IRC logs for #openttd on OFTC at 2021-01-03
⏴ go to previous day
00:00:33 <DorpsGek> [OpenTTD/OpenTTD] michicc merged pull request #8484: Fix: [OSX] Register fonts loaded directly from disk for text layout https://git.io/JL5xV
00:00:41 <DorpsGek> [OpenTTD/OpenTTD] mattkimber commented on pull request #8485: Codechange: improve performance for complex vehicle chains by resolving sprites less often https://git.io/JLdeX
00:00:50 <DorpsGek> [OpenTTD/OpenTTD] frosch123 commented on pull request #8452: Doc: Add labels to landscape grid description. https://git.io/JLdeM
00:05:10 <DorpsGek> [OpenTTD/OpenTTD] michicc commented on pull request #7000: Some NewGRF variables concerning railtypes https://git.io/JLdeN
00:08:29 <DorpsGek> [OpenTTD/OpenTTD] Eddi-z commented on pull request #7000: Some NewGRF variables concerning railtypes https://git.io/JLdvJ
00:08:43 <DorpsGek> [OpenTTD/OpenTTD] frosch123 commented on pull request #8452: Doc: Add labels to landscape grid description. https://git.io/JLdvT
00:13:16 <frosch123> michi_cc: if you want to know whether a vehicle is powered on the current tile, you can use var FE. you do not need a new variable for that at all
00:13:43 <frosch123> but well, again, main problem with PRs. they do not state their intention
00:13:53 <frosch123> you have to guess what those additions are meant for
00:14:21 <frosch123> now i have to guess why "returning false" for undefined types is intentional
00:14:29 <frosch123> when the PR does not describe any usecase
00:15:38 *** gelignite has quit IRC (Quit: Stay safe!)
00:18:11 <DorpsGek> [OpenTTD/OpenTTD] michicc commented on pull request #7000: Some NewGRF variables concerning railtypes https://git.io/JLdvW
00:20:48 *** frosch123 has quit IRC (Quit: be yourself, except: if you have the opportunity to be a unicorn, then be a unicorn)
00:21:11 <Eddi|zuHause> i'm fairly certain i described a usecase for detecting railtypes
00:24:10 <Eddi|zuHause> maybe i'm too deep in formal logic here, but comparing <undefined> to <undefined> should never return true
00:24:55 <DorpsGek> [OpenTTD/OpenTTD] Eddi-z commented on pull request #7000: Some NewGRF variables concerning railtypes https://git.io/JLdv5
00:34:59 *** nielsm has quit IRC (Ping timeout: 480 seconds)
00:41:44 <DorpsGek> [OpenTTD/OpenTTD] Eddi-z commented on pull request #7000: Some NewGRF variables concerning railtypes https://git.io/JLdfN
01:17:47 *** DasPoseidon has joined #openttd
01:28:21 *** DasPoseidon has quit IRC (Quit: DasPoseidon)
01:28:26 *** DasPoseidon has joined #openttd
02:42:01 *** Wormnest has quit IRC (Quit: Leaving)
02:44:33 *** DasPoseidon1 has joined #openttd
02:49:22 *** DasPoseidon has quit IRC (Ping timeout: 480 seconds)
02:49:22 *** DasPoseidon1 is now known as DasPoseidon
02:54:17 *** Progman has quit IRC (Remote host closed the connection)
03:24:13 *** debdog has quit IRC (Ping timeout: 480 seconds)
03:36:05 *** Gustavo6046 has quit IRC (Read error: Connection reset by peer)
03:36:26 *** Gustavo6046 has joined #openttd
03:47:29 *** Gustavo6046 has quit IRC (Ping timeout: 480 seconds)
03:48:38 *** Gustavo6046 has joined #openttd
03:49:25 *** DasPoseidon has quit IRC (Quit: DasPoseidon)
04:28:09 *** luaduck has joined #openttd
05:30:39 *** Gustavo6046 has quit IRC (Ping timeout: 480 seconds)
07:31:50 *** andythenorth has joined #openttd
07:38:06 *** WormnestAndroid has quit IRC (Remote host closed the connection)
07:38:19 *** WormnestAndroid has joined #openttd
07:40:36 *** WormnestAndroid has quit IRC (Read error: Connection reset by peer)
07:40:51 *** WormnestAndroid has joined #openttd
07:41:50 <andythenorth> I worry about Jeff not making enough money
08:13:29 *** sla_ro|master has joined #openttd
08:59:59 <FLHerne> "This is not the negation of condition 0x13, which makes it weird in corner cases, and hard to implement in NML." -- that's only consistent with the 'grfid is enabled' checks...
09:03:48 *** Progman has joined #openttd
09:13:21 <TrueBrain> LordAro: https://github.com/OpenTTD/OpenTTD/pull/8483 <- this makes town during map generation looks worse in many cases .. do we really want this? I can think of other solutions, if we want to fix this, like: do not level to sea-level, or something?
09:14:44 <TrueBrain> you are the one that approved it, so I am checking with you first :D
09:15:13 <LordAro> i didn't see any evidence that it made towns look worse
09:15:26 <TrueBrain> it is not flattening at all during world gen now
09:15:56 <LordAro> and samu's thing about the other bit of land levelling not done during world gen made sense
09:16:26 <LordAro> that is a bad sentence
09:17:56 <TrueBrain> lets find out why that terraform is special-cased :)
09:18:57 <TrueBrain> oef ... the file got renamed ... now that is tricky :)
09:19:37 <TrueBrain> awh, I end up with r1
09:21:11 <andythenorth> does town sign tile need to be flat?
09:21:38 <LordAro> TrueBrain: always fun :)
09:22:14 <TrueBrain> okay, fun "fact" .. that case he mentions in the PR, is only used in the original road layout
09:22:17 <TrueBrain> the other 3, do not use it :)
09:22:36 <TrueBrain> grid-based layouts do not touch it :P
09:23:04 <TrueBrain> never assume anything when it comes to town-code or aircraft-code :P
09:23:39 <FLHerne> How is the /* Note: Do not replace "^ SLOPE_ELEVATED" with ComplementSlope(). The slope might be steep. */ code relevant to this PR?
09:24:27 *** jottyfan has joined #openttd
09:24:34 <FLHerne> Oh, it's the `if (!generating_world [...]) {}` bit
09:25:29 <FLHerne> But...why is the reported bug only a problem for worldgen?
09:25:52 <FLHerne> Towns on sea-level could try and flatten land into the ocean at any time
09:25:57 <DorpsGek> [OpenTTD/OpenTTD] btzy commented on issue #8339: [Question] Rate limit password entry for multiplayer https://git.io/JkTYo
09:27:41 <TrueBrain> this has been annoying to me the last few weeks: I see a setting in the code, I go ingame to the setting, I type in the name: 0 matches
09:27:51 <TrueBrain> so now I have to guess how it is described ...
09:28:30 <TrueBrain> owh, it is not in the setting dialog
09:29:01 <Eddi|zuHause> TrueBrain: try that same thing in non-english :p
09:29:18 <TrueBrain> no tnx; if you play any game localized, you are only hurting yourself :D :D :D
09:31:57 <DorpsGek> [OpenTTD/OpenTTD] FLHerne commented on pull request #8483: Fix: Don't terraform during town generation https://git.io/JLdlc
09:32:38 <FLHerne> TrueBrain: Unless it's Zero Wing
09:33:46 <FLHerne> (all your base are belong to us!)
09:34:33 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #8483: Fix: Don't terraform during town generation https://git.io/JLdlC
09:34:44 <TrueBrain> having build_on_slopes off with that PR is hilarious :)
09:34:47 <TrueBrain> very small towns :P
09:37:51 <Eddi|zuHause> i specifically tested that the last time i messed with town terraforming
09:38:29 <TrueBrain> which is good, means you have a bit of QA in your heart :)
09:39:09 <Eddi|zuHause> (svn r14404) -Change [FS#2176]: don't make the town flatten land unconditionally when build on slopes is turned on. Based on a patch by Eddi.
09:41:01 <TrueBrain> improved hilly towns :P
09:42:27 <andythenorth> I introduced the idea of 'yield' at work, for tickets
09:42:45 <andythenorth> it's nothing clever, just made explicit what a few of us were doing in our head to prioritise
09:43:15 <andythenorth> some things are low yield or negative yield, they suck in thinking time for not enough benefit
09:43:47 <andythenorth> some things are very risky or hard, but very high yield, like building a new product
09:44:11 <Eddi|zuHause> i really like how old FS#numbers still correspond to github numbers :)
09:44:34 <TrueBrain> Eddi|zuHause: it took a lot of effort, but I just had to do it :P
09:44:41 <TrueBrain> sometimes it helps being a bit crazy :)
09:44:43 <andythenorth> _most_ of these "I found an esoteric bug" PRs are very low yield
09:44:47 <Eddi|zuHause> feel my appreciation :)
09:44:59 <andythenorth> they don't affect many people, and sometimes they're stacking edge case on edge case, to reach corner case
09:45:23 <andythenorth> similar is the endless chasing of "this cost isn't quite right"
09:46:08 <TrueBrain> personally, I don't really mind if people fix edge-cases. I always rate reviewing on: how much delta did this take. In other words: if I would have done this myself, would it be more or less time than the review I am about to do :P Which is a very harsh metric, I can tell you :D
09:48:13 <Eddi|zuHause> that's a very difficult thing to decide in some cases. like you could spend hours not finding the right place to change, but once someone tells you, you think "well, i could have found this"
09:48:41 <TrueBrain> yup; you need a bunch of experience to balance that
09:48:54 <TrueBrain> so it does require that you know the codebase, for sure :)
09:49:17 <TrueBrain> but important to note: I am not rating how long the other user took to get there; I mean, how-ever long, it is a good learning experience
09:49:27 <TrueBrain> and the metric only closes after review, not before :P
09:50:12 <TrueBrain> it really helps to see improvements in people btw .. new people tend to consume a lot of time, but you notice they very fast take less and less (compared to how long it would have taken you to do the same)
09:50:14 <TrueBrain> which is really nice :D
09:50:19 <TrueBrain> anyway, it is a shitty metric, just a fun one :P
09:50:47 <Eddi|zuHause> spoiler: most metrics are shitty. the trick is to find the least shitty metric :p
09:54:15 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on issue #8339: [Question] Rate limit password entry for multiplayer https://git.io/JkTYo
09:54:27 * andythenorth back to fixing the world tomorrow, last day of open source train lulz
09:56:30 <TrueBrain> for following up ;)
09:57:36 <andythenorth> I'd leave it open for a bit, Gwyd might reply
09:59:21 <andythenorth> once the unpowered train is stuck, the only option is to crash another train into it
10:07:02 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #8485: Codechange: improve performance for complex vehicle chains by resolving sprites less often https://git.io/JLdlM
10:10:34 <TrueBrain> I wonder if the 10x speed improvement hold on release builds :) Pretty sure it plummets a bit :) Still impressive, don't get me wrong :)
10:11:03 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain merged pull request #8481: Add: always set PERSONAL_DIR "/content_download" in search path https://git.io/JL556
10:17:18 <TrueBrain> now to review that silly airport fix PR ... owh boy
10:23:06 <Eddi|zuHause> i noticed an inconsistency... it's "RailtypeInfo" but "RoadTypeInfo"
10:24:46 *** snail_UES_ has quit IRC (Quit: snail_UES_)
10:25:15 <TrueBrain> oef .. that is unacceptable
10:31:16 <TrueBrain> okay, airport fixes looks fine, but the comments are ramblings of a mad man :) Guess my label was accurate :P
10:36:05 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #8458: Fixes for issues with rotated (non-rectangular) airports https://git.io/JLd8C
10:37:47 <andythenorth> there is also the nml spec P
10:38:59 <Eddi|zuHause> bah, i'm suddenly unsure... in one place i used railtype_map and in another i used railtype_list
10:39:05 *** jottyfan has quit IRC (Quit: jottyfan)
10:39:06 <Eddi|zuHause> they have subtly different meanings
10:39:18 <Eddi|zuHause> and i don't know if i picked the right one
10:47:13 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on issue #6618: Incorrect size bounds checks in vehicle viewport hash scan in ViewportAddVehicles https://git.io/JLd8E
10:51:34 *** gelignite has joined #openttd
10:52:14 <DorpsGek> [OpenTTD/OpenTTD] Eddi-z commented on pull request #7000: Some NewGRF variables concerning railtypes https://git.io/JLd8D
10:55:59 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on issue #7603: Town::cargo_accepted, Town::cargo_produced saveload and consistency issues https://git.io/fjB3A
10:59:37 <DorpsGek> [OpenTTD/OpenTTD] Eddi-z commented on pull request #7000: Some NewGRF variables concerning railtypes https://git.io/JLd8d
11:02:02 <michi_cc> andythenorth: Want some fun of your own validating #8479?
11:02:33 <DorpsGek> [OpenTTD/OpenTTD] Eddi-z commented on pull request #7000: Some NewGRF variables concerning railtypes https://git.io/JLd4f
11:02:44 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on issue #7651: Too many statues/fountains are built in town centers. https://git.io/fjXfH
11:02:47 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain closed issue #7651: Too many statues/fountains are built in town centers. https://git.io/fjXfH
11:03:05 <TrueBrain> owh, yes andythenorth , do it!
11:04:09 * andythenorth having adventures in .copy()
11:05:54 <TrueBrain> michi_cc: the PR is nearly impossible to review, but I guess a "I trust you, this is fine" will work just fine :D
11:06:58 <michi_cc> I think the only possible way would be to review commit by commit, as e.g. the first big on is only supposed to be pure copypaste.
11:07:21 <TrueBrain> yeah, I meant the first one is hard to review :P
11:07:27 <TrueBrain> I trust you did a copy/paste successful :D
11:07:47 <TrueBrain> I wish GitHub interface would allow this a bit better .. not sure how honestly
11:08:34 <michi_cc> Well, I did that copy/paste a long time ago, but I think I have caught and reapplied to whopping two changes that came later (incidentally both by me :)
11:11:12 <TrueBrain> btw, your commit messages are too long in many cases :P It goes against the idea of git!!!! :P (And I really do not care)
11:11:49 <DorpsGek> [OpenTTD/OpenTTD] JGRennison commented on issue #7603: Town::cargo_accepted, Town::cargo_produced saveload and consistency issues https://git.io/fjB3A
11:11:52 <DorpsGek> [OpenTTD/OpenTTD] JGRennison closed issue #7603: Town::cargo_accepted, Town::cargo_produced saveload and consistency issues https://git.io/fjB3A
11:12:27 <TrueBrain> Tnx for follow-up JGR :)
11:12:53 <TrueBrain> one more for your changelog LordAro :)
11:15:03 <michi_cc> TrueBrain: Yes, but of you go by that, a prefix of like "Codechange #1234: [OSX]" is deadly because it steals soooo many characters :p
11:15:15 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on issue #7675: Wires incorrectly drawn on top of pylons that should be in front https://git.io/fj9eq
11:15:29 <TrueBrain> michi_cc: that would not have changed anything at all in your case :D
11:15:47 <TrueBrain> I still have to find a good source that explains how commit messages should look, the first line and the lines after .. haven't found any yet
11:15:52 <TrueBrain> till that time, I am fine with what-ever :P
11:16:57 <michi_cc> The canonical way is a short < 70 chars summary on the first line (because terminals are stuck at 80 chars and screens still 13"), which a long explanation following after an empty line.
11:17:34 <michi_cc> Of course that makes only sense for us if the person/script generating the changelog would also look at the long explanation and not just the first line.
11:19:06 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on issue #7828: First time server registration ends up in multiple online servers instead of one https://git.io/JeVNi
11:19:09 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain closed issue #7828: First time server registration ends up in multiple online servers instead of one https://git.io/JeVNi
11:19:12 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on issue #7829: Unregister of MasterServer happens per IP, instead of per session_key https://git.io/JeVAz
11:19:15 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain closed issue #7829: Unregister of MasterServer happens per IP, instead of per session_key https://git.io/JeVAz
11:19:19 <TrueBrain> pfft, the OP of those two tickets can just stick it
11:19:54 <TrueBrain> michi_cc: I believe it is common to keep a 120 limit on the first line, but line length is one thing .. what you say in it is much more important
11:20:05 <TrueBrain> and this is where I am missing a good source to explain how to balance things
11:20:35 <TrueBrain> to take an example: "Codechange: [OSX] Re-arrange the OSX video driver code by combining all drawing code and moving the window/event handling to a different file."
11:20:42 <TrueBrain> there is a lot of redundant words there
11:21:02 <TrueBrain> it makes reading it lengthy and long .. and for the first 3 times, I just stopped reading the last 8 words or so :P
11:21:13 <michi_cc> Well, GitHub at least cuts it of at exactly 70 chars.
11:21:19 <TrueBrain> they still do? That sucks ..
11:21:36 <michi_cc> For display on the website I mean, not in the actual git of course.
11:21:56 <TrueBrain> Codechange: [OSX] split drawing code from window/event handling
11:22:03 <TrueBrain> for example might also cover the full length of the commit
11:22:50 <TrueBrain> but one thing I learnt over the years .. it is very hard to write good commit messages (and I often fail myself) .. especially as I cannot find a decent write-up how to balance what to put in there :)
11:23:11 <TrueBrain> at work I had an architect that would smack me in the face if I failed .. learnt a lot from that :D But that is from experience .. which is annoying :D
11:24:08 <michi_cc> Well, yes, that one might be a tad long :) Otherwise, I strive for the first line to be understandable as a single-line changelog entry.
11:24:19 <TrueBrain> I think that is a good goal, yes
11:24:41 <TrueBrain> we might need to make that a rule .. if it isn't already :)
11:25:58 <Wolf01> At work I make effort to at least write down the thing I changed, my coworker often doesn't do even that XD
11:26:23 <TrueBrain> I am often annoyed people tend to write what they changed .. like .. yes, I can see that in the diff
11:26:26 <TrueBrain> but wtf is it suppose to do?!
11:26:35 <TrueBrain> Fix: made blue into green
11:26:44 <TrueBrain> .. like .. yes .... thank you mister obvious ... but why?!
11:29:47 <DorpsGek> [OpenTTD/OpenTTD] JGRennison commented on issue #6618: Incorrect size bounds checks in vehicle viewport hash scan in ViewportAddVehicles https://git.io/JLd8E
11:30:35 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on issue #7872: Default openttd.cfg world gen settings don't allow Forests in Sub-Arctic https://git.io/JeFfG
11:30:38 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain closed issue #7872: Default openttd.cfg world gen settings don't allow Forests in Sub-Arctic https://git.io/JeFfG
11:34:42 <TrueBrain> slowly closing issues ... it is a slow process, but a lot of easy bugs in there :)
11:36:02 <andythenorth> michi_cc what are valid tests for 8479?
11:36:08 <andythenorth> it builds and starts :P
11:36:27 <michi_cc> Maybe also toggle fullscreen on and off.
11:36:57 <michi_cc> Otherwise, it is not supposed to change anything.
11:37:10 <andythenorth> did someone fix the smearing (random noise) on resizing window on mac?
11:37:16 <andythenorth> not happening right now
11:37:28 <andythenorth> 1.10.0 does it though
11:37:54 <michi_cc> Might be my change in ditching SetPortAlphaOpaque, which could even accidentally fix that crash report a few days ago.
11:38:24 <michi_cc> I made that change a few months ago because of that function being useless, not because of that report.
11:38:46 <andythenorth> oof I should build this on child #1's M1
11:38:49 <andythenorth> much faster probably
11:39:36 <michi_cc> Like, OTTD allocated a random storage buffer and then did trickery with SetPortAlphaOpaque on each show/hide instead of simply clearing the buffer on the get go.
11:40:27 <andythenorth> yeah 8479 prevents the smearing
11:40:32 <michi_cc> And it is even straight up copypaste from SDL1 used in quite a different context.
11:40:33 <andythenorth> plain black window during resize operations
11:40:55 <andythenorth> no idea if there's any performance changes, I don't have any valid tests for that
11:42:34 <michi_cc> It should be a bit slower as we clear the buffer now, but how often do you resize the window anyway?
11:43:24 <michi_cc> Oh, and one test missing of course is to minimize OTTD to the dock and then show it again.
11:43:41 <andythenorth> subjectively slow linker is subjectively slow :P
11:45:31 <andythenorth> I had a bug with the menu bar not appearing at top of full screen when mouse is pushed to top
11:46:08 <andythenorth> hmm maybe it's just a bit slow
11:46:24 <michi_cc> Unless I made a copy/paste mistake somewhere, I didn't change anything there.
11:46:25 <andythenorth> and the mouse sometimes stops at the boundary where the menu bar *will* show
11:47:26 <andythenorth> 1.10.2 does it too
11:47:33 <andythenorth> it's very inconsistent, so hard to judge
11:48:16 <andythenorth> ha the faster I move the mouse, the more likely to 'stick' it is
11:48:30 <andythenorth> I guess it's measuring velocity and trying to anticipate when to show menu bar
11:49:01 <andythenorth> we can ignore that one for this
11:49:11 <andythenorth> michi_cc do I need to test changing blitter or anything?
11:50:14 <michi_cc> No. I'm assuming I didn't fuck up the copy/paste totally if it runs :p and otherwise it is a complete no-change.
11:51:16 <DorpsGek> [OpenTTD/OpenTTD] JGRennison commented on pull request #8485: Codechange: improve performance for complex vehicle chains by resolving sprites less often https://git.io/JLdRt
11:51:17 <Eddi|zuHause> it could be a primary candidate for "if it compiles, ship it" :p
11:52:47 <michi_cc> And it is a total unrelated coincidence that we got a singular bug report that might be in SetPortAlphaOpaque earlier.
11:53:47 <andythenorth> now I have to find whatever branch I built for my big newgrf test savegame :)
11:53:52 * andythenorth only has one checkout P
11:54:04 <michi_cc> Eddi|zuHause: That's been the OSX state for years now :p
11:54:05 <andythenorth> and apparently my : key is failing
11:54:29 <andythenorth> hmm there is food..debris around my : key....related? :P
12:14:16 *** DasPoseidon has joined #openttd
12:24:46 <andythenorth> lot of waffle in that ticket
12:24:58 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain opened pull request #8486: Fix: [OSX] don't show top OS menu bar when in full screen https://git.io/JLd0R
12:25:07 <TrueBrain> you can also try ^^, and see if it works as advertised :D
12:28:09 <TrueBrain> sorry for all the MacOS requests at once :P
12:50:37 <Eddi|zuHause> maybe should tell him about click-and-hold buttons?
12:51:35 <Eddi|zuHause> it's only been 25 since the game came out, some features go unnoticed for that long :p
12:52:50 <andythenorth> I enjoy that it's not even about OpenTTD
12:52:56 <andythenorth> but a rant about times past
12:53:14 <andythenorth> I know 'but nobody asked for' is a core feature of reddit, but really, nobody asked for :)
12:53:51 <LordAro> andythenorth: ah, but OTTD is pbly the same
12:53:53 * andythenorth wishes Apple had faster linking
12:55:12 <andythenorth> the 'mouse stops when moving upwards' bug is now super annoying with that
12:55:17 <andythenorth> I guess I have to video it
12:55:28 <TrueBrain> andythenorth: did you try my PR?
12:56:00 <andythenorth> oh wait, what have I checked out here
12:56:38 *** DasPoseidon has quit IRC (Ping timeout: 480 seconds)
12:57:00 <andythenorth> yeah I'm testing wrong branch
12:57:17 <andythenorth> clicked wrong link :P
12:58:49 *** Progman has quit IRC (Remote host closed the connection)
13:01:27 <andythenorth> this is complicated
13:01:55 <TrueBrain> what is complicated?
13:02:35 <andythenorth> it needs a video TBH
13:02:46 <andythenorth> and the video will be...large file but eh
13:02:49 <TrueBrain> does the menu bar appear now in fullscreen? :)
13:03:20 <andythenorth> trying to get a reliable repro
13:04:25 <andythenorth> - mouse behaviour is horrible on fullscreen, it's "broken", but also in 1.10.x releases so not new
13:04:43 <andythenorth> - 'fullscreen' option in game options isn't checked in fullscreen mode
13:04:58 <andythenorth> - and toggling animation on / off has 'wtf' effects I haven't figured out yet
13:05:06 <TrueBrain> I am not asking you for a full rundown of all MacOS issues; for that we have an issue tracker :D
13:05:18 <andythenorth> yeah I have to keep comparing against 1.10.2
13:05:53 <TrueBrain> the menu bar still appears?
13:06:07 <andythenorth> I don't know yet )
13:06:12 <andythenorth> it's complicated
13:06:21 <TrueBrain> right ... *ignores andythenorth for a bit* :P
13:06:45 <andythenorth> patience grasshopper :)
13:06:51 <andythenorth> we need a reliable diagnosis
13:07:19 <DorpsGek> [OpenTTD/OpenTTD] mattkimber commented on pull request #8485: Codechange: improve performance for complex vehicle chains by resolving sprites less often https://git.io/JLdzy
13:08:10 <andythenorth> it's really fucking hard to diagnose :)
13:08:23 <andythenorth> because the mouse is broken at the top of the screen
13:08:31 <andythenorth> so it's really hard to know if the menu bar is appearing or not
13:08:42 <TrueBrain> I don't understand :P It either appears or not :D :)
13:08:49 <TrueBrain> you are speaking in riddles :P
13:08:55 <andythenorth> no, sometimes it appears, sometimes it doesn't
13:08:57 <TrueBrain> mainly I wonder if the patch does anything at all btw :)
13:10:55 <DorpsGek> [OpenTTD/OpenTTD] LordAro commented on pull request #8485: Codechange: improve performance for complex vehicle chains by resolving sprites less often https://git.io/JLdzp
13:14:30 <andythenorth> how do I build the bundle locally?
13:14:44 <andythenorth> just running ./openttd ain't gonna work as far as I can see
13:15:10 <TrueBrain> and that is mostly what I am afraid of, it seems this fix only works when starting the bundle
13:18:16 <michi_cc> andythenorth: Does the behaviour change if you switch focus to another app and back to OTTD again?
13:18:35 <michi_cc> I found that the menu bar sometimes doesn't react if started from command line if I don't do that.
13:23:00 <TrueBrain> I do not know if there is a non-bundle fix to make the bundle not appear, honestly
13:23:09 <TrueBrain> one of the two times :P
13:26:08 <DorpsGek> [OpenTTD/OpenTTD] mattkimber commented on pull request #8485: Codechange: improve performance for complex vehicle chains by resolving sprites less often https://git.io/JLdgX
13:26:31 <michi_cc> Hmm, there is presentationOptions in NSApp, maybe that can be set programmatically without the Info.plist
13:26:47 <TrueBrain> I got lost in MacOS docs :P
13:30:28 <DorpsGek> [OpenTTD/OpenTTD] mattkimber commented on pull request #8485: Codechange: improve performance for complex vehicle chains by resolving sprites less often https://git.io/JLdg5
13:31:48 <TrueBrain> andythenorth: you can just btw run "cpack" locally, which should give you a "dmg" file :)
13:31:53 <TrueBrain> which you can double-click ;)
13:32:51 <TrueBrain> it does fix the issue?
13:32:59 <LordAro> Timberwolf: oh wow, those trains are impressive
13:33:23 <DorpsGek> [OpenTTD/OpenTTD] mattkimber updated pull request #8485: Codechange: improve performance for complex vehicle chains by resolving sprites less often https://git.io/JL5pN
13:33:34 <DorpsGek> [OpenTTD/OpenTTD] andythenorth commented on pull request #8486: Fix: [OSX] don't show top OS menu bar when in full screen https://git.io/JLdgA
13:34:31 <TrueBrain> so .. MacOS is just being a twat, gotcha :)
13:34:55 <Timberwolf> LordAro: Thanks :)
13:35:00 <TrueBrain> I assumed it would only do this when it is in fullscreen
13:35:03 <andythenorth> might even just be local to my system
13:35:14 <LordAro> Timberwolf: super laggy though :p
13:35:20 <andythenorth> but 1.10.2 doesn't have the 'menu bar is broken' issue
13:35:59 <andythenorth> there's no frigging way to demonstrate this cursor bug, because you won't be able to see where my mouse actually is on a video :(
13:36:40 <andythenorth> I'd need to video my trackpad at same time
13:37:02 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #8486: Fix: [OSX] don't show top OS menu bar when in full screen https://git.io/JLd2v
13:37:05 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain closed pull request #8486: Fix: [OSX] don't show top OS menu bar when in full screen https://git.io/JLd0R
13:37:48 <TrueBrain> tnx for testing; can't believe how annoying MacOS is :P
13:38:10 <TrueBrain> hopefully michi_cc can find a software way of doing this, so we can do it on fullscreen yes/no .. if we fix the detection in our code whether you are fullscreen, that is :D
13:39:38 <michi_cc> VMware with some "patches" runs OSX quite okay on Intel hardware (AMD needs a bit more fiddling).
13:42:37 * andythenorth wonders about the mac-by-the-hour rentals
13:42:43 <andythenorth> they have full gui
13:43:02 <TrueBrain> might be nice for development, yeah .. attach the OpenTTD CC to it, and michi_cc can go nuts :P
13:44:10 <michi_cc> andythenorth: I tested one a few months ago with a free one hour trial or so. It was a lot shittier than the VMware variant.
13:45:40 <michi_cc> Frankly, if you don't want to run the absolutely latest, VMware runs quite good. You don't have accelerated graphics, which makes some things a bit sluggish (and some apps impossible to run), but OTTD doesn't care about that.
13:45:53 <michi_cc> Even sound and music work just fine.
13:46:58 <michi_cc> Yeah, I am on AMD. You just need to fiddle with some internal settings to get it running while on Intel you literally only need the unlocker and can go.
13:48:50 <DorpsGek> [OpenTTD/OpenTTD] mattkimber commented on pull request #8485: Codechange: improve performance for complex vehicle chains by resolving sprites less often https://git.io/JLd2z
13:53:33 <andythenorth> I did send some people macs some years ago, maybe fonso and peter
13:53:50 <andythenorth> we had piles of them to get rid of at work
13:53:54 <andythenorth> they weren't....good
13:56:35 *** frosch123 has joined #openttd
14:01:57 <TrueBrain> donate M1s to everyone you say? :P
14:02:01 <TrueBrain> (NO! I AM KIDDING!)
14:03:05 <DorpsGek> [OpenTTD/OpenTTD] mattkimber updated pull request #8485: Codechange: improve performance for complex vehicle chains by resolving sprites less often https://git.io/JL5pN
14:03:53 * andythenorth considers it briefly
14:05:15 <andythenorth> although for £700 Mac mini, you get the fastest single-core performance of any mac sold
14:05:55 <andythenorth> single core is nearly 50% higher than the £5k Mac pro
14:06:03 <TrueBrain> but .. it is still a mac :P
14:06:15 <TrueBrain> for work, Macs are awesome
14:06:19 <TrueBrain> for personal use ... less so :)
14:06:30 <andythenorth> stuff and things
14:06:54 * andythenorth mostly just jealous, the £700 is 10% faster in multicore than mine which was £2500
14:07:04 <andythenorth> moore's law restarted?
14:08:16 <milek7> does it ever stopped? it says about transistors, not core speed
14:08:51 <Samu> in a running game, terraform lowers a slope, then places a house immediately
14:09:49 <LordAro> Samu: it does happen in real life sometimes
14:09:50 <DorpsGek> [OpenTTD/OpenTTD] mattkimber commented on pull request #8485: Codechange: improve performance for complex vehicle chains by resolving sprites less often https://git.io/JLda4
14:09:57 <andythenorth> nice chart milek7
14:11:53 <DorpsGek> [OpenTTD/OpenTTD] mattkimber commented on pull request #8485: Codechange: improve performance for complex vehicle chains by resolving sprites less often https://git.io/JLda0
14:11:59 <Samu> i think im hunting for bugs way too much
14:12:34 <TrueBrain> finding bugs is easy; solving them cleanly is the hard part
14:13:53 <Eddi|zuHause> milek7: there's a theoretical upper limit, so it can't go on forever
14:14:54 <andythenorth> well I broke bananas again :P
14:15:13 <andythenorth> it really hates my FIRS bundle
14:15:19 * andythenorth has probably done a bad thing
14:15:28 <Samu> okay, the code wants to build a house in a tile which has a slope NW, then has a chance to levelland, first up, then down. it fails up, but succeeds down. then right immediately after, it builds the house there
14:16:26 <andythenorth> oh I see what happens
14:16:37 <andythenorth> Bananas can't auto fill information for FIRS
14:16:43 <andythenorth> so I open 'edit' in a new tab
14:16:57 <andythenorth> to copy information from previous upload
14:18:33 <Samu> could also be a tree on a coast
14:19:47 <Samu> DC_NO_WATER was of no help here :(
14:20:15 *** DasPoseidon has joined #openttd
14:22:33 <andythenorth> maybe some houses are just unlucky Samu?
14:28:33 * andythenorth considers automating port and starboard lights for ships
14:28:43 <andythenorth> have to manually swap them over when I flip the sprite currently
14:33:08 <frosch123> andythenorth: it can autofill stuff, if you upload the file first, and the hit "validate" or whatever the button is called
14:33:46 <andythenorth> frosch123 I've got a FIRS bundle that breaks it
14:33:50 <andythenorth> I suspect the origin is with me
14:33:52 <DorpsGek> [OpenTTD/OpenTTD] mattkimber updated pull request #8485: Codechange: improve performance for complex vehicle chains by resolving sprites less often https://git.io/JL5pN
14:34:03 <michi_cc> andythenorth: Cleaning up some more. Should hide menu/dock in fullscreen and also fixed quitting/starting while in fullscreen.
14:35:00 <michi_cc> I scrapped some things from other rotting things and it got a bit larger than just needed for the menu hide :)
14:35:35 <TrueBrain> if it works, I will review ;)
14:37:14 <DorpsGek> [OpenTTD/OpenTTD] mattkimber updated pull request #8485: Codechange: improve performance for complex vehicle chains by resolving sprites less often https://git.io/JL5pN
14:44:11 <milek7> oh, running unsigned apps on macos requires some scary hidden setting enabled?
14:45:43 <milek7> double-clicking binary just shows error
14:46:16 <milek7> but right click->open shows different dialog with option to run anyway
14:47:36 <DorpsGek> [OpenTTD/OpenTTD] michicc commented on pull request #8485: Codechange: improve performance for complex vehicle chains by resolving sprites less often https://git.io/JLdVi
14:48:19 <michi_cc> milek7: Welcome to unsigned app security made by Apple.
14:48:41 <DorpsGek> [OpenTTD/OpenTTD] mattkimber commented on pull request #8485: Codechange: improve performance for complex vehicle chains by resolving sprites less often https://git.io/JLdVX
14:54:26 <DorpsGek> [OpenTTD/OpenTTD] mattkimber updated pull request #8485: Codechange: improve performance for complex vehicle chains by resolving sprites less often https://git.io/JL5pN
14:56:53 <TrueBrain> frosch123: you are getting some love here :D Two approvals \o/ Wait, I need to get in on that action
15:05:08 <supermop_Home> andythenorth apparently i forgot to click the submit msg button last night
15:10:56 <andythenorth> lol Filton Abbey Wood services :P
15:11:20 <andythenorth> nvm wrong channel, FML
15:15:54 <FLHerne> What channel is that the right thing for?
15:16:24 <michi_cc> I'd guess Discord :p
15:17:00 <Eddi|zuHause> i'm not sure discord counts as a "channel" :p
15:17:03 <TrueBrain> in Discord they talk about the game, it is amazing :)
15:17:22 <andythenorth> when are we moving? :P
15:18:53 <andythenorth> supermop_Home got it thanks
15:34:47 <Samu> the fix is still... too restrictive
15:35:13 <Samu> basically, I forbid towns from terraforming down on slopes with a height of 0
15:44:01 <DorpsGek> [OpenTTD/OpenTTD] glx22 commented on pull request #8485: Codechange: improve performance for complex vehicle chains by resolving sprites less often https://git.io/JLdoL
15:44:21 * glx was lazy and didn't crop the screenshots
15:44:38 <glx> but the benefit is impressive
15:48:11 <TrueBrain> it is not the initial 10 times faster, but it sure is faster :D
15:48:15 <TrueBrain> especially for such a silly change :)
15:48:24 <TrueBrain> (as in, amount of code involved)
15:48:39 <Timberwolf> It's more effective the more complex the GRF set's sprite resolution change.
15:48:59 <Timberwolf> I'm quite happy it works that well even with more standard vehicles.
15:49:34 <glx> I included the save in my comment anyway
15:52:18 <Samu> now i'm fast forwarding to see if the town floods itself in the future
15:53:25 <FLHerne> Samu: That doesn't sound too restrictive
15:55:20 <FLHerne> You might want to add some more wiggles to the bank in that test, so it's tested with a variety of slope angles
15:59:05 *** Progman has joined #openttd
16:08:38 <Samu> nearly escaped the flood
16:09:42 *** jottyfan has joined #openttd
16:18:12 <Samu> need to test steep slopes
16:19:11 <Samu> there's some weird errors coming from the terraform command when working on steep slopes
16:20:59 <Samu> difficult to explain, it tries to terraform at least 2 corners at once of a steep slope, one of the corners forces the other corner to lower a height, before it's processed. When it is then processed, it already has the same height as that which was requested, and fail the command
16:22:01 <Samu> it could of continue executing the command though
16:22:10 <Samu> but it seems it was coded not to
16:25:43 <Samu> the "don't know why" isn't reassuring
16:26:08 <Samu> i also wanted to know why
16:28:27 *** Flygon has quit IRC (Quit: A toaster's basically a soldering iron designed to toast bread)
16:34:17 *** Gustavo6046 has joined #openttd
16:41:00 <FLHerne> That's a pretty cool screenshot
16:43:21 <supermop_Home> maybe towns should not be allowed to have their center tile at height 0
16:46:16 <michi_cc> andythenorth: You seen #8487?
16:55:52 <DorpsGek> [OpenTTD/OpenTTD] mattkimber commented on pull request #8485: Codechange: improve performance for complex vehicle chains by resolving sprites less often https://git.io/JLdij
16:55:55 *** Wormnest has joined #openttd
17:02:45 <frosch123> Timberwolf: did you test 8485 for effect vehicles (=vehicle smoke, etc.) and disaster vehicles? for them "sprite_seq.seq[0].sprite" is stateful
17:11:18 <DorpsGek> [OpenTTD/OpenTTD] LordAro updated pull request #8006: Codechange: Increase scrollbar length limit to UINT_MAX and make their length properly unsigned https://git.io/Jv8Vk
17:13:23 <andythenorth> michi_cc do you know what circumstances would cause the dock to appear?
17:13:33 * andythenorth trying to find a valid test, but no dock for me in 1.10.2
17:14:09 <michi_cc> I didn't have a dock before, it's just that the docs says that hiding the menu is only valid if you hide the dock, too.
17:14:21 <michi_cc> So I just set both to hidden.
17:15:08 *** snail_UES_ has joined #openttd
17:15:26 <andythenorth> michi_cc in 'game options' did you intend to fix the 'full screen' toggle indicator that is broken in 1.10.2? Or is it side-effect?
17:16:31 <michi_cc> Half-intended and fixed together with the whole quit/start in fullscreen mode.
17:16:47 <michi_cc> I.e. I can't fix the second without also fixing the first :)
17:17:41 <michi_cc> Area to test: Launch, Cmd+F to fullscreen, quit, relaunch, exit fullscreen, check if old window size.
17:17:43 <DorpsGek> [OpenTTD/OpenTTD] LordAro updated pull request #8458: Fixes for issues with rotated (non-rectangular) airports https://git.io/JLy1U
17:18:06 <DorpsGek> [OpenTTD/OpenTTD] LordAro updated pull request #8458: Fixes for issues with rotated (non-rectangular) airports https://git.io/JLy1U
17:18:12 <andythenorth> maybe that's just an Apple 'feature' now
17:18:57 <michi_cc> Well, my 10.14 doesn't show it.
17:19:05 <andythenorth> it's very odd though, because clicking on dock icon doesn't switch to the full-screen space for OpenTTD
17:19:22 <andythenorth> if you don't know the correct 3-finger scroll swipe magic, you can't get back to OpenTTD
17:19:47 <DorpsGek> [OpenTTD/OpenTTD] LordAro updated pull request #8458: Fixes for issues with rotated (non-rectangular) airports https://git.io/JLy1U
17:21:35 <andythenorth> maybe I screen record it
17:22:11 <michi_cc> andythenorth: Can you do one test and move src/os/cocoa/cocoa_v.mm line 497 to just before the "return true"?
17:23:01 <michi_cc> And for the dock icon, launched from command line or bundle?
17:23:27 <andythenorth> command line, I was about to test the bundle
17:23:47 <michi_cc> The dock thing might be because of the command line launch.
17:28:05 <andythenorth> and the menu bar
17:29:27 <andythenorth> michi_cc no src/os/cocoa here, do I need to build something?
17:30:42 <michi_cc> Oops, src/video/cocoa
17:33:25 *** jottyfan has quit IRC (Quit: jottyfan)
17:35:42 <michi_cc> But if launching from bundle fixes the problems, I don't think it will do anything.
17:36:28 <andythenorth> should I make a video of the cursor trap behaviour? It won't be very informative unless I do a voice track with "I am now moving the mouse up" etc :x
17:36:40 <andythenorth> (unrelated to this PR)
17:37:16 <michi_cc> Maybe. I can't get any cursor trap on my VM, but this is not always identical to the real deal.
17:38:46 <andythenorth> the change in the gist is worse :)
17:39:01 <andythenorth> it animates away the menu bar, but does not switch to the correct display space
17:39:15 <supermop_Home> how many construction stages are there?
17:47:49 *** nielsm has quit IRC (Ping timeout: 480 seconds)
17:56:11 <andythenorth> hmm I don't trust my mac testing right now, I have some annoying illness and my head is mush
17:56:31 <andythenorth> I'm not getting reproducible results, probably because I'm forgetting build steps
18:02:02 <supermop_Home> frosch123 houses and/or industries
18:07:24 <michi_cc> andythenorth: I added a second variant as a comment to your gist.
18:15:39 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain approved pull request #8458: Fixes for issues with rotated (non-rectangular) airports https://git.io/JLdMM
18:20:01 <frosch123> supermop_Home: houses have state 0-2 under construction and 3 finished
18:21:25 <frosch123> same for industries
18:21:56 <frosch123> so, 3 or 4 stages, depending on whether you include "finished"
18:23:58 <andythenorth> michi_cc anything particular to test with that gist?
18:24:13 <andythenorth> I tested the window size
18:25:02 <michi_cc> Did you have any problems regarding fullscreen and the menu bar with the inital PR when started from a bundle?
18:25:13 <DorpsGek> - Update: Translations from eints (by translators)
18:25:24 <michi_cc> I was hoping that the gist might improve behaviour if started from the command line.
18:25:34 <supermop_Home> frosch123 thanks
18:30:11 <andythenorth> michi_cc the gist does indeed improve behaviour when started from command line
18:30:19 <andythenorth> but the bundled package is worse I think
18:30:59 <andythenorth> I'm not convinced I'm correctly deleting previous bundles, so I'm testing again
18:31:04 <michi_cc> Okay, that is weird.
18:32:34 <andythenorth> nah the behaviour is completely inconsistent, macOS is doing something odd
18:33:51 <andythenorth> ok, it behaves differently if started windowed
18:33:54 <andythenorth> vs. started fullscreen
18:35:02 <andythenorth> nah I'm tempted to just declare macOS broken at this point
18:35:09 <andythenorth> the behaviour is entirely inconsistent
18:35:24 <andythenorth> sometimes clicking on the dock icon will switch to the correct display space
18:35:26 <andythenorth> sometimes it won't
18:38:28 <DorpsGek> [OpenTTD/OpenTTD] J0anJosep commented on pull request #8452: Doc: Add labels to landscape grid description. https://git.io/JLdDD
18:41:52 <DorpsGek> [OpenTTD/OpenTTD] J0anJosep commented on pull request #8452: Doc: Add labels to landscape grid description. https://git.io/JLdyv
18:42:05 <Samu> testing these requirements for lowering land: if (GetFloodingBehaviour(tile) != FLOOD_DRYUP && (z != 0 || !_generating_world))
18:44:53 *** J0anJosep has joined #openttd
18:48:14 <DorpsGek> [OpenTTD/OpenTTD] frosch123 commented on pull request #8452: Doc: Add labels to landscape grid description. https://git.io/JLdyZ
18:49:32 <J0anJosep> Is it ok to open a PR not for an issue but for a possible simplifacion of the DrawTree procedure? I don't have time to do it myself, but I would like other developers to check whether it is possible.
18:50:22 <frosch123> sounds like a "discussion"
18:50:37 <J0anJosep> Discussion then... Thanks
18:51:51 <frosch123> no idea how much you want to write, but discussions work better if you split them over multiple posts for each "chapter", so people can comment per "chapter"
18:53:52 <LordAro> J0anJosep: relatedly, *in general*, codechanges with no other purpose are frowned upon - there should be some sort of end result, performance improvement, or some other tangible change (though "improves readability" is valid)
18:55:45 <LordAro> changing stuff just for the sake of changing it (or even mildly improving it) just increases changes of introducing bugs
18:57:10 <TrueBrain> please upgrade to LordAro v1312.3
19:01:58 <frosch123> versions with 1, 2 and 3 in them are good
19:11:44 * andythenorth wonders why long purchase menu is considered such a crime
19:11:54 <andythenorth> compared to digging through complicated menus
19:12:00 <LordAro> and doesn't seem to make sense
19:17:34 <TrueBrain> why are you looking at highscore code :P
19:21:33 <DorpsGek> [OpenTTD/OpenTTD] mattkimber updated pull request #8485: Codechange: improve performance for complex vehicle chains by resolving sprites less often https://git.io/JL5pN
19:24:08 <Timberwolf> Nothing exciting, I realised I left a TODO in a coment even after I'd fixed what it referred to.
19:29:31 <DorpsGek> [OpenTTD/OpenTTD] LordAro approved pull request #8485: Codechange: improve performance for complex vehicle chains by resolving sprites less often https://git.io/JLdSE
19:32:42 <LordAro> Timberwolf: have you looked at #8279 at all, ooi?
19:40:08 <Timberwolf> Ooh, no - didn't see that there.
19:47:56 <michi_cc> andythenorth: I think this ^^^ is the canonical way to disable the menu in fullscreen. If that still misbehaves, I have no clue what to do.
19:53:25 <andythenorth> works as expected with ./openttd, now trying bundle package
19:54:17 <andythenorth> bundle package works
19:55:15 <michi_cc> So, the morale is, do it exactly like in the Apple example, and don't even think about changing a single letter somewhere :)
19:55:44 <michi_cc> You have to find the right example first, of course.
19:56:52 <andythenorth> I should video this mouse bug :|
19:57:03 <andythenorth> unrelated to that PR though
20:08:29 <DorpsGek> [OpenTTD/OpenTTD] J0anJosep commented on pull request #8452: Doc: Add labels to landscape grid description. https://git.io/JLd99
20:09:26 <DorpsGek> [OpenTTD/OpenTTD] J0anJosep updated pull request #8452: Doc: Add labels to landscape grid description. https://git.io/JLyT1
20:16:24 <andythenorth> no 8067 remains in fully comedy glory
20:16:49 <TrueBrain> but 8038 is gone with this?
20:17:00 <andythenorth> I wonder if 8067 is related to the mouse trap bug
20:17:11 <andythenorth> seems like the viewport height is incorrectly measured
20:21:31 <DorpsGek> [OpenTTD/OpenTTD] J0anJosep updated pull request #8452: Doc: Add labels to landscape grid description. https://git.io/JLyT1
20:23:34 <DorpsGek> [OpenTTD/OpenTTD] J0anJosep commented on pull request #8452: Doc: Add labels to landscape grid description. https://git.io/JLdHE
20:23:42 <andythenorth> mouse never stops moving
20:23:58 <andythenorth> you can see where pointer actually was when cursor catches up
20:24:13 <andythenorth> only happens when mouse movement contains an 'up' direction component
20:24:39 <andythenorth> the faster I move the mouse up, the more likely to trigger
20:25:28 <andythenorth> doesn't happen in windowed mode, and changing base set makes no difference (expected, but I tested anyway)
20:28:08 <michi_cc> andythenorth: I can't reproduce that at all, but VMware even in fullscreen mode itself is not totally isolated as it still has active border areas.
20:32:52 <DorpsGek> [OpenTTD/OpenTTD] J0anJosep updated pull request #8452: Doc: Add labels to landscape grid description. https://git.io/JLyT1
20:33:06 <andythenorth> can we attach logging to whatever handles cursor movement?
20:33:15 <andythenorth> I can only reproduce it in the upper 50% of the screen
20:33:22 <andythenorth> might even be upper 33%
20:34:07 <andythenorth> looks to me like it's trying to anticipate showing the menu bar by measuring mouse velocity
20:34:15 <andythenorth> but we...suppressed the menu bar so eh
20:34:55 *** Gustavo6046 has joined #openttd
20:36:01 <frosch123> andythenorth: try changing the call of UpdateCursorPosition in QZ_MouseMovedEvent: replace the "false" with a "true"
20:37:59 <Samu> I just invented a DC_NO_FLOOD
20:38:15 <Samu> let's see if I can make use of it properly
20:40:39 <andythenorth> frosch123 I can now see the system cursor in addition to cursor sprite...
20:42:30 <DorpsGek> [OpenTTD/OpenTTD] J0anJosep updated pull request #8452: Doc: Add labels to landscape grid description. https://git.io/JLyT1
20:44:40 <andythenorth> oof I thought maybe the OS traps the cursor for certain screen edges
20:44:59 <andythenorth> but I turned on 'hot corners' and ottd cursor isn't trapped by those
20:49:15 * andythenorth back to ships :P
20:49:47 <andythenorth> I might revive my railtype grf
20:56:45 <gregdek> Can anyone help me figure out why my server is loading null blitter and dedicated video driver even though I say both on the command line and in the config file to not do that?
21:04:04 <michi_cc> Starting with -D might force that, but I didn't check.
21:04:07 <frosch123> start "openttd -h", it will list the available blitters
21:04:42 <frosch123> if you only have "null" and "dedicated", you compiled without them
21:05:21 <frosch123> oh, and yes, -D forces the dedicated driver, that's the whole point of it, isn't it?
21:06:33 <DorpsGek> [OpenTTD/OpenTTD] michicc opened pull request #8491: Fix #8067: [OSX] Calculate title bar height instead of assuming a fixed value https://git.io/JLd7H
21:07:27 <michi_cc> TrueBrain: #8067 makes half of the previous PR #8487 superfluous, leave it in or partially revert?
21:07:32 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain approved pull request #8491: Fix #8067: [OSX] Calculate title bar height instead of assuming a fixed value https://git.io/JLd75
21:07:45 <michi_cc> I mean #8491 of course.
21:08:04 <gregdek> OK, so if -D forces null video driver and null blitter, how does one take a server-side screenshot?
21:08:06 <TrueBrain> you had me confused for a bit :D
21:08:13 <michi_cc> I looked at the wrong spot previously for the fullscreen startup grey bar.
21:08:26 <TrueBrain> what part do you want to revert?
21:08:46 <michi_cc> I don't really like the fullscreen switch in the first event loop iteration, feels hacky.
21:09:10 <TrueBrain> ah; yeah, if you can fix that now with this PR, I would fix it in 8491 honestly
21:09:12 <michi_cc> TrueBrain: Everything regarding fullscreen_on_mainloop
21:09:40 <michi_cc> fullscreen_on_mainloop is not useless, because I missed the real error.
21:10:00 <frosch123> gregdek: you can't. when coop made screenshots of their servers a few years back, they patched that
21:10:02 <TrueBrain> so yeah, remove it in this PR; why not :)
21:11:46 <gregdek> I don't suppose anybody has that patchset handy...
21:16:08 <DorpsGek> [OpenTTD/OpenTTD] michicc dismissed a review for pull request #8491: Fix #8067: [OSX] Calculate title bar height instead of assuming a fixed value https://git.io/JLd75
21:16:11 <DorpsGek> [OpenTTD/OpenTTD] michicc updated pull request #8491: Fix #8067: [OSX] Calculate title bar height instead of assuming a fixed value https://git.io/JLd7H
21:17:00 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain approved pull request #8491: Fix #8067: [OSX] Calculate title bar height instead of assuming a fixed value https://git.io/JLd5r
21:17:59 *** Gustavo6046 has joined #openttd
21:18:28 <DorpsGek> [OpenTTD/OpenTTD] michicc approved pull request #8466: Fix: GetCurveSpeedLimit needs railtype from current tile https://git.io/JLd5P
21:19:10 <DorpsGek> [OpenTTD/OpenTTD] michicc merged pull request #8466: Fix: GetCurveSpeedLimit needs railtype from current tile https://git.io/JLH4j
21:19:36 <LordAro> that's surely not your first ottd PR/patch?
21:22:31 <andythenorth> it's just the first one that frosch123 didn't write most of :P
21:23:46 <DorpsGek> [OpenTTD/OpenTTD] andythenorth commented on pull request #8491: Fix #8067: [OSX] Calculate title bar height instead of assuming a fixed value https://git.io/JLd5F
21:24:09 <DorpsGek> [OpenTTD/OpenTTD] michicc merged pull request #8491: Fix #8067: [OSX] Calculate title bar height instead of assuming a fixed value https://git.io/JLd7H
21:24:12 <DorpsGek> [OpenTTD/OpenTTD] michicc closed issue #8067: Toggling "Full animation" first glitches, then reduces vertical size https://git.io/JvAGj
21:32:10 <TrueBrain> at least many annoying ones are fixed today :D
21:32:13 <TrueBrain> nice job michi_cc :D
21:34:30 *** jottyfan has joined #openttd
21:40:20 *** J0anJosep has quit IRC (Quit: Konversation terminated!)
21:43:39 <Samu> I'm liking DC_NO_FLOOD, might need another name though
21:45:02 <Samu> it's a flag that is used for the town terraform command
21:45:35 <Samu> the affected tile types are trees and water
21:45:53 *** jottyfan has quit IRC (Quit: jottyfan)
21:46:42 <supermop_Home> wall at big window, or just in side, or off to side?
21:47:02 <Samu> at least I think trees on a shore are affected
21:47:23 <andythenorth> supermop_Home quite often I lie about chimneys
21:47:51 <andythenorth> b or d are the most likely locations to look good
21:48:10 <andythenorth> you can do small roof ridge chimneys but big ones look weird
21:49:47 <supermop_Home> a mountain lodge should have a big hearth, maybe double sided
21:50:15 <supermop_Home> but is it in the lobby or the restaurant, or between the two?
21:50:33 <andythenorth> you have to put your architect / floorplan brain aside
21:50:39 <andythenorth> draw it where it looks best as a sprite :)
21:51:03 <andythenorth> half the FIRS chimneys couldn't exist in reality, they're stuck through walls in an escher-esque way
21:51:32 <supermop_Home> most national park lodges have them inside a bit, so they warm two spaces
21:51:56 <supermop_Home> but crater lake and lake Quinault have them at an exterior lodge
21:52:07 <supermop_Home> as their style is more cottage-like
21:52:50 <supermop_Home> my motivation for putting it inboard is to not break up that big window
21:52:50 <andythenorth> they wouldn't fit on the tile
21:53:29 <supermop_Home> i like loc. A and C for that reason - can cheat
21:54:07 <andythenorth> I would put a big chimney were letter 'd' is, (not the circle)
21:56:36 <supermop_Home> im having too much fun with the construction stages and want to show the chimney going in and floors built around it
21:57:22 <andythenorth> huge fireplace, between lounge and bedrooms, chimney to rear wall
21:58:31 *** frosch123 has quit IRC (Quit: be yourself, except: if you have the opportunity to be a unicorn, then be a unicorn)
22:06:57 <LordAro> Timberwolf: you're obligated to make use of it, imo :p
22:08:39 <Timberwolf> If it was Roller Coaster Tycoon, I could add a loop around the walkway.
22:08:45 <Eddi|zuHause> yeah, it's a terrible bait :p
22:15:04 <supermop_Home> andythenorth ^ drive through hotel with platform in the middle of the resort
22:17:39 <andythenorth> sadly drive-through industries doesn't work very well :)
22:31:17 <DorpsGek> [OpenTTD/OpenTTD] SamuXarick opened pull request #8492: Fix: Don't allow towns to terraform floodable tiles https://git.io/JLdA0
22:32:18 *** Samu has quit IRC (Quit: Leaving)
22:32:34 <Eddi|zuHause> i always wanted an airport with a gap in the middle for a station
22:32:57 *** gelignite has quit IRC (Quit: Stay safe!)
22:44:18 <supermop_Home> could you fake extra construction stages with say a switch that changes the sprite for stage 2?
22:47:12 <supermop_Home> currently i have 0: hole dug, 1: two floor slabs built, 2: all floor slabs built, ground floor enclosed
22:48:05 <supermop_Home> but would be nice to have something between (2) and finished
22:50:07 <andythenorth> not sure there's any vars that could be used
23:04:34 *** sla_ro|master has quit IRC ()
23:05:34 *** andythenorth has quit IRC (Quit: andythenorth)
23:10:20 *** Progman has quit IRC (Remote host closed the connection)
23:14:34 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
23:47:18 <supermop_Home> well ive drawn construction stages... i guess all i have to do now is the poo
23:47:50 <supermop_Home> although an outdoor pool seems a bit silly in arctic climate
continue to next day ⏵
![https://i.imgur.com/orklzAH.png[/img]](https://i.imgur.com/orklzAH.png%5B/img%5D){kind=link}
