IRC logs for #openttd on OFTC at 2018-09-25
⏴ go to previous day
00:01:10 <glx> LordAro: searchpath looks like brute force ;)
00:01:42 <Samu> nevermind, i was edited the wrong file, grr
00:01:51 <glx> OSX is worse than MSYS/MSYS2
00:03:15 <glx> we needed to add many special cases for MSYS2
00:03:59 <lethosor_> Checking /usr/local/Cellar is too much work?
00:06:09 <lethosor_> Looks like more than necessary to me
00:06:36 <glx> I think it's for the next time something changes in OSX world ;)
00:07:04 <lethosor_> It's not a recent change by any means
00:07:22 <lethosor_> Homebrew has never installed iconv system-wide
00:09:28 <LordAro> i don't understand how it became an issue recently though
00:09:45 <LordAro> did older osx just leave an old iconv.h header lying around or something?
00:10:17 *** quiznilo has joined #openttd
00:10:31 <lethosor_> It's a system-specific issue somehow. It works fine for me, like I said
00:13:53 <andythenorth> it only started happening on a clean install of macOS for me
00:14:02 <andythenorth> old one probably cargo culted something
00:14:15 *** rocky113844 has joined #openttd
00:15:18 *** Gustavo6056 has joined #openttd
00:19:38 <Samu> 46/44/77/77/77/46/44/75/71/77/72 - _IsSlopedRoadEfficient
00:21:25 <LordAro> (wrote a load more text in #6917 for future people)
00:21:46 <Samu> 70/68/107/107/107/68/66/105/97/107/103 - _IsSlopedRoad (the original code)
00:22:32 *** Gustavo6056 is now known as Gustavo6046
00:23:17 <Samu> can't manage to make nielsm code to not assert
00:24:56 <Samu> what I was able to do without asserting
00:25:38 *** andythenorth has left #openttd
00:33:57 *** HeyCitizen has joined #openttd
00:41:21 *** rocky113844 has joined #openttd
00:45:07 <Eddi|zuHause> <LordAro> wait, crap, i can create branches on OpenTTD/OpenTTD as well? <- sounds like something that should be prevented?
00:51:59 *** Thedarkb-T60 has joined #openttd
01:15:30 *** Gustavo6056 has joined #openttd
01:21:30 *** haudrauf has joined #openttd
01:22:32 *** Gustavo6056 is now known as Gustavo6046
01:31:37 *** haudrauf has joined #openttd
01:45:15 *** Extrems has joined #openttd
02:15:44 *** Gustavo6056 has joined #openttd
02:21:54 *** snail_UES_ has joined #openttd
02:23:02 *** Gustavo6056 is now known as Gustavo6046
02:54:48 *** reldred` has joined #openttd
02:55:16 *** reldred` is now known as reldred
03:15:40 *** Gustavo6056 has joined #openttd
03:23:02 *** Gustavo6056 is now known as Gustavo6046
03:35:26 *** WWacko1976-work has joined #openttd
04:16:02 *** Gustavo6056 has joined #openttd
04:21:24 *** Supercheese has joined #openttd
04:23:03 *** Gustavo6056 is now known as Gustavo6046
05:15:59 *** Gustavo6056 has joined #openttd
05:23:03 *** Gustavo6056 is now known as Gustavo6046
05:37:54 *** haudrauf has joined #openttd
06:10:31 *** Sheogorath has joined #openttd
06:15:58 *** Gustavo6056 has joined #openttd
06:23:37 *** Gustavo6056 is now known as Gustavo6046
06:57:02 *** OsteHovel has joined #openttd
07:10:19 *** Supercheese has joined #openttd
07:16:24 *** Gustavo6056 has joined #openttd
07:22:07 *** Gustavo6056 is now known as Gustavo6046
08:13:04 *** gelignite has joined #openttd
08:16:17 *** Gustavo6056 has joined #openttd
08:22:43 *** Gustavo6056 is now known as Gustavo6046
09:16:35 *** Gustavo6056 has joined #openttd
09:23:18 *** Gustavo6056 is now known as Gustavo6046
10:03:45 *** Laedek_ has joined #openttd
10:16:47 *** Gustavo6056 has joined #openttd
10:18:52 *** Gustavo6- has joined #openttd
10:23:27 *** Gustavo6- is now known as Gustavo6046
10:23:35 *** rocky1138 has joined #openttd
10:26:57 *** Extrems has joined #openttd
10:28:44 *** ccfreak2k has joined #openttd
10:30:48 *** Lejving_ has joined #openttd
10:35:27 *** TheMask96 has joined #openttd
10:45:40 *** argoneus has joined #openttd
11:16:47 *** Gustavo6056 has joined #openttd
11:22:32 *** Gustavo6056 is now known as Gustavo6046
12:17:10 *** Gustavo6056 has joined #openttd
12:22:03 *** Gustavo6056 is now known as Gustavo6046
13:02:05 <Samu> yapf A formatter for Python files
13:03:57 <Samu> looks like I need to edit Graph AyStar to point it to the "Efficient" versions of my functions
13:04:11 <Samu> i was hoping i wouldn't have to
13:17:23 *** Gustavo6056 has joined #openttd
13:21:37 *** Gustavo6056 is now known as Gustavo6046
13:55:27 <Samu> when will mister efficience comes online?
14:01:07 <Samu> if (from - to == 1) return 1; if (from - to == -1) return 2; if (from - to == map_size_x) return 4; if (from - to == -map_size_x) return 8;
14:01:20 <Samu> would a switch/case be faster in here? less op,s i mean
14:17:26 *** Gustavo6056 has joined #openttd
14:17:28 <Eddi|zuHause> we should sue google for infringement...
14:23:15 *** Gustavo6056 is now known as Gustavo6046
14:29:39 <peter1138> Samu, profile it :p
14:52:07 *** snail_UES_ has joined #openttd
14:58:37 <Samu> local diff = from - to; switch (diff) { case 1: return 1; case -1: return 2; case map_size_x: return 4; case -map-size_x: return 8; default: assert(false); }
14:58:56 <Samu> why is a default always needed?
15:17:18 *** Gustavo6056 has joined #openttd
15:21:00 *** IcyPalm has joined #openttd
15:24:55 *** Gustavo6056 is now known as Gustavo6046
15:42:13 <Samu> attempt to negate a bool? uh what?
15:44:35 <Samu> case map_size_x: return 4;
15:44:44 <Samu> it's complaining about this line, i dont get it
15:47:07 <FLHerne> Cases have to be constexprs
15:49:34 <Samu> hello, i was waiting for u
15:50:22 <Samu> i couldn't make your yesterday code to work without asserting
15:56:14 *** IgnoredAmbience has left #openttd
16:05:57 *** IcyPalm has joined #openttd
16:09:47 *** IcyPalm has joined #openttd
16:17:35 *** Gustavo6056 has joined #openttd
16:24:39 *** Gustavo6056 is now known as Gustavo6046
16:59:51 <Samu> hmm, can't compare arrays
17:00:22 <Samu> assert(array1 == array2) always false :(
17:00:23 *** sla_ro|master has joined #openttd
17:04:26 <Samu> this function is probably the most used in the entire pathfinder. for every tile it test builds a bridge of every size, in all 4 directions
17:04:50 <Samu> ahem, every "allowed size"
17:08:05 <Samu> the most used is GetDirection, because every tile has a direction
17:08:17 <Samu> but that one is much cheaper
17:09:34 *** GroovyNoodle has joined #openttd
17:16:04 *** WWacko1976-work has quit IRC
17:17:50 *** Gustavo6056 has joined #openttd
17:18:25 <GroovyNoodle> Long text comming: I have a problem with my server. I had a raspberry pi running a dedicated server. It worked great! When I decided to upgrade the raspberry, I snapped the SD-card in half.. So I re-made the whole server, copied the old config (the only thing I had backed up) and started it. It can be reached if you use the IP, but it refuses to show up on the master server. Is this because my old one is
17:18:25 <GroovyNoodle> there (shown as offline) with the same IP? As it is a new raspberry the MAC-adress on the network is different.. The local adress is locked at the same as it was with the prevoius one.
17:20:22 <GroovyNoodle> I have tried with both the newly generated network_id I got when I recompiled the server, in the config, and with the old one from the previous config.
17:24:58 *** Gustavo6056 is now known as Gustavo6046
17:25:41 <planetmaker> GroovyNoodle, the presence of an old server in the server list does not have any influence anywhere whatsoever
17:26:09 <planetmaker> it only means that it was once seen. It will be updated if another server from the same IP contacts the master server
17:26:42 <planetmaker> Is your network config in all parts the same, especially do you allow the master server talk back to your server?
17:26:58 <planetmaker> i.e. are the ports open in all devices they need be?
17:27:08 <planetmaker> check also the router
17:27:40 <GroovyNoodle> Yes, the ports are the same, as I only changed the local ip, 192.168.1.26, to bind to the new mac-adress
17:27:54 <GroovyNoodle> I have no firewall on the pi blocking the ports
17:29:25 <GroovyNoodle> sorry, typed a bit confusing. All the required ports are open to 192.168.1.26. And I bound the new mac-adress of the new pi to that adress.
17:29:34 <GroovyNoodle> So no port config in the router has been changed
17:34:31 <Eddi|zuHause> you sure there is no ip conflict?
17:37:28 <Eddi|zuHause> check the following two things: 1) can you connect to the server (from inside the game) within your network, using the internal ip address. 2) can someone from outside the network join using your public ip address?
17:38:30 <GroovyNoodle> I can connect to the server, so can other people who use the ip-adress. I'm using a VPN, so the server even think I'm outside my own network..
17:39:23 <GroovyNoodle> "[2018-09-25 17:29:04] dbg: [net] [udp] we are not receiving the acknowledgement from the server" from the console
17:39:41 <GroovyNoodle> both udp and tcp are forwarded
17:39:56 <Eddi|zuHause> then the udp forwarding fails
17:40:29 <Eddi|zuHause> either outgoing or incoming udp packets are dropped
17:41:08 <GroovyNoodle> Can this be caused by anything in the config file?
17:41:20 <nielsm> Samu, you can perhaps do some caching of bridge types at the start of each year
17:41:33 <Eddi|zuHause> most likely by your router config
17:41:41 <nielsm> perhaps just pick a single preferred bridge type of each length every year
17:42:10 <Eddi|zuHause> or some firewall you didn't think you had
17:42:39 <GroovyNoodle> No firewall comes with raspbian, and it's a clean install (+ottd)
17:42:49 <GroovyNoodle> I will try rebooting the router later then
17:43:19 <GroovyNoodle> It SHOULD be correct, but something won't update then even if I save the settings..
17:46:44 *** Progman has joined #openttd
17:56:48 *** rocky113844 has joined #openttd
18:02:46 *** TheMask96 has joined #openttd
18:08:03 *** HerzogDeXtEr has joined #openttd
18:11:47 *** sushibear has joined #openttd
18:11:49 *** frosch123 has joined #openttd
18:17:51 *** Gustavo6056 has joined #openttd
18:24:04 *** Gustavo6056 is now known as Gustavo6046
18:40:21 <Samu> the index rawin does not exist
18:47:57 <acklen> squirrel tables or iptables?
18:50:41 <nielsm> or html tables or sql tables or...
18:51:47 <nielsm> or perhaps wooden tables???
18:56:44 <Samu> so i assume i'm doing something wrong
18:59:07 <Samu> it's also worse than my previous method
18:59:16 <Samu> 276 is best i got so far
19:06:20 <nielsm> those are tiny tiny gains, better would be seeing if you can simplify the algorithm itself
19:07:45 <nielsm> really, try caching the best bridge for all lengths at the start of each year
19:07:48 <Samu> well apparently i failed to do the caching
19:08:11 <nielsm> so you avoid the calls to AIBridgeList_Length and bridge_list.Begin
19:08:18 <nielsm> and bridge_list.Empty()
19:08:50 <Samu> i tried, but i either did it wrong, or it's worse
19:09:17 <Samu> i used in on Class Road thing
19:09:38 <nielsm> afaik the selection of bridges can change at year start and otherwise not
19:10:54 <Samu> I don't know how to do that :(
19:11:43 <nielsm> somewhere in the daily loop, if date is 1/1 then plan doing a bridge availability update some time soon
19:12:10 <Samu> so, that means outside the pathfinder
19:15:43 <Samu> ... the way i had it, it would print messages listing a type of bridge for every length. this was only printed once per pathfind instance, so it wasn't repeatedly recreating it, but... i dunno what delegation means
19:17:48 <Samu> if (Road._bridge_lengths.rawin(i) && AIBridge.Buildblabla (blabla, Road._bridge_lengths.rawget(i), bla bla
19:17:58 *** Gustavo6056 has joined #openttd
19:22:09 *** Gustavo6056 is now known as Gustavo6046
19:24:20 <Samu> max_bridge_length is 10 at default
19:24:37 <Samu> for (local i = 2; i < 10; i++) {
19:24:56 <Samu> local bridge_list = AIBridgeList_Length(3);
19:25:13 <Samu> local target = cur_node + 2 * (cur_node - last_node);
19:27:24 <Samu> for (local i = 2; i < this._max_bridge_length;) {
19:27:31 <Samu> local target = cur_node + i * (cur_node - last_node);
19:27:39 <Samu> local bridge_list = AIBridgeList_Length(i++);
19:29:54 <Samu> but, is it still doing the same thing?
19:31:34 <Samu> AIBridgeList_Length(i++); what is the value that is in there?
19:31:55 <nielsm> i++ returns the old value before incrementing the variable
19:32:07 <nielsm> so if i==2 before that line, the function is called with 2 as the parameter
19:32:17 <nielsm> and the variable is incremented after the function call
19:33:42 <Samu> well, the assert is comparing lengths
19:34:34 <Samu> assert(get_tunnels_bridges.len() == get_tunnels_bridges_efficient.len());
19:36:01 <Samu> wondering if the assert is working :p
19:36:08 *** andythenorth has joined #openttd
19:36:11 <Samu> gonna make it fail on purpose
19:36:43 <Samu> local bridge_list = AIBridgeList_Length(i); :p
19:37:28 <andythenorth> can't font size just be set by changing font?
19:38:36 <Samu> yep, i got an assert, so it seems ++i or i++ work the same
19:46:18 <Samu> i see these tiles are pushed twice
19:46:32 <Samu> would be better if they could be pushed only once :o
19:47:52 <Eddi|zuHause> andythenorth: what do you mean?
19:50:15 <Eddi|zuHause> that depends... font size of the ingame sprite font cannot be changed. and setting a truetype font can have a separate font size, but there is no GUI for that
19:50:59 <Eddi|zuHause> however, you should be able to override the sprite font with a (static) newgrf
19:51:23 <Samu> do you see the tiles being pushed twice?
19:51:30 <Samu> how would i avoid it from happening
19:52:31 <Samu> they're collected first, then they're being moved again inside that foreach
20:15:28 <andythenorth> such ticket bot?
20:15:46 <andythenorth> peter1138: ^ remind me what it got stuck on? o_O
20:16:55 <peter1138> My Hyper-V guest became unusably slow after the April Windows update, and so I have not had a development environment since then.
20:18:10 *** Gustavo6056 has joined #openttd
20:21:31 <andythenorth> not the expected answer :D
20:21:40 <andythenorth> I thought it was UI crap blocking it
20:21:44 *** Gustavo6056 is now known as Gustavo6046
20:26:58 *** Supercheese has joined #openttd
20:28:34 *** gelignite has joined #openttd
20:47:49 <Eddi|zuHause> hm, somehow they changed something on google maps, and now it is incredibly slow to load
20:48:21 <Eddi|zuHause> like, it does not keep up with slowly scrolling along a road
20:48:55 <Wolf01> They changed the api and again half of the calendars on the apps are borked
20:49:19 <Eddi|zuHause> not using any apps. just firefox
20:49:44 <Wolf01> Me too, web apps which use google calendar :P
20:50:34 <Wolf01> Also I had problems with FF on gmaps since they implemented the 3D version... but I must blame my slow connection too
20:52:29 <andythenorth> sometimes I wonder
20:52:37 <andythenorth> if stuff just stopped changing for a bit
20:52:42 <andythenorth> what might go wrong
20:52:58 <andythenorth> we spend so much engineering time just addressing upstream and downstream change
20:53:04 <andythenorth> that we hardly do any work on our products
21:07:02 <Wolf01> If something works fine, why change it?
21:08:25 <andythenorth> becuase upstream :P
21:08:32 <andythenorth> upstream of upstream
21:18:28 *** Gustavo6056 has joined #openttd
21:21:50 <Eddi|zuHause> thing is, things don't ever "work fine" (CVEs etc.), so not changing isn't a solution either
21:23:00 *** Gustavo6056 is now known as Gustavo6046
21:25:26 *** innocenat has joined #openttd
21:33:21 *** innocenat has joined #openttd
21:34:15 *** Supercheese has joined #openttd
21:35:03 *** rocky113844 has joined #openttd
22:00:15 *** Progman has joined #openttd
22:18:20 *** Gustavo6056 has joined #openttd
22:21:55 *** Gustavo6056 is now known as Gustavo6046
22:30:00 *** m3henry has joined #openttd
22:38:03 *** Tharbakim has joined #openttd
22:38:19 <peter1138> andythenorth, hmm, well, I set up a VNC server instead of using hyper-v's desktop mode and... it's back to normal performance.
22:39:22 <peter1138> Yeah I figured as much, the actual VM performance is fine, it's just the emulated graphics is shitty.
22:41:37 <peter1138> Even OpenTTD plays okay.
22:46:02 <peter1138> Maybe I should... consider... dual-booting again.
22:47:55 *** Wormnest has joined #openttd
22:52:57 <andythenorth> seems like a hassle :P
22:56:37 <peter1138> It is. Linux does weird things with an HTC Vive plugged in.
23:00:17 <andythenorth> one day, computers that work :P
23:13:01 <andythenorth> apple upgraded clang and my compile broke again
23:14:08 <andythenorth> src/economy.cpp:702:20: error: expected expression
23:14:09 <andythenorth> c->cur_economy = {};
23:18:22 *** Gustavo6056 has joined #openttd
23:19:04 <Samu> 305/658/642/100/576/569/584/617/653/695
23:21:22 <Samu> 257/519/516/89/468/466/-9528/505/505/530
23:21:39 <Samu> something doesn't make sense
23:24:32 *** Gustavo6056 is now known as Gustavo6046
23:26:03 <andythenorth> Apple LLVM version 10.0.0 (clang-1000.11.45.2)
23:30:25 <nielsm> does CXXFLAGS contain any --std=c++03 or other old version?
23:30:25 <nielsm> if not, tryh adding CXXFLAGS=--std=c++11 to the configure commandline
23:32:25 * andythenorth wonders whether to add a new issue :P
23:32:28 <andythenorth> or reuse the previous one
23:32:41 <andythenorth> nielsm: that compiles :)
23:32:53 <nielsm> add a note to the existing issue
23:33:21 <nielsm> gcc also understands that flag
23:34:13 <nielsm> so question really is whether to just officially adopt c++11 and reject any compiler not understanding the -std=c++11 flag, or thoroughly vacuum the code of any c++11 feature usage
23:35:33 <andythenorth> I've updated issue
23:35:50 <LordAro> andythenorth: oh dear
23:35:56 <LordAro> your version is now "10" again
23:36:24 <andythenorth> aren't upstream updates brilliant? :|
23:37:18 <LordAro> nielsm: personally i'd say the former
23:37:28 <LordAro> given that's what we're going for with the cmake stuff
23:37:34 <LordAro> but that's a way off being done
23:39:16 <LordAro> andythenorth: only CXXFLAGS is required, fwiw
23:39:46 <andythenorth> so it's resolvable eh
23:40:02 <nielsm> I'd warmly welcome official c++11 adoption
23:40:28 <LordAro> quite a lot of people would at this point :p
23:40:50 <nielsm> for initialization syntax, range-for loops, auto type inference, and lambdas
23:41:27 <nielsm> might break compilation on os'es that only have ancient compilers
23:41:32 <nielsm> that's their problem imo
23:42:26 <nielsm> so anyway yeah bedtime
23:42:42 <Samu> 257/519/516/89/468 vs 306/659/643/101/577
23:43:29 <Samu> @calc (257+519+516+89+468)/(306+659+643+101+577)
23:43:29 <DorpsGek> Samu: 0.808836395451
23:47:15 *** andythenorth has left #openttd
23:59:31 <m3henry> I'm about 60% of the way through re-nuking SmallVector now
continue to next day ⏵