IRC logs for #openttd on OFTC at 2025-06-22
⏴ go to previous day
00:27:27 *** Tirili has quit IRC (Quit: Leaving)
01:10:09 *** gelignite is now known as Guest18546
01:10:15 *** gelignite has joined #openttd
01:17:29 *** Guest18546 has quit IRC (Ping timeout: 480 seconds)
02:02:59 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
02:03:00 *** WormnestAndroid has joined #openttd
02:03:16 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
02:03:17 *** WormnestAndroid has joined #openttd
02:03:20 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
02:03:22 *** WormnestAndroid has joined #openttd
02:14:48 *** Tirili has quit IRC (Quit: Leaving)
02:27:25 *** WormnestAndroid has quit IRC (Remote host closed the connection)
02:27:28 *** WormnestAndroid has joined #openttd
02:29:31 *** gnu_jj_ has quit IRC (Ping timeout: 480 seconds)
02:32:55 *** firepup650 has joined #openttd
02:32:57 *** Wormnest has quit IRC (Quit: Leaving)
04:43:56 <DorpsGek> - Update: Translations from eints (by translators)
05:22:28 *** WormnestAndroid has quit IRC (Remote host closed the connection)
05:22:30 *** WormnestAndroid has joined #openttd
05:37:58 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
05:40:30 *** WormnestAndroid has joined #openttd
07:10:30 <truebrain> Nope, my fix did not fix it 😄
07:11:58 <truebrain> Okay, I also can't access the R2 via scripting atm
07:16:10 <truebrain> Right, this was another error than last time. Seems they had a weirder issue going through their network, and I was doing things on the bucket too.
07:19:17 <truebrain> Write actions on the bucket are also terribly slow, compared to earlier today
07:23:57 <truebrain> awh ... it was doing so well the second time
07:42:19 <truebrain> I think I know a way so we can retrigger these failed workloads .. sadly, I don't really have a way to test it other than testing it. So I might spam a few unsuccessfull while working that out.
07:53:46 <truebrain> And you can now retrigger a failed CDN upload. As long as the content is the same, it will be able to reupload.
07:53:53 <truebrain> Not a perfect solution, but one that will have to do 🙂
07:57:20 <truebrain> forgive the name; that will fix itself when we merge all this 🙂
07:58:25 <truebrain> minor issue found .. `CHANGELOG.md` vs `changelog.md` 🙂
07:58:50 <truebrain> I see that the website is in general just a bit weird what "changelog" it picks
08:25:20 *** keikoz has quit IRC (Ping timeout: 480 seconds)
08:34:13 <truebrain> Right. Now OpenGFX2 PRs just need reviews. But they work / are done 🙂
08:47:42 <truebrain> We were nearly out of "free" LFS, and these days it just aborts your clone. So I assigned a small budget for LFS usage 🙂 Should be fine!
09:54:41 *** gelignite has joined #openttd
12:51:17 <_zephyris> Looks good to me! But, I'll need another reviewer because I originally authored the release workflow PR.
12:51:47 <truebrain> all I needed to know is that you were okay with it 🙂
12:52:36 <truebrain> Right, biggest hurdle done! 😄
12:53:11 <locosage> omfg, I remember steel mills having pax acceptance but coal mines having it at 2/8 is probably new to me
12:53:20 <locosage> so 4 mines accept passengers %)
12:53:27 <truebrain> There are some optimizations we can do, like caching etc. Or what _glx_ suggested, trying MacOS runners.
12:53:27 <truebrain> When I have some time I will check that out, but no promises time-wise 😄
12:54:19 <truebrain> Going to kick off a nightly run, just so there is one. As the one there is now, is a bit weird to have 😛
12:54:48 <locosage> actually, a lot of vanilla industries have partial pax acceptance 😵💫
12:56:20 <pickpacket> locosage: does that mean that grouping a bunch of these industries together makes them accept pax?
12:56:51 <_zephyris> truebrain: Looks great!
12:57:05 <pickpacket> _zephyris: YOU look great!
13:05:07 <peter1138[d]> Someone would probably report that original behaviour as a bug...
13:05:34 <locosage> didn't they already?
13:08:14 <locosage> and farm actually does have like 2/8 or smth
13:12:40 <locosage> wonder if this is too confusing
13:17:31 <peter1138[d]> Yeah, you don't need to know that it's more 1.
13:18:19 <locosage> it's nice to have more than 8 when placing goods station in town as they grow and sometimes lose acceptance
13:19:37 <locosage> hm, maybe only do that for house cargo...
13:35:05 *** akimoto has joined #openttd
13:47:44 <locosage> funnily cargo with the most unreliable acceptance is called always_accepted in the code 😅
13:58:11 <digitalfox> truebrain: + _zephyris awesome job getting automatic builds of “classic”
13:58:11 <digitalfox> Any change for HD releases? Please pretty please? 🙂
13:59:24 <truebrain> Without someone spending some proper time on caching etc, I doubt it is reasonable to run that on GitHub CI 🙂
13:59:53 <digitalfox> Oh I see, well someday 🙂👍
13:59:58 <truebrain> But it fully depends on someone spending the time on it, I guess
14:00:47 <truebrain> no, my "is" was correct; I shouldn't re-read my sentences and think I got it wrong 😛
14:10:17 <peter1138[d]> Can't andy's MBP do it?
14:24:02 <andythenorth> one of my kids has one to sell
14:24:05 <andythenorth> can we use that? 😛
14:49:35 <_zephyris> Needs to be more complete too, IMO
14:58:51 *** Wormnest has joined #openttd
16:08:26 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
16:08:31 *** WormnestAndroid has joined #openttd
16:08:35 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
16:08:36 *** WormnestAndroid has joined #openttd
16:08:56 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
16:08:57 *** WormnestAndroid has joined #openttd
18:35:20 <andythenorth> player-configurable speed limit for each installed railtype? o_O
18:56:45 <andythenorth> lol theatre bug manifests 😄
19:02:09 <peter1138[d]> Isn't that just starting the game too early for the default houses?
19:02:19 <peter1138[d]> Which is fixed by loading NewGRF houses.
19:08:09 <peter1138[d]> It's been... raining.
19:10:23 <andythenorth> seems I closed a bug report about it in 2013 with 'use newgrf'
19:10:32 <andythenorth> I forgot to load a house grf in my game 😛
19:37:38 <peter1138[d]> I can't help feeling that station cargo graphs are not that useful.
19:38:00 <andythenorth> do they do any harm? 🙂
19:38:36 <peter1138[d]> And is town population and number of houses worth graphing?
19:39:08 <andythenorth> over what time period?
19:39:27 <peter1138[d]> Same... 24 months, 24 quarters, 24 years.
19:39:44 <andythenorth> I suspect we have players who would enjoy studying that
19:42:21 <locosage> peter1138[d]: probably more useful than cargo
19:42:54 <peter1138[d]> Needs separate y-axis again, as they're not related...
19:42:57 <locosage> though need to be careful with the amount of data as there can be a lot of towns
19:43:20 <locosage> yeah, I had town graphs somewhere as an example
19:44:26 <peter1138[d]> Sadly town population can be huge so needs a uint32_t per data point :/
19:45:40 <peter1138[d]> JGR's RxCompress allows up to ~ 3963072
19:45:43 <locosage> it's not very useful for towns that have no service in
19:46:17 <peter1138[d]> Maybe that's enough for normal game play?
19:47:09 <peter1138[d]> Pretty sure this is just an fiddled-with anomoly that can be disregarded.
19:48:16 <locosage> you mean town so big? may be doable with game script
19:49:09 <locosage> oh, yeah, gs have expand command that ignores normal growth rules
19:49:30 <locosage> max normal growth with gs peaks around 1.75m houses I think
19:49:38 <locosage> which isn't far off to be fair
19:49:54 <locosage> without gs it's about 25k housse
19:52:15 <locosage> green fill under the chart is funding
19:54:12 <peter1138[d]> Well, using RxCompress drops the population/houses history to 244 bytes per town.
19:54:20 <peter1138[d]> Which is basically nothing.
19:57:40 <locosage> I kinda missed what is RxCompress 😅
19:58:31 <peter1138[d]> We store production and transported as separate `uint32_t` fields. Could probably use 1, use 24 bits for production, and 8 bits for transported %age.
19:58:46 <peter1138[d]> Not 100% accurate though 🙂
19:59:15 <dh1> not 100% accurate is realistic though
19:59:24 <peter1138[d]> RxCompress is something in JGRPP to range compress a value larger than 65535 to fit into a uint16_t.
19:59:27 <dh1> how many transport companies have perfect accounting and stocktake
19:59:51 <dh1> all values in the game should be put through a filter of true to life inaccuracy
20:00:12 <locosage> peter1138[d]: oh, is that in memory compression or savegame only?
20:00:43 <dh1> is there a weather factor for speed?
20:01:02 <peter1138[d]> If I use it for town history, I can store population / num houses up to 3.9 million in a uint16_t each.
20:01:30 <peter1138[d]> Which is why I'm wondering if caring about more than 3.9 million is necessary... even though I have this savegame with one much larger.
20:01:37 <peter1138[d]> (And yes, it is just one town.)
20:02:15 *** WormnestAndroid has quit IRC (Remote host closed the connection)
20:02:17 *** WormnestAndroid has joined #openttd
20:02:45 <locosage> well, I have 1mil city-builder server and consider that ridiculously high goal already
20:02:50 <locosage> even with gs schenenigans
20:03:13 <locosage> no-gs town peak at about 2mil population iirc
20:03:56 <locosage> on the other hand, people are more likely to care about largest towns
20:07:24 <peter1138[d]> Yellow line is clamped 😮
20:08:41 <locosage> peter1138[d]: it's not uncommon for a game to have about a thousand towns so it's still quite a lot though
20:09:09 <peter1138[d]> If you have 1000 towns you already have a large game.
20:11:19 <locosage> last Hellish event had 1400 towns and it's still 5Mb save at the end with all the infrastructure
20:11:32 <locosage> 5mb compressed though
20:14:10 <locosage> 244 bytes for what time range though?
20:14:15 <peter1138[d]> Link the save game?
20:15:42 <locosage> let me see if there is a link on mh website
20:15:47 <locosage> I just did the save myself
20:18:52 <peter1138[d]> Hmm, master save is smaller 🙂
20:19:03 <locosage> it's cmclient save so there is some extra
20:19:32 <peter1138[d]> Oh fuck, tainted.
20:22:21 <peter1138[d]> Doesn't matter, I saved it in master to get a useful comparision anyway.
20:22:21 <locosage> he uses vanilla so you'd be happy :P
20:23:15 <locosage> it's still better to get vanilla save though for clean result
20:23:31 <locosage> I'm not 100% sure cmclient doesn't still affect it after re-save
20:23:40 <peter1138[d]> Probably but I cba.
20:23:47 <locosage> It did before but maybe not anymore
20:27:16 <peter1138[d]> -rw-r--r-- 1 petern petern 5340476 Jun 22 21:26 'Freaky Express, 2091-05-08b.sav'
20:27:16 <peter1138[d]> -rw-r--r-- 1 petern petern 5265592 Jun 22 21:26 'Freaky Express, 2091-05-10.sav'
20:27:41 <peter1138[d]> Couldn't be bothered to run it much longer 😮
20:28:24 <peter1138[d]> So with 10 months of extra data...
20:28:40 <peter1138[d]> This isn't clean though, as it also includes extra industry history data.
20:29:31 <locosage> those filenames don't even say which one has more data xD
20:29:38 <locosage> I guess the larger one?
20:29:47 <peter1138[d]> b is, er, my branch :p
20:30:12 <peter1138[d]> Ah yea, I let the game run for 10 months in master and in my branch
20:30:37 <peter1138[d]> master grew by 60kb.
20:30:40 <locosage> so like extra 80k compressed for all the town data?
20:31:29 <peter1138[d]> No, 80K compressed for all the town data, and all the extra industry data.
20:31:46 <peter1138[d]> But only 10 months, so a lot of it is still zeros.
20:33:34 <locosage> though, as usual, question is not only the save size but server connection time
20:34:31 <locosage> I didn't get to investigate it yet but there is something fishy with map download speed when connecting
20:34:34 <peter1138[d]> 80kb is not going to make much of a difference on a 5.2MB file.
20:34:46 <locosage> it seems to either be bottlenecked by compression speed or smth else entirely
20:35:28 <peter1138[d]> Who cares. Run stupid large maps, expect longer connection time.
20:36:10 <locosage> I care, I sometimes can't even connect to MH events reliably :p
20:36:58 <locosage> though I guess compared to tree noise it's all irrelevant xD
20:41:04 <locosage> oh, you closed my tree pr 😢
20:41:23 <locosage> but only the better one for some reason %)
20:41:36 <locosage> well, no trees in multiplayer then I guess :P
20:41:56 <locosage> did you at least take a look at tileloop speed improvement there?
20:42:02 <locosage> it's like 20% or smth
21:14:41 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
21:38:09 *** keikoz has quit IRC (Ping timeout: 480 seconds)
21:46:40 *** tokai|noir has joined #openttd
21:46:40 *** ChanServ sets mode: +v tokai|noir
21:53:39 *** tokai has quit IRC (Ping timeout: 480 seconds)
23:55:08 *** Flygon has quit IRC (Read error: Connection reset by peer)
continue to next day ⏵
