IRC logs for #openttd on OFTC at 2022-10-23
⏴ go to previous day
00:03:42 *** wallabra_ has joined #openttd
00:07:37 *** wallabra has quit IRC (Ping timeout: 480 seconds)
00:07:37 *** wallabra_ is now known as wallabra
00:11:34 *** wallabra[m] has joined #openttd
00:17:46 <TallTyler> petern: I found a bug with the infrastructure window which is in master but is more apparent in a scrolling window. The clipping region is too short and cuts off part of the bottom. Here it's just the drop shadow of the P in Airports but in a scrolling window it cuts text in half. Any idea where I should look to fix this?
00:26:55 <glx[d]> wrong height declared somewhere ?
00:43:56 <glx[d]> as if dxdiag would help us
00:56:26 <glx[d]> in debugger `size->height` seems right (it's the sum of individual bounding box heights
00:59:53 *** Eddi|zuHause has quit IRC (Remote host closed the connection)
01:00:38 *** Eddi|zuHause has joined #openttd
01:12:59 <TallTyler> Yeah I don’t think that’s it
01:13:36 <TallTyler> JGR’s patch which I’m adopting rewrites the height calculation entirely and I still get the bug
01:23:11 <glx[d]> g of signals and y of tramway are affected too
01:24:15 <glx[d]> and it's the same in company window
01:27:19 <glx[d]> I guess it's because the shadow is drawn over next widget, and it's erased when the widget is cleared
01:53:55 *** wallabra_ has joined #openttd
01:55:03 *** wallabra has quit IRC (Ping timeout: 480 seconds)
01:55:03 *** wallabra_ is now known as wallabra
02:35:00 *** Wormnest has quit IRC (Quit: Leaving)
02:38:38 <glx[d]> definitely the widget size, string bounding box doesn't count shading
02:52:45 <Arastais> what part of compilation auto-generates the enums in `script_window.hpp`? it's including my new widget file
02:53:43 <Arastais> and when i change something it removes what i manually included
02:54:46 <glx[d]> it's all done by a CMake script
02:56:04 <Arastais> do you know which one? i can't find it in the `src/` folder one
02:56:31 <glx[d]> it's not in src, it's in cmake/script
02:59:09 <glx[d]> using a custom font the issue is non existent, I'd blame `_default_font_height`
03:01:17 <Arastais> glx[d]: found it, but i dont understand why its not including the widget header. It's extracting the enum from it but not actually including the file itself?
03:03:08 <glx[d]> you modified src/script/api/script_window.hpp.in ?
03:04:32 *** D-HUND has quit IRC (Ping timeout: 480 seconds)
03:05:04 <glx[d]> that's the generated file source
03:05:35 <glx[d]> the script auto adds stuff to it
03:06:27 <Arastais> glx[d]: i did but i misspelled the header to include 🤦
03:07:57 <glx[d]> // @enum .*Widgets ../../widgets/*_widget.h@ENUM_WIDGETS@ <-- this line adds the enums based on a regex
03:08:58 <glx[d]> hmm I should look to make it autofill the #include list
03:16:33 <Arastais> another thing: when i try to call `DrawWidgets()` in one of my custom windows, i fail the `_grf_text[stringid].grfid != 0` assertion since the _grf_text array is empty, but its only in this wonder, even tho I've added other similar windows. I've never seen this befoer and the call stack/debugger doesnt help too much here. Any ideas?
03:20:34 <Arastais> I have a window almost exactly like this one and it doesnt fail the assertion, and shares a lot of StringIDs
03:41:35 <glx[d]> it's trying to draw a string coming from a newgrf
03:44:20 <glx[d]> you should find the DrawString() call in the stack
04:06:11 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
04:44:01 <Arastais> glx[d]: Thats what i assumed, but that doesn’t make much sense since like i said its just this window. Is it possible that i left out a StringID?
04:44:25 <Arastais> glx[d]: I looked at that briefly, but couldnt find much. Ill take a better look. Thanks
05:07:37 *** WormnestAndroid has joined #openttd
05:30:12 *** sla_ro|master has joined #openttd
05:35:14 *** MaverickFischer has quit IRC (Quit: Due to budget cuts, the light at the end of the tunnel has been turned off.)
05:37:38 *** MaverickFischer has joined #openttd
06:15:24 *** sla_ro|master has quit IRC ()
06:43:32 *** bryjen has quit IRC (Quit: Leaving)
07:20:40 <petern> TallTyler: Text shadows aren't included in the bounding box, so always get cropped if a drawpixelinfo clipping region is used.
07:22:35 <petern> Every single widget and list would end up with more spacing if it was, so it's probably better to just include an extra pixel in the clipping reguon
07:22:55 <petern> And yes this happens already in a lot of places
07:31:19 <petern> So basically if it's just the shadow that's clipping, don't worry about it.
07:32:09 <petern> If it's the giant gap at the bottom, that's the giant widget padding at the bottom that is easy enough to fix (as I have done in my scaling branch)
08:16:27 <andythenorth> what shall we do today Brain?
08:17:56 <TrueBrain> being utterly confused about some code ..
08:18:13 <TrueBrain> I check if a pointer is not a nullptr, and free it .. it crashes, because I am freeing a nullptr ..
08:18:16 <TrueBrain> HOW DOES THAT MAKE SENSE?!
08:18:41 <andythenorth> are you checking the correct pointer? 😛
08:18:54 <andythenorth> remember, `make` is not `make install` etc 😛
08:19:20 <andythenorth> I found 'cabbage' and 'cabbbage' in code
08:19:32 <andythenorth> supposed to be 'cabbage'
08:19:38 <andythenorth> but happened to work until it didn't
08:22:27 <andythenorth> what if C++ is broken? 😮
08:22:51 <andythenorth> hmm what if checking the pointer causes it to become null 😛
08:23:13 <andythenorth> or it could be a bug in your CPU
08:25:11 *** gelignite has joined #openttd
08:26:11 <TrueBrain> nah, it is much simpler .. it is a thread-issue instead
08:26:37 <Eddi|zuHause> thread issues are simple. right.
08:29:47 <andythenorth> a friend of my produced a diagnostics tree once
08:30:11 <andythenorth> if all other things have been ruled out, then the only remaining cause is that there is a swarm of magnetic bees nearby
08:34:53 *** Tirili has quit IRC (Quit: Leaving)
08:37:18 <petern> andythenorth: difficult to find a 10 pixel high truetype font with the clarity of TTD's sprite font :/
08:37:47 <TrueBrain> in this case it is just as simple as: create a thread in your ctor body, not earlier .. the object might not been fully initialized yet 😄 So silly
08:38:16 <andythenorth> petern: also 'free and license compatible'?
08:39:02 <petern> Fonts seem to come in SIL Open Font License, I have no idea if that is compatible.
08:39:26 <nielsm> the classic fonts included with windows and macos are truetype fonts designed for low res displays, but yeah those aren't really free
08:39:42 <andythenorth> controversially, we could license a font 😛
08:39:43 <petern> Windows does even use those any more 🙂
08:39:47 <nielsm> Helvetica, Arial, Verdana, Tahoma, (Times)
08:40:24 <petern> Even Arial doesn't work, it loses its boldness at small sizes.
08:40:51 <nielsm> I remember reading that one of the major design goals of Tahoma specifically was readability/clarity on lower res displays, and comfortable for reading stuff on web pages
08:41:45 <petern> Yes. If you see Verdana/Tahoma on a website these days it means it was designed in the late 90s/early 2000s, and hasn't been updated since.
08:42:13 <petern> Verdana Bold looks okay at 10px, but it looks bad when scaled up.
08:43:17 <nielsm> personally I use Tahoma for small and normal game font, and Times New Roman for the newspaper font
08:43:40 <nielsm> actually maybe it was Georgia for newspapers
08:54:57 <petern> Also I just modified `zoom_func.h` so now waiting for recompile 😦
08:55:03 <nielsm> I have small_size=9, medium_size=12, large_size=18, mono_size=12
08:55:18 <nielsm> that is somewhat larger than default I guess
08:56:18 <petern> Kinda. 6, 10, 18, 10, but in the sprite font there is no space above the glyphs.
08:56:40 <petern> Truetypes all have some space above, not even consistently.
08:58:06 <petern> I'd be happy if I could remove that 🙂
09:01:14 <andythenorth> hmm searching for fonts is meh
09:01:20 <andythenorth> so much of the web is now broken
09:02:01 <petern> Good job we left, it's much better now.
09:02:14 <andythenorth> if the referendum had solely been on cookie law, I think it would have been 96% Brexit
09:02:31 <petern> It was mostly on racism.
09:03:43 <andythenorth> well I am just glad Dominic's strategy has worked
09:03:58 <andythenorth> his idea was that Europe was dividing the Tories and making it hard for them to govern
09:04:03 <andythenorth> so get it out the way
09:04:21 <petern> They are very good at it now, yes.
09:04:45 <Eddi|zuHause> good thing the tories are not divided anymore...
09:06:15 <Eddi|zuHause> so concluding this morning: threads are simple, fonts are simple, politics are simple
09:08:48 <petern> Oh right, templates don't work like that in C++ 😄
09:12:33 <TrueBrain> ah, okay, this test-framework forks test-cases ... which is interesting when testing a TCP-based library .. lol
09:13:12 <Eddi|zuHause> sounsd like a DDoS recipe
09:13:58 <Rubidium> Eddi|zuHause: simple is the new difficult? ;)
09:19:23 *** HerzogDeXtEr has joined #openttd
09:27:44 <Eddi|zuHause> andythenorth: now implement ships consisting of multiple sprites with separate bounding boxes
09:28:29 <andythenorth> they already consist of multiple sprites 😛
09:28:48 <Eddi|zuHause> great, that means you are half done :p
09:29:01 <andythenorth> just the hard 50% to do
09:29:22 <Eddi|zuHause> *note* bounding boxes: simple.
09:30:39 <andythenorth> sorry EU, but how was this ever a good idea?
09:30:58 <andythenorth> oh wait, you say what, the big adtech vendors deliberately influenced the legislation?
09:31:27 <andythenorth> so that most people would just start clicking explicit 'accept' to everything?
09:31:44 <petern> The lawmakers didn't anticipate that the advertisers would make as it difficult as possible to opt-out, for some reason.
09:31:45 <andythenorth> and so that most of the web became horrific, compared to e.g. facebook
09:32:30 <Eddi|zuHause> that's what happens when you got malicious actors
09:32:40 <petern> Just use Consent-O-Matric.
09:33:14 <Bouke> Well the law requires that declining should be as easy as accepting. So this banner from Yahoo is not conformant.
09:33:35 <Bouke> Google recently changed their consent to be conforming.
09:34:20 <Eddi|zuHause> andythenorth: blaming this on the EU lawmakers (as bad as they probably are) is like shooting the messanger
09:34:38 <andythenorth> I'm not reserving the blame exclusively
09:34:44 <andythenorth> there's enough for everyone
09:34:56 <andythenorth> even we can have some, because very few of us want to pay for information
09:35:14 <andythenorth> but information does not want to be free, even though it sounds good to say so
09:35:47 <Bouke> Also (and this is the tragic part here), the intent was probably that website owners wouldn't want to put up such a banner and would change their website such that consent wasn't needed. But adtech and data grabbing prevailed.
09:36:33 <andythenorth> the actual Stewart Brand position is misquoted
09:36:34 <petern> Alright, not keen on Tahoma either 😦
09:36:40 <andythenorth> information wants to be free and expensive
09:36:50 <petern> How long to convert the bitmap font to truetype...
09:37:09 <andythenorth> it's almost impossible to run a modern website without cookies
09:37:27 <andythenorth> (non-essential cookies)
09:37:51 <Bouke> How so? Why do you need to set cookies?
09:38:05 <andythenorth> they leak in from e.g. embedded media services etc
09:38:19 <andythenorth> so unless you want to...run your own video platform etc
09:38:48 <petern> I self-host web 'assets' these days instead of using CDNs.
09:38:51 <Bouke> Well yeah when you're including free services from adtech and data grabbing providers, that's what happens.
09:39:05 <andythenorth> it's a constant battle
09:39:18 <andythenorth> some of these vendors are paid for on business accounts, but they constantly set cookies
09:39:26 <andythenorth> even when the account is configured not to
09:39:26 <petern> (Linking to Google to get a font may require consent...)
09:39:32 <Bouke> With streaming-over-http supported almost everywhere (thanks to Apple), hosting videos shouldn't be that hard anymore?
09:39:54 <andythenorth> it's so much more than bandwidth
09:40:11 <andythenorth> if you have a SaaS tool for staff to create, edit and auto-transcript videos
09:40:34 <andythenorth> putting friction in the way where we have to hire a web dev to download those and upload them to a CDN is ...
09:40:46 <michi_cc[d]> LordAro: You totally look like you want to start a changelog PR for a potential Halloween beta release 🙂
09:41:22 <andythenorth> Halloween = local authority no longer cares about rating
09:41:34 <andythenorth> that feature alone is worth shipping yesterday
09:42:39 <Bouke> That means we no longer get to plant trees?! I guess we'll need a special Christmas edition as well then. 🌲
09:42:47 <andythenorth> christmas tree farm
09:42:53 <andythenorth> I think there's a grf for that
09:44:32 <Bouke> petern: Sure, you're using "free" services from the biggest advertising business, they'll want something in return.
09:44:49 <andythenorth> nah these are paid B2B SaaS
09:44:56 <andythenorth> but the industry is so used to slurping data
09:45:17 <andythenorth> so they think nothing of sticking some cookies into their embeds
09:45:23 <andythenorth> even if they claim GDPR compliance etc
09:45:36 <andythenorth> the default position is 'so what, just add us to your cookie consent'
09:45:47 <andythenorth> or 'why doesn't your cookie consent automate this?'
09:45:52 <Bouke> The free services here being the fonts google provides, I don't think they offer a business account?
09:46:28 <andythenorth> google fonts didn't set cookies last time I checked, they just potentially use the referrers 🙂
09:46:36 <Bouke> Sure... such B2B Saas are a different kind of evil.
09:46:47 <petern> And they have the client's IP address.
09:49:15 <Bouke> Yeah I wondered about that too. By default Google Analytics is set to store IP addresses. It is claimed that GA cannot be used in the EU in its default configuration. Turning off IP address collection in GA fixes that. But... how can you (as an external party) even check if that happens? And the IP address is going to Google regardless.
09:49:59 <petern> Trust good, they will never do Evil...
09:51:32 <andythenorth> but how else will Google sell AdSense?
09:51:46 <andythenorth> what's good for Google is good for the Google
09:51:49 *** nielsm has quit IRC (Ping timeout: 480 seconds)
09:53:30 <andythenorth> why can't Dall-e draw these boats for me?
09:54:37 <pickpacket> andythenorth: you need to seed it with enough of your own boats first. A few thousand should probably do
09:54:43 <petern> Alright, snapping the slider to 10% means you can easily get to round numbers, but rule out 25% increments. Maybe snap to 5%....
09:55:05 <petern> (Or maybe even 25% is enough)
09:57:42 <Eddi|zuHause> 25% steps probably enough
09:57:57 <Eddi|zuHause> use a modifier key to disable snapping
09:59:57 <petern> I still need to set the tooltip, it's currently still the music/sound one 😄
10:00:04 <petern> (And perhaps that one should be split too)
10:00:04 <Rubidium> just snap to any multiple of 1/10, 1/4 and 1/3 ;)
10:00:41 <Eddi|zuHause> and what about 1/sqrt(2)?
10:01:39 <petern> I could put in 25% scale markers, but limit to 10%, so you can't reach half of them 😄
10:02:21 <Rubidium> you can reach 60% of them ;)
10:02:24 <Eddi|zuHause> "oddly unsatisfying" :p
10:02:55 <Eddi|zuHause> only make the 5 ones reachable, instead of the 0 ones :p
10:05:12 <Eddi|zuHause> also, live-rescale the slider while you slide it, so you can't ever slide it correctly with the mouse :p
10:05:15 <petern> 50% + 1, no? The full range is not 0-100.
10:05:33 <andythenorth> can we have a live slider for UI zoom?
10:05:35 <petern> Haha, that is what it did originally
10:05:52 <petern> andythenorth: I'm Discord-streaming it
10:07:16 <petern> _ctrl_pressed modifier ftw
10:07:39 <andythenorth> chunky bevels should be mandatory
10:07:55 <petern> I'm also feeling that tbh
10:08:28 <Eddi|zuHause> remove all settings, make the game only playable "as the programmer intended"...
10:09:52 <petern> That label colour inconsistency...
10:10:17 <petern> Should probably remove the colon too.
10:10:59 <petern> The thing with this branch is, once it's done, it's pretty underwhelming
10:11:43 <petern> Hah, I made RTL slider scale work
10:15:54 <LordAro> michi_cc[d]: i'll see what i can come up with :)
10:16:20 <petern> Hmm, maybe I should remove the bevel option. Dunno.
10:16:53 <petern> And also the slider is missing chunky ness, as it's on a gradient, Hmm.
10:16:56 <LordAro> petern: i want you to keep both 11x & the bevel option
10:17:25 <petern> 11x is ridiculous, I've put it to 5x now.
10:18:32 <michi_cc[d]> I only approve of a chunky bevel option if there is also an anorexic bevel option 🙂 (Translation: go it throw out of the window the option).
10:20:31 <petern> Oh, simply drawing the slider with thick lines mostly works.
10:21:05 <petern> Although the non-right-angle corners look a little off.
10:27:30 <petern> Hmm, `std::map<int, std::string>` can have nullptr assigned as the string, but .second can't be compared with nullptr
10:27:56 <Bouke> Rivers should force water flow, even after terraforming
10:28:00 <petern> Probably assigning nullptr isn't valid.
10:48:49 <LordAro> TrueBrain: your backport PRs never added the 'backported' label :(
10:49:06 <TrueBrain> they still have the other label?
10:49:15 <TrueBrain> what was the PR that backported them, do you know?
10:49:16 <LordAro> nope, removes the backport requested label
10:49:39 <LordAro> #9602 & #9611 are the ones i've found so far
10:49:49 <LordAro> i've fixed the PRs manually though
10:50:09 <TrueBrain> those PRs don't have the backported label no, as they are not the ones being backported
10:50:17 <TrueBrain> they contain PRs that are backported, which are labeled correctly
10:50:21 <TrueBrain> not sure what you are sad abou tnow 😄
10:50:36 <LordAro> those are the backport PRs
10:50:42 <TrueBrain> yes, but they are not backported 😄
10:50:52 <TrueBrain> we label PRs that we did backport as "backported" 🙂
10:51:00 <TrueBrain> "backport requested" -> "backported" 🙂
10:51:20 <LordAro> they were not done by whatever automation you were using
10:51:25 <TrueBrain> owh, you should have said before you did .. I have a script that does that 😛 Clearly it wasn't run 🙂
10:51:26 <LordAro> they only removed the backport requested label
10:52:01 <TrueBrain> either way, those PRs contain enough information to run the script again
10:53:49 <LordAro> looks like subsequent backport PRs did add the label
10:54:04 <TrueBrain> guess I did fix the bug already then 😛
10:55:42 <pickpacket> I know you've all been wondering what a teahouse looks like. Well, ladies and gentlemen, the results after my research are in
10:55:59 <pickpacket> They look like *drumroll* any other house!
10:56:18 <pickpacket> Consequently I have no idea where to draw inspiration from :P
11:11:56 <Eddi|zuHause> that PR description is either a one-liner or a giant refactoring :p
11:12:32 <michi_cc[d]> It's unlikely that you'd do a giant refactoring to fix one specific git commit.
11:13:19 <Eddi|zuHause> dunno, i have seen some really euphemistic descriptions here :p
11:13:38 <LordAro> michi_cc[d]: congrats on PR #21
11:16:35 <andythenorth> oof drawing pixels without my glasses
11:16:41 <andythenorth> but is it better? 😛
11:18:02 <pickpacket> towns have noise limits? never seen that before
11:19:23 <Eddi|zuHause> yes, those are the nimbys that don't want an airport
11:28:03 <glx[d]> For some unknown reason I can't merge with GitHub mobile app (something went wrong)
11:36:05 <pickpacket> Eddi|zuHause: it's a non-default setting?
11:40:52 <andythenorth> can we merge faster PBS from JGRPP?
11:42:03 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
11:42:08 *** WormnestAndroid has joined #openttd
12:06:27 <petern> Ok, removing bevels button...
12:22:28 <petern> That's 1 yes and 1 no :p
12:22:47 <petern> (2 yes including andythenorth but we shouldn't listen to him)
12:30:02 <andythenorth> chunky bevels discord theme?
12:30:33 <petern> No bevels in Discord, just rounded rectangles.
12:31:03 <petern> I did wonder if a "flat" widget layout would work, but seems unlikely as we stuff buttons together with no spacing.
12:35:38 <FLHerne> andythenorth: "faster"?
12:36:20 <andythenorth> trains reserve further ahead
12:36:34 <andythenorth> so they're not always driving up to a red signal
12:36:39 <andythenorth> clearly "faster"
12:36:45 <LordAro> isn't that implemented really hackily?
12:37:01 <andythenorth> it's really effective in my limited tests
12:37:11 <andythenorth> it's not actually "faster", I think it's probably "slower"
12:37:18 <andythenorth> JGR: will know 🙂
12:37:37 <andythenorth> trains will reserve ahead through junctions I think, blocking them for other trains
12:37:44 <andythenorth> and it may be slower in the implementation, no idea
12:38:05 <FLHerne> oh, the multi-aspect thing
12:38:16 <andythenorth> yeah, although you don't need to use the MAS
12:38:23 <andythenorth> you can just use PBS red/green
12:44:04 <TallTyler> I really like the multi-aspect thing 🙂
12:45:21 <TallTyler> The big improvement is that trains finding a yellow signal slow down so trains of different speeds can run on the same track without constantly catching up to each other and fully stopping. They just slow down and keep running on yellows without ever fully stopping.
12:46:58 <EmperorJake> TallTyler: There's a separate feature that allows trains to speed match with the one before it, it doesn't even need realistic braking I think
12:50:58 <andythenorth> TallTyler: faster
12:51:13 <andythenorth> PBS must be slow, because trains frequently stop and wait
12:51:53 <petern> I find it amusing how the reasoning behind "PBS is slow" morphed...
12:52:22 <andythenorth> might not be entirely accidental 😛
12:52:22 <petern> Slower code performance vs slower train speed
12:52:48 <petern> You and your memeing it
12:53:04 <andythenorth> when there are lolz, they must be harvested
12:54:26 <andythenorth> hmm should I actually watch Gold Rush?
12:54:30 <andythenorth> or just clips of Tony Beets?
12:54:36 <andythenorth> I think I tried watching the actual show once
12:54:48 <andythenorth> I wish I could subscribe to an alternative Discovery offering
12:55:15 <andythenorth> "Ice Road Truckers, but without the manufactured drama and competition"
12:55:36 <andythenorth> Deadliest Catch is actually ok for this, there is enough drama without them inventing much
12:59:18 <LordAro> i've been watching compilations of old HIGNFY series
12:59:48 <LordAro> "ooh look, isn't Boris so amusing"
13:01:07 <TallTyler> EmperorJake: I thought that was incompatible with realistic braking and was automatically disabled but I could be wrong
13:02:52 <EmperorJake> Hmm the settings don't say anything like that
13:16:56 <petern> It's dried up, so I guess I could go out for a bike ride.
13:17:08 <petern> Apart from my weird aversion mid-afternoon rides.
13:20:11 <andythenorth> worst body-clock time of the day for exercise no?
13:20:18 <petern> It just feels like I should be coming home again already. Energy levels are different. Or something.
13:20:26 <andythenorth> stay in, merge faster PBS 😛
13:21:15 <andythenorth> no PBS on school nights?
13:29:14 *** ChanServ sets mode: +v tokai
13:29:53 <JGR> TallTyler: They are compatible, the speed matching thing is from Joker's parchpack
13:31:27 <JGR> The intent of the realistic braking feature is not about the "speed" of PBS at all
13:35:59 *** tokai|noir has quit IRC (Ping timeout: 480 seconds)
13:39:50 *** WormnestAndroid has quit IRC (Remote host closed the connection)
15:30:30 <andythenorth> Venga boys are back in town
15:33:06 *** Wormnest has joined #openttd
15:33:18 *** WormnestAndroid has joined #openttd
15:33:52 *** Extrems has quit IRC (Remote host closed the connection)
15:34:36 *** Extrems has joined #openttd
15:53:31 *** supermop_Home has joined #openttd
16:13:05 *** sla_ro|master has joined #openttd
16:38:25 <supermop_Home> I wonder if anyone has ever modified an F or FM to half-frame
16:40:03 <supermop_Home> feels like you'd just need to cut a custom gear for the film advance lever, re-screenprint the frame counter wheel, and add some blinders behind the shutter?
16:40:52 <supermop_Home> i guess you'd want to etch some new lines on the focusing screen
16:45:33 <supermop_Home> ok i guess that concludes that thought experiment. the answer is yes
17:23:36 <andythenorth> oof 2 more boats to draw
17:47:06 *** Lord_Scruby has joined #openttd
17:47:06 <Lord_Scruby> the shaping does look pretty good
18:19:24 <Bouke> JGR: I find it strange that “realistic acceleration” is in vanilla (and on by default?) while “realistic breaking” is not. I think vanilla should also have the latter.
18:19:57 <JGR> It is too out of scope for vanilla
18:20:06 <TallTyler> Realistic acceleration makes the game easier, not harder. Realistic braking makes signaling harder.
18:20:19 <TallTyler> Original acceleration is a bit garbage 🙂
18:22:11 <Bouke> Well maybe not enabled by default then if you think it’s bad for casual players. But at least available as a setting.
18:22:51 <andythenorth> well we need to do something about PBS 🙂
18:23:02 <andythenorth> we have at least one candidate
18:23:19 <Bouke> In general, or with realistic braking?
18:23:41 <nielsm> routing restrictions for path signals should be in the main branch though
18:23:42 <andythenorth> well realistic braking is one fix for PBS
18:23:55 <JGR> Bouke: The implementation is more involved than a drop down with two items would make it appear
18:24:05 <nielsm> to deprecate the last use cases for block signals
18:24:45 <Bouke> jfs-: is this explained somewhere in more detail? What kind of restrictions?
18:24:58 <nielsm> "routing restrictions" means programmable signals
18:25:20 <nielsm> add programs to signals that restrict what routes they are allowed to set
18:25:35 <Bouke> As BBS is currently used to build priority merges for example?
18:25:43 <FLHerne> have an action2 chain decide whether it's green or red
18:26:03 <FLHerne> with the surrounding signal states/distances as variables
18:26:12 <nielsm> no that won't be useful at all
18:26:13 <FLHerne> I'm sure there's no way that could become a nightmare
18:26:58 <FLHerne> nielsm: I think you could use it to implement a couple of signal types for priority merging, which is the main thing I use block signals for
18:27:51 <andythenorth> or we could bring back block, but not pre-signals
18:27:57 <nielsm> priority signals with routing restrictions is simply something like setting one signal to "attempt to reserve the next 3 blocks ahead"
18:28:06 <FLHerne> also, the GRF should be allowed to create and read user-configurable parameters per signal
18:28:29 <FLHerne> I mean, the behaviour would probably be similar to progsigs
18:28:49 <nielsm> now you're just making programmable signals but worse (much more confusing and harder to use)
18:29:12 <nielsm> what would make more sense would be a way to copy programs in/out of the game
18:29:14 <FLHerne> but it would possibly allow more advanced logic conveniently
18:29:25 <FLHerne> and/or without people insisting on adding it
18:30:14 <nielsm> you'd still need the exact same engine for evaluating and acting on the logic inside the game code (rail pathfinder), the difference is just if the user has direct or only indirect control of it
18:31:59 <FLHerne> do you *want* to give the entire OTTD user base direct control over a programmable logic thing?
18:32:04 <FLHerne> the bug reports will be endless
18:32:33 <nielsm> they already have programmable vehicle orders
18:32:40 <nielsm> I don't see any endless stream of bug reports with those
18:32:41 <FLHerne> and the bug reports are endless :p
18:32:52 <FLHerne> tbf, not so much with the conditional ones
18:32:55 <andythenorth> /me just wanted faster PBS 😛
18:32:58 <FLHerne> but only because no-one uses them
18:33:38 <nielsm> I think it could make sense to ship the game with some "templates" for common signal programs
18:33:42 <FLHerne> if it's behind NewGRF, then there'll be like one grf that adds a couple of useful signal types that most people use, and they can complain to whichever sucker makes it
18:34:28 <nielsm> you'd just see an endless stream of reports about NewGRF development tools being hard to use then
18:34:29 <FLHerne> and people who know what they're doing can add more but it'll be more obviously their own fault when it's broken
18:34:39 <JGR> Setting GRF authors up for failure doesn't seem that helpful
18:34:42 <FLHerne> there isn't really one now
18:34:53 <FLHerne> it's so obviously hard to use that people give up before complaining
18:35:10 <nielsm> I don't remember when I last saw someone reporting that conditional orders were broken either
18:35:16 * FLHerne might have a slight BoFH attitude today
18:36:03 <JGR> Conditional orders do work, but there aren't that many scenarios where they're really useful
18:38:31 <andythenorth> conditional orders aren't broken, they're just stupid
18:38:39 <andythenorth> the only one that really should be there is missing
18:38:48 <andythenorth> which is 'wait for cargo x%'
18:39:52 <andythenorth> 'other use cases may exist' 😛
18:41:11 <Bouke> andythenorth: Honest question: what is slow?
18:41:26 <andythenorth> they're always red until the train approaches
18:42:27 <TallTyler> Time to merge all the things? 😛
18:42:44 <TrueBrain> if that is how you wish to start things off 😛
18:43:10 <TrueBrain> anyway, gratz dude, well deserved 🙂
18:43:10 <JGR> andythenorth: There is a load percentage conditional order
18:43:17 <Bouke> andythenorth: Why are they red by default?
18:43:24 <TallTyler> TrueBrain: Thank you 😊
18:43:24 <andythenorth> what else could they be? 🙂
18:43:47 <andythenorth> but then trains would stop at yellows 😛
18:44:04 <TallTyler> Distant signals are fake
18:44:10 <andythenorth> it's the literal implementation of PBS 🙂
18:44:11 <Xaroth8> My condolences, TallTyler
18:44:57 <andythenorth> PBS is fixed in JGRPP with reserve-through (I don't know the official JGR terminology) 🙂
18:45:27 <JGR> PBS in vanilla is fine for normal players
18:46:16 <JGR> It doesn't need fixing in that sense
18:48:55 <TallTyler> Path-based signaling
18:50:21 <TallTyler> If the only problem with path signals is that they’re red until a train gets to them… Don’t path signals know if there’s only one possible route to the next signal? Could they automatically act like block signals in that case?
18:50:58 <TallTyler> Besides priority merges I mostly see comments from people here using block signals between junctions because having them green is “more realistic”
18:51:00 <nielsm> I don't understand why you want to see a signal be green when there is no train
18:51:00 <andythenorth> afaik, all they know is if a path is reserved through them
18:51:20 <andythenorth> that's why the JGR solution is so good 🙂
18:51:25 <andythenorth> it doesn't just make all signals green
18:51:56 <andythenorth> if we could make them yellow by default and clear them as a train reserves a path, that would work
18:51:59 <andythenorth> but we can't afaict
18:52:49 <andythenorth> there's a couple of JGR features that are very hard to unsee once you've seen them
18:52:59 <andythenorth> auto-separate is the other one
18:53:09 *** HerzogDeXtEr has quit IRC (Read error: Connection reset by peer)
18:53:30 <TallTyler> With real signals you only want to drop them to red when a train passes them thereby turning it red by detecting itself. Yellow by default would be odd because a train could come from the other direction and drop a signal in front of a driver, potentially causing a crash.
18:54:36 *** felipedobri has joined #openttd
18:55:47 <michi_cc[d]> TallTyler: Remember which side you are on now when talking about how PRs go stale 😇
18:57:52 <TrueBrain> maybe the worst part of this all, I never asked if TallTyler was okay with this 😛
18:58:26 <DorpsGek> - Update: Translations from eints (by translators)
19:00:55 *** Xaroth8 is now known as Xaroth
19:01:02 <TallTyler> I had my opportunity to object 🤷
19:01:07 *** felipedobri has quit IRC (Quit: Page closed)
19:01:19 <TallTyler> I guess I’ll have to learn to review PRs now 😛
19:01:27 <Xaroth> TrueBrain is like a bus... you don't go stand in front of a bus.
19:01:33 <TrueBrain> TallTyler: That comes with the job description, yes 😛
19:02:06 <TallTyler> Challenge accepted, I guess
19:02:26 <michi_cc[d]> You are at least safe in the sense that self-approval isn't allowed.
19:02:49 <TallTyler> I still feel quite like a novice but I’ve definitely learned a lot in the past couple years of contributing
19:03:07 <andythenorth> I approve of myself
19:03:16 <TrueBrain> TallTyler: pretty sure most of us felt like that when we started out 😄
19:03:18 <petern> We just create a second sock-puppet account to get around self-approvals...
19:03:35 <andythenorth> hmm did I do more grf commits than OpenTTD has yet?
19:03:36 <TallTyler> Which sock puppet are you?
19:03:37 <michi_cc[d]> Well, we might be doing a release soon. Usually pressing the proper buttons for that was happily given to newest "volunteer". 🙂
19:03:52 <petern> We're all sock puppet. There is only 1 actual developer...
19:04:01 <TrueBrain> with many many personalities
19:04:36 <TallTyler> That’s where your username comes from…you’re the one true brain?
19:04:49 <andythenorth> I have not caught up 😦
19:04:54 <andythenorth> frigging translator commits 😦
19:05:01 <andythenorth> I used to have those, but we nerfed them
19:05:17 <andythenorth> FIRS is only 6,881 commits
19:05:35 <andythenorth> Iron Horse is 6,081 commits
19:05:51 <petern> The scary thing is when VS Code tells me that I authored something 14+ years ago..
19:06:07 <TrueBrain> petern: I disabled that in VSCode ... I couldn't work like that 😛
19:07:48 <TallTyler> michi_cc[d]: I’m fine to be volunteered for that, but somebody would definitely have to walk me through it 🙂
19:08:40 <michi_cc[d]> Well, there are only two complicated things with a release, the changelog PR and thinking of a proper announcement message.
19:08:58 <TrueBrain> the rest is totally and utterly automated 😄
19:09:03 <michi_cc[d]> Everything else is simpler than you'd expect, thanks to some hard work by TrueBrain.
19:16:27 *** frosch has quit IRC (Quit: User went offline on Discord a while ago)
19:21:07 <TallTyler> Hmm, do we need a new title game?
19:21:33 <michi_cc[d]> Not for a beta anyway, as that isn't branched.
19:21:55 <TallTyler> And in the past I’ve also seen a request for translators to get their languages completed before release
19:21:58 <Rubidium> maybe consider thinking about #9101 before branching?
19:22:01 <TrueBrain> michi_cc[d]: owh, beta first, duh, nice 🙂
19:22:49 <michi_cc[d]> TallTyler: Indeed, hence why I personally would do a beta now.
19:23:21 <TrueBrain> Rubidium: yeah, otherwise we never do it 😄
19:23:24 <michi_cc[d]> There's also #9953, which seems to have stalled somewhere.
19:23:52 <michi_cc[d]> BTW, do the release CI actions need some fixing too, or was that done in one go with the other stuff?
19:24:31 <TrueBrain> glx[d]: fixed the nightly, so most of it should work
19:24:40 <TrueBrain> the one thing we are never sure of, are things like ubuntu deb files
19:24:55 <TrueBrain> those can break in mysterious ways, and are only checked upon release
19:25:35 <TallTyler> If we do a beta now when would the release be? Maybe there would be enough time for a new title save, if that’s important to people. Unfortunately turnout last year was pretty poor, I suspect because the cool new moving title scared people away 😦
19:25:57 <TrueBrain> make it a xmas release?
19:26:27 <TallTyler> Beta on Halloween, release on Christmas sounds fun 😄
19:26:32 <michi_cc[d]> Depends if we want to do any RCs before or not. But IMHO a month gap minimum, otherwise there won't be much feedback.
19:27:27 <Rubidium> RC with thanksgiving?
19:27:47 <TrueBrain> seems we found a new tradition
19:28:13 <TallTyler> RIP April Fool’s Day tradition
19:28:25 <TrueBrain> and I regret NOTHING
19:28:34 <TallTyler> New OpenTTD for Christmas will be an excellent gift
19:28:55 <TrueBrain> align with the xmas stream of .. darn it, forgot their name
19:29:04 <TrueBrain> make it land on their stream
19:29:15 <TrueBrain> give them the excluse 😛
19:29:20 <TrueBrain> andythenorth: will know the name I am looking for
19:33:14 <andythenorth> I am going to try and meet Lewis soon
19:33:19 <andythenorth> we talked about it last week
19:33:33 <andythenorth> can't do it this week, kids are off school
19:33:48 <Rubidium> I guess you can't combine the RC with the #10k party at the Southern Alps on Thanksgiving
20:03:54 <petern> Which somehow means waiting for another full compile.
20:05:10 <andythenorth> more thread units 😛
20:06:34 <petern> Is it not possible to paste text? Hmm
20:28:48 <petern> Do `network_chat_box_height` and `network_chat_box_width_pct` settings need to exist?
20:29:19 <petern> And indeed `network_chat_timeout`
20:30:54 *** virtualrandomnumber has joined #openttd
20:31:10 *** virtualrandomnumber has quit IRC ()
20:38:02 <TrueBrain> no; comes from a time where everything was a setting 😄
20:43:22 <TallTyler> So, if we do a title game competition, I can post the announcement today and the deadline could be December 1st? That gives people just over a month to create their entry with an easy deadline to remember, and a couple weeks to vote. Thoughts?
20:44:12 <petern> Is the deadline the end of the 1st or the beginning of it...
20:47:02 <TallTyler> I don’t think anything is so pressing that it matters, but we could say 23:59:00 UTC to be specific 🙂
20:53:17 <TrueBrain> welcome to #uk-special 🙂
20:59:44 * Flygon confused Australia noises
21:00:07 <Flygon> How the hell did the UK top Australia in terms of Prime Ministerial cluster... ducks :D
21:00:57 <FLHerne> he ruled himself out (almost certainly because he wasn't going to win)
21:01:23 <FLHerne> I was almost looking forward to the carnage if he won
21:02:47 <Flygon> I thought you were like
21:06:45 <TallTyler> The title game competition post is ready to go. Should I mention the release date of December 25th? What about the beta on Halloween? Or are those separate announcements?
21:21:11 <TallTyler> Okay, doesn’t make sense to mention the beta, but can I promise a Christmas release?
21:21:50 <petern> I mean, who's going to be making released on the 25th? 😄
21:22:21 <TallTyler> How about “before Christmas?”
21:23:29 <TallTyler> If there’s a title game competition people will be wondering when the release will be
21:23:45 <petern> You also don't really want a Christmas-themes competition game.
21:23:57 <petern> Yeah but then again, let them wonder?
21:24:15 <TallTyler> “OpenTTD 13 coming soon” 😛
21:32:53 *** nielsm has quit IRC (Ping timeout: 480 seconds)
21:34:49 <TrueBrain> Nicely done, looks good 😄
21:37:11 <Arastais> I have a window with a `WWT_RESIZEBOX`, but can't resize horizontally, only vertically. I've set every widget to `SetResize(1, y)` where y is either 1 or 0. Am i missing osmething?
21:45:12 <petern> You probably need a `SetFill()` somewhere too.
21:51:03 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
22:05:21 <Arastais> petern: i have that as well for everything except `WWT_FRAME`s, still no luck
22:14:27 *** sla_ro|master has quit IRC ()
22:37:02 <Arastais> Arastais: did it for the frames too, no luck
22:37:40 <Arastais> in fact, I noticied that it can't even stretch to its default height based on its `WindowDesc`, it gets drawn in the center as if it were bigger but then gets adjusted to its fixed vertical size
22:55:45 *** gelignite has quit IRC (Quit: Stay safe!)
23:07:41 *** Montana has joined #openttd
23:12:23 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
23:15:09 *** WormnestAndroid has joined #openttd
23:21:06 <Arastais> sorry, theres gonna be a lot of updates bc there was a conflitch and i have to rebase now. I'll also need to comment the changes
23:23:13 <glx[d]> VS is configurable but it's not always simple
23:24:10 <glx[d]> and default configuration for if/else doesn't follow our style
23:35:52 <Arastais> glx[d]: I tired to stick to the coding style as always, but if you find something I missed let me know.
23:36:53 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
23:38:57 *** WormnestAndroid has joined #openttd
23:40:07 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
23:40:12 *** WormnestAndroid has joined #openttd
23:41:43 *** Montana has quit IRC (Quit: Leaving)
23:45:04 <Arastais> glx[d]: oh sorry, I think I misunderstood. I meant I need to put a comment in the PR saying what changes I've made since the last force push.
23:47:09 <glx[d]> I know VS can be annoying with coding style
23:47:20 <LordAro> Arastais: we can see what changes are made, a comment isn't (usually) necessary
23:48:58 <Arastais> glx[d]: yea, all ur suggestions are from VS doing it without me noticing lol. Basically if you paste something or edit lines in a certain way (such as putting a semicolon), it auto-indents everything it can to the formatting in the VS settings. I'll correct them now
23:49:17 <Arastais> LordAro: lol sorry, i meant in terms of the UI. I made a comment saying how the UI looks in-game now
23:49:22 <LordAro> i'm talking about the act of purely force-pushing, making additional changes to the PR could well need a comment
23:50:00 <glx[d]> yeah usually after a paste you can do a ctrl-z to undo the auto formatting
23:51:10 <glx[d]> and it's possible to configure VS to format as you want
23:52:18 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
23:53:19 <glx[d]> because it will auto format every time a block is closed (very nice feature, except when it uses to wrong format)
23:53:57 <Arastais> glx[d]: yea I do but i miss some formatting rules and forget to ctrl-z sometimes
23:57:07 <Arastais> foudn a few more mistakes, correcting them now
23:58:13 *** wallabra_ has joined #openttd
23:59:11 <glx[d]> we can be annoying with coding style but it's for the benefit of everybody 🙂
continue to next day ⏵
