IRC logs for #openttd on OFTC at 2021-04-05
⏴ go to previous day
00:20:05 *** Progman has quit IRC (Remote host closed the connection)
00:23:16 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
00:23:34 *** WormnestAndroid has joined #openttd
00:41:04 *** gelignite has quit IRC (Quit: Stay safe!)
00:54:52 *** HerzogDeXtEr has quit IRC (Read error: Connection reset by peer)
01:20:19 *** spnda has quit IRC (Quit: Page closed)
01:32:24 *** jottyfan has joined #openttd
01:36:19 *** Rainy has quit IRC (Remote host closed the connection)
01:53:48 *** gregdek has quit IRC (Ping timeout: 480 seconds)
01:54:39 *** reldred has quit IRC (Ping timeout: 480 seconds)
01:56:02 *** gregdek has joined #openttd
01:56:14 *** reldred has joined #openttd
02:15:12 *** glx is now known as Guest576
02:21:38 *** Guest576 has quit IRC (Ping timeout: 480 seconds)
02:41:02 *** Gustavo6046 has quit IRC (Read error: Connection reset by peer)
02:41:16 *** Gustavo6046 has joined #openttd
02:53:53 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
02:54:39 *** WormnestAndroid has joined #openttd
03:27:33 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
03:27:56 *** WormnestAndroid has joined #openttd
04:37:27 *** Wuzzy has quit IRC (Quit: Wuzzy)
05:47:31 *** snail_UES_ has quit IRC (Quit: snail_UES_)
05:59:34 <DorpsGek> [OpenTTD/OpenTTD] kneekoo commented on pull request #8936: Add: Plural support for Romanian translations https://git.io/JY9iQ
06:07:18 <DorpsGek> [OpenTTD/OpenTTD] kneekoo commented on pull request #8936: Add: Plural support for Romanian translations https://git.io/JY9Xf
06:15:13 *** sla_ro|master has joined #openttd
07:23:32 *** heffer has quit IRC (Quit: heffer)
07:29:32 *** jottyfan has joined #openttd
07:34:29 *** supermop_Home has quit IRC (Remote host closed the connection)
07:37:53 *** andythenorth has joined #openttd
07:54:33 <peter1138> Why am I up so late?
07:55:26 *** Progman has joined #openttd
07:58:10 <peter1138> Ah, OzTrans is also a WinXP user...
08:04:23 <andythenorth> we used to have hold-out customers on IE6
08:04:39 <andythenorth> I seriously considered buying them £300 laptops to replace their government-issued ones
08:05:03 <andythenorth> would have been a lot cheaper overall than supporting IE 6 was
08:09:18 <TrueBrain> michi_cc: you think we can include zstd support in 1.11.1 too? Well, PR still hasn't been reviewed .... but maybe this poke helps someone to do just that :P
08:10:09 <TrueBrain> lol @ WinXP .. tempted to reply: it would be wildly irresponsible for us to support such an old system :P
08:10:33 <TrueBrain> I am seriously worried that people still use it .. scary
08:11:08 <LordAro> what was the reason we dropped it? was it just that the CI/VS didn't support it?
08:12:13 <_dp_> interestingly my first zstd tests on citymania didn't go as well as in the theory
08:12:39 <_dp_> didn't have time to investigate but somehow zstd does 2-3x savegame size
08:12:59 <_dp_> and dowload speed seems to be limited by something else than just connection speed
08:13:57 <LordAro> 2-3x savegame size was expected, no?
08:15:08 <andythenorth> we can't replace trees with something else because savegame?
08:15:22 <TrueBrain> replace trees with hot air balloons?
08:15:33 <_dp_> LordAro, no, it was 120-145% in tests compared to lzma:2 and on citymania i got 4.5MB vs 1.21
08:16:32 <peter1138> andythenorth, I logged into a government website the other day which did not work in Firefox, Edge or Chrome. It did work in IE11, though.
08:16:43 <andythenorth> peter1138 tax dollars at work
08:16:49 <andythenorth> replace trees with pre-generated tree combinations
08:16:55 <LordAro> peter1138: not a gov.uk site, i hope?
08:16:58 <andythenorth> 128 would be plenty :P
08:17:06 <_dp_> btw, if anyone wants to try it you can patch available formats in cmclient and join citymania, it will use zstd if forced to
08:17:18 <_dp_> I won't be able to do anything this week unfortunately
08:17:40 <LordAro> didn't factorio post something about optimising a similar tree problem? perhaps could copy them
08:18:19 <peter1138> TrueBrain, it's still there in my Windows 10 install.
08:18:23 <andythenorth> I did the maths once on how many combinations we could get for 3 trees per tile
08:18:30 <andythenorth> I dunno where I put it
08:18:41 <_dp_> but so far I'm using this order for compressions: "lzma:0 zlib:2 lzo:0 zstd:1"
08:18:57 <peter1138> buried under defra.gov.uk
08:20:30 <andythenorth> how much tree storage is there currently?
08:21:21 <andythenorth> counter and stuff
08:21:49 <TrueBrain> I am mostly curious what the "tree problem" is :P
08:22:05 <TrueBrain> it seems there is some consensus about it, but .. what is the problem? :D
08:22:15 <peter1138> Apparently trees are not very compressable, it's almost random data.
08:22:33 <andythenorth> allegedly it (1) bloats savegame size (2) may contribute to map gen time (3) not very newgrf-able
08:22:44 <TrueBrain> I would guess none of the _m data is compressable really, as most is "random" (from the perspective of the compression)
08:23:02 <TrueBrain> we use bits for everything :P
08:23:05 <peter1138> I think the rest of it largely follows patterns, though.
08:23:09 <_dp_> in lzma:2 about 3/4 of savegame size is due to trees
08:23:18 <andythenorth> discussions here about (3) tend to end up talking about (1) and (2) due to interests of participants here
08:23:27 <andythenorth> i.e. it's more interesting to talk about map and compression than grf
08:23:58 <TrueBrain> I also wonder about 2) .. as mapgen does tileloop on every tile; are tiles with trees slower to tileloop?
08:24:23 <TrueBrain> or is placing the part that is slow? :D
08:24:26 <TrueBrain> so many questions .. sorry :P
08:24:52 <peter1138> Just placing, and I don't think it's slow, it's just slower than not doing it, logically.
08:24:58 <_dp_> well, they do something so should be slower but not significantly
08:25:18 <_dp_> also if they're allowed to spread they do some neighbor checks
08:25:32 <peter1138> Generating industries contributes to map gen time. It's not exactly a problem. Until using ECS :D
08:25:46 <andythenorth> newgrf trees would allow cool stuff like slope-sensitivity or forests vs small woods without patching core game
08:25:49 <TrueBrain> rivers are horribly slow in mapgen too :P
08:25:59 <andythenorth> could we stop at "rivers are horrible"?
08:26:13 * andythenorth not pissing on anyone there, I helped do rivers
08:26:31 <andythenorth> took fricking ages relative to quality of result
08:26:40 <peter1138> Every time I tried making map gen more interesting I gave up because a generalised perlin noise is actually quite a bit slower.
08:26:56 <andythenorth> "just use small maps" :D
08:26:58 <TrueBrain> that is now possible :P
08:27:15 <TrueBrain> and in general, people don't seem to mind slow mapgen (when looking at other games), just the lack of preview is annoying
08:27:20 <TrueBrain> as it might mean you need to generate a few new maps
08:27:23 <andythenorth> I tried generating 4k x 4k to test mac performance to answer a forum question, it's SOOOOOOO slow
08:27:34 <andythenorth> generally 512x512 is near instant
08:27:53 <TrueBrain> and if TGP was a decent perlin noise, you could just show a low-res preview
08:27:54 <_dp_> fft noise seems quite fast
08:28:31 * andythenorth plays Arctic map
08:28:48 <TrueBrain> peter1138: honestly, we could cheat by embedding a hard-coded noise table, which is shifted around by randomness
08:28:51 <TrueBrain> should be very quick :D
08:28:55 <TrueBrain> (just takes more disk-space)
08:29:15 <peter1138> Reticulating Splines
08:29:41 <andythenorth> fetch the noise from bananas as content :P
08:29:52 <TrueBrain> 4k by 4k is at least a 16M noise sprite :P
08:30:30 <_dp_> didn't benchmark it much though
08:30:38 <peter1138> Random-As-A-Service
08:30:47 <peter1138> Or XKCD method. -4.
08:31:02 <TrueBrain> "return 4; // Fair dice roll"
08:31:10 <peter1138> Dunno where the - came from
08:33:37 <peter1138> Heh, IE11's "About Internet Explorer" actually just gives you an old-school About Window... for Windows 10.
08:40:13 <peter1138> milek7, XP-build maintainer for life, sorry you did this to yourself...
08:52:13 <TrueBrain> reminds me how you used to do Minecraft Servers .. start local client, download modpack, zip the whole folder, copy to server, run server
08:52:19 <TrueBrain> people got so lazy with OpenTTD :)
08:56:36 <andythenorth> they are stoopid and frequent :)
08:56:54 <andythenorth> I can code generate some horrible tile check that will be really slow and have errors in
08:59:09 <TrueBrain> it is not difficult if we use relay-network + lobby system .. just possibly a bit expensive for us
08:59:33 <TrueBrain> orudge: AWS budget warning triggered, current estimate is 70 dollar more than last month. How are the donations going? :D
09:19:08 *** Zuu has quit IRC (Ping timeout: 480 seconds)
09:20:39 *** gelignite has joined #openttd
09:32:09 <andythenorth> far as I can see, when using transfers, cargo packet age isn't reset for each leg of a journey?
09:32:58 *** FLHerne has joined #openttd
09:45:19 <nielsm> yes that's intentional
09:46:28 <nielsm> the income for the transport is calculated by direct distance from source to destination, not as distance traveled
09:46:51 <nielsm> so you also need to consider how long the total trip was, not each individual leg
09:50:48 <Eddi|zuHause> andythenorth: transfer credit is a lie. the only thing that matters is the age and distance for the complete journey
09:51:18 <andythenorth> so the current strategy is to massively nerf 'local' vehicles like metro
09:51:43 <andythenorth> by reducing the cargo aging period horribly
09:52:03 <andythenorth> so with cdist, on a 3-leg route between 2 big cities
09:52:11 <andythenorth> i.e. metro -> intercity train -> metro
09:52:22 <andythenorth> what's gonna happen by the time the passengers are on the intercity train?
09:54:52 <DorpsGek> [OpenTTD/OpenTTD] michicc commented on pull request #8934: Change: NewGRF window apply button to disable if changes not made https://git.io/JYH3j
09:55:15 <DorpsGek> [OpenTTD/OpenTTD] ghisvail commented on pull request #8948: Fix: Honor default soundfont for FluidSynth on Debian https://git.io/JYHsJ
09:55:32 * andythenorth thinks that the pax will already be aged
09:56:15 <andythenorth> so cargo aging differences between different types of intercity trains will have no effect
09:56:40 <andythenorth> 'complex systems are fun'
09:57:13 <nielsm> cdist doesn't have a good solution for it, since cargo doesn't have destinations, so there are no plans for the total route
09:57:14 <Eddi|zuHause> what exactly do you mean? example: "local" has an aging speed of 4 per day, and "intercity" 1 per day. the trip takes 2 days local, 10 days long distance, and 3 days local. total trip time is 15 days, total cargo age will be 2*4+10*1+3*4 = 30
09:57:54 <Eddi|zuHause> effectively, the cargo is treated as having travelled twice as long
09:58:15 <andythenorth> where's the floor under aging take effect?
09:58:45 <Eddi|zuHause> that sentence does not parse
09:59:12 <andythenorth> cargo has a maximum age
09:59:19 <andythenorth> for the payment algorithm
09:59:29 <TrueBrain> michi_cc: I was way ahead of you :P :P
09:59:47 <TrueBrain> by some seconds ... :D
09:59:53 <Eddi|zuHause> yeah, not sure. probably something in the order of 2 years travel time
10:00:00 <andythenorth> Eddi|zuHause context might help, I think Iron Horse breaks cdist networks
10:00:02 <michi_cc> TrueBrain: Double comment replies now.
10:00:24 <JGR> Days in transit is capped at 255, as it is stored in a byte
10:00:31 <michi_cc> TrueBrain: All cgNATs I know are not symmetrical as otherwise you wouldn't save public IPs.
10:00:32 <TrueBrain> anyway, us needing TCP for STUN is the biggest drawback
10:00:33 <peter1138> Hmm, if JGRPP has better map gen, can we borrow it?
10:00:44 <TrueBrain> michi_cc: hmm, the ones I know are :P Funny :D
10:00:49 <Eddi|zuHause> JGR: so 255*2.5 days
10:00:51 <TrueBrain> why wouldn't it save public IPs?
10:01:10 <nielsm> TrueBrain stop using TCP, switch to UDP, support classic NAT punching
10:01:12 <Eddi|zuHause> @calc 255*2.5/365
10:01:12 <DorpsGek> Eddi|zuHause: 1.7465753424657535
10:01:13 *** HerzogDeXtEr has joined #openttd
10:01:28 <TrueBrain> isn't symmetrical only assigning an unique (ip, port) pair?
10:01:34 <TrueBrain> nielsm: I await your PR for that :D
10:02:44 <michi_cc> TrueBrain: Hmm, it seems that term is absolutely not properly defined and various sources mean different things.
10:02:57 <TrueBrain> I am not surprised :P
10:03:18 <JGR> peter1138, might be worth be worth liaising with reldred. He's added all the recent mapgen related stuff in my branch.
10:03:54 <FLHerne> added all the recent mapgen stuff in his branch, apparently
10:04:02 <Eddi|zuHause> andythenorth: "ridiculously" lowering the aging period means you reach the cap earlier
10:04:42 <andythenorth> Horse ages metro cargo every 32 ticks
10:05:01 <Eddi|zuHause> default is 185 ticks
10:05:12 <DorpsGek> Eddi|zuHause: 5.78125
10:05:20 <Eddi|zuHause> so roughly 6 times faster
10:05:26 <DorpsGek> [OpenTTD/DorpsGek] TrueBrain opened pull request #47: Add: translate Discussion events into IRC messages https://git.io/JYHGz
10:05:45 <andythenorth> but still 3 months
10:06:08 <Eddi|zuHause> @calc 255*2.5*32/185/365
10:06:08 <DorpsGek> Eddi|zuHause: 0.3021103295075898
10:06:28 <Eddi|zuHause> @calc 255*2.5*32/185/365*12
10:06:31 <DorpsGek> Eddi|zuHause: 3.625323954091078
10:06:43 <Eddi|zuHause> 3 months is... short
10:06:59 <michi_cc> From googling around, UPD hole punching in generall seems simpler than TCP hole punching, simply because NAT gateways can't really do connection tracking with UDP.
10:07:13 *** HerzogDeXtEr1 has joined #openttd
10:07:24 <TrueBrain> there is nothing to track :P
10:07:33 *** HerzogDeXtEr1 has quit IRC ()
10:07:34 <TrueBrain> in AWS if an UDP stream goes silent for 60 seconds, it is terminated
10:07:37 <TrueBrain> which was fun to find out :D
10:07:42 * andythenorth considering removing all use of cargo aging
10:07:51 <andythenorth> either by setting it all to 185
10:07:55 <andythenorth> or setting it all to 0
10:08:18 <FLHerne> reldred: Anyway, the original question was "does JGRPP have better map generation, and can it be pulled upstream?"
10:08:32 <TrueBrain> nginx does similar things, to track UDP streams
10:08:47 <Eddi|zuHause> andythenorth: i think the original idea was to allow making the aging period longer
10:08:56 <TrueBrain> (and yes, I had to relay UDP through AWS for the master-server .. was ... enjoyable :D)
10:09:07 <andythenorth> some kind of balancing for large maps
10:09:19 <andythenorth> could have been done differently, but eh
10:09:29 <reldred> FLHerne: it’s not that different to what Truebrain has done for 1.11, I just allowed a few things normally tied up in fixed values to new settings
10:09:49 <andythenorth> setting cargo aging to 0 is just too weird, and breaks too much else in the game?
10:11:18 <Eddi|zuHause> i don't think that even works
10:11:19 <michi_cc> There are various libs for emulating TCP over UDP, most notably QUIC :)
10:11:35 <TrueBrain> the Steam relay system is also fully UDP
10:11:39 <TrueBrain> and has a "TCP-like" mode
10:11:43 <TrueBrain> they did a really nice job
10:12:02 <reldred> FLHerne, peter1138 there were four main patches I contributed to JGRPP; 1st: Rainforest height line, now made redundant by Truebrain's work on 1.11 terrain gen. 2nd: Town zone generation customization. Basically there's a fixed table up used to determine the size of zones up until there's 92 houses, then it's a flat calculation, I just broke that out so you can control it with fixed numbers.
10:13:50 <reldred> FLHerne, peter1138, 3rd: River settings; basically just allowed tweaking a few things, min length of rivers, lake probability, lakes allowed in desert, tropic band surrounding rivers in desert, and lastly this one weird condition that required rivers to generate around the tops of hills (within 2'ish tiles) which combined with another condition that requires rivers to also neighbor a higher tile result in not many generating
10:14:07 <orudge> TrueBrain: that won't be a problem I think! Donations are higher certainly just now (not extravagently so, but will cover the extra usage I'm sure)
10:14:28 <reldred> FLHerne, peter1138, and the last patch I submitted was controlling rock density. Flat number for size, or scale size/density by height.
10:15:12 <reldred> Really all the stuff I did was super low hanging fruit because I'm not a C++ programmer. Just had time to burn. JGR fixed all my screwups with saves, fixed some bugs in the settings menu, etc.
10:15:31 <reldred> It's Yet More Settings, which not everyone likes
10:16:16 <reldred> But I happen to like it, and I made it, and JGR accepted my rubbish PR's so... ¯\_(ツ)_/¯
10:16:20 <peter1138> And I still want to replace "variety" with something that works a lot better.
10:16:55 <reldred> Yeah, I was looking at poking around in TGP but then TrueBrain just coughed up his PR and I basically breathed a sigh of relief.
10:17:18 <andythenorth> variety does kinda work, I just used it in map gen, got what I expected
10:17:31 <reldred> I do want to take a look at Improved Trees though, I think some simple number tweaks could really make it shine.
10:17:44 <andythenorth> it does make weird squares at certain map settings
10:17:57 <reldred> Yeah, it's really prevalent on Tropic
10:18:48 <DorpsGek> [OpenTTD/OpenTTD] michicc merged pull request #8944: Fix #8935: [OSX] Crash when clicking 'Save' due to wrongly-threaded OS call. https://git.io/JYycG
10:19:03 <DorpsGek> [OpenTTD/OpenTTD] michicc merged pull request #8945: Fix: Apply master effect volume during mixing instead of sound start. https://git.io/JYyVO
10:26:11 <reldred> But anyway, yeah, I'm happy to backport my patches to OpenTTD trunk just don't expect miracles. I don't necessarily have a clue what I'm doing.
11:13:13 <TrueBrain> "It's apparent that you have used Ad-words marketing to promote your business in the past" <- say what now? We did what? When? Stupid scam mails to info@
11:14:15 <DorpsGek> [OpenTTD/OpenTTD] michicc commented on pull request #8845: Feature: zstandard compression for network games https://git.io/JYHRe
11:14:37 <DorpsGek> [OpenTTD/OpenTTD] michicc commented on pull request #8845: Feature: zstandard compression for network games https://git.io/JYHRU
11:14:51 <FLHerne> TrueBrain: Are they also trying to sell you quadcopters in bulk?
11:14:59 <FLHerne> I've been getting some pretty odd spam recently
11:15:28 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #8845: Feature: zstandard compression for network games https://git.io/JYHRG
11:16:47 <peter1138> Plugged a USB keyboard in. Windows 10. Please reboot.
11:20:01 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #8845: Feature: zstandard compression for network games https://git.io/JYHRx
11:20:21 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #8845: Feature: zstandard compression for network games https://git.io/JqnwH
11:21:54 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #8845: Feature: zstandard compression for network games https://git.io/JYH03
11:46:39 <DorpsGek> [OpenTTD/OpenTTD] JGRennison opened pull request #8951: Fix graphical issues with dragging measurement tooltips in the viewport https://git.io/JYHzN
11:47:15 *** FLHerne has quit IRC (Ping timeout: 480 seconds)
11:49:39 <nielsm> I smashes "backport requested" on a bunch more PRs
11:49:54 <nielsm> someone else should go them over and check everything looks reasonable
11:50:27 *** JGR has quit IRC (Quit: Back later)
12:01:49 <andythenorth> idea seems cool, implementation I am the wrong person to comment
12:02:19 <TrueBrain> implementation is simple :P
12:05:03 <milek7> what's the problem with TCP?
12:05:41 <nielsm> very difficult to connect two machines behind each their own NAT directly via TCP
12:12:42 *** frosch123 has joined #openttd
12:13:25 <peter1138> Who doesn't have their own IP address space anyway?>
12:17:42 <nielsm> ugh... during an interactive rebase, is there a way to go one step back... I put two commits in the wrong order
12:17:54 <nielsm> if I just swap them it will work
12:20:10 <peter1138> Don't think so, I think you'll need to restart
12:20:31 <peter1138> Maybe with git reflog but easy to go wrong.
12:23:58 *** snail_UES_ has joined #openttd
12:29:08 <TrueBrain> nielsm: git rebase --abort :P
12:29:20 <TrueBrain> the only way I know ..
12:31:13 <LordAro> complete rebase & swap them back again?
12:32:51 <nielsm> it was a situation of converting checkpoint commits into fixups, and putting two in wrong order so both would cause huge conflicts
12:33:17 <nielsm> and the entire thing was 50 commits long so aborting would be hell too
12:33:33 <nielsm> I ended up just eating the conflicts and resolving them in the stupidest way possible
12:33:44 <nielsm> since they'd get fixup'd later again anyway
12:36:41 <nielsm> argh who ordered SNOW IN APRIL
12:40:47 *** jellyknight has joined #openttd
12:41:52 <EER> Hi all, taking a bit of a look at bananas-api today and while trying to get up and running with docker I noticed that --cdn_url is mandatory and not passed in the Dockerfile.
12:42:46 <EER> So either I'll add a default value to the dockerfile and improve the error message when it's not specified, or I can try and make it not mandatory?
12:43:03 *** Gustavo6046 has quit IRC (Ping timeout: 480 seconds)
12:43:25 <EER> My python is not great (I'm experimenting) so I've gone for the default value for now so I can get it up and running at least :)
12:47:28 *** gelignite has quit IRC (Ping timeout: 480 seconds)
12:52:14 *** FLHerne has joined #openttd
12:55:25 <andythenorth> so delete the luxury pax cars?
12:57:18 <andythenorth> they have lower capacity than standard pax cars
12:58:25 <TrueBrain> EER: when starting the docker, you can add arguments; so adding it works for now
12:58:41 <TrueBrain> I will check why it is not there / why it is mandatory soon. Can't remember which is the correct approach :)
12:59:33 <TrueBrain> owh, you already have it running, sorry, missed that part
13:00:14 *** supermop_Home_ has joined #openttd
13:00:20 <peter1138> andythenorth, "for aesthetics only"
13:01:20 <nielsm> but isn't horse meant to only have actually useful things?
13:01:21 <EER> TrueBrain, thanks, yup got it to run at least, I remember also seeing some example basic bananas repository somewhere to help developers get set up easier, but can't seem to find it?
13:01:39 <EER> I cloned the existing Bananas repo, which was clearly a mistake :')
13:01:55 <TrueBrain> for the API, cloning the BaNaNaS repo is fine :)
13:02:02 <TrueBrain> only for bananas-server that is a bit of a problem
13:02:25 <andythenorth> peter1138 I could just rename them to something else
13:02:30 <andythenorth> 'alternative pax cars'
13:02:33 <andythenorth> and level the capacity
13:02:48 <andythenorth> I did that maybe 100 times already for freight wagons :P
13:03:20 <andythenorth> because 'luxury' implies they do something
13:03:52 <reldred> I mean, I honestly don't see any problem with how they're working now.
13:04:06 <EER> Got it, now to start tinkering :)
13:06:54 <andythenorth> reldred why would you use them at all?
13:07:02 <andythenorth> they make no sense
13:07:29 <peter1138> Lots of this game doesn't make sense.
13:08:05 <reldred> Eh, I dunno there seems to be cases in some of the maps I play where they end up being more/less profitable for certain routes.
13:09:15 <reldred> Same as restaurant cars, I never really had a problem with how you implemented them.
13:09:41 <reldred> There were edge cases where they seemed to make money, there were other cases where they caused trains to bleed money.
13:10:28 <andythenorth> did you try and work out why they were profitable?
13:10:28 <reldred> I dunno man, you seem very hard on yourself sometimes, that if you're ideas aren't 100% perfect all the time you have to purge them from ever existing.
13:10:47 <FLHerne> andythenorth: Players do all kinds of weird superstitious things because they're "better" according to some private hunch
13:11:07 <FLHerne> No-one cares whether it actually helps, because who cares, it's fun
13:11:27 * andythenorth playing a different game TBH
13:11:48 <andythenorth> I've been playing against the newgrf spec for 10 years
13:11:58 <andythenorth> the game is beating the spec
13:12:22 <reldred> I mean, that's the thing, this game is a lot of different things to a lot of different people
13:16:37 <andythenorth> I could rename them 'crap pax cars' :)
13:16:42 <andythenorth> as they are much much worse on every measure
13:16:50 <TrueBrain> michi_cc: just cheaper :P
13:19:29 <michi_cc> It's open source and you can self-host as far as I know.
13:19:45 <TrueBrain> told me another story :D
13:20:16 <TrueBrain> one doesn't rule out the other btw :P
13:21:35 <TrueBrain> funny how they do not mention the open source option on Pricing
13:21:42 <TrueBrain> michi_cc: see the gist :P Already in there :D
13:22:01 <michi_cc> Well, pricing seems to be for cloud hosting, not self-hosting.
13:23:26 <TrueBrain> you would think you also say on top: Free if you self-host!
13:23:33 <TrueBrain> but this might be the better way to make a buck more :)
13:24:01 <TrueBrain> in general I have a hard time reading these websites
13:24:16 <TrueBrain> just small pieces of information you ahve to glue together :D
13:24:22 <andythenorth> can we expose the cargo payment props (10, 11, 12) to vehicle newgrf, as global vars?
13:24:47 * andythenorth considering reimplementing the payment algorithm per vehicle, and using it to adjust running costs
13:24:48 <TrueBrain> but yeah, michi_cc , I will check it out
13:24:52 <TrueBrain> if someone already wrote all the stuff for us
13:24:55 <TrueBrain> why bother doing it again :P
13:25:04 <TrueBrain> do you have experience with this product, or how did you find it?
13:25:49 <andythenorth> I don't really want to copy the pax payment rates for all industry grfs into Iron Horse
13:25:55 <andythenorth> seems a bit overkill
13:26:46 <Eddi|zuHause> you should probably back out like 3 levels and rethink your original problem
13:27:15 <michi_cc> TrueBrain: I know that Paradox uses it for at least some of their games (probably the Enterprise version :)
13:27:40 <TrueBrain> I am still not much wiser what it does exactly .. docs are not .. browse-friendly
13:27:59 <TrueBrain> seems to be HTTPS based
13:28:05 <andythenorth> Eddi|zuHause I think it's simple, if I can just get the payment rates, and a var for days since cargo was loaded, and the current age of the loaded cargo
13:28:20 <andythenorth> I can adjust the run costs of pax car vs. luxury pax car
13:28:28 <michi_cc> Initially they just used Steam I believe, but as soon as you start adding other platforms/consoles into the mix, Steam alone doesn't cut it anymore.
13:28:30 <andythenorth> so they both earn the same for the same trip
13:28:42 <andythenorth> despite having different capacity
13:28:55 <Eddi|zuHause> that sounds wrong on so many levels
13:29:06 <andythenorth> I might need to be able to calculate run cost on every tick or so also
13:29:24 <andythenorth> we don't have perm. storage on vehicles, I might need that
13:29:35 <andythenorth> no it's probably not needed actually
13:30:16 * andythenorth wonders how much this will break with daylength
13:30:23 <andythenorth> I might need to know the daylength factor also
13:30:25 <TrueBrain> michi_cc: it seems to be a bit overkill for our needs :P
13:30:53 *** jottyfan has quit IRC (Quit: jottyfan)
13:31:47 <TrueBrain> the thing I try to avoid, is us having our own accounting system, honestly :P
13:31:49 <milek7> there's also possiblity of using webrtc library
13:31:51 <milek7> then it would have emscripten build interop :D
13:33:28 <TrueBrain> and what I like about "join-keys", is that it works with all Rich Presence systems too
13:33:34 <TrueBrain> you just have to automatically exchange it
13:33:37 <TrueBrain> and someone can join your server
13:33:43 <TrueBrain> via what-ever network connected method is available
13:34:06 <TrueBrain> giving the master-server a coordination role
13:34:21 <frosch123> the other day someone wrote here, that discord api is MIT licensed, but i cannot back that up with facts
13:35:25 <TrueBrain> contains several fixes for memleaks etc
13:35:30 <TrueBrain> against their old MIT licensed SDK
13:37:29 <frosch123> ok, but that API does not include the networking/relay part, right?
13:37:47 <TrueBrain> no, and that part of Discord is currently not "available" to us
13:38:00 <TrueBrain> for over a year now Discord doesn't accept new games
13:38:03 <TrueBrain> haven't read up on how/why
13:39:01 <nielsm> I actually have a registration of OpenTTD as a game... I think unless they deleted it for inactivity
13:39:01 <peter1138> I'm not even sure what that refers to, "commerce"?
13:39:43 <frosch123> so we have to wait until ms buys discord, and see what happens next :p
13:43:30 <nielsm> TrueBrain: I invited you to the OpenTTD team on discord developer stuff
13:44:00 <TrueBrain> I saw the mail, will check in a bit :)
13:44:03 <TrueBrain> now first, DOTA!!!!! :D
13:49:59 <Timberwolf> andythenorth: I think that might be a little too cynical.
13:50:39 <andythenorth> if you can think of improvement...I won't start an edit war :)
13:51:43 <Timberwolf> I think the caution I'd go for is, "this has the effect of scaling the cargo payment decay graph in the horizontal axis, and may not necessarily result in more (or less) being earnt for certain combinations of cargo, route length, vehicle speed and other contributory factors such as transfer credits."
13:51:55 <Timberwolf> Or something along those lines.
13:51:55 <EER> I'm trying to get the bananas-cli to connect to my local server, but I think something is wrong with my authentication. First I got the error that audience was not one of "developer", the audience in CLI is hardcoded to be github in the url, so I updated that to say developer, next I am getting a html response where the CLI expects json. Maybe this is not the intended interface for testing?
13:53:52 <EER> At least I'm learning a lot of Python today ;-)
13:55:08 <andythenorth> cargo aging - another case of apparently-true-but-misleading information in forums with a long tailwind
13:55:20 <andythenorth> the days I've wasted on that crap :D
13:57:45 <TrueBrain> EER: I am guessing you are trying the dart version of bananas-cli, not bananas-frontend-cli we provide?
13:57:58 <TrueBrain> anyway, if you want to do local authentication based on GitHub, you need to register your own app at GitHub
13:58:06 <supermop_Home_> andythenorth just play goldeneye
13:58:20 <TrueBrain> I have no clue what you are trying btw, I am just reading snippets :D
13:58:25 <TrueBrain> but you can use staging we provide fine for testing
13:59:06 <andythenorth> supermop_Home_ warcraft 1 orcs and humans
13:59:38 <peter1138> Is it too cold for cycling: 1) yes, 2) also yes
13:59:53 <andythenorth> do some lat stretches or something
14:00:12 * andythenorth failed at lunch, and is eating cornflakes
14:00:16 <supermop_Home_> c&c modem game?
14:00:27 <andythenorth> supermop_Home_ Doom coop?
14:00:58 *** Wormnest has joined #openttd
14:01:38 <peter1138> I "made" some bread, so lunch was ok
14:01:54 * andythenorth somewhat jealous
14:02:20 <EER> TrueBrain: I tried using the web version, and found I had to do github registration, which I couldn't be bothered about at first, so then tried the CLI tool to list packages instead, which gives me the above
14:02:33 <andythenorth> oh DB Set beats the newgrf spec by setting 1st class pax cars to higher capacity than 2nd class
14:02:52 <EER> I'd like to pick up one of the beginner issues in bananas_server, but first trying to set up a working environment so I know what I am looking at :)
14:03:21 <EER> so right now the status is "get it up and running and solve any problems I find on the way there"
14:05:57 <TrueBrain> if you want to developer for the server, don't bother with github, basically :)
14:06:04 <TrueBrain> just set the authentication method to "developer"
14:06:08 <TrueBrain> much easier to deal with :D
14:06:29 <EER> Do I do that on the API side?
14:06:39 <EER> I think that's already the case
14:07:40 <EER> I had to manually patch the cli code to ask for developer audience, which leads me to believe this is not really the way to go?
14:08:28 <EER> I think the server should return a 302 with a redirect to that url, instead of printing html output
14:08:40 <EER> (based on how the cli handles the http response)
14:09:33 <EER> (note that while I'm plenty familiar with programming and HTTP, I have almost no experience with python, but hey I know that it cares about spaces and what more do you need really)
14:11:03 <EER> The api already selects developer, because in the docker file these parameters are given: `"--user", "developer", "--client-file", "clients-development.yaml"`
14:12:25 <DorpsGek> [OpenTTD/OpenTTD] BloodyIron commented on issue #8947: [Feature Request] Key binds, changing all of them (also, WASD movement) https://git.io/JYSUC
14:14:20 <andythenorth> there must be a way to mathematically fake higher cargo income for some vehicles
14:14:26 <andythenorth> it can't be impossible
14:15:04 <andythenorth> the pax payment rate is known in FIRS, the algorithm is known, the vehicle can do maths
14:15:30 * andythenorth can't figure out the implementation
14:17:05 <TrueBrain> EER: which cli are you using? bananas-frontend-cli? I believe he is onlyt designed for github
14:24:55 <peter1138> I wonder if town-density should also scale by water amount...
14:30:32 <andythenorth> pls send me your newsletter :)
14:30:55 * andythenorth can't remember the rest of the quote, Simpsons
14:33:17 <EER> TrueBrain: hacked in "developer" flow support \o/
14:36:39 <andythenorth> hmm, what if I give luxury pax cars some power value?
14:37:06 <andythenorth> I might be able to calculate how much power is needed so that the train earns more
14:37:18 <andythenorth> kinda depends on the player's route though
14:39:17 <andythenorth> can I measure slope?
14:41:11 <andythenorth> can GS track individual trains? :)
14:41:32 <andythenorth> and can GS modify the payment calculation when it runs?
14:42:53 <TrueBrain> EER: that would btw be a first good issue to solve in frontend-cli, that the developers flow also works :D
14:42:59 <nielsm> should I do a draft PR of my social plugin patch?
14:43:01 <TrueBrain> I only tested it against staging, I think
14:44:11 <EER> I'll tidy it up and send a PR for it :)
14:44:48 <TrueBrain> nielsm: go for it; at least gives a point to start talking from honestly
14:47:39 <TrueBrain> I believe LordAro had a friend that already started the Steam part, so maybe that helps get a bit of perspective nielsm :)
14:48:25 <FLHerne> peter1138: So you get constant towns per land area? I like that idea
14:49:23 <peter1138> Currently if I use 50-75% sea level I have to manually adjust the number of towns, because Very Low becomes too much.
14:56:38 <peter1138> Hmm, of course, generating towns is uncoupled from the terrain generator.
14:57:39 <peter1138> And also, industry scaling as well.
15:02:38 <andythenorth> how about novelty pax coaches, instead of luxury
15:02:52 <andythenorth> playground coach
15:03:09 <TrueBrain> nielsm: give or take what I would expect from the API indeed
15:03:24 <TrueBrain> just how to register plugins etc, that will be the interesting part now :D
15:03:51 <TrueBrain> not a fan doing it via registry, tbh
15:04:53 <DorpsGek> [OpenTTD/bananas-frontend-cli] erenes opened pull request #20: Support developer authentication flow in bananas-cli https://git.io/JYQTb
15:05:24 <EER> I always forget about the commit message rules :/
15:05:31 <nielsm> TrueBrain yeah I'm thinking maybe better just some files following some name pattern in the program folder
15:05:35 <TrueBrain> don't worry too much about it EER , we can fix that :)
15:05:43 <TrueBrain> nielsm: its own folder, I would say
15:05:57 <TrueBrain> location of SocialJoinRequestedGame feels weird, in network_gui.cpp :)
15:06:03 <TrueBrain> but okay, minor details for a code review
15:06:06 <TrueBrain> the concept is more important atm :D
15:07:54 <andythenorth> supermop_Home_ what is the crazy idea with changing restaurant car capacity using luxury cars?
15:08:01 <andythenorth> apart from it's mad...there might be something in it
15:08:04 <TrueBrain> LordAro: the repo you linked for Steam integration is no more? :(
15:09:04 <TrueBrain> the commit hashes still work :P
15:09:16 *** Compu has quit IRC (Remote host closed the connection)
15:09:57 <TrueBrain> not the client_name I think; that is also a nice addition :D
15:10:15 <nielsm> those two aren't really compatible with the idea of having multiple social plugins
15:11:09 <TrueBrain> hmm .. partially, I guess
15:11:15 <TrueBrain> Discord doesn't do browser, I would guess
15:11:23 <TrueBrain> and if I start via Steam, I would expect to use the username from Steam
15:11:42 <TrueBrain> guess some callbacks are broadcast
15:11:50 <TrueBrain> others are first come first serve?
15:11:58 <nielsm> that would also mean we need to recognize that the player has never manually entered a player name
15:12:06 <TrueBrain> that code is already there :)
15:12:19 <TrueBrain> owh, no, it overwrites it
15:12:35 <TrueBrain> even less of a problem, just set it to which ever comes first
15:12:40 <TrueBrain> just to get ride of "Player"
15:13:52 <TrueBrain> but honestly, now I think about it a bit more, is having 2 plugins active sensible .. depends a bit what this will do towards the future
15:14:08 <TrueBrain> some parts are some parts are not :D
15:14:11 <TrueBrain> this is a weird world :P
15:14:39 <TrueBrain> you also only want 1 overlay, for example
15:16:21 <peter1138> LordAro, does "East Yorkshire" exist?
15:18:23 *** FLHerne has quit IRC (Read error: No route to host)
15:18:26 *** FLHerne_ has joined #openttd
15:19:01 <LordAro> peter1138: generally known as "east riding"
15:20:47 <peter1138> Trying to get some data validation going on and my colleagues (yeah, bank holidays, lol) are fairly useless.
15:21:28 <peter1138> The latest file *does* mostly validate, but only because 3000 records have had their county changed to "Greater London" which is blatantly wrong.
15:22:23 <peter1138> Bristol, Greater London. Cardiff, Greater London. Yeah.
15:23:13 <andythenorth> I think I've been to East Yorkshire
15:23:22 <andythenorth> where is Robin Hood's Bay?
15:25:24 <LordAro> most people would describe that as North Yorkshire, tbh
15:25:33 <LordAro> "East Yorkshire" is more Hull area
15:26:56 <peter1138> Hmm, I should try selling my cycle carrier roof bars
15:27:14 <andythenorth> they're surprisingly hard to sell when I tried
15:27:26 <peter1138> Yes, people don't trust them.
15:27:39 <andythenorth> roof box sold easily
15:27:49 <andythenorth> the bars...I think I left them on the street
15:27:51 <peter1138> Also these are Thule, so good and expensive and unlikely to get anywhere near what I paid for them.
15:27:57 <andythenorth> standard method round here, most things go if left out
15:29:40 <peter1138> One issue is because I'm 6'6" they are not actually long enough for two of my bikes...
15:30:25 <peter1138> Also the tourer is steel, and weighs 22kg by itself, so getting it up on the roof is somewhat a challenge.
15:32:10 <andythenorth> if it helps, my electric bike is about 99kg
15:32:18 <andythenorth> probably doesn't help :P
15:32:33 <LordAro> i'd be surprised if it was actually that much :p
15:33:01 <supermop_Home_> andythenorth is it a motorcycle?
15:33:12 <andythenorth> it's a legal grey area
15:33:54 <supermop_Home_> apparently Harrogate has a five guys now?
15:34:46 <andythenorth> seems about 25kg for my electric bike
15:42:19 <peter1138> Feels like 40 though...
15:42:39 <peter1138> 25kg isn't much more than my tourer :/
15:43:31 <FLHerne_> peter1138: That's quite a chonky bike
15:43:59 <FLHerne_> Even my old steel tourer is only 16kg, all-up with the rack and mudguards and such
15:44:30 <FLHerne_> I guess 6'6" means a big frame
15:44:58 <peter1138> I could measure it again but cba right now
15:45:09 <peter1138> It does have a dynamo though.
15:45:53 <FLHerne_> Rob's carbon thing is, like, 8
15:46:02 <FLHerne_> Sometimes I make excuses to borrow it :p
15:46:22 *** Gustavo6046 has joined #openttd
15:46:27 <peter1138> Even my carbon bike is chunky, although it was relatively cheap. Also it's about 5 years old now, needs some replacement parts.
15:47:16 <peter1138> But the wrong wheel bearing is wonky, despite being replaced, so it probably needs new wheels, and they are annoyingly, not cheap.
15:48:34 <DorpsGek> [OpenTTD/OpenTTD] ghisvail dismissed a review for pull request #8948: Fix: Honor default soundfont for FluidSynth on Debian https://git.io/JYSRj
15:48:37 <DorpsGek> [OpenTTD/OpenTTD] ghisvail updated pull request #8948: Fix: Honor default soundfont for FluidSynth on Debian https://git.io/JYSRe
15:49:49 <andythenorth> goes it throw out not having a cargo payment multiplier?
15:50:06 <andythenorth> just because mb doesn't like it, does that mean we can't do it?
15:50:07 <DorpsGek> [OpenTTD/OpenTTD] ghisvail commented on pull request #8948: Fix: Honor default soundfont for FluidSynth https://git.io/JYQZ6
15:51:08 *** Flygon has quit IRC (Quit: A toaster's basically a soldering iron designed to toast bread)
15:52:54 <FLHerne_> The *bearing* is wonky?
15:53:03 <andythenorth> return BigMulS(dist * time_factor * num_pieces, cs->current_payment, 21);
15:53:13 <FLHerne_> Doesn't just need the spokes re-trueing?
15:53:31 <andythenorth> 21 is a bit magical?
15:53:45 <peter1138> No, it's the bearing.
15:54:06 <peter1138> Or rather, whatever in the hub the bearing sits on.
15:54:43 * andythenorth adventures in C++
15:57:00 <andythenorth> well apparently this doubles payments
15:57:01 <andythenorth> return BigMulS(dist * time_factor * num_pieces, 2 * cs->current_payment, 21);
15:57:29 <andythenorth> I think this scope is far from any scope that would have the vehicle properties in it
15:57:40 <andythenorth> but that's just a matter of function params yes/no?
15:58:00 * andythenorth is in GetTransportedGoodsIncome
16:02:10 <andythenorth> wait wat, nogo can modify cargo payment?
16:07:17 <andythenorth> look what I did, I hope we're all proud
16:10:09 <peter1138> Are you trying to stuff cargo payment stuff into vehicle properties? o_O
16:10:58 <andythenorth> depends how good my C++ is
16:11:07 <andythenorth> but we already can guess...not good
16:11:43 <andythenorth> it might be funny
16:12:23 <andythenorth> if the multiplier could be cb36, then we could have extra income at christmas for santa specials
16:12:30 <andythenorth> might have to go to depot to trigger that :P
16:12:40 <michi_cc> It might help to actually use the multiplier in GetTransportedGoodsIncome :)
16:13:07 <andythenorth> oh you read the diff? :)
16:13:22 <andythenorth> also I lied to nogo about the cargo payment
16:13:32 <michi_cc> Yes, but then a parameter to GetTransportedGoodsIncome?
16:13:48 <TrueBrain> w00p, we have a new peak of 3254 \o/
16:14:04 <andythenorth> TrueBrain is there a bell we can ring or something?
16:14:13 <TrueBrain> would be fun if we could
16:14:14 <andythenorth> how about the crossing bell sound?
16:14:19 <TrueBrain> still more people play Portal2 btw than OpenTTD
16:14:19 <andythenorth> in a discord audio channel
16:14:19 <michi_cc> But you are not using that parameter :)
16:15:15 <michi_cc> TrueBrain: Either we got a lot more "buys" or SteamSpy is very optimistic right now :)
16:15:25 <TrueBrain> I was just looking at that
16:15:26 <andythenorth> michi_cc I definitely didn't just silently update the diff to remove that 2
16:15:28 <TrueBrain> as it surprised me too :P
16:15:37 <TrueBrain> so not that much more since yesterday :)
16:16:22 <TrueBrain> but stats are slacking behind
16:16:36 <TrueBrain> in 4 hours or so the stats update again to show a more recent picture, so who knows
16:16:59 <TrueBrain> can't wait till the month is over, gives us access to the Hardware Survey
16:17:04 <TrueBrain> gives some insights in what people use
16:17:52 <andythenorth> so it seems I need to get the vehicle in scope for CargoPayment::PayFinalDelivery and CargoPayment::PayTransfer
16:18:08 <TrueBrain> so ... what am I doing tonight
16:18:15 <TrueBrain> I have no active projects, it seems :o
16:18:55 <andythenorth> CargoPayment::CargoPayment seems a bit magical
16:20:28 * andythenorth wonders if VehicleCargoList has the vehicle in scope trivially
16:21:29 *** FLHerne_ has quit IRC (Quit: There's a real world out here!)
16:21:58 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
16:22:01 <andythenorth> "your friend's children are not your friends" too true, IRL
16:22:06 *** WormnestAndroid has joined #openttd
16:24:41 <EER> friend in ye olde VB6 days was what we call internal in C# nowadays (iirc)
16:25:10 <EER> I hope you're not inflicted with vb :p
16:25:38 <andythenorth> I think we can assume I have never touched VB
16:25:41 <nielsm> in C++ a friend is a clas or function that's allowed to touch your privates
16:25:55 <andythenorth> I'm not sure THAT applies in RL
16:26:20 <nielsm> being friends in C++ comes with benefits
16:26:50 <andythenorth> hmm I can't see any way for the VehicleCargoList to get the vehicle it belongs to?
16:27:27 <andythenorth> typedef CargoList<VehicleCargoList, CargoPacketList> Parent; gives the parent class, not the parent object?
16:31:03 <andythenorth> so possibly VehicleCargoList would need to store the multiplier on init?
16:31:16 <andythenorth> then be updated when cb36 triggers
16:31:23 * andythenorth has no clue how to C++
16:37:27 * andythenorth thinks this just fragments OpenTTD
16:37:44 <andythenorth> what's the magic cargo profit CB do, wouldn't that be cleaner?
16:39:07 <andythenorth> would still need the vehicle :|
16:43:07 * _dp_ never have been happier getting a game to crash
16:43:58 <_dp_> means I don't have to install a freaking Visual Studio now that I can debug it in linux xD
16:46:06 <tommy> i've got a lot of money and no ideas, what should i do?
16:47:23 *** tommy has quit IRC (Quit: leaving)
16:49:31 <_dp_> write a negative review on steam
16:49:40 <_dp_> at least that seems to be a popular solution
16:49:54 <_dp_> (for anyone still here don't actually do that, I'm joking)
16:50:11 <DorpsGek> [OpenTTD/bananas-frontend-cli] erenes opened pull request #21: Fix: do not try to authenticate when requesting help (#4) https://git.io/JYQzj
16:51:55 <DorpsGek> [OpenTTD/bananas-frontend-cli] erenes updated pull request #21: Fix: do not try to authenticate when requesting help (#4) https://git.io/JYQzj
16:54:13 <andythenorth> is there some way I can iterate all vehicles to find out which vehicle a VehicleCargoList instance belongs to?
16:55:11 <_dp_> hmm.. why is it spamming [Detaching after fork from child process xxx] in gdb on the main screen?
17:00:04 <peter1138> Maybe it's forking...
17:00:34 <nielsm> I think the autosave process on unix uses fork?
17:03:58 <frosch123> maybe that external midi player
17:04:09 <frosch123> but regular threading does not fork
17:04:39 <andythenorth> hmm CargoPayment::CargoPayment caches a reference to front vehicle?
17:04:48 <andythenorth> can it cache a reference to current vehicle?
17:04:50 <peter1138> When you set everything up to scale by gui... and then you need to not scale at one point... bah :/
17:05:32 <_dp_> it seems to fork like every second
17:07:49 <TrueBrain> nielsm: btw, Rich Presence or Social Presence?
17:07:54 <TrueBrain> the first seems to be used mostly, not?
17:08:55 <nielsm> rich presence via a social plugin?
17:09:00 <peter1138> Social implies you have friends.
17:09:21 <TrueBrain> I haven't read the word "social" in any of the docs I read about this at least :)
17:09:30 <peter1138> Is there an easy opposite for "ScaleGUITrad"?
17:09:33 <TrueBrain> lol @ copyright notice .. you have an old version it seems :D
17:10:05 <TrueBrain> but it states a year, that was funny to me :)
17:10:48 <nielsm> oh, the plugin api header
17:11:01 <_dp_> yeah, it forks in MusicDriver_ExtMidi::DoPlay
17:17:38 * andythenorth is incapable of making choices :x
17:21:42 <peter1138> Hmm, this is nicer.
17:23:15 <peter1138> Hmm, accidentally fixes the control widgets too. "Oops"
17:25:32 <peter1138> Or perhaps breaks. Hmm.
17:26:16 *** jottyfan has joined #openttd
17:27:24 * andythenorth wonders how people write actual code
17:28:08 <LordAro> andythenorth: copious amounts of stackoverflow posts
17:28:25 <LordAro> i supose we should do a 1.11.1 soon, given the macos saving issue
17:28:28 <andythenorth> it means holding 9 or 18 things in your head, where I can hold 3
17:28:43 <andythenorth> I could write flash games because it's all one loop, and it's just scripting
17:29:03 <andythenorth> I hate writing js because it's state strewn all over the DOM, it's events, and there's no real loop
17:29:20 <andythenorth> I can script python compiles because it's only 3 dimensional
17:30:33 <peter1138> I see. Currently with "double" gui zoom, buttons are not actually double size, they are double sprite size. Subtle but distinct difference.
17:32:45 <DorpsGek> [OpenTTD/OpenTTD] perezdidac updated pull request #8934: Change: NewGRF window apply button to disable if changes not made https://git.io/JYPk3
17:32:49 <peter1138> Scrollbars go from 12px to 20px, not 24px. Hmm.
17:32:58 <DorpsGek> [OpenTTD/OpenTTD] perezdidac commented on pull request #8934: Change: NewGRF window apply button to disable if changes not made https://git.io/JYQ6n
17:33:38 <DorpsGek> [OpenTTD/OpenTTD] perezdidac updated pull request #8934: Change: NewGRF window apply button to disable if changes not made https://git.io/JYPk3
17:34:01 <DorpsGek> [OpenTTD/OpenTTD] perezdidac commented on pull request #8934: Change: NewGRF window apply button to disable if changes not made https://git.io/JYQ6B
17:34:05 <peter1138> Any idea how to do 1.5x sprite scaling? :/
17:34:42 <didac> yeah arbitrary even better :D that would be really neast
17:36:44 *** jottyfan has quit IRC (Quit: jottyfan)
17:41:59 <_dp_> I've seen some interesting upscale libs but don't remember where now
17:42:39 <peter1138> Hmm, I guess it would be sufficient to scale padding even if the sprites are not arbitrarily scaled.
17:43:39 <_dp_> nah, not quite, before 4k I was very much missing 1.5 sprite scale
17:46:07 <nielsm> I still occasionally think about adding a vector-based sprite format to GRF
17:46:18 <nielsm> for GUI and font-like things
17:49:05 <andythenorth> we could use something totally crazy like....svg
17:58:27 <peter1138> Have beer, will ponder
17:59:03 *** otetede has joined #openttd
18:07:38 *** FLHerne has joined #openttd
18:07:46 <DorpsGek> [OpenTTD/bananas-api] perezdidac opened pull request #85: Change: add upload-date to packet size validation https://git.io/JYQMM
18:08:55 <DorpsGek> [OpenTTD/bananas-server] perezdidac commented on pull request #43: Add: pass upload date to OpenTTD client https://git.io/JYQDJ
18:12:05 <didac> has anyone tried making it possible to overlap bridges? building one that goes on top of another?
18:12:37 <LordAro> didac: graphical issues is the main blocker, i believe
18:14:32 <didac> yeah I am pretty sure, I tried a while ago by removing the condition that prevents bridges from getting built if there is an existing one, but an assertion kicked. Found out the logic to determine whether there is a bridge is by iterating in every direction until hitting a tunnelbridge head
18:14:54 <peter1138> Yeah, it was done long ago, but not merged.
18:21:54 *** y2kboy23 has joined #openttd
18:29:19 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
18:29:32 *** WormnestAndroid has joined #openttd
18:35:51 <andythenorth> peter1138 so capping stations, would kinda fix my Iron Horse pax coach madness, in the most roundabout and weird way :P
18:36:22 <andythenorth> (station cargo waiting amount)
18:39:01 <DorpsGek> [OpenTTD/OpenTTD] michicc commented on pull request #8934: Change: NewGRF window apply button to disable if changes not made https://git.io/JYQHU
18:39:38 <michi_cc> TrueBrain: I really like the preview stuff, much faster than fetching and compiling everything locally :D
18:44:01 <DorpsGek> [OpenTTD/OpenTTD] perezdidac commented on pull request #8934: Change: NewGRF window apply button to disable if changes not made https://git.io/JYQHM
18:55:44 <peter1138> andythenorth, then you get onto "should newgrf stations be able to change the cap"
18:56:06 <peter1138> and also "what's a reasonable cap"
18:56:32 <peter1138> and also stupid things like "is this cap cargo specific"
18:57:37 <andythenorth> then we have regrets
18:57:56 <andythenorth> but then I cap stations and I don't need 2 kinds of 8/8 pax with high and low capacity :P
18:58:14 <andythenorth> currently I have those, and it causes 'grf is fucked and unbalanced' reports
18:58:57 <andythenorth> the only antidote to cdist pax is VERY HIGH CAPACITY VEHICLES
19:02:30 *** jottyfan has joined #openttd
19:05:20 <peter1138> As some Steam review said, if cargo actually had destinations then it could work better.
19:05:42 <peter1138> But also it's not entirely wrong that people go where the transport goes.
19:05:54 <andythenorth> we can rebuild them!
19:06:17 <peter1138> It's almost nogo terrority though.
19:06:22 <andythenorth> we can 'just' manipulate the weighting
19:06:35 <peter1138> If nogo could assign a destination instead... hmmm....
19:06:42 <andythenorth> it would need to be per cargo, I don't know if JGR solved that yet
19:06:45 <peter1138> Also rebalance, again.
19:06:54 <andythenorth> balance isn't anyway
19:07:18 <peter1138> Isn't that the joke...
19:07:26 <peter1138> Also my beer ran out.
19:07:45 <peter1138> Maybe I should get another, I wouldn't want it to... go off.
19:08:24 <andythenorth> so New Vehicle Running Costs
19:08:36 <andythenorth> they're vehicle running costs, but just a % of earnings
19:08:43 <andythenorth> much better all round
19:08:47 <peter1138> Running costs have long been newgrf tweakable no?
19:08:57 <andythenorth> they even have a setting!
19:09:00 <andythenorth> and base cost mods!
19:09:07 <peter1138> % of earnings... that's unfair, like taxation!
19:09:14 <andythenorth> frosch123 is that a good enough 'auto' ^^
19:09:21 <andythenorth> some vehicles are just more efficient than others :P
19:09:56 <peter1138> % of earnings is a bit meh. Guaranteed to make profit regardless of, well, anythig
19:10:14 <andythenorth> it avoids having to play-test 'balance' and stuff
19:10:24 <andythenorth> which is currently the most tedious part of newgrf dev
19:10:28 <andythenorth> and not one I can automate
19:10:35 <andythenorth> and I can't rely on playerbase for it
19:10:40 <andythenorth> for $reasons about people
19:10:44 <peter1138> Just ask Samu to do it.
19:11:04 <andythenorth> ideally I just paint trains, assign power and stuff
19:11:15 <andythenorth> game does the rest
19:11:32 <andythenorth> no more thinking about maps from 64^ to 4096^2
19:11:41 <peter1138> Never thing about that.
19:11:45 <peter1138> Make it work for 256x256.
19:11:50 <andythenorth> I try to only think about that
19:11:54 <andythenorth> the ultimate size
19:11:59 <andythenorth> all others are worse
19:12:05 <TrueBrain> michi_cc: I might have found a clever way to use STUN with TCP after all :D
19:12:07 <TrueBrain> I have to test it out
19:12:08 <peter1138> No, but consider it the baseline.
19:12:22 <TrueBrain> but I think I can utilize AWS ALB to offload the TCP connections sufficiently :)
19:12:25 <peter1138> If it's out of whack for other sizes, so is everything else...
19:14:11 <andythenorth> what are synonyms for 'luxury' in Luxury Passenger Coach
19:14:19 *** HerzogDeXtEr has quit IRC (Read error: Connection reset by peer)
19:14:24 <andythenorth> that basically mean 'this one looks like Mk2A not Mk3 coach'
19:14:31 <andythenorth> or 'this one has 12 wheels for amusing spriteness'
19:14:43 <andythenorth> it's just like...an alternative coach
19:15:01 <andythenorth> and it happens to have crap capacity also and is entirely non-balanceable
19:15:09 <andythenorth> 'Worse Passenger Coach"
19:15:17 <andythenorth> "Just Relax Passenger Coach"
19:15:44 <andythenorth> "Passenger Coach In The Time of Covid"
19:15:54 <andythenorth> "Socially Distanced Passenger Coach"
19:16:52 <peter1138> Leave it unbalanced. People interested in efficiency will not use it. People who are just playing with a trainset will use it for the visuals.
19:18:43 <andythenorth> it was 'balanced' by cargo aging, and nobody noticed that was irrelevant until LC proved it
19:18:57 <andythenorth> I was aware and quietly filing it down the back of the sofa
19:19:06 * andythenorth has gone full irrational about this now though
19:23:21 <andythenorth> Iron Horse is accidentally completely realistic for coach capacities
19:23:52 <andythenorth> a BR standard MK2 would carry 64 or 48, depending on whether it had arm rests on the chairs or not
19:24:11 <andythenorth> arm rests were not fitted in regions with heavy commuter traffic
19:24:20 <andythenorth> in all other respects coach was identical
19:25:19 <FLHerne> andythenorth: I think the fact that no-one noticed proves peter1138's point
19:26:24 <andythenorth> everyone has proved the point :)
19:26:32 <FLHerne> But that's not the point?
19:26:33 <andythenorth> I have just fallen into a temporary insanity hole
19:26:56 <FLHerne> Can't you just make it, like, *more* anti-cargo-aging?
19:27:13 <FLHerne> Or did I miss the bit where you proved that was impossible because the game is broken
19:27:46 <andythenorth> it's the need to nerf standard vehicles that's weird
19:27:51 <andythenorth> and everything else
19:28:03 <FLHerne> nerf everything, OTTD is way too easy
19:28:22 <andythenorth> to balance two vehicles with same role, but different capacity, the only solution is to make one of them impossible to use above a certain length
19:28:30 <FLHerne> I liked the thing in NARS where it was actually quite easy to slip up and go bankrupt in the first few years
19:28:38 <andythenorth> except that dp proves cargo aging doesn't matter anyway, only distance
19:28:44 <FLHerne> come to think of it, that was some server so there might have been different basecosts too
19:28:48 <andythenorth> NARS used to kill me often
19:30:05 <FLHerne> Is industry funding/prospecting cost a callback?
19:30:16 <FLHerne> i.e. can you make the first ones cheaper?
19:30:43 <andythenorth> not listed as a cb
19:31:02 <andythenorth> currently I have a commit deleting the non-standard pax cars, but I'm not convinced
19:32:18 <DorpsGek> [OpenTTD/bananas-frontend-cli] erenes opened pull request #22: Fix: Use urllib to combine urls (#5) https://git.io/JYQbM
19:33:08 <andythenorth> there must be some amusing solution, there's always a way
19:33:49 <peter1138> Just make pretty graphics. Let users figure out it's not worth using.
19:35:18 <peter1138> Hmm, interesting, not all titlebars are the same height currently.
19:39:29 <TrueBrain> simulating local NAT is very tricky
19:41:11 <peter1138> Easy, just configure one of your routers properly.
19:41:30 <TrueBrain> I need to simulate it, not have it :P
19:41:42 <peter1138> Easiest way to simulate... is to do it :D
19:41:58 <TrueBrain> best flight simulator is sitting in a real plane, gotcha :)
19:42:04 <PhaseV> Hi there, can i ask for help in german?
19:42:20 <peter1138> I have... hmm, 5 routing capable devices at home, although 3 are configured for L2 switching/bridging.
19:42:22 <TrueBrain> is the German for help also help? :D
19:42:38 <TrueBrain> I dunno what the German is "for help"
19:42:46 <TrueBrain> sorry, I am not helping :D
19:43:52 <peter1138> Urgh, where's the code for the graph "keys" windows...
19:44:37 <peter1138> Ah. Legend. Of course.
19:46:12 <PhaseV> I have set up a dedicated server under Linux with my buddies today (Ver 1.11.0) unfortunately we have a few problems. 1. every time the server is restarted, it generates a new map 2. although autosave is off, an autosave is performed and you get a timeout after 20 seconds. How can we fix these problems?
19:48:19 <TrueBrain> 1) you have to save and load that savegame on restart
19:48:34 <TrueBrain> 2) possibly this is an emergency save that is being done, but I am not sure
19:49:13 <peter1138> Dimension GetSpriteSize(SpriteID sprid, Point *offset = nullptr, ZoomLevel zoom = ZOOM_LVL_GUI);
19:49:21 <peter1138> How does that even work, when ZOOM_LVL_GUI isn't a constant?
19:49:24 <PhaseV> 1) Ok, then we have to script this
19:49:50 <TrueBrain> might be nice if dedicated servers could just reload the last save .. hmm .. something to consider for next version
19:50:39 <peter1138> Okay, apparently in C++ they don't have to be constant. Hmm.
19:50:55 <LordAro> peter1138: is it not a constant? what is it?
19:51:11 <peter1138> It's a define, to a static variable.
19:51:23 <peter1138> Worse, it was probably me.
19:52:01 <PhaseV> but how we can prevent the disconnect?
19:52:16 <TrueBrain> PhaseV: what is the error given on the server?
19:53:08 <PhaseV> aftern 20 seconds no reasson from server and the ottd cliend is frozen and disconnects from server
19:53:15 <TrueBrain> LordAro: possible the build folder had a faulty CMake run and is not pointing to the wrong folders? Dunno .. feels weird :P
19:53:19 <michi_cc> peter1138: Default arguments are evaluated at the call site and not during function declaration, which is why they don't need to be strictyl constants.
19:53:40 <TrueBrain> PhaseV: I have no clue.
19:53:42 <peter1138> Yeah, I'm kinda used to C# which is more restrictive.
19:53:53 <michi_cc> This is different to languages like C#/.net, where the default arguments are acutally backed into the function metadata.
19:54:31 <peter1138> Anyway, I basically want to get GetSpriteSize and FONT_HEIGHT_NORMAL but without GUI zoom applied. Hmm.
19:54:56 <glx> TrueBrain: possible cache issue yes
19:55:31 *** nielsm has quit IRC (Ping timeout: 480 seconds)
19:55:31 *** frosch123 has quit IRC (Quit: be yourself, except: if you have the opportunity to be a unicorn, then be a unicorn)
19:55:33 <peter1138> I think it's easy for GetSpriteSize, but FONT_HEIGHT_NORMAL, less so.
19:55:50 <peter1138> Although actually. Hmm.
19:56:13 <peter1138> FONT_HEIGHT_NORMAL isn't scaled by GUI zoom anyway, that's text zoom.
19:56:33 <DorpsGek> [OpenTTD/OpenTTD] perezdidac updated pull request #8934: Change: NewGRF window apply button to disable if changes not made https://git.io/JYPk3
20:03:38 <michi_cc> glx: Sanity check question: Is anything confused by the path starting with '/build'?
20:04:01 <glx> no, cmake doesn't care about the path
20:04:29 <glx> and we have an happy customer :)
20:04:36 <LordAro> glx: is it possible to delete that file when it has that particular error?
20:04:44 <LordAro> or at least clarify the error message
20:05:44 <peter1138> Maybe if I got rid of separate font zoom... it's dumb anyway.
20:06:07 <glx> it should be possible to add a hint in the error message
20:06:09 <TrueBrain> hmm .. can you have a socket for listening and a socket for connecting under Windows? You can under linux with SO_REUSEADDR, but not sure that also works on Windows ..
20:06:56 <didac> hey folks, has anyone tried to add a button to reset all settings to default? I am considering doing it, with a query confirmation popup of course, but I wonder anyone will strongly oppose to that.
20:07:40 <LordAro> TrueBrain: reuseaddr is pretty new, isn't it?
20:07:56 <TrueBrain> but I remember Windows did something different with it
20:08:04 <LordAro> what am i thinking of then...
20:08:08 <TrueBrain> well, it is a security nightmare to begin with, but that is a different store :D
20:08:12 <TrueBrain> SO_EXCLUSIVEADDRUSE ?
20:08:18 <TrueBrain> (still 10 years old :P)
20:08:44 <peter1138> You wouldn't want to kill off Windows XP...
20:08:45 <LordAro> there was something that allowed multiple things listening on the same port, iirc
20:09:22 <TrueBrain> but docs say it should work on Windows
20:09:26 <TrueBrain> so .. I am just going to try, I guess
20:09:38 <LordAro> you're gonna make me google this, aren't you? :p
20:10:08 <TrueBrain> if you can't leave it alone what ever pop'd in your mind, I guess you have to
20:10:13 <TrueBrain> my mind-reading-machine is broken
20:10:17 <TrueBrain> they keep saying: next month
20:10:20 <TrueBrain> but that is clearly a lie :D
20:11:19 <DorpsGek> [OpenTTD/OpenTTD] michicc approved pull request #8934: Change: NewGRF window apply button to disable if changes not made https://git.io/JYQja
20:15:03 <spnda> From LordAro's comment on 8934: What are the plans for the new windows for 1.12, are they documented anywhere?
20:15:20 <spnda> I only saw some mockups from TrueBrain on a new main menu
20:17:12 *** EER has quit IRC (Remote host closed the connection)
20:22:25 <TrueBrain> 28 curators reviewed OpenTTD .. lets see what they say :D
20:24:07 <DorpsGek> [OpenTTD/OpenTTD] michicc merged pull request #8934: Change: Disable NewGRF window apply button if no change was made https://git.io/JYPk3
20:26:26 <TrueBrain> what a horrible interface does Steam have for that :P
20:27:38 <peter1138> Can't be as bad as OpenTTD's interface.
20:27:51 <peter1138> andythenorth, this other beer ran out too. Defective :(
20:30:06 <LordAro> TrueBrain: having reviewed some release notes, i *think* i'm thinking of SO_REUSEPORT, but i'm sure it was more recent than 3.9...
20:31:09 <TrueBrain> how does REUSEPORT differ from REUSEADDR ..
20:33:56 <TrueBrain> now the question .. how am I going to test out if this works, without implementing it completely first :P
20:34:43 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #8955: Change: [CMake] Improve In-source error message https://git.io/JY7fu
20:34:49 <glx> tried to auto delete cache, but it seems cmake recreates it on exit
20:35:03 <TrueBrain> never auto delete :P You will hurt someone somewhere in the future :D
20:35:44 <DorpsGek> [OpenTTD/OpenTTD] glx22 updated pull request #8955: Change: [CMake] Improve In-source error message https://git.io/JY7fB
20:36:07 <TrueBrain> finally someone that fixed my scratch in that file :P
20:39:48 <andythenorth> peter1138 shameful
20:40:00 <andythenorth> but I loaded my Warcraft 1 savegame that I kept losing
20:41:49 <peter1138> Bah, why is this window not resizing...
20:42:29 <andythenorth> it's had too many beers
20:42:49 <andythenorth> become obstreperous
20:43:13 <peter1138> Okay then I'll open the whisky.
20:43:16 <glx> <spnda> I only saw some mockups from TrueBrain on a new main menu <-- that's about all we have for now I think, but it was a starting point
20:43:18 <andythenorth> that will fix it
20:43:37 <andythenorth> I mean...we could just add a newgrf cargo payment multiplier and be done with it eh :)
20:43:48 <andythenorth> it's about 10 simple lines, a savegame bump
20:43:57 <andythenorth> and then something that completely mystifies me :P
20:44:09 <glx> and there may be a gist too spnda
20:44:27 <peter1138> What, for all newgrf?
20:45:55 <andythenorth> do I mean savegame bump?
20:45:59 <spnda> Actually, no, that's not where I saw them
20:46:10 <andythenorth> I mean a cache would have to be primed when loading an old save
20:46:17 <andythenorth> and I might be wrong anyway
20:46:20 <TrueBrain> LordAro: I am happy you found SO_REUSEPORT :P
20:46:36 <glx> and sorry for all the free HLs ;)
20:46:38 <TrueBrain> under Windows, SO_REUSEADDR works, but under Linux it doesn't .. but SO_REUSEPORT does :D
20:51:13 <glx> cross platform is always fun ;)
20:51:28 <andythenorth> we have very different ideas of fun :)
20:56:06 *** Samu has quit IRC (Quit: Leaving)
20:58:00 <TrueBrain> " Only one usage of each socket address (protocol/network address/port) is normally permitted" but it does work :P
20:58:16 <TrueBrain> Python seems to act up a bit
20:59:35 <TrueBrain> can we just drop Windows support?
20:59:40 <TrueBrain> and shows we are not MS fanboys!
21:00:14 <TrueBrain> omg .. and imagine that we only support MS compiler .....
21:00:21 <TrueBrain> this has gone to far
21:00:47 <peter1138> Release on Windows Store oinly
21:03:41 <TrueBrain> okay, SO_REUSEADDR is annoying as .... :P
21:06:18 <TrueBrain> so this trick works fine on Linux, but Windows needs more ... motivation :P
21:06:40 <TrueBrain> you need a TCP socket that is both accepting and connecting, which of course is a bit iffy
21:11:22 <peter1138> You need to make an outbound connection from a specific port?
21:12:27 <TrueBrain> for UDP that is trivial
21:12:31 <TrueBrain> for TCP .. a bit more trickery is required
21:12:38 <peter1138> Magic that may or may not work anyway :D
21:12:42 <TrueBrain> but rumors are it is possible
21:16:56 *** sla_ro|master has quit IRC ()
21:27:00 <TrueBrain> Windows has a really hard time mixing a listening socket with a connecting socket :D
21:28:05 <DorpsGek> [OpenTTD/OpenTTD] LC-Zorg opened issue #8956: Disasters: wrong location shown in factory/refinery destroyed message https://git.io/JY7L2
21:29:58 <LordAro> looks like location of where the vehicle enters the map, rather than the industry
21:31:51 <TrueBrain> luckily there now is a FF that makes it easier to get these kind of disasters :P
21:33:42 <peter1138> Damn, heating turned off, freezing house :/
21:33:55 <peter1138> Maybe I should invest in some insulation
21:35:42 <peter1138> Why do some windows resize based on minimum sizes, and others don't... hmm.
21:36:23 <heffer> SO_REUSEPORT is newer than SO_REUSEADDR and for UDP is kinda broken in older Kernels.
21:36:55 <heffer> SO_REUSEPORT has the benefit of limiting the reusing of a port to the user that initially started the process that bound the socket first.
21:37:46 <heffer> One advantage or SO_REUSEADDR is that you can set this on a port after binding it. So you can bind the port first, set SO_REUSEADDR after and then bind more sockets.
21:38:44 <heffer> This way you can prevent a user from inadvertently starting another, different process (i.e. not openttd) on the same port.
21:39:31 <heffer> Haven't worked with either using TCP but with UDP incoming traffic is evenly distributed across the bound sockets which usually isn't a problem if all connections use the same recv method.
21:40:01 <TrueBrain> if you use SO_REUSEADDR on TCP on Windows, it is undefined where SYN/ACKs end up :P
21:40:06 <TrueBrain> if you bind 2 sockets on the same address
21:42:06 <heffer> The headache of handling that must be mind-bending.
21:43:00 <heffer> I only worked with these settings implementing L2TPv3 tunnels which carry the ID of the connection they belong to in every packet header. So it's kinda obvious what belongs where.
21:44:02 *** Progman has quit IRC (Remote host closed the connection)
21:44:21 <TrueBrain> right, Python proof-of-concept done .. now to test it in the real world .. fun
21:44:30 <peter1138> Oh, because ReInit() tries to keep close to the original size. Hmm.
21:50:43 <peter1138> Probably there's too many ways to set widget size up. Hmm.
21:50:51 <DorpsGek> [OpenTTD/OpenTTD] LC-Zorg opened issue #8957: Sounds of game - zooming out the view in fact doesn't reduce the volume of sounds https://git.io/JY7qV
21:51:51 <TrueBrain> that moment you realise you no longer have any place to SSH to to start a simple test server on a public ip :D
21:52:00 <TrueBrain> sweet .. the day is finally here :D
21:52:42 <TrueBrain> tried the BaNaNaS VPSes, but someone did what I asked him to .. and firewalled the VPSes :P
21:53:14 <TrueBrain> and I don't understand those new firewalls enough to know how to poke a hole :P
21:54:00 <TrueBrain> so how am I going to test this now :P
21:54:32 <DorpsGek> [OpenTTD/OpenTTD] PeterN commented on issue #8957: Sounds of game - zooming out the view in fact doesn't reduce the volume of sounds https://git.io/JY7qV
21:55:08 <TrueBrain> LordAro: yes, that is the correct response from you to me :D
21:56:01 <LordAro> tbh, i'm not sure how to poke a hole either without some (brief) googling
21:56:39 <TrueBrain> awh, http x-forwarded-for header doesn't contain source port , hmm .. there goes that idea
21:57:15 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
21:57:29 *** WormnestAndroid has joined #openttd
22:03:09 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
22:03:32 <TrueBrain> one of the worst things about vim
22:03:37 <TrueBrain> you copy/paste a blob Python
22:03:47 <TrueBrain> and it adds all kind of weird tabs and spaces
22:03:55 <DorpsGek> [OpenTTD/OpenTTD] LC-Zorg commented on issue #8957: Sounds of game - zooming out the view in fact doesn't reduce the volume of sounds https://git.io/JY7qV
22:04:32 *** jellyknight has quit IRC (Quit: Stay safe!)
22:04:47 <heffer> TrueBrain: : set noai fixes that quickly :P
22:06:32 <LordAro> TrueBrain: :set paste
22:07:02 <TrueBrain> LordAro: yeah, found it after calling some search results names
22:07:07 <TrueBrain> people that go fully in depth explaining vim
22:07:12 <TrueBrain> instead of just .. telling me .. wtf I have to do :P
22:07:15 <TrueBrain> like I care about vim
22:07:28 <heffer> Not as much as it cares about you
22:08:40 <glx> peter1138: oh it's from the current opendune maintainer
22:08:57 <TrueBrain> so that is what his name comes from :P
22:10:56 <peter1138> Heh. Anyway, maybe that's something to look at, maybe it's not.
22:11:16 <TrueBrain> "At the moment it is mainly supported by Apple software and hardware."
22:11:21 <TrueBrain> that is my main issue with that solution :(
22:11:33 <peter1138> My hardware supports it and is nothing to do with Apple.
22:11:45 <TrueBrain> LIES on that website!
22:11:51 <TrueBrain> we know who to tell that it has lies! :D
22:11:58 <peter1138> It's not universal, that's for sure.
22:12:14 <peter1138> And some consider it a security issue so disable it.
22:12:28 <peter1138> It won't work for provider-based NAT either, but that's evil shit anyway.
22:12:34 * andythenorth should not read the internet
22:12:38 <TrueBrain> hmm .. I am testing some NATs .. and this NAT uses the same port on the public side as the internal port
22:12:42 <TrueBrain> that is not expected
22:12:43 <peter1138> Why doesn't everyone just have IPv6 yet?
22:13:46 <FLHerne> andythenorth: I don't think they're actually wrong, it's just lacking some tact
22:13:52 <FLHerne> Possibly lost in translation?
22:14:11 <FLHerne> I must admit, I've been finding the new 'Location' toolbar button annoying
22:14:15 <peter1138> I have 9 real IPv4 and a /48 in IPv6.
22:14:22 <FLHerne> More consistent, but it's tiny and hard to click on a laptop
22:14:23 <peter1138> Well, not mine, it's the providers, but routed to me.
22:14:26 <andythenorth> FLHerne it's just unrealistic
22:14:32 <andythenorth> either we can have progress
22:14:36 <andythenorth> or we can have atrophy
22:14:51 <peter1138> My main qualm with the new icon is that I keep sending vehicles to depots by mistake.
22:14:55 <FLHerne> Well, yeah, but that's what feedback is for :p
22:14:57 <TrueBrain> I like "some tact", that is a nice description :D
22:15:06 <FLHerne> They said they liked some things
22:15:27 <DorpsGek> [OpenTTD/OpenTTD] 2TallTyler commented on issue #8956: Disasters: wrong location shown in factory/refinery destroyed message https://git.io/JY7L2
22:16:05 <andythenorth> the idea of soliciting feedback from the player base falls at the first hurdle
22:16:17 <andythenorth> because the playerbase has NFI what it wants
22:16:47 * andythenorth says this in full disclosure, my day job is basically getting public feedback on government policy
22:17:00 <andythenorth> but the public knows what it wants on policy decisions
22:17:32 <andythenorth> players of a free game have no fucking clue
22:18:13 <andythenorth> public knows what it wants
22:18:21 <andythenorth> result is...result
22:18:24 <andythenorth> that's democracy
22:18:30 <andythenorth> game design isn't democracy
22:18:46 <andythenorth> 'feedback' is making commits or reviews
22:19:19 <DorpsGek> [OpenTTD/OpenTTD] perezdidac opened pull request #8958: Feature: button to reset game settings to their default values https://git.io/JY7Yr
22:20:50 <peter1138> Damn this rabbit whole.
22:21:39 <andythenorth> more rabbit than Sainsburys
22:22:22 <FLHerne> andythenorth: I feel like government policy should be like that too :p
22:22:28 <DorpsGek> [OpenTTD/OpenTTD] LordAro commented on pull request #8958: Feature: button to reset game settings to their default values https://git.io/JY7YN
22:22:57 <FLHerne> peter1138: "Button to delete openttd.cfg", then
22:23:18 <LordAro> peter1138: i asked for you
22:24:55 <andythenorth> FLHerne what, made with the people it affects, by people who have skin in the game? ;)
22:27:41 <peter1138> I was actually joking. I assume "reset game settings" means just the, er, game settings, rather than all settings.
22:28:26 <FLHerne> I think that proves that "settings" and "game settings" are bad names for different things
22:28:37 <FLHerne> (not that anyone's proposed better ones yet)
22:29:04 <supermop_Home_> can we be rid of the grey menu yet?
22:29:12 <TrueBrain> hmm .. I think one of the two NATs I am testing with is symmetric NAT :P It does nothing what my own NAT is doing :D
22:30:29 <supermop_Home_> why is the goldeneye cradle music over 4 minutes long?
22:30:40 <supermop_Home_> that level take like 45-90 seconds
22:31:21 <supermop_Home_> i never got 00 agent on Aztec or control...
22:36:50 <andythenorth> I only played it on multiplayer
22:36:54 <andythenorth> classic student game
22:37:05 <andythenorth> one character was consistently OP, can't remember which
22:37:59 <supermop_Home_> its actually the short moonraker soldier
22:38:39 <supermop_Home_> most house rules were 'no oddjob' but no one cared about the moonraker lady that was smaller
22:43:27 <DorpsGek> [OpenTTD/OpenTTD] perezdidac commented on pull request #8958: Feature: button to reset game settings to their default values https://git.io/JY73K
22:51:06 <andythenorth> something something hiding behind crates
22:57:58 <TrueBrain> okay, on linux I can NAT punch the hell out of a TCP connection now
23:00:26 <TrueBrain> also works on Windows :o
23:01:38 <TrueBrain> on one of my test boxes I sometimes get "No route to host", which is really odd ..
23:03:09 <TrueBrain> but it punches the hole just fine
23:08:06 <TrueBrain> andythenorth: it is past 12:00, what are you still doing up?
23:08:14 <TrueBrain> anyone still alive wanting to help test my NAT punching tool? :D
23:09:42 * andythenorth feeling self-destructive and weird :P
23:09:54 <TrueBrain> well, we can try if this works on Mac too, I guess? :P
23:09:55 <andythenorth> wife is getting up at 5.30am so tomorrow will be shit
23:10:08 <andythenorth> what button do I press?
23:13:18 <DorpsGek> [OpenTTD/OpenTTD] perezdidac commented on pull request #8941: Fix: fix nullptr deletion in DeleteWindowById https://git.io/JY7ZZ
23:26:44 <TrueBrain> we had some success, but also some failure :P
23:30:25 <TrueBrain> it seems to confuse NATs when you keep using the same port :D
23:31:11 <TrueBrain> ha, now it works :D :D :D :D
23:31:17 <DorpsGek> [OpenTTD/OpenTTD] michicc commented on pull request #8958: Feature: button to reset game settings to their default values https://git.io/JY7ng
23:32:22 <TrueBrain> now just to do it in C++ :P
23:32:53 <TrueBrain> this really is the low-effort method :)
23:32:55 <andythenorth> oh is this http websockets?
23:33:18 <TrueBrain> so you can join a game with your friend without opening any ports
23:33:22 <andythenorth> websockets are like some ancient history thing that Flash games did
23:40:05 <TrueBrain> cannot believe this works with firewalls tbfh :P
23:41:26 <TrueBrain> time to get some zzzz now :D
23:47:48 * andythenorth staying up like an idiot
23:54:23 <FLHerne> spnda: Is that based on the upstream VS highlighting file, or have you managed to reinvent it?
23:54:48 <spnda> I am the one who PR'd that highlighting file, so yes it is
23:55:02 <spnda> Though I have re-run and edited it slightly for this release, so it's not quite up to date
23:55:40 <FLHerne> Ok, no wasted effort then :-)
continue to next day ⏵
