openttd 05:48:31 *** andythenorth has joined #openttd 05:48:45 <andythenorth> is FIRS expected to work with 64x64 and just one town? 05:48:53 <andythenorth> or any map size and just one town? 05:52:18 <Rubidium> I think the more interesting question is, how is it supposed to work when creating a scenario? Then you (temporarily) got no towns 05:56:10 <Rubidium> I first thought you could do a check when loading the NewGRF, but that idea quickly made no sense as the NewGRFs will be loaded before any of the towns in a new game are made. So that wouldn't be a nice solution, which effectively makes it that you ought to support it. On the other hand, if you create an industry that requires watery/snowy/desert/rainforest tiles to be built upon and those do not exist 05:56:16 <Rubidium> the player might be screwed in a similar manner as when your set cannot place some industries 05:58:55 *** nielsm has quit IRC 06:01:41 *** andythenorth has quit IRC 06:04:29 *** andythenorth has joined #openttd 06:08:10 <andythenorth> it's a bit lolz currently 06:09:04 <andythenorth> industries doing any town checks seems unwise 06:09:07 <andythenorth> currently 06:09:44 <andythenorth> unrelated: seems map gen tries to place industries in order of ID 06:10:02 <andythenorth> so if a lower ID industry checks for presence on map of a higher ID, it won't be built 06:10:05 <andythenorth> at least on small maps 06:21:49 <andythenorth> maybe I can manipulate the probability until the dependencies exist 06:27:18 <andythenorth> nope 06:27:28 <andythenorth> ok this means I've just shipped a broken FIRS :D 06:27:30 <andythenorth> oh dear :) 07:07:56 <peter1138> https://blog.davetcode.co.uk/post/21st-century-emulator/ 07:09:36 <TrueBrain> someone went a bit overboard :P 07:09:40 <TrueBrain> looks like fun : 07:09:41 <TrueBrain> D 07:09:42 <andythenorth> it's not April 1? 07:10:10 <andythenorth> "Naturally I ran into a number of new issues with this approach. " 07:18:17 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain reopened issue #9386: Building on MacOS fails with "bus error: 10" https://git.io/Jngox 07:18:20 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on issue #9386: Building on MacOS fails with "bus error: 10" https://git.io/Jngox 07:19:44 <TrueBrain> no clue why it is that commit, and not one up 07:19:51 <TrueBrain> owh well, time to do other stuff first 07:20:48 *** HerzogDeXtEr has joined #openttd 07:53:05 *** jottyfan has joined #openttd 08:24:13 *** WormnestAndroid has quit IRC 08:28:42 *** WormnestAndroid has joined #openttd 09:03:00 *** jottyfan has quit IRC 09:07:21 <peter1138> Hmm, SQL server supports encrypted data fields. Seems pointless to have the server able to decrypt... 09:33:21 *** tokai has joined #openttd 09:33:21 *** ChanServ sets mode: +v tokai 09:40:40 *** Samu has joined #openttd 09:53:09 *** iSoSyS has joined #openttd 10:18:02 *** WormnestAndroid has quit IRC 10:18:15 *** WormnestAndroid has joined #openttd 11:16:04 *** iSoSyS has quit IRC 11:31:11 *** tokai|noir has joined #openttd 11:31:11 *** ChanServ sets mode: +v tokai|noir 11:38:05 *** tokai has quit IRC 12:11:26 *** J0anJosep has quit IRC 12:19:55 *** glx has joined #openttd 12:19:55 *** ChanServ sets mode: +v glx 12:30:23 *** andythenorth has quit IRC 12:50:54 *** andythenorth has joined #openttd 13:01:47 <DorpsGek> [OpenTTD/OpenTTD] embeddedt commented on pull request #9411: Replace master-server (UDP-based) with Game Coordinator (TCP-based) https://git.io/JcKik 13:40:27 *** nielsm has joined #openttd 14:07:27 <peter1138> https://twitter.com/alexjc/status/1411966249437995010 14:07:28 <peter1138> ... 14:14:42 *** belu has joined #openttd 14:15:14 <belu> Hi, 14:15:44 <belu> anyone knows how to disable shadowing under the trains on the railroad in the latest version ? 14:16:07 <LordAro> "show path reservations" setting 14:16:07 <peter1138> That'll be part of the graphics you are using. 14:17:05 <belu> ...so the train is going on the track, and as it passes tile by tile is like footprinting each and every tile on it's way 14:17:24 <belu> like being a darker shadow under the train 14:17:31 <peter1138> Oh, path reservations, lol 14:17:41 *** WormnestAndroid has quit IRC 14:17:41 <peter1138> Yeah, that... isn't new. 14:17:45 <belu> ok thank you, i''ll try now 14:18:06 <belu> new for me as I was playing in the old 0.4.7 till now :)) 14:18:23 <LordAro> peter1138: the default did change in the latest release though 14:18:37 <LordAro> belu: that is impressively old 14:18:48 <DorpsGek> [OpenTTD/OpenTTD] glx22 commented on issue #9386: Building on MacOS fails with "bus error: 10" https://git.io/Jngox 14:19:00 <belu> yes i know, i had to update as it couldn't be played online ... 14:19:17 <peter1138> LordAro, to on? Strange, that's ... undesirable. 14:19:52 <glx> it's useful for beginners 14:26:59 *** WormnestAndroid has joined #openttd 14:27:24 <belu> "show path reservations" setting worked, many thanks, 14:27:49 <belu> and is there an option to retain older trains, instead of the list getting 'rid" of the old ones ? 14:28:03 <glx> yes, "never expire" 14:28:09 <belu> and having only the newest ones ? 14:28:16 <belu> thank you ! 14:32:56 *** gelignite has joined #openttd 14:33:35 *** andythenorth has quit IRC 14:37:51 *** belu has left #openttd 14:41:05 <FLHerne> 0.4.7, released 26th March 2006 14:41:07 <FLHerne> wow 14:58:31 <glx> pff trying things requiring release build is so slow 14:59:15 <Samu> AIDepotList(AITile.TRANSPORT_WATER), which tile? north, south or both? 14:59:39 <Samu> must test 15:02:12 <Samu> im going to assume north 15:02:23 *** roadt_ has quit IRC 15:02:43 <glx> Isn't it documented ? 15:04:15 <Samu> https://docs.openttd.org/ai-api/classAIDepotList.html 15:08:17 <glx> tile is the one used when building it 15:11:32 <DorpsGek> [OpenTTD/OpenTTD] fsimonis commented on pull request #9405: Reduce allocations in string handling https://git.io/Jc6sN 15:11:37 <DorpsGek> [OpenTTD/OpenTTD] fsimonis closed pull request #9405: Reduce allocations in string handling https://git.io/Jcmd2 15:25:20 <glx> so yes north tile Samu (as "The depot will be built towards the south from 'tile', not necessarily towards 'front'.") 15:25:48 <Samu> nice, that makes things simpler 15:26:16 *** roadt has joined #openttd 15:28:50 <glx> funny, when building a water depot you can use any tile for 'front', as long as it's on one of 'tile' axis 15:29:00 *** iSoSyS has joined #openttd 15:39:49 *** iSoSyS has quit IRC 15:48:18 <Rubidium> glx / TrueBrain: might https://github.com/rubidium42/OpenTTD/tree/split_off_settings_table help with the bus 10 error? Or rather, can one of you test whether it helps? I haven't been able to reproduce the issue yet, probably due to a too old compiler or something stupid like not passing the right settings 16:19:18 *** WormnestAndroid has quit IRC 16:19:41 *** yeoldetoast has quit IRC 16:20:10 *** WormnestAndroid has joined #openttd 16:23:00 *** Progman has joined #openttd 16:28:13 *** WormnestAndroid has quit IRC 16:34:23 *** snail_UES_ has joined #openttd 16:53:10 <glx> Rubidium: just started a build, will take some time for it to complete :) 17:01:32 <glx> now in the longest part, linking 17:16:49 *** andythenorth has joined #openttd 17:16:51 <andythenorth> so at map gen are industries just built in ID order, lowest first? https://github.com/OpenTTD/OpenTTD/blob/28e90769f74e55ea7c8f75146d5b33e0aa777da8/src/industry_cmd.cpp#L2326 17:17:25 <glx> there's a first round of one of each type in type order 17:17:33 <glx> then other are placed randomly 17:17:39 <nielsm> it tries to build one of each first, then does random until the target is hit 17:18:27 <andythenorth> excellent :) that means I can read src even if I can't write it :) 17:18:33 <andythenorth> and it matches observed behaviour 17:18:54 <andythenorth> it does mean I need to manage IDs 17:19:06 <andythenorth> so that industries which depend on another industry have higher relative ID 17:19:24 <andythenorth> and I can't automate that (because IDs need to be stable) 17:19:30 <andythenorth> so I'll do it manually 17:20:07 <andythenorth> I can't co-locate different industries by town or region, so I have to check distance to type a from type b 17:22:21 <glx> Rubidium: gcc is happy with your proposal 17:36:18 <Rubidium> glx: nice. Then I'll clean it up and make a PR out of it 17:43:01 * andythenorth wonders if newgrf should gain some general purpose map scripting :P 17:43:26 <andythenorth> industry placement would be a use case 17:43:33 <andythenorth> and industry closure 17:49:07 *** Flygon has quit IRC 18:10:15 <nielsm> andythenorth: what about that suggestion I had a while back where industrytypes have a callback for "min/max desired number in this town" callback? 18:24:27 <DorpsGek> [OpenTTD/OpenTTD] rubidium42 opened pull request #9414: Split off settings.cpp into several separate files https://git.io/Jc6h3 18:25:27 <Rubidium> only "problem" with that PR is that it now conflicts with glx's PR 18:27:56 *** jottyfan has joined #openttd 18:30:11 <andythenorth> nielsm _could_ work 18:30:26 <Samu> oh no, my script took too long to save 18:30:28 <andythenorth> towns are problematic 18:30:36 <andythenorth> because the player can create a map with only 1 town 18:30:42 <andythenorth> and newgrf can't know the number of towns 18:30:54 <andythenorth> so creating 'types' of town is dangerous 18:31:02 <peter1138> Can't or doesn't? 18:31:16 <andythenorth> has been said twice that it shouldn't 18:31:31 <andythenorth> doesn't is the factual truth 18:31:38 <andythenorth> the rest is just open source 10 people 10 opinions 18:34:22 <andythenorth> peter1138 patchpack? :) 18:37:25 <peter1138> The IHaveAPatchForThatPack 18:40:15 <nielsm> also revisiting the regions idea, they could perhaps be defined by newgrf: a feature that only one grf is allowed to implement (loading a second grf that tries to implement it could then fail loading that grf). the feature only has callbacks, one callback is after mapgen has finished placing all towns, it receives the number of towns created/in existence, and returns the number of regions to 18:40:15 <nielsm> create. 18:41:26 <nielsm> the game then divides the map geographically into that number of regions by some kind of grid or whatever, and inside each region it picks one (roughly centered) town that becomes the capital of the region. 18:42:13 <andythenorth> yes 18:42:17 <nielsm> other towns might belong to a region either by the geographical subdivision, or by which capital is closest, the grf callback might be able to choose that or maybe it's just hardcoded 18:42:24 <andythenorth> regions could just be partitioning n adjoining towns 18:42:26 <andythenorth> or something 18:42:45 <andythenorth> and ideally a regional capital would be parent scope of town or similar 18:42:56 <andythenorth> then parent vars could be accessed 18:43:06 <andythenorth> or new explicit vars alternatively 18:43:15 <nielsm> if the subdivision makes a subdivision with zero towns in it, then it might try making a different pattern or something, or maybe sub-subdivide the other subdivisions with a large number of towns to make sure the requested number of capitals are selected 18:44:02 <nielsm> after the capitals are chosen and towns have a regional designated, the grf gets to visit every town (in some order) to initialise their town registers 18:44:14 <andythenorth> there is probably some case for multiple definitions of regions (with IDs for each set), but might be best avoided 18:44:33 <nielsm> and if new towns are created during gameplay, that init town cb is also run, of course 18:44:51 <andythenorth> unrelated, but now that we have town registers an industry removed cb might be useful 18:44:58 <andythenorth> like javascript on unload 18:45:13 <nielsm> probably yes 18:45:32 <andythenorth> probably the easiest case to envisage for regions is power stations 18:45:40 <nielsm> also if you want to have "min/max desired count in this town (or region?)" cb for industrytypes 18:45:42 <andythenorth> it would be easy to write electricity production to a register 18:45:52 <andythenorth> but it's silly to have a power station in every town 18:46:08 <nielsm> and the game has to close down industries to get below the max desired, it needs to ask the defining grf which industry to close 18:47:15 <nielsm> also I hope you're taking notes here because I'm not, I'm just braindumping an idea that popped up 18:47:27 <andythenorth> we have webster :) 18:48:46 <andythenorth> so the map gen cb 'towns finished' cb gets a result for number of region partitions to create? 18:48:54 <andythenorth> and we might know the number of towns at that point? 18:49:23 <andythenorth> we can found towns in game also? 18:50:04 *** tokai has joined #openttd 18:50:04 *** ChanServ sets mode: +v tokai 18:50:12 <nielsm> yeah I guess it should know the map dimensions, and number of towns created 18:50:39 <nielsm> the "init individual town" cb might also want to be able to toggle the "is city" flag 18:50:52 <DorpsGek> [OpenTTD/OpenTTD] DorpsGek pushed 1 commits to master https://git.io/Jcif9 18:50:53 <DorpsGek> - Update: Translations from eints (by translators) 18:56:55 *** tokai|noir has quit IRC 19:02:16 <andythenorth> that would be helpful 19:02:39 <andythenorth> if e.g. a town is 'mostly farms' 19:02:51 <andythenorth> it shouldn't be flagged as a city for growth 19:06:24 <glx> <@Rubidium> only "problem" with that PR is that it now conflicts with glx's PR <-- oh I'm used to fix conflicts in this PR :) 19:08:29 <glx> hmm regression seems to fail for VS2017, let's try locally 19:08:41 *** y2kboy23_ has quit IRC 19:09:10 <nielsm> and, I guess there should be some kind of global variable indicating the number of regions present, and the grfid of who defined the regions 19:09:17 *** y2kboy23 has joined #openttd 19:11:41 <Rubidium> glx: yeah, the regression failure kinda baffles me. I got no idea why it would fail 19:17:07 <glx> hmm I guess no output and no crash, maybe failed to open the savegame 19:17:56 <glx> I really should improve CI to upload the output on failure :) 19:18:20 <glx> because comparison is done after all the filtering 19:19:33 <glx> ok regression failed locally too (for VS2017) 19:19:39 <glx> I can check the output 19:20:48 <glx> ok both regression test give an empty file 19:21:15 <andythenorth> nielsm probably 19:21:39 <andythenorth> also is there a map capital? :P 19:23:01 <nielsm> multiple administrative levels gets annoying 19:25:54 <andythenorth> I can't see a gameplay reason for it 19:26:03 <andythenorth> but it would give the grf map-wide storage :P 19:26:03 <glx> seems to be a crash Rubidium 19:26:17 <andythenorth> so I am mostly interested in the side-channel exploit it offers 19:27:01 <glx> I'll do more digging later, I have an overcooked2 party waiting :) 19:27:27 <glx> oh it's a div/0 19:40:30 <Samu> how do i solve "This script took too long to Save" 19:40:50 <nielsm> making saving faster, or maybe play on smaller maps 19:41:18 <Samu> strange that I never got it before 19:41:26 <Samu> now that i added ships, i get it 19:42:34 <Samu> *cough* can I suggest increase MAX_SL_OPS *cough* 19:43:16 <Samu> but ok, i guess my townManager code is the problem 19:43:54 <Samu> this code here was actually from LuDiAI, the original 19:44:46 <Rubidium> glx: might that be due to some ordering of initialization the SettingTable variables are not initialized correctly (i.e. too early)? If so, then that might mean existing main is broken as well for the private settings tables 19:45:06 *** Gustavo6046 has quit IRC 19:49:57 *** bro has quit IRC 19:50:13 *** bro has joined #openttd 19:51:19 <nielsm> Samu: the best choice for save-load for AI and GS to be fast would be to structure the data used by the script in such a way that you don't *need* to run any expensive data collection during save but can just dump the existing data right in, and similar for loading you want to be able to basically consume the data as they were saved 19:52:22 <andythenorth> nielsm if we did regions for christmas, we could have north pole and santa :P 19:53:16 <nielsm> stop reminding me of the winter holidays themed industry set I never got very far with 19:53:19 *** Gustavo6046 has joined #openttd 20:00:40 <nielsm> this thing: https://0x0.st/s5CK.txt 20:02:37 <Samu> https://i.imgur.com/13tODxK.png - the last autosave where it managed to save, and what is loaded when loaded back 20:14:08 <Samu> https://github.com/SamuXarick/LuDiAI-AfterFix/blob/master/TownManager.nut 20:14:10 <Samu> halp 20:15:24 <Samu> saveTownManager iterating over the m_nearCityPairArray takes about 85% of the save time 20:16:39 <andythenorth> santa claus is coming to town 20:16:46 *** nielsm has quit IRC 20:19:38 *** Compu has joined #openttd 20:31:26 <Samu> i really need to investigate what's in that huge array 20:31:36 <Samu> seems fishy that it's so large, but... 20:32:03 <Samu> 512x512, tons of towns 20:32:27 <Samu> it's an array of town pairs 20:32:56 <Samu> each pair is a route to try building 20:37:31 <Samu> but it's strange, i tested the AI in 4096x4096 maps before, reaching 5k road vehicles 20:37:39 <Samu> 12k towns 20:37:56 <Samu> maybe i didn't have autosave enabled to test 20:41:07 <Samu> when i did those 100 years tests, i had autosave 20:41:28 <Samu> the ai didn't crash with a too long to save... this is very odd 20:46:45 *** jottyfan has quit IRC 20:55:50 *** gelignite has quit IRC 21:14:15 *** andythenorth has quit IRC 21:25:30 *** Samu has quit IRC 21:31:16 *** andythenorth has joined #openttd 21:33:54 *** Progman has quit IRC 21:46:53 *** andythenorth has quit IRC 21:49:08 *** andythenorth has joined #openttd 21:51:17 <peter1138> M1 Macbook? 22:04:50 <andythenorth> well 22:05:19 <andythenorth> i9 Macbook is definitely fail 22:05:27 * andythenorth must to sleep 22:05:29 *** andythenorth has quit IRC 22:05:48 <glx> i9 in anything not well ventilated 22:19:05 <peter1138> Problem with a Macbook is it comes with Mac OS. 22:24:44 <DorpsGek> [OpenTTD/team] SlipperyDuck777 opened issue #232: [en_AU] Translator access request https://git.io/JciVM 22:48:20 *** yeoldetoast has joined #openttd 23:07:27 <DorpsGek> [OpenTTD/OpenTTD] James103 commented on pull request #8285: Rewrite `OverflowSafeInt` to avoid undefined behaviour https://git.io/Jci6N 23:28:09 <glx> Rubidium: it fails when loading intro game, and indeed the 3 SettingTable are not yet initialised when GetSettingFromName() is called