IRC logs for #openttd on OFTC at 2020-07-02
⏴ go to previous day
01:08:25 <DorpsGek_III> [OpenTTD/OpenTTD] techgeeknz updated pull request #8217: Port GUI rendering improvements from JGRPP https://git.io/Jf9yI
01:21:08 <DorpsGek_III> [OpenTTD/OpenTTD] techgeeknz commented on pull request #8217: Port GUI rendering improvements from JGRPP https://git.io/JJUZw
01:29:46 <DorpsGek_III> [OpenTTD/OpenTTD] techgeeknz commented on pull request #8217: Port GUI rendering improvements from JGRPP https://git.io/JJUZF
02:20:28 <DorpsGek_III> [OpenTTD/OpenTTD] techgeeknz commented on pull request #8217: Port GUI rendering improvements from JGRPP https://git.io/JJUnd
02:40:33 *** tokai|noir has joined #openttd
02:40:33 *** ChanServ sets mode: +v tokai|noir
02:45:56 <DorpsGek_III> [OpenTTD/OpenTTD] techgeeknz commented on pull request #8217: Port GUI rendering improvements from JGRPP https://git.io/JJUcF
02:58:22 *** Wormnest has joined #openttd
03:43:17 <DorpsGek_III> [OpenTTD/OpenTTD] techgeeknz opened issue #8252: Possible duplication of functionality in `screenshot.cpp` https://git.io/JJUWw
03:51:03 <DorpsGek_III> [OpenTTD/OpenTTD] James103 opened issue #8253: Very large company values are not graphed correctly https://git.io/JJUWy
03:52:30 <DorpsGek_III> [OpenTTD/OpenTTD] techgeeknz updated pull request #8217: Port GUI rendering improvements from JGRPP https://git.io/Jf9yI
03:53:34 <DorpsGek_III> [OpenTTD/OpenTTD] techgeeknz commented on pull request #8217: Port GUI rendering improvements from JGRPP https://git.io/JJUW7
04:07:33 <DorpsGek_III> [OpenTTD/OpenTTD] techgeeknz commented on pull request #8217: Port GUI rendering improvements from JGRPP https://git.io/JJUlJ
05:58:54 *** Wormnest has joined #openttd
06:06:01 <DorpsGek_III> [OpenTTD/OpenTTD] techgeeknz opened pull request #8254: Add: Extend DECLARE_ENUM_AS_BIT_SET with extra operations https://git.io/JJUBh
06:26:06 <DorpsGek_III> [OpenTTD/OpenTTD] techgeeknz updated pull request #8254: Add: Extend DECLARE_ENUM_AS_BIT_SET with extra operations https://git.io/JJUBh
06:48:43 *** Gustavo6046 has joined #openttd
06:57:04 <DorpsGek_III> [OpenTTD/OpenTTD] techgeeknz updated pull request #8254: Add: Extend DECLARE_ENUM_AS_BIT_SET with extra operations https://git.io/JJUBh
07:06:57 <DorpsGek_III> [OpenTTD/OpenTTD] techgeeknz updated pull request #8254: Add: Extend DECLARE_ENUM_AS_BIT_SET with extra operations https://git.io/JJUBh
07:08:46 <DorpsGek_III> [OpenTTD/OpenTTD] techgeeknz updated pull request #8254: Add: Extend DECLARE_ENUM_AS_BIT_SET with extra operations https://git.io/JJUBh
07:19:20 <DorpsGek_III> [OpenTTD/OpenTTD] techgeeknz updated pull request #8254: Add: Extend DECLARE_ENUM_AS_BIT_SET with extra operations https://git.io/JJUBh
07:19:44 *** Wormnest has joined #openttd
07:35:01 <DorpsGek_III> [OpenTTD/OpenTTD] techgeeknz updated pull request #8217: Port GUI rendering improvements from JGRPP https://git.io/Jf9yI
08:02:52 *** sla_ro|master has joined #openttd
08:18:46 <DorpsGek_III> [OpenTTD/OpenTTD] LordAro approved pull request #8251: Fix: 8250: NRT: Company infrastructure window always omits last road/… https://git.io/JJUuK
08:19:21 <DorpsGek_III> [OpenTTD/OpenTTD] LordAro approved pull request #8247: Fix #8162: Improve Error Message When Converting Town Owned Road https://git.io/JJUu6
08:21:21 <DorpsGek_III> [OpenTTD/OpenTTD] LordAro commented on pull request #8254: Add: Extend DECLARE_ENUM_AS_BIT_SET with extra operations https://git.io/JJUu1
08:21:40 <DorpsGek_III> [OpenTTD/OpenTTD] LordAro approved pull request #8236: Fix #7772: Show destination on mouseover when vehicle stopped (and not in depot) https://git.io/JJUuM
08:23:31 <DorpsGek_III> [OpenTTD/OpenTTD] LordAro approved pull request #8227: Fix: Thread unsafe use of SendPacket for PACKET_SERVER_MAP_SIZE https://git.io/JJUu5
08:23:47 <DorpsGek_III> [OpenTTD/OpenTTD] LordAro approved pull request #8218: Add: [CMake] More improvements for install https://git.io/JJUud
08:39:42 <DorpsGek_III> [OpenTTD/OpenTTD] techgeeknz commented on pull request #8254: Add: Extend DECLARE_ENUM_AS_BIT_SET with extra operations https://git.io/JJUzZ
08:45:29 *** andythenorth has joined #openttd
09:28:19 *** andythenorth has joined #openttd
09:33:50 *** WormnestAndroid has quit IRC
09:34:03 *** WormnestAndroid has joined #openttd
09:45:49 *** WormnestAndroid has quit IRC
09:46:06 *** WormnestAndroid has joined #openttd
10:37:26 <andythenorth> 'we can rebuild him!'
10:42:16 *** Flygon_ has joined #openttd
12:10:07 *** arikover has joined #openttd
12:35:31 <DorpsGek_III> [OpenTTD/OpenTTD] nielsmh merged pull request #8251: Fix: 8250: NRT: Company infrastructure window always omits last road/… https://git.io/JJJq2
12:35:33 <DorpsGek_III> [OpenTTD/OpenTTD] nielsmh closed issue #8250: NRT: Company infrastructure window always omits last road/tramtype https://git.io/JJfRr
12:38:24 *** andythenorth_ has joined #openttd
12:48:55 <DorpsGek_III> [OpenTTD/OpenTTD] JGRennison commented on pull request #8217: Port GUI rendering improvements from JGRPP https://git.io/JJUDe
12:51:13 <DorpsGek_III> [OpenTTD/OpenTTD] techgeeknz updated pull request #8217: Port GUI rendering improvements from JGRPP https://git.io/Jf9yI
12:51:49 *** gelignite has joined #openttd
12:57:45 <DorpsGek_III> [OpenTTD/OpenTTD] techgeeknz commented on pull request #8217: Port GUI rendering improvements from JGRPP https://git.io/JJUDr
12:59:45 <DorpsGek_III> [OpenTTD/OpenTTD] techgeeknz opened pull request #8255: Prepare: Update core function headers for refactor of GUI rendering https://git.io/JJUDi
13:03:57 <DorpsGek_III> [OpenTTD/OpenTTD] techgeeknz opened pull request #8256: Refactor GUI rendering (based on JGRPP) https://git.io/JJUDh
13:06:12 <DorpsGek_III> [OpenTTD/OpenTTD] techgeeknz opened pull request #8257: Refactor GUI rendering: Add documentation and name `Viewport` consistently https://git.io/JJUyq
13:09:19 <LordAro> that's not what i wanted at all
13:14:10 <DorpsGek_III> [OpenTTD/OpenTTD] nielsmh commented on pull request #8255: Prepare: Update core function headers for refactor of GUI rendering https://git.io/JJUyr
13:16:17 <DorpsGek_III> [OpenTTD/OpenTTD] nielsmh commented on pull request #8217: Port GUI rendering improvements from JGRPP https://git.io/JJUyP
13:16:45 <DorpsGek_III> [OpenTTD/OpenTTD] ldpl commented on issue #8219: Inexplicably low redraw rate on multiplayer server https://git.io/JfHiF
13:17:10 <DorpsGek_III> [OpenTTD/OpenTTD] LordAro commented on pull request #8255: Prepare: Update core function headers for refactor of GUI rendering https://git.io/JJUyX
13:18:34 <DorpsGek_III> [OpenTTD/OpenTTD] LordAro commented on pull request #8217: Port GUI rendering improvements from JGRPP https://git.io/JJUyH
13:20:01 <DorpsGek_III> [OpenTTD/OpenTTD] techgeeknz commented on pull request #8217: Port GUI rendering improvements from JGRPP https://git.io/JJUy5
13:21:01 <DorpsGek_III> [OpenTTD/OpenTTD] LordAro commented on pull request #8257: Refactor GUI rendering: Add documentation and name `Viewport` consistently https://git.io/JJUyA
14:00:30 <nielsm> the savegame version should be the current after loading has finished
14:00:35 <Eddi|zuHause> i think that was a hack so it could be backported to release branch
14:02:58 <LordAro> Eddi|zuHause: can't see it in the 1.1 branch
14:05:30 <dP> anyway, it's seemingly safe to remove now
14:06:21 <andythenorth_> seems the AWS options are mutually exclusive
14:06:22 <Eddi|zuHause> i wouldn't make any assumptions about "safeness" before we cleared up what the intention was
14:06:55 <andythenorth_> either (a) setup for http straight to S3 XOR (b) use a cloudfront distribution to get https
14:07:07 <andythenorth_> I did (a), now I need to reverse the instructions and start again
14:07:39 <Eddi|zuHause> didn't we clean up a lot of these CONDNULLs a while ago?
14:08:05 <dP> Eddi|zuHause, yeah, I've seen them removed in some places but this one is still there
14:08:14 <dP> can remove it as well while I'm at it
14:09:04 <Eddi|zuHause> where in this damn github interface can i see a list of commits to a branch?
14:10:57 <andythenorth_> it's fiddly to find
14:11:13 <andythenorth_> '24, 174 commits'
14:11:27 <andythenorth_> [replace branch as needed]
14:13:55 *** arikover has joined #openttd
14:30:33 <DorpsGek_III> [OpenTTD/OpenTTD] ldpl opened pull request #8258: Codechange #8258: Remove unused town cargo caches from the savegame https://git.io/JJUHp
14:32:50 <dP> also #backport_requested for #8159 ;)
14:41:20 <LordAro> nothing is backport requested at present
14:41:38 <LordAro> if we decide to do a 1.10.3, then we'll go through all PRs
14:43:08 <dP> I though the whole point of that label was to mark prs so they don't get forgotten accidentally
14:43:09 <DorpsGek_III> [OpenTTD/OpenTTD] LordAro approved pull request #8258: Codechange #8258: Remove unused town cargo caches from the savegame https://git.io/JJUQX
14:43:32 <LordAro> dP: yeah, but that's only when they're scattered in amongst PRs that are marked as such
14:43:37 <LordAro> here, none of them are marked
14:45:15 <dP> it kind of already happened with that #8159 and 1.10.2 :p
14:45:25 <LordAro> i probably shouldn't have approved that PR, given it fails the CI :p
14:47:37 <dP> that CI doesn't seem to be feeling well :p
14:48:07 <LordAro> it seems to have broken loading one of the regression test saves
14:52:31 <dP> hm, its output looks jibberish to me but I did find one bug...
14:55:09 <DorpsGek_III> [OpenTTD/OpenTTD] ldpl dismissed a review for pull request #8258: Codechange #8258: Remove unused town cargo caches from the savegame https://git.io/JJUQX
14:55:09 <DorpsGek_III> [OpenTTD/OpenTTD] ldpl updated pull request #8258: Codechange #8258: Remove unused town cargo caches from the savegame https://git.io/JJUHp
14:55:45 <LordAro> dP: azure's output isn't very clear, you need to click a lot to get the actual output
14:56:35 <dP> LordAro, what actual output do you mean? I clicked a lot but still only got some nonsense
14:56:42 <dP> while there was indeed a regression
15:01:13 <dP> oh, found it, just didn't expect the actual error be in grey when there is a huge block of red "errors" nearby.
15:23:47 <CornsMcGowan[m]> hello, asking before i start work on this feature
15:24:03 <CornsMcGowan[m]> would something like the Factorio railplanner be accepted into trunk?
15:25:01 <CornsMcGowan[m]> a build tool more powerful than polyline where you can set a start/end point and uses path search algorithms to find the best solution
15:25:47 <CornsMcGowan[m]> i've made WIP feature branches before but never got around to finishing, below are implementations for canals and rails respectively
15:26:40 <nielsm> personally I'd say it's against the spirit of the game
15:26:51 <CornsMcGowan[m]> i'd intend to include bridges/tunnels as well
15:27:33 <CornsMcGowan[m]> and also implement a new DoCommand for building non-straight rails (so I don't have to chain DoCommand together)
15:27:59 <CornsMcGowan[m]> i think it would be beneficial to a lot of players, especially on mobile devices where building is often tedious
15:39:25 <CornsMcGowan[m]> i just remembered i've asked the same question before
15:50:37 <CornsMcGowan[m]> from last year: LordAro says "i'm not going to complain if someone else decides it is desirable"
15:50:37 <CornsMcGowan[m]> FLHerne says "I don't think I'd /use/ it, but I have no objection to it existing"
15:51:44 <andythenorth_> I really didn't get on with polyline
15:52:04 <andythenorth_> but Railroad Tycoon 2 / 3 had auto-route between start and end point
15:55:57 <CornsMcGowan[m]> oh i didn't know that :0
16:02:39 <CornsMcGowan[m]> oh man the RRT2 interface is quite something
16:07:34 <Eddi|zuHause> i see a handful of issues with a feature like this: a) tweaking it to make "sane" routes", b) live-switching between preferring tunnels, bridges or cuttings, c) setting minimum curve radius and slope distances, d) adding a second track to an existing rail, etc.
16:08:47 <Eddi|zuHause> if i were using a feature like this, i'll see myself constantly fighting against it because it doesn't build the exact route i have in mind
16:11:12 <Eddi|zuHause> what is an "optimal" route varies widely during gameplay, depending on what engines you have available, or how much money you have
16:11:30 <CornsMcGowan[m]> thats true indeed (re the optimal route)
16:12:17 <CornsMcGowan[m]> we could make the weighting of slopes adjustable with an interface similar to the signal auto spacing GUI
16:26:27 <CornsMcGowan[m]> Eddi|zuHause: can you elaborate on those points?
16:26:55 <CornsMcGowan[m]> like, what would need to change in order for you be less opposed to using the feature
16:32:32 <FLHerne> CornsMcGowan[m]: I'm not sure I agree with myself :p
16:32:51 <FLHerne> I still wouldn't use it
16:33:44 <FLHerne> Players who care at all about aesthetics won't use it, because no-one's figured out an algorithm to build nice-looking routes
16:35:37 <FLHerne> But for the efficiency-uber-alles people it still won't beat "perfectly straight line over every mountain, and buy all the locos"
16:36:19 <CornsMcGowan[m]> i understand re: aesthetics
16:37:00 <CornsMcGowan[m]> but for efficiency wouldn't it be well suited? given we can indeed tweak it to produce the fastest route between points
16:37:28 <CornsMcGowan[m]> also i imagine mobile users would enjoy the feature since it relives the tedium of building rails, roads and canals
16:37:58 <FLHerne> Well, as you can see on some competitive servers, the optimal route is normally a perfectly straight line disregarding terrain :p
16:38:40 <CornsMcGowan[m]> lmao thats true, and its possible to give slopes the same weight as flat tracks so it would construct such a path
16:38:48 <CornsMcGowan[m]> unless you mean terraforming as well
16:39:03 <FLHerne> Locos to go up continous slopes at full speed are cheaper than the lost income from a longer route, which is silly, but eh
16:39:29 <FLHerne> Sure, but we have a tool to build straight lines already ;-)
16:39:29 <CornsMcGowan[m]> yeah its not the nicest look
16:39:47 <FLHerne> I can see the appeal for mobile
16:40:07 <FLHerne> Well, I can't, because I can't see how anyone plays on mobile at all, but...
16:40:30 <CornsMcGowan[m]> well i think the addition of this feature makes it slightly more palatable for them
16:40:45 <FLHerne> My phone is fairly large-screened, and I still found it incredibly frustrating
16:41:03 <CornsMcGowan[m]> FLHerne: is there anything you would change about the railplanner to make it more appealing to you?
16:41:20 <FLHerne> I guess I don't see the point
16:41:36 <FLHerne> The economic model of OTTD is ridiculous and uninteresting
16:42:12 <FLHerne> Timetabling is awkward and un-fun except for the results you can produce
16:42:38 <FLHerne> So...the part I find fun is designing and building routes
16:42:48 <FLHerne> If you automate that, why are you playing?
16:43:12 <Eddi|zuHause> <CornsMcGowan[m]> like, what would need to change in order for you be less opposed to using the feature <-- i don't think the feature is in a state where i could make a useful contribution
16:43:15 <FLHerne> Oh, and playing model-railway with scenery and objects
16:43:36 <Eddi|zuHause> i mean, i could make a dozen suggestions that will be unimplementable
16:43:43 <Eddi|zuHause> but nobody would be helped by that
16:43:48 <CornsMcGowan[m]> i want to hear them :p
16:43:52 <FLHerne> Which is also defeated by automation unless you achieve the Holy Grail of a track-planner that builds non-hideous double-track railways
16:44:20 <nielsm> at the absolute minimum I think there needs to be an enforced construction delay, and it must construct in segments, rather than giving players a way to build 1000 tiles of complex-routed rail in a single tick
16:44:50 <CornsMcGowan[m]> a suggestion Alberth made in the TT-forums amounted to: limit the amount of track that could be built at once
16:46:17 <CornsMcGowan[m]> e.g. some arbitrary limit like 100 tiles can be built with each click
16:47:01 <CornsMcGowan[m]> nielsm: "there needs to be an enforced construction delay, and it must construct in segments" can't that be implemented via the current burst/time period construction limits?
16:47:15 <Eddi|zuHause> for a useful track planner i need: a) a preview that lets me tweak it before doing any actual construction, b) tweakable settings (curve radius, slope steepness, tunnel/bridge percentage, ...), c) movable intermediate nodes
16:47:26 <Eddi|zuHause> that should keep you busy for a while
16:47:54 <CornsMcGowan[m]> a) is already done, b) amounts to implementing a gui similar to the signal spacing adjustments
16:48:09 <CornsMcGowan[m]> c) can be acheived by: building shorter sections at a time i guess
16:48:23 <Eddi|zuHause> no, c) is the opposite of that
16:48:40 <Eddi|zuHause> top-down vs. bottom-up
16:48:56 <CornsMcGowan[m]> Eddi|zuHause: re: point a), do you consider the current ghost preview (i.e. the same style as current track construction) sufficient?
16:49:22 <nielsm> yes extending the current landscaping limits system to also apply to track building, and then have track construction somehow queue up segments, would be okay
16:49:44 <nielsm> the main thing being that players who prefer to build manually aren't disadvantaged
16:49:51 <CornsMcGowan[m]> FLHerne: i also enjoy track building, but i find building longer stretches tedious - my main entertainment comes from building junctions and stations
16:50:30 <Eddi|zuHause> CornsMcGowan[m]: the graphical representation is not the point. the point is that you set the nodes down, and can move them, and only when everything is right, you press the "now do the construction" button
16:50:57 <CornsMcGowan[m]> i see, so a Confirm/Cancel window like what the mobile app does
16:51:04 <CornsMcGowan[m]> and yeah the node placement stuff
16:52:34 <Eddi|zuHause> CornsMcGowan[m]: think like the google maps route planner where you can drag the suggested route around
16:52:42 <CornsMcGowan[m]> ah yeah that one
16:53:19 <Eddi|zuHause> you can add/remove intermediate nodes if you don't agree with the suggested route
16:53:43 <CornsMcGowan[m]> Eddi|zuHause: i feel like i'm getting the short end of the stick - what if i implement all these features and your feedback is "ah i still don't like it"
16:53:50 <Eddi|zuHause> and it has alternative routes for certain optimisation criteria (e.g. speed vs. length)
16:54:26 <CornsMcGowan[m]> nielsm: what do you mean by disadvantaged?
16:55:31 <Eddi|zuHause> CornsMcGowan[m]: i don't think i can dispel these concerns of yours
16:55:35 <nielsm> a player building manually would probably spend several months of game time on building a route that a player using the autobuilder could build in half a month of game time
16:56:27 <CornsMcGowan[m]> i see, but i feel the two players you propose aren't playing to the same goal
16:56:56 <andythenorth_> the most yak-shaving-reduction tool for me would be building 1, 2, 3, or 4 tracks at once
16:57:06 <andythenorth_> with some magic slope adjustments
16:58:04 <LordAro> CornsMcGowan[m]: not all patches are meant for "vanilla"
16:58:11 <LordAro> tis the way of things
16:58:27 <nielsm> if you make such a tool available to players, AI writers will also want it
16:58:39 <nielsm> and then suddenly the entire pathfinding part of AI writing is shunted away
16:58:46 <Eddi|zuHause> well, i'd say AI would be the primary beneficiary of this tool
16:58:51 <CornsMcGowan[m]> um, doesn't AI already have access to the A* pathfinding library
16:59:18 <nielsm> not really no, AIs technically have to implement pathfinding themselves
16:59:26 <Eddi|zuHause> no, AI authors reimplemented A* in their own libraries
16:59:30 <nielsm> and it just happens that most use the same library for it
17:00:19 <nielsm> and as far as I understand, that was one of the design goals for NoAI, to explicitly _not_ provide any built-in pathfinding
17:00:49 <CornsMcGowan[m]> oh spooky, i didn't know that
17:00:54 <CornsMcGowan[m]> is there a reason for that?
17:01:26 <CornsMcGowan[m]> nvm you said already what could possibly happen
17:02:22 <CornsMcGowan[m]> i really do think mobile users would benefit from this
17:02:37 <CornsMcGowan[m]> and PR'ing to trunk is the only way to get it there i think
17:04:12 <nielsm> give the mobile version a virtual touchpad to move the mouse cursor around and it'll play (almost) the same as for most laptop users
17:06:23 *** virtualrandomnumber has joined #openttd
17:06:47 *** virtualrandomnumber has quit IRC
17:09:10 <CornsMcGowan[m]> wasn't there a time before Autorail and Autoroad
17:10:12 <nielsm> yes originally there were only directional build tools
17:10:56 <nielsm> and originally in Transport Tycoon the diagonal rail tools were not draggable
17:11:11 <Eddi|zuHause> i remember playing the original game and always pressing 1-4 for directions
17:11:32 <nielsm> I personally never use autorail, it's too finicky to hit the direction I want to build, but always use autoroad
17:11:54 <nielsm> yep left hand on 1234 keys and right hand on mouse
17:12:02 <CornsMcGowan[m]> omg non draggable diagonal rail
17:12:03 <Eddi|zuHause> i always use autorail, but only with dragging
17:12:08 <nielsm> and QWE keys for switching to landscaping tools
17:12:30 <CornsMcGowan[m]> i use autorail/road and yeah QWE landscaping
17:12:42 <Eddi|zuHause> once you start dragging, it's not important where you started the click
17:12:49 <CornsMcGowan[m]> also the ctrl-F6/7/8 for the build tools and the rest of the number row
17:13:53 <nielsm> hm maybe I should derail a valley or two, for a change
17:14:25 <nielsm> they claim to have improved stability a lot with last week's patch
17:14:39 <CornsMcGowan[m]> what do you mean derail a valley :0
17:15:02 <nielsm> Derail Valley, a train driving simulator
17:15:17 <CornsMcGowan[m]> oh yeah that game
17:19:27 <Eddi|zuHause> that's a horrible page
17:19:37 <Eddi|zuHause> it misses some basic introductions
17:19:54 <nielsm> the pathfinder it talks about is a library
17:32:20 <CornsMcGowan[m]> i guess we'll start with step 1: implementing a burst/64k frame limit for construction?
17:32:54 <CornsMcGowan[m]> would y'all accept that into trunk
17:35:20 <nielsm> yeah that'd be a good feature imo, since it would also limit what modified clients can do
17:36:06 <nielsm> tuning the default limits could be a bit time consuming, it would probably involve benchmarking a bunch of skilled players
17:37:57 <dP> imo if there is automatic rail tool it should do a bloody good job
17:38:10 <dP> last thing I want to see is people building spaghetti just because it's faster
17:39:27 <CornsMcGowan[m]> nielsm: how would we go about benchmarking players
17:40:50 <nielsm> implement all the construction limits and just set the limits to infinite, but also add some logging of what values the limits hit
17:41:40 <nielsm> so it's possible to pull statistics like "95% of the time veteran players never go above X burst"
17:44:01 <CornsMcGowan[m]> actually i'll see how straightforward it is to implement this first
17:44:51 <dP> also I still think it's better to make a polyrail smarter than do a fully automatic rail planner
17:45:07 <dP> it's still about as fast for most practical purposes but leaves player with a much more control
17:48:29 <CornsMcGowan[m]> what about limiting the max path length for the rail planner? e.g. max 50 track sections built/planned at once
17:48:40 <CornsMcGowan[m]> also im not sure how to make the polyrail smarter
17:48:52 <andythenorth_> I can't remember what the usability issue was with polyrail
17:48:59 <andythenorth_> something was 10% wrong with the implementation
17:50:10 <dP> starting new polyrail is a bit tricky to figure out but once you do it's great
17:51:36 *** jottyfan has joined #openttd
17:58:54 <CornsMcGowan[m]> nielsm: what actions would construction limit include?
17:59:52 <CornsMcGowan[m]> i don't think stations should be included
18:02:12 <CornsMcGowan[m]> since i think they're the one element that can be built in a rectangle area rather than a line
18:09:17 <CornsMcGowan[m]> okay i'm going to
18:09:32 <CornsMcGowan[m]> implement this feature on my fork
18:09:46 <CornsMcGowan[m]> y'all can get a feel of how it works when its in a playable state
18:10:33 <CornsMcGowan[m]> and if its agreed it shouldn't be in trunk then it can be kept as a patch or proposed to 3rd party clients
18:23:03 <CornsMcGowan[m]> is the maintainer of the mobile app on here?
18:23:58 *** HerzogDeXtEr has joined #openttd
18:24:36 *** Progman has joined #openttd
18:30:45 *** frosch123 has joined #openttd
18:51:08 *** sla_ro|master has joined #openttd
19:29:26 *** b_jonas has joined #openttd
19:40:12 <Gustavo6046> Oh, so that one tune in I think it was Modern Motion... turns out it was a Hotel Giant track. Gee, I never knew it, until I saw it in a video completely unrelated to OpenTTD. I was like, "whaaa"
20:28:06 <DorpsGek_III> [OpenTTD/OpenTTD] techgeeknz closed pull request #8255: Prepare: Update core function headers for refactor of GUI rendering https://git.io/JJUDi
20:28:06 <DorpsGek_III> [OpenTTD/OpenTTD] techgeeknz commented on pull request #8255: Prepare: Update core function headers for refactor of GUI rendering https://git.io/JJTT8
20:28:32 <b_jonas> Minningwell Ridge local authority doesn't allow this. Hey town council, did you know that I own all the roads here, I built them for you before I made you rich from oil, so I can just destroy enough of them to place my railway tracks. Joke's on you, local authority, with your insatiable hunger of trees! I love being a tycoon with lots of cash and no morales.
20:31:32 <DorpsGek_III> [OpenTTD/OpenTTD] techgeeknz commented on pull request #8255: Prepare: Update core function headers for refactor of GUI rendering https://git.io/JJTTg
20:32:52 <DorpsGek_III> [OpenTTD/OpenTTD] techgeeknz commented on pull request #8255: Prepare: Update core function headers for refactor of GUI rendering https://git.io/JJTT6
20:35:47 <nielsm> what do you do when your diesel loco is low on fuel and you know the route will be mostly downhill for the next several km?
20:36:13 <nielsm> you of course switch off the engine to save fuel, and pray to have enough brake reservoir for managing the speed downhill
20:36:59 <andythenorth_> you need regen :P
20:40:10 <DorpsGek_III> [OpenTTD/OpenTTD] techgeeknz commented on pull request #8257: Refactor GUI rendering: Add documentation and name `Viewport` consistently https://git.io/JJTTh
20:40:54 <b_jonas> another one built? just how many oil refineries does this map need?
20:42:21 <Wolf01> Bah, redeemed only 3 games on this month's humble choices
20:42:43 <milek7> nielsm: brake reservoir?
20:42:59 <DorpsGek_III> [OpenTTD/OpenTTD] LordAro commented on pull request #8255: Prepare: Update core function headers for refactor of GUI rendering https://git.io/JJTkT
20:44:27 <nielsm> pressurised air tank filled by the locomotive's compressor, used to recharge the brake pipe pressure when brakes are released
20:50:43 <DorpsGek_III> [OpenTTD/OpenTTD] techgeeknz commented on pull request #8257: Refactor GUI rendering: Add documentation and name `Viewport` consistently https://git.io/JJTk8
20:51:37 <DorpsGek_III> [OpenTTD/OpenTTD] techgeeknz closed pull request #8257: Refactor GUI rendering: Add documentation and name `Viewport` consistently https://git.io/JJUyq
20:54:02 <DorpsGek_III> [OpenTTD/OpenTTD] LordAro commented on pull request #8257: Refactor GUI rendering: Add documentation and name `Viewport` consistently https://git.io/JJTk2
20:55:47 <frosch123> the brexit story from yesterday is finished
20:56:12 <DorpsGek_III> [OpenTTD/OpenTTD] techgeeknz commented on pull request #8255: Prepare: Update core function headers for refactor of GUI rendering https://git.io/JJTkw
20:56:46 <DorpsGek_III> [OpenTTD/OpenTTD] techgeeknz closed pull request #8256: Refactor GUI rendering (based on JGRPP) https://git.io/JJUDh
20:58:10 <LordAro> i think i broke him :(
20:58:30 <LordAro> i don't think i was unreasonable?
21:00:27 *** otetede has joined #openttd
21:01:45 <DorpsGek_III> [OpenTTD/OpenTTD] techgeeknz commented on pull request #8257: Refactor GUI rendering: Add documentation and name `Viewport` consistently https://git.io/JJTkS
21:01:51 <frosch123> i don't know the whole story. but when outside people reformat/apply style, they have always run into a wall here
21:02:46 <frosch123> i hope none encouraged them to do so
21:03:28 <nielsm> here the issue is mainly piggybacking large-scale "cosmetic" changes on functional changes
21:03:38 <LordAro> i wanted them to split a couple of unrelated changes out into separate PRs
21:03:50 <LordAro> but they've just built 3 PRs on top of each other
21:03:50 <nielsm> and separating them out is a huge chunk of work
21:03:54 <LordAro> which defeats the point
21:04:15 <nielsm> one approach would be just to submit one of those PRs at a time
21:04:27 <nielsm> and submit the next when the first is merged
21:05:03 <LordAro> that would work, except their "first" PR has stuff that goes in the second PR
21:05:09 <LordAro> (adding unused functions and the like)
21:05:38 <nielsm> yea, and juggling code chunks around between commits can be a massive chore
21:10:58 <glx> he seems motivated enough to do it right
21:12:54 <nielsm> we need to chunk out the entire graph thing and do statistics over from scratch
21:13:05 <LordAro> that is such a James103 issue
21:15:52 <LordAro> and not that i'm trying to beat him or anything, but it took me all of 5 minutes and 3 sed commands to do the viewport renames
21:16:07 <LordAro> it's trivial to split those changes out
21:17:09 <andythenorth_> 8253 is perfectly reproduced, accurate, perceptive, creative diagnostics
21:17:17 <andythenorth_> shame it's just not really in any way significant
21:18:17 <DorpsGek_III> [OpenTTD/OpenTTD] glx22 closed issue #8204: cmake: make install DESTDIR="something" ignores GLOBAL_DIR option https://git.io/JfM9j
21:28:39 <dP> andythenorth_, yeah, if only people reported crashes and desyncs like that xD
21:32:06 <glx> at least for crash we usually have the log and the dump (nice bonus when we also get the reproducing steps)
21:32:35 <glx> but yeah james' reports are nicely made
21:42:04 <DorpsGek_III> [OpenTTD/OpenTTD] glx22 commented on issue #8249: CMake: Not possible to cross-compile due to settinsgen/strgen https://git.io/JJvbJ
23:40:49 <DorpsGek_III> [OpenTTD/OpenTTD] techgeeknz opened pull request #8259: Fix: Globally apply preprocessor directive coding style https://git.io/JJTYb
23:42:26 <DorpsGek_III> [OpenTTD/OpenTTD] techgeeknz opened pull request #8260: Codechange: Spell 'Viewport' consistently https://git.io/JJTYj
23:59:01 <DorpsGek_III> [OpenTTD/OpenTTD] LordAro commented on pull request #8259: Fix: Globally apply preprocessor directive coding style https://git.io/JJTOP
23:59:59 <DorpsGek_III> [OpenTTD/OpenTTD] LordAro commented on pull request #8260: Codechange: Spell 'Viewport' consistently https://git.io/JJTOM
continue to next day ⏵