Back to Home / #openttd / 2020 / 05 / Prev Day | Next Day
#openttd IRC Logs for 2020-05-23

---Logopened Sat May 23 00:00:15 2020
00:22-!-keoz [~keikoz@2a01:e35:2fd5:51e0:f8f9:92f8:8ddf:16d3] has quit []
02:10-!-snail_UES_ [] has quit [Quit: snail_UES_]
02:11-!-cHawk- [~cHawk@] has joined #openttd
02:11-!-cHawk- is "realname" on #debian #openttd #debian-next
02:14-!-cHawk [~cHawk@] has quit [Ping timeout: 480 seconds]
02:52-!-andythenorth [] has joined #openttd
02:52-!-andythenorth is "andythenorth" on #openttd
03:08-!-adikt [~adiktz@2601:280:5a80:49e0:b036:ba0b:2f17:e218] has quit [Remote host closed the connection]
03:10-!-sla_ro|master [] has joined #openttd
03:10-!-sla_ro|master is "slamaster" on @#sla #openttd
03:30-!-Progman [] has joined #openttd
03:30-!-Progman is "Peter Henschel" on #openttd
03:33-!-adikt [~adiktz@2601:280:5a80:49e0:fc79:9b:909d:5db6] has joined #openttd
03:33-!-adikt is "realname" on #openttd #debian-next #debian
03:33-!-andythenorth [] has quit [Read error: Connection reset by peer]
03:37-!-andythenorth [] has joined #openttd
03:37-!-andythenorth is "andythenorth" on #openttd
03:39-!-Wolf01 [] has joined #openttd
03:39-!-Wolf01 is "Wolf01" on #openttd
03:39-!-adikt [~adiktz@2601:280:5a80:49e0:fc79:9b:909d:5db6] has quit [Remote host closed the connection]
03:42-!-adikt [~adiktz@2601:280:5a80:49e0:545f:406e:f2cc:d17a] has joined #openttd
03:42-!-adikt is "realname" on #openttd #debian-next #debian
03:42-!-Samu [] has joined #openttd
03:42-!-Samu is "realname" on #openttd
03:58-!-keoz [~keikoz@2a01:e35:2fd5:51e0:f8f9:92f8:8ddf:16d3] has joined #openttd
03:58-!-keoz is "Grmph" on #openttd
04:05<Samu>Yexo, do you
04:08<Samu>the WaterPF requires a different builder than mine
04:23-!-Gustavo6046 [~Gustavo60@2804:14d:4cd8:8380::2] has quit [Remote host closed the connection]
04:24-!-Gustavo6046 [~Gustavo60@2804:14d:4cd8:8380::1] has joined #openttd
04:24-!-Gustavo6046 is "Gustavo6046" on #openttd
05:57-!-keoz [~keikoz@2a01:e35:2fd5:51e0:f8f9:92f8:8ddf:16d3] has quit []
05:58-!-WormnestAndroid [] has quit [Remote host closed the connection]
05:58-!-WormnestAndroid [] has joined #openttd
05:58-!-WormnestAndroid is "WormnestAndroid" on #openttd
06:05-!-nielsm [] has joined #openttd
06:05-!-nielsm is "Niels Martin Hansen" on #openttd
06:17<andythenorth>if a newgrf procedure stores to a temp register, will the register value persist after the procedure returns?
06:24<nielsm>the temps are allocated per callback invocation, as I understand it
06:24<nielsm>so when the callback completes, the temps are lost, not before
06:32<andythenorth>that sounds hopeful
06:33<andythenorth>I'm going to try and make FIRS compile faster
06:33<andythenorth>it's only 18 seconds for me from clean, but still
06:33<andythenorth>faster is nice
06:33<andythenorth>pypy makes an insane difference to FIRS compile time
06:59<andythenorth>macos won't compile
07:01<andythenorth>older rev compiles
07:02<andythenorth>I haven't done a full bisection, just went to last commit before recent std:: changes
07:11<+michi_cc>andythenorth: make clean
07:11<+michi_cc>Seems to be a failure in our dep detection.
07:11<andythenorth>thanks :)
07:12<andythenorth> I did configure, didn't think about clean
07:15*andythenorth wonders if spritelayout can use procedures
07:15<andythenorth>already uses expressions
07:19<andythenorth>I particularly want to get rid of expressions like this
07:20<DorpsGek_III>[OpenTTD/OpenTTD] michicc updated pull request #8091: Add: [Script] Native priority queue; useful e.g. for pathfinders.
07:20<andythenorth>they are generated hundreds of times into firs.nml
07:20<DorpsGek_III>[OpenTTD/OpenTTD] michicc commented on pull request #8091: Add: [Script] Native priority queue; useful e.g. for pathfinders.
07:48<andythenorth>the nml expressions here break my brain :)
07:48<andythenorth>I didn't write them
07:54<DorpsGek_III>[OpenTTD/nml] FLHerne opened pull request #150: Fix #147: rounding issues
07:57<LordAro>FLHerne: something in the regression test wouldn't go amiss
07:57<FLHerne>LordAro: ...there is?
07:57<LordAro>FLHerne: i should scroll down the whole diff
07:58<DorpsGek_III>[OpenTTD/nml] FLHerne commented on pull request #150: Fix #147: rounding issues
08:28-!-glx [] has joined #openttd
08:28-!-mode/#openttd [+v glx] by ChanServ
08:28-!-glx is "Loïc GUILLOUX" on +#openttd
09:31-!-Progman [] has quit [Remote host closed the connection]
10:08<andythenorth>hi :)
10:11<andythenorth>is it worth using an intermediate grfcodec step for FIRS
10:12<andythenorth>only saves 5s max
10:12<andythenorth>probably not
10:12<FLHerne>andythenorth: No
10:12<andythenorth>typical compile is < 14s now
10:13<andythenorth>that is really quite good
10:13<FLHerne>andythenorth: What would make FIRS a lot faster would be if you directly built an NML AST
10:13<FLHerne>Currently you spend about half the time templating all your numbers into a wall-of-text, and then parsing said wall-of-text
10:13<andythenorth>is that a practical thing to do? o_O
10:14<FLHerne>I mean, it's all Python
10:14<FLHerne>So yes
10:14<FLHerne> will make it less ugly
10:17<FLHerne>I suppose it would be a bit more likely to get broken by nml changes
10:17<FLHerne>But the syntax isn't a stable interface either
10:18<andythenorth>if I was a bit smarter :P
10:18<andythenorth>FIRS *does* have a tree of industry objects, and tile objects
10:18<andythenorth>and some of the callbacks are in a tree structure
10:19<andythenorth>there's a bit too much logic in actual nml right now though
10:20<FLHerne>You'd pretty much go through all the templates and replace them with functions that returned a chunk of AST
10:23*andythenorth pretends he knows what an AST is, instead of just what the definition means :P
10:23<DorpsGek_III>[OpenTTD/nml] glx22 approved pull request #148: Version changes.
10:23-!-WormnestAndroid [] has quit [Read error: Connection reset by peer]
10:23-!-WormnestAndroid [] has joined #openttd
10:23-!-WormnestAndroid is "WormnestAndroid" on #openttd
10:24<FLHerne>andythenorth: Look at
10:27<FLHerne>andythenorth: My suggestion is to pretty much do what that does, except without the intermediate "write everything and parse it" step
10:27*andythenorth wonders whether to look for more 0.1s to take off FIRS
10:27<andythenorth>or just buy a computer that is not slow ass
10:27<andythenorth>I downgraded my CPU significantly in March :|
10:28<andythenorth>how would we get it into nml to write out the result? pickle? Or import nml modules to FIRS?
10:28<DorpsGek_III>[OpenTTD/nml] FLHerne merged pull request #148: Version changes.
10:29<FLHerne>andythenorth: The latter
10:29<nielsm>import the nml modules into FIRS and generate the parse tree directly
10:29<nielsm>then you can have nml output the parse tree to nml format if you want an .nml file
10:31-!-D-HUND [~debdog@2a00:79c0:637:e600:7a24:afff:fe8a:d04d] has quit [Quit: Initiating getting-the-hell-out-of-here maneuver!]
10:32<FLHerne>That might need a bit of improvement if he wants a *nice* nml file :p
10:36<andythenorth>currently I'm favouring the 'buy a not slow computer' theory :P
10:38<FLHerne>glx: Were you planning to post that unused-parameter optimization as a PR?
10:39<andythenorth>hmm should Intel Turbo Boost apply for single-threaded Python?
10:40<FLHerne>In principle
10:40-!-debdog [~debdog@2a00:79c0:637:e600:7a24:afff:fe8a:d04d] has joined #openttd
10:40-!-debdog is "Wowbagger" on #openttd
10:40<FLHerne>They have turbo for any number of cores now, but highest freq if only one loaded
10:40<FLHerne>But only if it's not too hot
10:41<DorpsGek_III>[OpenTTD/nml] FLHerne commented on pull request #150: Fix #147: rounding issues
10:50-!-sla_ro|master [] has quit []
10:59<DorpsGek_III>[OpenTTD/OpenTTD] michicc updated pull request #8091: Add: [Script] Native priority queue; useful e.g. for pathfinders.
11:13<DorpsGek_III>[OpenTTD/nml] glx22 opened pull request #151: Add: Ignore unreferenced named parameters
11:28-!-snail_UES_ [] has joined #openttd
11:28-!-snail_UES_ is "Jacopo Coletto" on #openttd
11:33-!-Smedles [] has quit [Quit: No Ping reply in 180 seconds.]
11:33-!-Smedles [] has joined #openttd
11:33-!-Smedles is "Paul Smedley" on #openttd
11:37<FLHerne>> return ((input * this->multiplier) + (round && this->shift != 0 ? 1 << (this->shift - 1) : 0)) >> this->shift;
11:38<FLHerne>Given that `round && <anything truthy>` is always 1, I feel like this is broken
11:40<FLHerne>No, I misunderstood the precedence
12:11<+michi_cc>planetmaker: Performance is usually used as the reason for not doing "dynamic" track sprites, but if you look at what a crapload of stuff the catenary drawing does to get crossings right, I don't see any performance problem with a few more vars.
12:11<+michi_cc>Of course, somebody still has to code it.
12:30-!-sla_ro|master [] has joined #openttd
12:30-!-sla_ro|master is "slamaster" on @#sla #openttd
12:32<DorpsGek_III>[OpenTTD/nml] FLHerne updated pull request #150: Fix #147: rounding issues
12:35-!-Progman [] has joined #openttd
12:35-!-Progman is "Peter Henschel" on #openttd
12:38<DorpsGek_III>[OpenTTD/nml] FLHerne commented on pull request #150: Fix #147: rounding issues
12:59-!-Flygon [~Flygon@2001:44b8:411e:4e00:607d:4c58:d259:3b78] has quit [Quit: A toaster's basically a soldering iron designed to toast bread]
13:12<DorpsGek_III>[OpenTTD/nml] FLHerne updated pull request #150: Fix #147: rounding issues
13:22<andythenorth>hmm FIRS makefile might be clown shoes
13:23<andythenorth>the Iron Horse one was fixed for me
13:29<andythenorth>hmm different kind of clown shoes :D
13:29<andythenorth>that's saved 10% of compile time
13:36-!-Progman [] has quit [Remote host closed the connection]
13:58-!-snail_UES_ [] has quit [Quit: snail_UES_]
14:15-!-patrick[m]1 [~patrickne@2001:470:1af1:101::5be2] has joined #openttd
14:15-!-patrick[m]1 is "" on #openttd #biz
14:47<DorpsGek_III>[OpenTTD/nml] FLHerne commented on pull request #151: Add: Ignore unreferenced named parameters
14:53<andythenorth>I cut 1MB off firs.nml
15:11-!-adikt [~adiktz@2601:280:5a80:49e0:545f:406e:f2cc:d17a] has quit [Remote host closed the connection]
15:12-!-gelignite [] has joined #openttd
15:12-!-gelignite is "realname" on #openttd
15:13-!-adikt [~adiktz@2601:280:5a80:49e0:fc79:5d38:a078:347a] has joined #openttd
15:13-!-adikt is "realname" on #openttd #debian-next #debian
15:41-!-Progman [] has joined #openttd
15:41-!-Progman is "Peter Henschel" on #openttd
15:43<andythenorth>reduced redundancy
15:43<andythenorth>eliminated repeated spritesets
15:43-!-snail_UES_ [] has joined #openttd
15:43-!-snail_UES_ is "Jacopo Coletto" on #openttd
15:44<andythenorth>also a bunch of repeated code for setting terrain tile
15:44<andythenorth>it's snail_UES_ :)
15:58-!-adikt [~adiktz@2601:280:5a80:49e0:fc79:5d38:a078:347a] has quit [Remote host closed the connection]
16:01-!-adikt [~adiktz@2601:280:5a80:49e0:3907:321e:ef1c:7de3] has joined #openttd
16:01-!-adikt is "realname" on #openttd #debian-next #debian
16:11-!-jback [] has joined #openttd
16:11-!-jback is "Jasper Backer (jhq)" on #openttd #linode #debian-next #debian #virt
16:16-!-snail_UES_ [] has quit [Quit: snail_UES_]
16:16-!-snail_UES_ [] has joined #openttd
16:16-!-snail_UES_ is "Jacopo Coletto" on #openttd
16:17-!-snail_UES_ [] has quit []
16:17-!-snail_UES_ is "Jacopo Coletto" on #openttd
16:17-!-snail_UES_ [] has joined #openttd
16:17-!-snail_UES_ [] has quit []
16:24<andythenorth>sub 10s FIRS compile with primed caches :)
16:24<andythenorth>that's really not bad
16:25<andythenorth>grfcodec would not be much faster
16:25<andythenorth>ok grfcodec is sub 2s, but still
16:52<Samu>@seen yexo
16:52<@DorpsGek>Samu: yexo was last seen in #openttd 22 hours, 58 minutes, and 13 seconds ago: <Yexo> GN
17:14<supermop_Home>andythenorth is the hst car like a luxury car?
17:25<andythenorth>supermop_Home intermediate
17:25<andythenorth>it has a small payment decay bonus
17:28-!-Laedek [~quassel@] has quit [Quit: Laedek]
17:36-!-sla_ro|master [] has quit []
17:39-!-Samu [] has quit [Quit: Leaving]
17:48<supermop_Home>cant decide whether these 5 tile wyvern hauled luxury cars should become HSTs or something banal
17:49<andythenorth>what's the class 50 thing?
17:49<andythenorth>what version are you playing? :)
17:50<andythenorth>wait for the Onslaught I reckon
17:52<supermop_Home>I've got the onslaught
17:52<supermop_Home>it seems overkill for 5 tiles?
17:53<supermop_Home>vanguard seems best incremental upgrade
17:53<andythenorth>it is
17:53<supermop_Home>but i think i might need to switch to regular cars to increase capacity
17:53<andythenorth>but it's deliberately a bit vanilla and dull
17:53<supermop_Home>maybe electrify the line and use the Olympic emu thing?
17:53<supermop_Home>not a super busy line though
17:55<andythenorth>use a snowplough
17:55<andythenorth>silly andythenorth
17:56<supermop_Home>bit busy here
17:57<supermop_Home>currently DMUs turn back in the center platform
17:57<andythenorth>such road types :)
18:00<supermop_Home>those freights going to the harbor really should be electric
18:03-!-Wormnest [] has quit [Ping timeout: 480 seconds]
18:09*andythenorth sleeping time
18:09<andythenorth>I just started a new game, but ma eyes are closing
18:10-!-andythenorth [] has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
18:16-!-gelignite [] has quit [Quit: Stay safe! Stay at home! Stop the chain reaction!]
18:39-!-Laedek [~quassel@] has joined #openttd
18:39-!-Laedek is "Laedek" on #openttd
18:48-!-Wolf01 [] has quit [Quit: Once again the world is quick to bury me.]
19:18-!-nielsm [] has quit [Ping timeout: 480 seconds]
19:19<Speeder>how hard is to create a new economy for FIRS?
19:19<Speeder>I am making a Map about Brazil
19:19<Speeder>and kinda sad that half of the relevant industries are in one economy, and the other half is in the other
19:19<Speeder>don't even need new industries, just merge existing ones
19:20<Speeder>(although I would love to have a nuclear powerplant... Brazil has only one but it is wildly famous, maybe because there is only one)
19:20<Speeder>(that said Brazil export a lot of nuclear fuel, so would be good to be able to do that too)
19:21<FLHerne>Speeder: Code-wise, trivial
19:22<supermop_Home>Speeder "a lot" of nuclear fuel is still usually pretty small in terms of train loads per year,
19:23<FLHerne>Add a new file under src/economies with the list of cargos, add `industry.economy_variations['NEW_ECONOMY'].enabled = True to the industries you want to be in it
19:23<FLHerne>And a couple of strings
19:23<FLHerne>AIUI, the hard part is getting the balance right so it's fun
19:24<FLHerne>And drawing all the pixels, of course
19:24<supermop_Home>which is the main reason andy hasn't included any nuclear industry chains.. it's just not that interesting at Openttd scale
19:24<Speeder>well, for nuclear fuel maybe join all of the minerals as one, "Rare Earths"
19:24<Speeder>Brazil is one of the biggest exporters of "rare earths"
19:25<Speeder>FLHerne, my idea is merge a bunch of "Hot Country" into "Extreme"
19:25<Speeder>specially cassava and coffee
19:25<Speeder>since those are staple foods here
19:27<FLHerne>Makes sense to me
19:28<Speeder>where is firs source?
19:28<FLHerne>Although Extreme is pretty big already, so you might want to drop something...
19:28<Speeder> <<< gives 404
19:29<FLHerne>(master won't build with current nml, you need the v4 branch or one of the other recent ones)
19:29<FLHerne>(or older nml)
19:29<Speeder>also would have to change "beet"
19:31<Speeder>how I figure out what v4 does different?
19:31<Speeder>or how useful, or not... it is?
19:33<FLHerne>Ask andy tomorrow is probably easiest
19:33<FLHerne>Rephrasing more clearly:
19:34<FLHerne> - The 'master' branch of FIRS can't be compiled with the current version of nml
19:34<FLHerne>So you can compile the 'v4-development-track' or another recent branch with current nml
19:34<Speeder>just saw FIRS 4 will remove extreme and hot country
19:35<Speeder>that... kinda is totally opposite of what I want :P
19:35-!-snail_UES_ [] has joined #openttd
19:35-!-snail_UES_ is "Jacopo Coletto" on #openttd
19:35<Speeder>so I won't use v4 as base
19:35<FLHerne>Or one of the tagged FIRS releases with an older nml version
19:35<Speeder>need to choose another branch
19:35<FLHerne>sorry, v4-release-track is the current branch
19:40-!-snail_UES_ [] has quit [Quit: snail_UES_]
19:49-!-snail_UES_ [] has joined #openttd
19:49-!-snail_UES_ is "Jacopo Coletto" on #openttd
20:12-!-adikt [~adiktz@2601:280:5a80:49e0:3907:321e:ef1c:7de3] has quit [Remote host closed the connection]
20:25-!-Progman [] has quit [Remote host closed the connection]
21:17-!-adikt [~adiktz@2601:280:5a80:49e0:60a8:622b:25ad:c85b] has joined #openttd
21:17-!-adikt is "realname" on #openttd
21:22-!-Wormnest [] has joined #openttd
21:22-!-Wormnest is "Wormnest" on #openttd
21:35-!-gnu_jj [] has quit []
21:43-!-snail_UES_ [] has quit [Quit: snail_UES_]
22:02-!-gnu_jj [] has joined #openttd
22:02-!-gnu_jj is "jj" on #ceph-devel #ceph #openttd
22:27-!-tokai [] has joined #openttd
22:27-!-tokai is "Christian Rosentreter" on #openttd
22:27-!-mode/#openttd [+v tokai] by ChanServ
22:31-!-Flygon [~Flygon@2001:44b8:411e:4e00:b5ef:b499:cd1:4cdb] has joined #openttd
22:31-!-Flygon is "Flygon" on #openttd
22:34-!-tokai|noir [] has quit [Ping timeout: 480 seconds]
22:45<Speeder>I can't f igure how to make my msys2 install pillow :(
22:45<Speeder>whenever I tyr it complains there is no libjpeg
22:45<Speeder>but I installed all the ones I could find
22:53-!-D-HUND [~debdog@2a00:79c0:66b:6700:7a24:afff:fe8a:d04d] has joined #openttd
22:53-!-D-HUND is "Wowbagger" on #openttd
22:54<Speeder>msys2 is just nuts
22:57-!-debdog [~debdog@2a00:79c0:637:e600:7a24:afff:fe8a:d04d] has quit [Ping timeout: 480 seconds]
23:05-!-glx [] has quit []
23:22-!-Wormnest [] has quit [Ping timeout: 480 seconds]
23:31<Speeder>why building FIRS is so hard?
23:31<Speeder>old nml doesn't run
23:31<Speeder>new nml refuses to compile it
23:40-!-snail_UES_ [] has joined #openttd
23:40-!-snail_UES_ is "Jacopo Coletto" on #openttd
23:50<Eddi|zuHause>well, $someone once told me i shouldn't bother making nml backwards compatible when i attempted to do so
23:50<Eddi|zuHause>"maintaining two branches is easy" they claimed
---Logclosed Sun May 24 00:00:17 2020