IRC logs for #openttd on OFTC at 2015-06-12
⏴ go to previous day
01:33:16 *** Supercheese has joined #openttd
02:37:22 *** supermop has joined #openttd
02:45:12 *** Stimrol has joined #openttd
06:03:17 *** Supercheese has joined #openttd
06:56:16 *** Eddi|zuHause has joined #openttd
07:53:12 *** sla_ro|master has joined #openttd
08:41:49 *** andythenorth has joined #openttd
08:42:31 <andythenorth> how you taunt me
08:48:36 *** Hiddenfunstuff has joined #openttd
09:12:49 <andythenorth> so cdist and distributing from secondary industries
09:12:56 <andythenorth> seems the best way is to build a station per destination
09:13:05 <andythenorth> and let OpenTTD handle the cargo allocation according to station rating
09:31:44 <peter1138> Bah, audio CD with cracks in it, has read errors. NOT at the cracks...
09:34:53 <andythenorth> ‘unlike vinyl, CDs are virtually indestructible, and will never skip’
09:35:18 <peter1138> It plays in a regular player.
09:35:42 <peter1138> But when ripping them I prefer to get them as correct as possible.
09:52:15 <supermop> hmm should i build a deltic or 47 for this mail train
09:52:30 <supermop> 1987 so both seem a bit old
10:30:42 *** NumberNoid has joined #openttd
10:33:18 *** Supercheese has joined #openttd
10:34:12 *** JacobD88 has joined #openttd
10:52:26 *** Prof_Frink has joined #openttd
10:56:05 *** Myhorta has joined #openttd
10:59:36 *** Alberth has joined #openttd
10:59:36 *** ChanServ sets mode: +o Alberth
11:05:31 *** NumberNoid has joined #openttd
11:05:31 *** TinoDidriksen has joined #openttd
11:05:31 *** tycoondemon has joined #openttd
11:05:31 *** NGC3982 has joined #openttd
11:22:26 *** Born_Acorn has joined #openttd
11:48:24 *** andythenorth has joined #openttd
11:48:32 <andythenorth> has anyone got a working daylength patch?
11:52:59 <peter1138> Days are already longer. Wait til December.
12:04:06 *** Klanticus has joined #openttd
12:08:15 <Eddi|zuHause> warning: untested :p
12:09:57 * andythenorth ponders a 1x, 2x, 4x, 8x intro date multiplier for Iron Horse
12:53:35 *** jottyfan has joined #openttd
12:53:44 <andythenorth> what industries do you Australian types need?
12:54:15 <jottyfan> hi again, about subsidy in busybee- game script
12:54:35 <jottyfan> I've a working solution but could not translate all the language files except english and german
12:55:11 <jottyfan> where can I put my git repo with all the content so that the developer of busybee can include it if accepted?
13:00:27 <Pikka> Australians need... um...
13:00:29 <Eddi|zuHause> you don't have to translate to every language known to man...
13:01:10 <Eddi|zuHause> i assume there's a devzone entry for busy bee, where you can open a ticket
13:02:43 *** Klanticus has joined #openttd
13:02:44 <jottyfan> do you think I can just clone this repo, make a branch, add my changes and push it?
13:02:56 <jottyfan> or is this too rude?
13:05:27 <Eddi|zuHause> you won't be able to push
13:05:51 <Eddi|zuHause> make a patch, and attach it to a ticket
13:05:55 <planetmaker> jottyfan, you need to apply for a project and give us your public key.
13:06:14 <planetmaker> I'm pretty sure also that busy bee repo is a mercurial one
13:06:25 <planetmaker> as translation services work exclusively with mercurial
13:07:03 <jottyfan> this turns out to become complex
13:07:13 <planetmaker> 20therefore%20legitimate%20to%20be%20hosted%20from%20you.%20More%20infos%20follows:
13:07:43 <Eddi|zuHause> could use an url shortener :p
13:08:15 <Eddi|zuHause> iirc there were shortened urls for some of these things
13:08:35 <planetmaker> I created them somewhen... and forgot :D
13:08:46 <jottyfan> ok, I'll have a look, thanks
13:09:17 <Eddi|zuHause> Flygon: none of these are industries
13:10:46 <jottyfan> so I have to register before I throw out a new branch?
13:11:00 <Alberth> jottyfan: do you have a patch?
13:11:18 <Flygon> Are we allowed to designate Collingwood's Apartment Towers as an Industry that generates Bogans, Refugees, and Asians?
13:11:24 <planetmaker> jottyfan, a page which allows pushing anything without registration probably still needs invention. Or is a malware site
13:11:27 * Flygon runs away. VERY quickly.
13:11:37 <planetmaker> or a paste service
13:12:26 <peter1138> You might get a maintainer to pull from you without registration
13:14:27 <planetmaker> peter1138, though that would be probably similar hassle as walk him through registration and providing a key to push himself :)
13:14:35 <jottyfan> so if I put the tar with the new version to my homepage, planetmaker, you could load it, check it and make it available to the community?
13:15:12 <planetmaker> I'm site admin at devzone, but not one of the busy-bee maintainers
13:15:36 <planetmaker> Alberth, to his own repo. Not yours :)
13:15:54 <planetmaker> I would not dare allow that w/o your permission
13:16:40 <Alberth> I am confused how that helps tbh
13:17:23 <Alberth> apparently there is a git repo already
13:17:33 <planetmaker> jottyfan, in order to understand what you provide: is it a change to busybee which you want to see implemented there? Is it your own version of busy bee which you want to give to players? Start a fork of busy bee to continue development?
13:17:40 <jottyfan> @Alberth: I've extended the busy bee game script to recieve rewards after delivering cargo to come closer to subsidies
13:18:26 <jottyfan> I wonder if it should be a fork or extend busy bee - let the decision up to the developers
13:18:32 <jottyfan> it's just a very small extension
13:19:04 <jottyfan> I don't want anything for it, just publish it so that others may use it too
13:19:09 <jottyfan> I could make a patch
13:19:19 <jottyfan> of all the files I changed
13:19:22 *** JacobD88 has joined #openttd
13:19:23 <planetmaker> I can give you a repo at the DevZone, no problem. Maintaining that repo... not something I fancy (i.e. won't happen). Publishing to players: best do yourself via bananas
13:20:10 <planetmaker> DevZone can provide git repos, but they're not supported in any manner for translations or bundling
13:20:21 <planetmaker> nor build services
13:20:25 <jottyfan> and if I just put it to my homepage, give you the link and you can do whatever you want with it?
13:22:05 <Alberth> if you publish anything, please give it a different name first
13:22:39 <andythenorth> Busy Bee is explicitly no reward ;)
13:22:40 <planetmaker> if you ask me: well, sure, you can do that. I then recommend to share that link in the appropriate forum thread. And be so kind towards the original maintainer of busy-bee (aka Alberth) to rename it thus :) )
13:22:42 <andythenorth> forks are invited
13:23:52 <Alberth> jottyfan: if you want BB-devs (me & andy) to have a look, please provide a patch. Otherwise, you can fork, and publish yourself
13:24:19 <jottyfan> ok, so I create a patch of all the files I changed and publish it to my homepage?
13:24:48 <jottyfan> doing this under linux by patch file.original file.changed or differently?
13:25:10 <jottyfan> for the patches? I could...
13:25:19 <planetmaker> jottyfan, didn't you use git for your development? then use its diff command
13:25:32 <jottyfan> git diff? I can do also
13:25:52 <Alberth> that's why you want to have a VCS :)
13:27:04 <Alberth> but otherwise, you need an original and a change directory with everything in it (and nothing else), and run a diff -u --recursive orig_dir new_dir
13:27:17 <Alberth> git diff is an order of magnitude simpler though
13:28:51 *** zeknurn has joined #openttd
13:28:52 <jottyfan> git diff original-repo-ID
13:29:02 <jottyfan> but I've found the original first... :-D
13:29:11 <Alberth> hopefully you tagged the branch :)
13:29:37 <jottyfan> git log shows the changes
13:29:50 <jottyfan> but what's the first git id before I did any change
13:29:53 <Alberth> yeah, but you have to search manually :)
13:30:12 <Alberth> the one at the bottom of git log :p
13:30:32 <jottyfan> no, seems not to be the right one
13:30:34 <Alberth> git hashes are just random, no pattern in them
13:30:40 <jottyfan> I must have made some mistakes using git
13:31:06 <planetmaker> well, or you didn't start with an unmodified version :)
13:31:19 <jottyfan> so again: get the original sources, copy my changes into them and commit and make git diff afterwards?
13:31:21 <planetmaker> after all, the original cannot have been git
13:31:47 <planetmaker> simply get a checkout of the original source and then use the command alberth gave to compare the two dirs
13:32:12 <Alberth> use an export or archive thingie to prevent .git stuff
13:32:38 <Alberth> unless you want a patch with all the .git files :p
13:32:52 <jottyfan> I used the busy bee tar file inside .openttd
13:33:09 <jottyfan> then made a git repo out of it
13:33:20 <Alberth> that's not entirely source, but fair enough
13:33:24 <jottyfan> this diff is the right one?
13:34:59 <planetmaker> you can better judge that: is it the entirety of thins you changed?
13:35:18 <jottyfan> and then make it git repo
13:35:22 <jottyfan> add my changes and commit
13:35:34 <Alberth> if you started with the tar, do that again
13:36:44 <Alberth> unpack the tar, and export your copy to a new directory to get rid of the .git files etc
13:37:12 <Alberth> then you can run a diff between both directories
13:38:05 <Alberth> diff --recursive --brief unpacked_dir git_copy gives a list of changed files, so you can see whether the expected files are there
13:39:00 <Alberth> that diff output should only contain files that you want in the patch
13:39:37 <Alberth> the patch itself is created with diff -u --recursive unpacked_dir git_copy > mypatchfile.patch
13:40:22 <Alberth> and you can open that mypatchfile.patch to see if it is what you expect
13:40:30 <jottyfan> ok, it'll take a bit of time
13:40:45 <Alberth> when happy, post that patch at a paste service
13:41:10 <Alberth> yeah, a git branch tag is really useful :)
13:41:38 <Alberth> something for the next time :)
13:45:55 <jottyfan> the diff is in german - is this a problem for you?
13:47:19 <Alberth> eventually, yes, project should have a single language, imho
13:47:36 <Alberth> but code should work, shouldn't it?
13:48:11 <jottyfan> diff -u --recursive unpacked_dir git_copy > mypatchfile.patch
13:48:22 <jottyfan> gives only the name of files that changed, not the changes itself
13:48:50 <jottyfan> maybe I'd better go the git way?
13:49:08 <Alberth> perhaps you have an alias with --brief in it?
13:49:52 <jottyfan> that's the folders: BusyBee-RC2M (the original one) and BusyBeeSubsidy (the changed one)
13:50:01 <Alberth> you can try using the executable directly
13:50:03 <jottyfan> this is the command: diff -u --recursive BusyBee-RC2M BusyBeeSubsidy > BusyBeeSubsidy.patch
13:50:20 <jottyfan> the result is just the files that changed, not the content of the files
13:51:42 <Alberth> peter1138: he doesn't have an original branch tag
13:52:00 <peter1138> If you were working in master... shame on you.
13:52:15 <jottyfan> no, I did a branch but merged back to master...
13:53:42 <andythenorth> if you have the changes, and you have Busy Bee, this is trivial no?
13:53:54 * andythenorth is missing something
13:54:17 <Alberth> you can try /usr/bin/diff instead of just 'diff', ie the absolute path to the 'diff' program
13:55:23 <Alberth> andythenorth: you're missing 'experience' :p
13:55:57 <jottyfan> ok, so I put the diff (thats 7688 bytes long) to?
13:56:38 <Alberth> or some other paste service
13:58:45 <jottyfan> hope that's good enough for you to have a look at it...
13:59:43 <Alberth> only one problem, you're not preserving old news if rewards are disabled
14:00:08 <Alberth> oh, you can't even disable it
14:00:55 <jottyfan> I didn't check every option
14:01:05 <Alberth> I would recommend you add value 0 as well, and in that case revert to original behavior and news
14:01:09 <jottyfan> just tried it and it worked
14:01:35 <Alberth> it can't work, you changed the messages
14:02:07 <Alberth> the " and earned {STRING}{CURRENCY_LONG}" part should be fully optional
14:02:14 <jottyfan> but I could offer if reward > 0 GSText(label1) else GSText(label2)
14:02:49 <Alberth> also for the goals of course
14:02:50 <jottyfan> but I can just provide the german and english version
14:04:50 <Alberth> I only provide the english version normally, the other languages are handled by translators
14:05:24 <jottyfan> but what happens if I add a new language key that doesn't exist in others translations?
14:05:34 <jottyfan> shouldn't the game script crash?
14:05:54 <Alberth> it reverts to the base language, ie english
14:06:15 <Alberth> and the translator is notified of missing translations
14:06:29 <jottyfan> so english is must have, all other languages are nice to have?
14:07:00 <jottyfan> when I'm ready, I'll come back to irc and tell about
14:07:09 <Alberth> {GOLD} in STR_COMPANY_GOAL looks wrong
14:07:19 <jottyfan> what's wrong about that?
14:07:37 <Alberth> the 'for' tex colour should be the same as the leading text
14:08:31 <Alberth> {GOLD}{CARGO_LONG} <-- that changes the colour of the cargo amount, relative to the leading text
14:08:56 <Alberth> don't know exactly, source should tell that, and otherwise, just try :)
14:09:55 <jottyfan> it's not in the english.txt
14:10:00 <jottyfan> what's default color
14:10:43 <jottyfan> so how to name the new language keys?
14:10:51 <Alberth> it's not in the .nut file?
14:10:56 <jottyfan> is it betetr to include the keys in others or clone the keys?
14:11:18 <Alberth> andythenorth: do you have any problem with adding optional reward money to BB ?
14:12:02 <Alberth> probably better make a new string, so translators have freedom to re-arrange wording
14:12:28 <jottyfan> so STR_COMPANY_GOAL cloned to STR_COMPANY_GOAL_SUBSIDY ?
14:12:48 <jottyfan> STR_SUBSIDY_REWARD to be included in STR_COMPANY_GOAL
14:13:20 <jottyfan> it's 3 language keys I changed, so I do 3 clones
14:13:39 <jottyfan> I could also do what cargo_long does
14:14:28 <Alberth> no idea what you mean, just add 3 new strings
14:14:36 <jottyfan> create STR_GOAL_REWARD as for {WHITE}{STRING}{CURRENCY_LONG} and include that in STR_COMPANY as {STRING}
14:15:12 <jottyfan> but it's done that way several times
14:15:24 <Alberth> translators want to move stuff around in the sentence to fit their language
14:15:41 <Alberth> and by making sub-strings, you make that very difficult
14:15:42 <jottyfan> ok, so better real clones
14:15:57 <jottyfan> and have some duplicates in translation
14:16:20 <Alberth> well, they get 'related strings', it's copy/paste from the web page, mostly :)
14:17:01 <jottyfan> STR_COMPANY_GOAL_REWARD :Deliver {GOLD}{CARGO_LONG} {ORANGE}to {STRING}{ORANGE} for {WHITE}{STRING}{CURRENCY_LONG}
14:17:01 <jottyfan> STR_COMPANY_GOAL :Deliver {GOLD}{CARGO_LONG} {ORANGE}to {STRING}{ORANGE}
14:17:28 *** TheMask96 has joined #openttd
14:17:33 <Alberth> yes, but drop "GOAL_ in the first one?
14:17:47 <Alberth> you want to align all :
14:18:29 <jottyfan> even with STR_COMPANY_GOAL_REWARD_NEWS
14:18:39 <jottyfan> need no spaces before :
14:19:19 <jottyfan> doesn't work for STR_COMPANY_GOAL_WON_NEWS
14:19:25 <jottyfan> so either I add more spaces to all lines
14:19:38 <jottyfan> or (better?) I shorten REWARD by R?
14:19:45 <jottyfan> STR_COMPANY_GOAL_R_WON_NEWS
14:20:01 <Alberth> better, add REWARD, and I'll fix the alignment
14:20:32 <jottyfan> it's just spaces, isn't it?
14:20:44 <Alberth> alignment fixes make a long patch
14:20:57 <Alberth> it's harder to see what has changed
14:21:13 <Alberth> so it 's easier to make a 2nd commit with only whitespace fixes
14:21:30 <jottyfan> so I just break the length on that one line
14:23:45 <andythenorth> Alberth: I am not big on rewards, but if you want to
14:23:52 <andythenorth> I like the idea of no reward
14:24:24 <jottyfan> I like rewards because of the subsidy problem for the cargodist goods
14:24:31 <Alberth> I want to preserve not having a reward
14:24:43 <jottyfan> so I can have a kind of subsidy for cargodist goods too
14:24:58 <jottyfan> default be no subsidy factor?
14:26:01 <Alberth> jottyfan: BB is not fixing that problem, it's just ignoring it
14:26:17 <Alberth> a better solution may be to disable cdist for such cargoes
14:26:18 <jottyfan> but It feels like fidex
14:26:33 <jottyfan> no, i like cargodist very much
14:26:40 <jottyfan> but I also like subsidies
14:27:49 <jottyfan> can I use c code in nut files?
14:28:08 <jottyfan> x < 0 ? do sth. : do sth. different?
14:28:26 <Alberth> the language is 'squirrel', version 2, to be precise
14:28:28 <jottyfan> instead of: if (x < 0) { do sth. } else { do sth.}
14:31:17 <jottyfan> so for "local goal_text = ..." I need to replace it by if (...) { local goal_text = ... } else { local goal_text = ... }"?
14:31:35 <jottyfan> or do I have do declare goal_text in front of the if clause?
14:33:41 <Alberth> me too, you ask about ?:, then conclude it has to be replaced by an if statement :)
14:34:54 <jottyfan> If I'd do it in java, I'd prefer local goal_text = reward > 0 ? GSText... : GSText...
14:35:37 <Alberth> I prefer the snippet, except for the string name that should be changed
14:36:08 <Alberth> and I'd write the call at one line :)
14:36:31 <Alberth> you use GSText.STR_COMPANY_GOAL twice, with a different number of parameters
14:37:52 <jottyfan> what do you mean with "I'd write the call at one line"
14:38:13 <Alberth> merge lines 3 & 4, and 6 & 7
14:40:15 <Alberth> lines 3 and 5 are supposed to start there?
14:40:15 <jottyfan> the line breaks come from original sources
14:40:37 <jottyfan> I corrected the spaces
14:41:00 <Alberth> original sources have very little meaning, imho
14:42:32 <Alberth> I'd add an empty line between the assignments, or put each call at one line vertically aligning the arguments
14:42:57 <Alberth> the primary point is readabilty
14:43:09 <Alberth> you want to see how things change in one look
14:43:26 <Alberth> now it's one big blob of text
14:43:44 <jottyfan> so what shoud I do exactly?
14:44:03 <Alberth> make it as good as possible
14:45:06 <jottyfan> that's why I ask here in irc so much
14:45:16 <jottyfan> like a stupid script kiddie
14:47:42 <Alberth> play with alignment, try a few alternatives, until you find one that looks ok
14:48:23 <Alberth> the trick is not to read, just glance at it, and check whether it shows changes visually
14:48:50 <jottyfan> but this makes the diff more complex
14:49:20 <Alberth> yes, that's another trade-off :)
14:49:38 <Alberth> but eventually, code layout quality wins
14:49:42 <jottyfan> ok, I'd give you what I have until now because of an upcoming meeting
14:50:00 <jottyfan> just a minute, after that, I'm busy
14:50:26 <Alberth> ha, one more minute BB coding :p
14:51:12 <Alberth> jottyfan: no need to give partial stuff, I'll wait
14:51:45 *** andythenorth has left #openttd
14:52:49 <jottyfan> hope this can make it to busybee
15:03:20 *** Supercheese has joined #openttd
15:55:48 *** sla_ro|master has joined #openttd
17:25:00 *** HerzogDeXtEr has joined #openttd
17:38:27 *** TheMask96 has joined #openttd
18:39:08 *** Progman has joined #openttd
18:50:32 *** JacobD88 has joined #openttd
19:07:08 *** andythenorth has joined #openttd
19:15:46 *** Stimrol has joined #openttd
19:20:51 *** Pereba_ has joined #openttd
19:27:22 *** Pereba_ is now known as Pereba
19:33:19 *** Supercheese has joined #openttd
19:47:15 *** CompuDesktop has joined #openttd
19:58:26 *** frosch123 has joined #openttd
20:19:22 *** Myhorta[1] has joined #openttd
20:27:33 *** Suicyder has joined #openttd
20:33:12 <andythenorth> so how about extending newgrf spec with torque curve? o_O
20:33:31 <Eddi|zuHause> what does that do?
20:33:50 <andythenorth> massively over-realistic vehicle physics
20:34:04 <andythenorth> primarily it reflects available gear ratios
20:34:11 <andythenorth> and / or engine characteristics
20:34:17 <Rubidium> then lets start with modeling hunting oscillations and wear and tear of tracks
20:34:39 <andythenorth> and the tyre inflation of RVs
20:34:56 <andythenorth> also road surface
20:35:13 <Alberth> bit silly without road types
20:36:10 * andythenorth was looking at the horsepower of Road Hog
20:36:56 <andythenorth> it’s getting quite divorced from reality :)
20:41:18 <Alberth> nah, reality is buggy :)
20:44:18 *** andythen_ has joined #openttd
20:48:14 <andythenorth> who is andythen_ ?
20:52:24 *** JacobD88 has joined #openttd
20:53:28 <Eddi|zuHause> but really, who is andy, then?
20:54:53 *** JacobD88 has joined #openttd
20:56:05 <andythenorth> important question
21:08:49 <andythenorth> eh, so multiple stations does work much better with cargodist at secondary industry
21:08:54 <andythenorth> interesting change of strategy :P
21:09:08 <andythenorth> usually the best approach is a single large station for pickup
21:11:12 <Alberth> aren't just 2 or 3 stations getting anything?
21:11:18 <Alberth> hmm, perhaps with cdist not
21:21:51 *** tokai|noir has joined #openttd
21:21:51 *** ChanServ sets mode: +v tokai|noir
21:23:51 <andythenorth> with cdist, distributing outbound cargo from a secondary industry is pretty much broken for a single station case
21:24:13 <andythenorth> nearly all cargo is assigned to one route, leaving thousands of tons awaiting pickup, and all other routes waiting for cargo
21:24:25 <andythenorth> which also causes station blocking :P
21:27:07 <Alberth> ah, yes, you need waypoints to sort trains
21:27:20 <Alberth> and buffer storage of early trains
21:28:04 <andythenorth> don’t think that affects cdist much though :)
21:28:18 <Alberth> so using stations as waypoints basically?
21:28:46 *** smoke_fumus has joined #openttd
21:47:33 *** Progman has joined #openttd
21:55:42 *** NumberNoid_ has joined #openttd
22:07:01 <peter1138> FSX at 4 fps is not fun :(
22:16:22 <_dp_> is there some good way of storing some extra date in ttd savegame that doesn't prevent it from being loaded by unmodified game?
22:18:10 <glx> but it will be discarded by unmodified game
22:20:05 *** NumberNoid__ has joined #openttd
22:21:26 <_dp_> hm, to me it looks like any unexpected chunks cause an error
22:21:43 <_dp_> ch = SlFindChunkHandler(id);
22:21:45 <_dp_> if (ch == NULL) SlErrorCorrupt("Unknown chunk type");
22:23:03 <Eddi|zuHause> peter1138: moar power!!
22:24:43 <Eddi|zuHause> _dp_: if you dig in the history, you might find stuff where releases got added data without changing the savegame version. not sure if that helps your cause
22:25:08 <frosch123> Eddi|zuHause: but that does not work in both ways :)
22:26:29 <frosch123> _dp_: anyway, i guess the best method is to use gamescript data
22:27:05 <frosch123> you can likely just store whatever there
22:27:35 *** minexew has joined #openttd
22:27:53 <frosch123> esp. in the language files of the gamescript
22:28:10 <frosch123> maybe you can even store a game script translation for a "new" translation
22:28:33 <minexew> hey, how do i bring attention to an issue in the bugtracker?
22:28:38 <minexew> i submitted a patch but nobody seems to have noticed :D
22:28:56 <_dp_> frosch123, hm, interesting idea, I'll try it, thanks
22:29:08 <frosch123> minexew: what makes you think noone noticed?
22:29:23 <_dp_> also was thinking of utilizing persistent storages
22:30:11 <minexew> frosch123: the fact that i got no reply
22:31:07 <frosch123> ah, it's a windows patch
22:31:33 <frosch123> well, yeah, likely noone noticed
22:33:08 <frosch123> i am not sure whether we had an active windows dev this year
22:33:48 *** Myhorta has joined #openttd
22:35:10 *** CompuDesktop is now known as Compu
22:35:26 <minexew> so there's no point in submitting windows-specific patches?
22:35:46 <Eddi|zuHause> EU IV, which i played a lot recently, allows me to choose between "Fullscreen", "Windowed" and "Windowed Fullscreen", where the latter is basically what is described in that issue. don't know if that is a valid path for this game
22:36:33 <minexew> yeah, i didn't want to mess with the UI, so i made it implicit for the native resolution
22:37:15 <Eddi|zuHause> well, that sounds reasonable, but i'm not one that can help you getting this along...
22:37:23 <frosch123> it's the same on linux :) fullscreen is also just full-desktop without borders
22:37:38 <frosch123> well, at least on my machine
22:38:06 <minexew> even if you use a low-resolution fullscreen mode?
22:38:08 <Eddi|zuHause> well, i have notable differences in behaviour between "fullscreen" and "windowed fullscreen", especially regarding alt+tab
22:38:23 <Eddi|zuHause> with the same resolution as native
22:38:58 <frosch123> minexew: i don't think switching video modes is allowed for the average application
22:39:28 <minexew> frosch123:who does the scaling then?
22:39:36 <frosch123> so, you may say, there is only borderless maximised, no fullscreen :p
22:40:03 <Eddi|zuHause> namely, "fullscreen" will un-fullscreen to a window, and then minimize, upon alt+tab, whereas "windowed fullscreen" will stay fullscreenized in the background (making the switch less interruptive and faster)
22:40:40 <Eddi|zuHause> whereas "windowed fullscreen" fails to actually set the fullscreen flag in the window manager, which i have to do manually on each start
22:40:50 <minexew> Eddi|zuHause:that's exactly why this is an issue in the first place :) alt-tab behavior for "fullscreen" is awful
22:42:53 <minexew> i guess i'll just roll my own fork then
22:43:02 <Eddi|zuHause> minexew: sure. but you now need people to test this on all sorts of windows versions out there
22:43:17 <Eddi|zuHause> and a dev to do a code review
22:43:28 *** andythenorth has left #openttd
22:45:03 <frosch123> minexew: well, at some point the osx port turned so bad, that patches got committed unreviewer, assuming that they couldn't possibly make it worse :)
22:45:16 <frosch123> so, maybe the same will happen to the windows port
22:45:49 <minexew> i have poor experience with the os x port
22:45:59 <minexew> periodic spikes of CPU usage
22:46:09 <minexew> but that's been 1-2 years ago
22:46:40 <minexew> though i'm surprised there are no active windows devs
22:47:15 <frosch123> windows was always the minority here
22:48:12 <frosch123> currently there are hardly 4 active linux devs
22:48:52 <minexew> using github for collaboration could help ;)
22:50:08 <minexew> the barriers to entry seem needlessly high here
22:50:57 <Rubidium> how are the barriers for github any less, when the main problem is someone doing a proper review?
22:52:01 <minexew> well for one, if more people can see your project, more people are likely to step up to do reviews
22:52:26 <Eddi|zuHause> how would switching to github expose the project to more people than it already is?
22:52:57 <Eddi|zuHause> i don't think exposure is the problem at all...
22:53:30 <minexew> github has a lot of social bullshit that helps with this
22:53:40 <Eddi|zuHause> it's like you going to the greek government and tell them "how about you go on kickstarter?"
22:53:42 <minexew> e.g. if you follow somebody and they star a project, you'll see that
23:05:39 <Rubidium> the main problem with OpenTTD's development is its heritage. We don't really want to break things greatly as doing that means OpenTTD becomes something that is not based on Transport Tycoon. This makes the possible avenues of changes relatively small
23:06:33 <Rubidium> besides that, the issues that crop up aren't really interesting. It's mostly fighting with the subtle differences between different drivers and/or different versions of the same operating system, or just minor niggles
23:08:01 <Rubidium> furthermore OpenTTD has become quite stable, making merging unvetted things something that is rather frowned upon. As such someone with more or less proven knowledge of the affected components should review changes before merging them
23:08:55 <minexew> but with huge maps and massive networks that were not possible in TT, some usability improvements would be surely welcome
23:08:57 <Rubidium> after all, I can easily review complex encryption algorithms that come from the NSA and not spot any problems with it, but that is mostly because I am to unfamiliar with the intricacies of the field and such
23:10:25 <minexew> i can't imagine a reason why somebody would object to features 1-3 from that thread
23:12:24 <Eddi|zuHause> it's not really about objecting to features
23:12:31 <Eddi|zuHause> it's checking implementations
23:12:54 <Eddi|zuHause> is it error-free? is it maintainable by someone other than the author?
23:13:26 <Eddi|zuHause> does it disrupt anything that was there before?
23:13:34 *** nanoha-chan has joined #openttd
23:13:50 <Rubidium> the second chunk of the first patch looks iffy
23:14:36 <Rubidium> and it seems to be done, so in the second patch exactly the wrong thing can be done
23:19:43 <minexew> Eddi|zuHause:what would you suggest as a solution to this lack of reviewers?
23:21:50 <Eddi|zuHause> there are generally 2 ways to solve lack of staff: reactivate old staff or introduce new staff. the latter is not just done by exposing the project to more people, those new people must be active enough to prove valuable and capable of being introduced to the core staff to make such reviews
23:22:45 <Eddi|zuHause> bringing in too many new unexperienced people may even prove counterproductive, as there are not enough veteran people to weed things out
23:24:02 <minexew> but how are those new people supposed to prove themselves? by submitting patches that end up gathering dust? :P
23:25:26 <Rubidium> hint: solving existing bugs in an elegant manner will yield more points and faster "graduation" than cranking out many features but never finishing them. That is the basic functionality is there, but all the corner cases are a gaping hole. For example day length patch(es)
23:28:53 <minexew> from reading the forums i got the impression that nobody will bother reviewing a big patch, such as one needed for a significant change like that
23:30:25 <Eddi|zuHause> that's not really it.
23:30:47 <Eddi|zuHause> huge patches did get reviewed, but it requires A LOT of patience and dedication on both sides
23:31:24 <Eddi|zuHause> in the end, this is still a hobby project, which doesn't fit these two attributes very well
23:32:30 <Eddi|zuHause> and no amount of (throw buzzword in here) will change that
23:33:53 <Rubidium> not to mention the amount of times the answer of the author of the reviewed patch(set) was something amongst the like of: explitive, take it or leave it... you suck all the fun out of this patch
23:35:25 <Rubidium> and patch packs are somewhat detrimental to getting authors working on patches, because as soon as it is in a patch pack they often think they're done
23:36:48 <minexew> well that should be no surprise, as it's much easier to get your patch into a patch pack than into trunk
23:37:04 <minexew> especially if it's of poor quality
23:43:26 <minexew> how difficult would it be to get something like the industry tooltips patch through, if the code was OK?
23:44:04 <Rubidium> and that patch set is quite bad as it requires 40 times more memory for network packages, meaning sending a 10 MiB map in MP requires 400 MiB of memory server side
23:44:23 *** nanoha-chan has joined #openttd
23:44:55 <minexew> Rubidium:i get it, i didn't write that code and can't argue for its quality
23:45:10 <Rubidium> oh, it's even a factor 44.887
23:45:42 <Eddi|zuHause> baker's 40, because you're such a good customer
23:47:04 <Rubidium> if the code was sane and elegant, then it mostly depends on when a core developer with knowledge of the area has time to review it but that should probably not take a hell of a lot of time
23:47:13 <Rubidium> and the latter is the main problem
23:52:47 <Rubidium> one might say that "everyone" wants to bolt new things onto OpenTTD, but nobody wants to maintain OpenTTD
continue to next day ⏵