IRC logs for #openttd on OFTC at 2021-08-28
⏴ go to previous day
00:18:41 *** Progman has quit IRC (Remote host closed the connection)
00:56:23 *** Tirili has quit IRC (Quit: Leaving)
02:41:14 *** debdog has quit IRC (Ping timeout: 480 seconds)
02:59:07 *** Wormnest has quit IRC (Quit: Leaving)
03:06:45 *** HerzogDeXtEr has joined #openttd
03:11:52 *** D-HUND is now known as debdog
03:12:52 <DorpsGek> [OpenTTD/OpenTTD] 2TallTyler commented on pull request #9513: Change: Use colourblind-friendly gradient for linkgraph https://git.io/JEavN
04:35:11 *** tokai|noir has joined #openttd
04:35:11 *** ChanServ sets mode: +v tokai|noir
04:42:10 *** tokai has quit IRC (Ping timeout: 480 seconds)
06:57:45 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
06:58:43 *** WormnestAndroid has joined #openttd
07:00:35 *** andythenorth has joined #openttd
07:06:46 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
07:07:08 *** WormnestAndroid has joined #openttd
07:16:20 *** andythenorth has quit IRC (Quit: andythenorth)
07:28:46 *** andythenorth has joined #openttd
07:30:33 *** virtualrandomnumber has joined #openttd
07:30:58 *** virtualrandomnumber has quit IRC ()
08:15:47 *** sla_ro|master has joined #openttd
08:30:42 *** Progman has joined #openttd
08:54:59 <LordAro> why am i getting youtube ads for (open)suse?
08:55:04 <LordAro> who's paying for that?
08:58:46 <TrueBrain> Sorry, I had some money to spare
09:25:09 <Samu> hi. gonna try something before giving up on black line
09:31:51 *** nielsm has quit IRC (Ping timeout: 480 seconds)
09:42:29 <DorpsGek> [OpenTTD/game-coordinator] TrueBrain opened pull request #52: Fix: one client could attempt multiple simultaneous connections to the same server https://git.io/JEV8T
09:42:40 <TrueBrain> let's see if this is the reason for the high amount of timeouts I see
09:45:07 <DorpsGek> [OpenTTD/game-coordinator] TrueBrain merged pull request #52: Fix: one client could attempt multiple simultaneous connections to the same server https://git.io/JEV8T
09:52:15 <Samu> truebrain it's during the download phase
09:52:47 <Samu> i tried hosting a 2048x2048 map, the download is 10 MiB, and i can't just join myself
09:53:09 <Samu> the dl speed is just too slow even though I have a 10 Mbps upload
09:53:28 <Samu> times out during map transfer
09:54:38 <Samu> I time out at about 2 MB transfered
09:54:55 <TrueBrain> I am happy you talk about oranges, but I was talking about apples :) That has absolutely nothing to do with what I am trying to tackle. So be careful with stating facts ;) It might not hold true :)
09:55:11 <LordAro> Samu: is 10Mbps upload what you pay for, or what you actually get?
09:55:25 <TrueBrain> in your case, the server kicks you as you take too long to join
09:55:35 <TrueBrain> this is something you can already fix yourself, by allowing more time to join
09:55:57 <TrueBrain> (or, you know, don't host a 2kx2k on a 10 Mbps connection :P)
09:56:35 <LordAro> 10Mbps would take ~8s to download/upload 10MiB
09:56:38 <LordAro> what's the default timeout?
09:57:31 <LordAro> TrueBrain: that's a thought though, why would Samu end up connecting via TURN/GC/whatever, rather than communicating directly?
09:57:53 <TrueBrain> LordAro: crappy NAT router being crappy?
09:58:13 <TrueBrain> many NAT routers do not allow connecting to their own IP
09:58:16 <TrueBrain> which is against RFC
09:58:20 <TrueBrain> but .. yeah .... what can we do
09:58:25 <TrueBrain> that is a very common problem :)
09:58:50 <TrueBrain> (and a very old one, even ... I had that fight already years ago with my router :P)
09:59:36 <TrueBrain> "ValueError: 40 is not a valid PacketGameType"
09:59:39 <TrueBrain> euhm .. okay ... wth
10:00:27 <TrueBrain> ah .. I did not implement PACKET_SERVER_SHUTDOWN :D
10:00:45 <Samu> i dont have any other way to test joining my servers unless someone volunteers
10:01:20 <LordAro> Samu: use a smaller map then
10:01:44 <LordAro> 1024x1024 is already 1/4 of the size, much easier to download
10:01:54 <LordAro> or, as TB says, increase your local server timeout
10:04:22 <DorpsGek> [OpenTTD/py-protocol] TrueBrain opened pull request #16: Add: support for SERVER_SHUTDOWN in Game protocol https://git.io/JEV0w
10:04:24 <Samu> not enough time to join 1024x1024 either :o
10:04:41 <LordAro> how long is it trying to connect for?
10:05:07 <LordAro> seriously though, you need to increase the timeout
10:05:16 <DorpsGek> [OpenTTD/py-protocol] TrueBrain merged pull request #16: Add: support for SERVER_SHUTDOWN in Game protocol https://git.io/JEV0w
10:05:40 <Samu> about 17 seconds self-measured, so it could be wrong
10:06:28 <Samu> with a clock, timed out after 18 secs
10:06:48 <LordAro> network.max_download_time is the setting you want, i think
10:06:54 <Samu> Network - Game Connection Lost
10:07:55 <TrueBrain> meh, GitHub no longer auto-updates action-status .. that is annoying :P Need to refresh .. :D
10:08:09 <DorpsGek> LordAro: 6.756756756756757
10:08:43 <Samu> "Network-Game connection lost." I've been saying time out, but the actual error message is that
10:08:58 <LordAro> max_join_time (default 500) + max_download_time (default 1000) could be 18s
10:12:21 <Samu> i set all xxxx_time to 32000
10:12:42 <Samu> max_download_time 32000 didn't solve it though
10:12:43 <LordAro> now remove them one by one until you find the one that was actually needed to be changed
10:12:54 <DorpsGek> [OpenTTD/game-coordinator] TrueBrain opened pull request #53: Fix: clean up verify tokens once we are done with them https://git.io/JEVuI
10:12:57 <LordAro> sounds like you have some strange networking, at any rate
10:14:53 <DorpsGek> [OpenTTD/game-coordinator] TrueBrain opened pull request #54: Fix: crash if a server left during registration https://git.io/JEVuz
10:15:02 <DorpsGek> [OpenTTD/game-coordinator] TrueBrain merged pull request #53: Fix: clean up verify tokens once we are done with them https://git.io/JEVuI
10:15:31 <glx> What happens if you manually connect to 127.0.0.1 samu ?
10:16:11 <DorpsGek> [OpenTTD/game-coordinator] TrueBrain merged pull request #54: Fix: crash if a server left during registration https://git.io/JEVuz
10:16:33 <Samu> when i search LAN, my server is displayed there
10:17:14 <Samu> i see it listed twice, one with the internet IP address, the other with 192.168.0.10:3979
10:17:33 <Samu> i join blazing fast instantly to 192.168.0.10
10:17:43 <LordAro> TrueBrain: should there be some protection for the GC in the event of an not-understood packet?
10:17:54 <LordAro> it shouldn't crash regardless of the packet, no?
10:18:05 <TrueBrain> owh, "crash" is wrongly used here .. oops
10:18:09 <TrueBrain> the connection is forcefully closed
10:18:24 <TrueBrain> it generated a crashreport :P
10:18:29 <TrueBrain> I mixed up my lingo :D
10:19:00 <TrueBrain> the GC has many layers of protection against completely crashing; protocol is wrapped, tasks are wrapped, etc :)
10:19:10 <TrueBrain> it only really crashes if the redis server is going away, basically :P
10:19:20 <Samu> when i click search internet, the address changes to an invite code, no longer displaing the ip address, something's strange
10:20:17 <DorpsGek> [OpenTTD/game-coordinator] TrueBrain opened pull request #55: Fix: if a server leaves while a client is connecting, abort the attempt https://git.io/JEVz2
10:20:36 <glx> Use the local address to join then
10:21:22 <Samu> 127.0.0.1:3979 also works, blazing instafast join
10:21:57 <DorpsGek> [OpenTTD/game-coordinator] TrueBrain merged pull request #55: Fix: if a server leaves while a client is connecting, abort the attempt https://git.io/JEVz2
10:22:37 <TrueBrain> okay, now for the issue where 1.11 clients receive the IP of a 12.x server, but of course cannot query them .. hmm
10:22:50 <TrueBrain> the initial idea was to leave them in the list, so you can see there are servers, just not ones that are compatible with you
10:23:09 <TrueBrain> the other solution is to remove them, so over the years the server-list for pre-12.x gets smaller and smaller
10:24:25 <TrueBrain> I think I prefer the first, as seeing an empty-list on an old client gives the wrong vibe, so to say
10:26:16 <Samu> the culprit is the max_join_time, it's the last one remaining at 32000
10:26:28 <Samu> all others are back to their defaults and I can still join
10:26:40 <glx> The culprit is your router I'd say
10:27:38 <glx> You can check with -d net6 to see what connection method is used
10:28:32 <DorpsGek> [OpenTTD/game-coordinator] TrueBrain commented on issue #51: Pre 12.x clients receive IP of 12.x servers with no way to reach them https://git.io/JEZNy
10:28:35 <DorpsGek> [OpenTTD/game-coordinator] TrueBrain closed issue #51: Pre 12.x clients receive IP of 12.x servers with no way to reach them https://git.io/JEZNy
10:28:43 <glx> (not directly stated, but you can see which of openttd URL are resolved)
10:31:33 <glx> So direct connection, something is weird on your router
10:32:14 <Samu> it's only 1 computer connected to the router
10:36:39 <DorpsGek> [OpenTTD/game-coordinator] TrueBrain opened pull request #56: Add: keep stats about the openttd-version used for listing https://git.io/JEVVe
10:37:03 <glx> If downloading is fast for localhost but not through the router, there's probably something interfering
10:37:58 <DorpsGek> [OpenTTD/game-coordinator] TrueBrain merged pull request #56: Add: keep stats about the openttd-version used for listing https://git.io/JEVVe
10:39:34 <Samu> what's the difference between 127.0.0.1 and 192.168.0.10?
10:39:51 <TrueBrain> right, time to annoy people by closing all TURN connections .. not much I can do about that :(
10:40:41 <glx> No diff between both, it's your computer IP
10:41:30 <glx> But 127.0.0.1 exists when your not connected to a network
10:42:00 <Samu> 127.0.0.1 is my computer -> my computer
10:42:17 <Samu> 192.168.0.10 is my computer -> router -> my computer?
10:44:01 <Samu> 84.91.251.215 is my computer -> router -> internet -> router -> my computer? I like to think that's it, but unsure
10:45:58 <Samu> max join time was the issue, i just tried joining a 10 MB map with it set to 32000 and i joined it
10:46:00 <glx> Router is skipped in 192.168.0.10 to 192.168.0.10 connection
10:46:34 <Samu> max_download_time seems not doing what's supposed to be doing
10:48:07 <Samu> let me try a different savegame format
10:48:36 <glx> You can use tracert in a command window to see the steps
10:50:24 *** roadt__ has joined #openttd
10:54:39 <Samu> i just tested something, i set a max_download_time to 1
10:55:11 <Samu> tried joining the server, and i still joined it fine, because max_join_time is 32000
10:55:41 <Samu> max_download_time seems to be doing nothing?
10:56:55 *** roadt_ has quit IRC (Ping timeout: 480 seconds)
11:02:06 <Samu> the server log says "processing map took too long" about myself failing to join
11:16:05 *** WormnestAndroid has quit IRC (Remote host closed the connection)
11:17:35 *** WormnestAndroid has joined #openttd
12:18:16 <andythenorth> what's for lunch also?
12:18:45 <andythenorth> cheese and chicken toastie?
12:20:32 <DorpsGek> [OpenTTD/game-coordinator] TrueBrain opened pull request #57: Fix: exception when a server leaves while a connect was just initiated https://git.io/JEVd6
12:21:38 <DorpsGek> [OpenTTD/game-coordinator] TrueBrain merged pull request #57: Fix: exception when a server leaves while a connect was just initiated https://git.io/JEVd6
12:31:23 *** glx is now known as Guest5650
12:32:01 *** Guest5650 has quit IRC (Quit: Quit)
12:37:16 *** Progman has quit IRC (Remote host closed the connection)
12:38:07 <glx> oh I found a funny bug, on multiplayer window, when trying to join a server without setting a player name if fails and disable the join button (expected), but I needed to close the window after setting a name for the button to be enabled again
12:38:37 <TrueBrain> I look forward to the PR that fixes that issue ;)
12:39:03 <Samu> im currently experimenting different colours for linkgraph
12:39:56 <glx> Samu: I quickly joined your server (it was a normal download speed)
12:40:18 <glx> but it's not a too big map now
12:41:07 <Samu> how much speed, would you be able to quanify?
12:41:12 <DorpsGek> [OpenTTD/OpenTTD] glx22 merged pull request #9515: Fix b6a116a: convert server_advertise to server_game_type in config file https://git.io/JE2fM
12:42:03 <glx> almost instant download I'd say, less than 10s probably
12:42:58 <Samu> would u like to test the 213 MB save?
12:46:18 <Samu> my ethernet says something around 15.9 mbps
12:47:06 <Samu> that's definitely faster than 1.1 mbps for when i join myself
12:48:33 <Samu> thx, you're in, that's what I needed to know
12:48:46 <glx> yeah but it's unusable for me :)
12:49:17 <TrueBrain> w00p, seems I am finally hitting true-positives on "unused strings" ... STR_CONFIG_SETTING_DISABLE_UNSUITABLE_BUILDING seems unused :P
12:49:19 <Samu> oh, client dropped because 501 ticks behind, frame_freq etc
12:49:39 <glx> yes my computer can't handle this map size
12:50:03 <TrueBrain> owh, I can blame LordAro for that, so it is fine :)
12:50:06 <TrueBrain> (commit 188bf0fbc9)
12:50:59 <glx> ok let's look at the GUI issue
12:51:47 <glx> at least I don't have to draw a new window ;)
12:57:30 <Samu> i wish i could be colorblind for a day
12:59:24 <Samu> testing shades of milk-coffee brown with shades of pale green
13:12:29 <Flygon> So I've been learning how to make complex circuits in Logisim.
13:12:55 <Flygon> Thank you OpenTTD for making it mentally possible for me to wrap my head around this network of logic gates and circuit lines.
13:13:08 <Flygon> Also I think I've created a monster.
13:13:35 <Flygon> And now really wish Logisim had macros because having to send serial sets of commands through manually clicking the data pins IS REALLY TEDIOUS
13:14:08 <Flygon> Obviously an issue OpenTTD doesn't suffer. :D
13:14:30 <glx> hmm I can't reproduce in master, need to check things
13:16:54 <TrueBrain> I find so much crap that we did with strings, it is incredible ...
13:16:55 <TrueBrain> STR_FORMAT_DEPOT_NAME_TRAIN + 2 * vt + (d->town_cn == 0 ? 0 : 1)
13:17:02 <TrueBrain> implied is that town_cn is never 1 for aircrafts
13:18:01 <TrueBrain> well, vt is never "aircraft", to be more correct
13:18:03 <glx> and now beta works too, weird
13:18:12 <TrueBrain> yet there is a string name VERY identical to the rest
13:18:19 <TrueBrain> boy .. we made a real mess out of this :P
13:18:55 *** roadt__ has quit IRC (Ping timeout: 480 seconds)
13:22:03 <Samu> there's always a colour that looks bad with other colour, so I'm gonna give up on this
13:23:29 *** roadt__ has joined #openttd
13:24:36 <glx> yeah maybe a different color scale is needed depending on map type
13:38:12 <DorpsGek> [OpenTTD/OpenTTD] SamuXarick commented on pull request #9513: Change: Use colourblind-friendly gradient for linkgraph https://git.io/JEwtj
13:47:49 <TrueBrain> STR_JUST_DATE_SHORT seems unused; lol
13:48:19 <glx> depends on locale setting maybe
13:49:04 <TrueBrain> let me rephrase: I cannot find anything in our code that can point to STR_JUST_DATE_SHORT :)
13:50:09 <glx> but I don't see STR_JUST_DATE_TINY + something
13:50:18 <glx> so maybe it's really unused
13:50:18 <TrueBrain> src/saveload/saveload.cpp:3378: case 1: SetDParam(1, STR_JUST_DATE_TINY); break;
13:50:25 <TrueBrain> that is all where TINY is used
13:50:33 <TrueBrain> as I said: I cannot find any reference to it ;)
13:51:02 <TrueBrain> sorry, let me be more precise: in a way that might hint to how _SHORT can be used
13:51:55 <TrueBrain> guess it was just added "to be complete"
13:52:34 <glx> uint8 date_format_in_default_names; ///< should the default savegame/screenshot name use long dates (31th Dec 2008), short dates (31-12-2008) or ISO dates (2008-12-31)
13:53:01 <TrueBrain> the "short" is TINY :P
13:54:24 <glx> in cfg it's long, short, iso
13:55:03 <glx> I guess tiny was added later and replaced short internally
13:58:00 <TrueBrain> I am surprised by how few strings I can find that are unused
13:58:02 <TrueBrain> I expected a lot more
14:21:17 <TrueBrain> can you forward-blame in git? :P Find out when a line is being removed :D
14:27:19 <TrueBrain> 3 mysterious to solve ...
14:34:22 <TrueBrain> lol, some are unused since 2009 :P
14:38:19 <TrueBrain> right, can explain all except for the STR_NETWORK_ERROR ones
14:40:47 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain opened pull request #9517: Codechange: reference a string by its name, not with "another + 1" https://git.io/JEw0l
14:42:00 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #9517: Codechange: reference a string by its name, not with "another + 1" https://git.io/JEw0l
14:43:44 <DorpsGek> [OpenTTD/OpenTTD] glx22 approved pull request #9517: Codechange: reference a string by its name, not with "another + 1" https://git.io/JEwEO
14:58:16 *** andythenorth has quit IRC (Quit: andythenorth)
14:59:46 <TrueBrain> yeah, I updated the PR with that :)
15:01:23 <TrueBrain> seems somewhere in the branch it got added but in the end, nothing was using it :)
15:02:48 *** Wormnest has joined #openttd
15:08:05 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain merged pull request #9517: Codechange: reference a string by its name, not with "another + 1" https://git.io/JEw0l
15:11:11 <glx> ok tramway was never use, road was
15:11:21 <glx> (and has been removed since)
15:25:26 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain opened pull request #9518: Codechange: validate what strings are defined but not used https://git.io/JEwKM
15:25:41 <TrueBrain> let's see if the CI part works
15:28:40 <TrueBrain> right .. turns out that what I had was mostly completely, just an annoying job to go through all the things it detected .. but hopefully this helps towards the future to prevent strings becoming unused :)
15:28:46 <TrueBrain> still surprised how few there were :)
15:29:30 <TrueBrain> "unknown pragma" .. hmm .. why does it work locally ? Weird
15:29:54 <TrueBrain> and now it doesn't .. lol .. well, let me fix that
15:32:24 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #9518: Codechange: validate what strings are defined but not used https://git.io/JEwKM
15:38:15 <glx> I can already see ###lenght won't be updated when the number of item change :)
15:38:31 <TrueBrain> hence the safeguards :)
15:38:52 <TrueBrain> if you add an entry without changing the number, the script will complain :)
15:39:50 <TrueBrain> let me textual hint to that in the error
15:39:51 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #9518: Codechange: validate what strings are defined but not used https://git.io/JEwKM
15:40:05 <TrueBrain> I was also thinking just having a marker "###list", and that we use the newline to indicate "end of list"
15:40:35 <TrueBrain> too bad it is really difficult to access the values of enum constants, as otherwise we could just do things like "###length MAX_COMPANIES"
15:40:42 <TrueBrain> which would be a lot more useful
16:02:10 *** andythenorth has joined #openttd
16:09:07 *** ChanServ sets mode: +v tokai
16:16:05 *** tokai|noir has quit IRC (Ping timeout: 480 seconds)
16:20:38 *** Progman has joined #openttd
16:58:21 *** tokai|noir has joined #openttd
16:58:21 *** ChanServ sets mode: +v tokai|noir
17:05:24 *** tokai has quit IRC (Ping timeout: 480 seconds)
17:18:52 <DorpsGek> [OpenTTD/game-coordinator] TrueBrain opened pull request #58: Fix: don't close client connection when the invite-code is invalid https://git.io/JErUV
17:23:28 <DorpsGek> [OpenTTD/game-coordinator] TrueBrain merged pull request #58: Fix: don't close client connection when the invite-code is invalid https://git.io/JErUV
18:16:50 *** Flygon has quit IRC (Quit: A toaster's basically a soldering iron designed to toast bread)
18:20:25 *** andythenorth has quit IRC (Quit: andythenorth)
18:40:34 <TrueBrain> okay ... user on Discord reported an issue with networking ..
18:40:42 <TrueBrain> turns out it is a big boo-boo
18:41:31 <DorpsGek> [OpenTTD/game-coordinator] TrueBrain opened pull request #59: Fix: broadcast server connect-failed / stun-result to other GC instances https://git.io/JErnb
18:41:42 <TrueBrain> I cannot believe not more people reported this issue
18:42:19 <TrueBrain> multi instance biting me in the ass :P
18:42:21 <TrueBrain> and I did so well .....
18:43:22 <DorpsGek> [OpenTTD/game-coordinator] TrueBrain updated pull request #59: Fix: broadcast server connect-failed / stun-result to other GC instances https://git.io/JErnb
18:45:16 <DorpsGek> [OpenTTD/game-coordinator] TrueBrain merged pull request #59: Fix: broadcast server connect-failed / stun-result to other GC instances https://git.io/JErnb
18:46:14 <TrueBrain> at least that most likely explains the huge amount of timeouts :P
18:47:17 <DorpsGek> - Update: Translations from eints (by translators)
18:48:55 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain opened pull request #9519: Fix: when DNS didn't resolve any IPs, the debug message was misleading https://git.io/JErcD
18:50:56 <TrueBrain> lets terminate all those pesky TURN connection in the middle of the evening .. that is going to make me popular :P
18:56:50 <Rubidium> just time it with some netsplit and blame it on that;)
18:58:00 *** andythenorth has joined #openttd
19:02:47 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
19:03:12 *** WormnestAndroid has joined #openttd
19:12:37 <DorpsGek> [OpenTTD/OpenTTD] glx22 approved pull request #9519: Fix: when DNS didn't resolve any IPs, the debug message was misleading https://git.io/JEr4T
19:16:35 *** gelignite has joined #openttd
19:19:11 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #9514: Change: Cap docking tile occupancy penalty to a max of one ship https://git.io/JErBY
19:19:49 <TrueBrain> turning a nob from left all the way to the right seems like a bad balance to strike :)
19:19:57 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain merged pull request #9519: Fix: when DNS didn't resolve any IPs, the debug message was misleading https://git.io/JErcD
19:20:00 <LordAro> yeah, definitely fixingthe wrong problem
19:20:14 <TrueBrain> I had to find a lot of context to understand what it was fixing
19:20:19 <TrueBrain> as the motivation is ... not a motivation :D
19:20:30 <TrueBrain> but I found a bug report stating the problem :P
19:22:00 <LordAro> ...the fact that the PR does not even reference #8001 does not help
19:25:57 <TrueBrain> but okay, I guess the sqrt(count) or something similar might solve the issue too
19:26:26 <TrueBrain> still keeps multidocks useful, even if they get more busy, yet prevents runaway PFs
19:26:34 <TrueBrain> although I haven't tried it, so who knows
19:26:36 <LordAro> still feels like the wrong solution
19:26:38 <glx> sqrt ? in something important for network sync ?
19:27:00 <LordAro> ships shouldn't be reporting lost in the first place
19:27:10 <TrueBrain> just in general, they shouldn't :D
19:27:14 <TrueBrain> they are in a constant state of being lost :P
19:27:34 <glx> they are lost even with a path
19:27:46 <LordAro> yeah, that's the weird bit
19:28:15 <LordAro> the very small amount of debugging i did indicated that it was hitting the pathfinder limit before reaching the destination, so was just using the partial path instead
19:28:27 <LordAro> but that was many months ago, and i might have been wrong
19:28:37 <TrueBrain> guess we should apply modern PF solutions :)
19:29:05 <TrueBrain> always surprised, even back in YAPF days, nobody did that
19:29:10 <TrueBrain> even Factorio does it :P
19:30:24 <glx> anyway the penalty being linear with the number of ship feels too much
19:30:53 <TrueBrain> I think both problems should be addressed
19:31:18 <glx> and it should probably also use the number of docks to decide
19:31:31 <TrueBrain> ln(ships) / ln(docks)?
19:31:56 <TrueBrain> guess the main issue is that the penalty needs to be high enough that the next dock tile is considered
19:32:17 <glx> yes that was the original idea I think
19:32:19 <TrueBrain> so I kinda agree with LordAro , that this isn't a proper solution, and possibly only solves a few edge-cases
19:33:08 <TrueBrain> you kinda want a ship to go to a focal point of the docks, and after that pick the one with the least amount of ships (and closest to him)
19:33:25 <TrueBrain> that possibly would give the most pleasant solution to the player
19:35:28 <glx> road/bus stops use something like that IIRC, but doing the same on open water may be difficult
19:37:13 <TrueBrain> you kinda would want to give the user the ability to set that point :P
19:43:25 <andythenorth> also ships that could route around peninsulas :P
19:45:38 *** Wormnest has quit IRC (Ping timeout: 480 seconds)
19:46:06 <TrueBrain> not even a real crash
19:46:29 <TrueBrain> I expected something better
19:46:43 <TrueBrain> I am disappointed in Win95 :P
19:46:49 <TrueBrain> it could have gone out with a bigger bang :)
19:46:57 <andythenorth> I mean...that's nothing
19:46:58 <Rubidium> I expected Rick Astley ;(
19:47:04 <LordAro> TrueBrain: you made me do this
19:47:06 <TrueBrain> now that would have been funny :)
19:47:21 <andythenorth> I had a 1998 iMac that could crash every 5-10 minutes
19:47:22 <TrueBrain> LordAro: and I regret NOTHING
19:47:29 <andythenorth> Apple was much faster
19:47:37 <TrueBrain> well, lets do this again in 49 days
19:48:07 <andythenorth> I could crash 7000 times in that time on an iMac, if I stayed up 24 hours a day
19:50:09 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #9518: Codechange: validate what strings are defined but not used https://git.io/JEwKM
19:51:27 *** Wormnest has joined #openttd
19:52:03 <glx> well it was not a crash, but it stopped responding and required a reboot
19:52:29 <Samu> I have presented other PRs in #8001 in the past, but they've been closed
19:57:41 <Samu> actually just one other PR
20:25:24 *** gelignite has quit IRC (Quit: Stay safe!)
20:30:41 *** sla_ro|master has quit IRC ()
20:34:00 <Samu> dayum, town growth and oil rigs spawning, these tiny little things also influence the num of search nodes
20:34:15 <Samu> on a route that are entirely unrelated to them
20:45:43 <DorpsGek> [OpenTTD/OpenTTD] JGRennison opened issue #9520: [Bug]: Wrong unreservation behaviour when reversing a train inside a depot when an incoming reservation exists https://git.io/JErKD
20:55:24 <Samu> no penalty + 0 ships = good
20:55:40 <Samu> the only result where ship is lost won't appear
20:56:22 <Samu> anything above 10000 means "ship is lost"
21:12:32 <glx> and tested with more than one docking tile ?
21:16:06 <Samu> testing that is gonna be tricky
21:16:39 <Samu> or maybe not, just got an idea
21:17:02 <Samu> will try something tomorrow, im out of time today
21:26:54 *** Samu has quit IRC (Ping timeout: 480 seconds)
22:27:44 *** Progman has quit IRC (Remote host closed the connection)
22:56:10 *** andythenorth has quit IRC (Quit: andythenorth)
23:20:46 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
23:35:27 *** colde has quit IRC (Read error: Connection reset by peer)
23:52:43 *** twpol has quit IRC (Read error: Connection reset by peer)
continue to next day ⏵
