--- Log opened Sat Jul 22 00:00:49 2017
20170722 00:03:58-!- DDR [~David@S0106f0f249839863.vc.shawcable.net] has joined #wesnoth-dev
20170722 00:06:45-!- celmin [~celticmin@unaffiliated/celticminstrel] has joined #wesnoth-dev
20170722 00:33:33-!- irker282 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20170722 00:50:44-!- Bonobo [~Bonobo@61.68.206.22] has joined #wesnoth-dev
20170722 00:54:36< vultraz_iOS> celmin: ping
20170722 00:54:37< vultraz_iOS> celticminstrel: ping
20170722 00:55:40< vultraz_iOS> celmin: IIRC this was for the old command primpt? https://github.com/wesnoth/wesnoth/blob/accelerated_rendering/src/controller_base.cpp#L73-L78
20170722 00:55:42< vultraz_iOS> primpt
20170722 00:55:44< vultraz_iOS> pROMPT
20170722 01:02:16-!- mkdroid [~null@unaffiliated/matthiaskrgr] has quit [Ping timeout: 276 seconds]
20170722 01:03:41-!- mkdroid [~null@unaffiliated/matthiaskrgr] has joined #wesnoth-dev
20170722 01:10:53< celmin> Uhh.
20170722 01:11:11< celmin> Yes and no?
20170722 01:11:22< celmin> Don't forget hotkeys can also come in as TEXTINPUT events.
20170722 01:11:50< vultraz_iOS> what?????
20170722 01:12:04< celmin> ...
20170722 01:18:18-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection]
20170722 01:18:25-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev
20170722 01:21:32< vultraz_iOS> celmin: how is this possible
20170722 01:23:57-!- mkdroid [~null@unaffiliated/matthiaskrgr] has quit [Ping timeout: 240 seconds]
20170722 01:25:25-!- mkdroid [~null@unaffiliated/matthiaskrgr] has joined #wesnoth-dev
20170722 01:31:06< vultraz_iOS> celmin: btw did you ever make any progress on the viewport zoom math?
20170722 01:31:07< vultraz_iOS> (https://github.com/wesnoth/wesnoth/issues/1845)
20170722 01:32:04< celmin> I did not.
20170722 01:33:28< vultraz_iOS> I think the problem with it is it does not correctly take into account hex shapes
20170722 01:33:37< vultraz_iOS> hexes are not places end-to-end in a grid
20170722 01:33:52< vultraz_iOS> therefor, if zoom is 72 and you have 3 hexes, the width is not 216
20170722 01:33:57< vultraz_iOS> but rather, 180
20170722 01:34:17< vultraz_iOS> the position offsets are 1/4 for horizontal and 1/2 for vertical
20170722 01:35:37-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Ping timeout: 255 seconds]
20170722 01:35:47< vultraz_iOS> i feel like an idiot, though, since I'm not seeing the appropriate formula
20170722 01:35:56-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170722 01:37:52-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Client Quit]
20170722 01:43:10< celmin> As for "how is this possible", it's because Aginor used TEXTINPUT events to get hotkeys to reflect your keyboard layout, or something.
20170722 01:53:50-!- mkdroid [~null@unaffiliated/matthiaskrgr] has quit [Ping timeout: 258 seconds]
20170722 01:54:36< vultraz_iOS> celmin: this confuses me
20170722 02:11:13-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection]
20170722 02:11:19-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev
20170722 02:36:42-!- irker011 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20170722 02:36:42< irker011> wesnoth: Ignacio R. Morelle wesmere:master 4a630b9d00f7 / wesmere/js/ (retina.js retina.min.js): Fix Retina image selection not kicking in across wesnoth.org subdomains https://github.com/wesnoth/wesmere/commit/4a630b9d00f7dc2c2ff4187d124ffb0610045dae
20170722 02:38:39< irker011> wesnoth: Ignacio R. Morelle wesmere:master 954d474d1e5f / wesmere/.htaccess: CORS is now handled in the server configuration https://github.com/wesnoth/wesmere/commit/954d474d1e5fb3e10a975a9e66925219233f6b74
20170722 02:52:21< celmin> Confuses, huh. :/
20170722 02:52:52< celmin> TBH, my feeling is that SDL2's keyboard handling is not well-designed.
20170722 02:53:22< celmin> So maybe it's okay for it to be confusing. >_>
20170722 02:55:02< celmin> Then again, I suspect there may be literally no way to query the physical keyboard layout...
20170722 02:55:55< celmin> An AZERTY layout (for example) is effectively equivalent to plucking off the keycaps and putting them back on it different positions.
20170722 02:56:11< celmin> AFAIK (there could be something I don't know about).
20170722 03:01:31< irker011> wesnoth: Ignacio R. Morelle wesmere:master c6c24a21b218 / wesmere/sass/mw/_links.scss: sass/mw: Remove link icons from files.w.o and SF.net links https://github.com/wesnoth/wesmere/commit/c6c24a21b218f4f30b5c0ed8e6adaf457b1bb162
20170722 03:03:53< vultraz_iOS> why would you want that?
20170722 03:08:05< vultraz_iOS> the physical layout
20170722 03:08:32< celmin> Because when you're binding keys, the most logical name for the key is what's written on the physical device.
20170722 03:13:46-!- celmin [~celticmin@unaffiliated/celticminstrel] has quit [Quit: And lo! The minstrel departs, to spread the music to the masses!]
20170722 03:53:39-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection]
20170722 03:53:45-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev
20170722 03:55:56-!- horrowind [~Thunderbi@p2003008E6C0B26B6964452FFFE0220ED.dip0.t-ipconnect.de] has quit [Ping timeout: 255 seconds]
20170722 03:55:58-!- horrowind1 [~Thunderbi@p2003008E6C0B2623964452FFFE0220ED.dip0.t-ipconnect.de] has joined #wesnoth-dev
20170722 03:58:22-!- horrowind1 is now known as horrowind
20170722 04:01:26-!- horrowind [~Thunderbi@p2003008E6C0B2623964452FFFE0220ED.dip0.t-ipconnect.de] has quit [Quit: horrowind]
20170722 04:29:47-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has quit [Quit: And lo! The computer falls into a deep sleep, to awake again some other day!]
20170722 04:46:33-!- Kwandulin [~Kwandulin@p200300760F12B3DC20ADEAC2A91F0980.dip0.t-ipconnect.de] has joined #wesnoth-dev
20170722 05:41:26-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection]
20170722 05:41:34-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev
20170722 06:04:04-!- JyrkiVesterinen [~JyrkiVest@85-23-197-3.bb.dnainternet.fi] has joined #wesnoth-dev
20170722 06:04:55-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has quit [Read error: Connection reset by peer]
20170722 06:05:04-!- bumba [~bumbadada@wesnoth/developer/bumbadadabum] has joined #wesnoth-dev
20170722 06:05:26-!- bumba [~bumbadada@wesnoth/developer/bumbadadabum] has quit [Remote host closed the connection]
20170722 06:16:30-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20170722 06:31:47-!- timotei_ [~timotei@wesnoth/developer/timotei] has joined #wesnoth-dev
20170722 06:40:23-!- timotei_ [~timotei@wesnoth/developer/timotei] has quit [Ping timeout: 268 seconds]
20170722 07:02:31-!- JyrkiVesterinen [~JyrkiVest@85-23-197-3.bb.dnainternet.fi] has quit [Quit: .]
20170722 07:04:35-!- zookeeper [zookeeper@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20170722 07:24:47-!- boucman [~rosen@wesnoth/developer/boucman] has left #wesnoth-dev []
20170722 07:29:35-!- Coffee_irc [~david@61.68.206.22] has joined #wesnoth-dev
20170722 07:38:08-!- JyrkiVesterinen [~JyrkiVest@85-23-197-3.bb.dnainternet.fi] has joined #wesnoth-dev
20170722 07:42:59-!- Kwandulin [~Kwandulin@p200300760F12B3DC20ADEAC2A91F0980.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20170722 08:20:48-!- Kwandulin [~Kwandulin@p200300760F12B3DCAC8159F81658B184.dip0.t-ipconnect.de] has joined #wesnoth-dev
20170722 08:41:58-!- mjs-de [~mjs-de@x4db6df19.dyn.telefonica.de] has joined #wesnoth-dev
20170722 08:47:00-!- Duthlet [~Duthlet@dslb-188-101-088-243.188.101.pools.vodafone-ip.de] has joined #wesnoth-dev
20170722 08:55:30-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection]
20170722 08:55:36-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev
20170722 09:09:09-!- stikonas [~stikonas@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170722 09:18:13-!- stikonas [~stikonas@wesnoth/translator/stikonas] has quit [Ping timeout: 255 seconds]
20170722 09:18:33-!- stikonas [~stikonas@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170722 09:25:05-!- stikonas [~stikonas@wesnoth/translator/stikonas] has quit [Ping timeout: 240 seconds]
20170722 09:37:55-!- irker011 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20170722 09:49:53-!- DDR [~David@S0106f0f249839863.vc.shawcable.net] has quit [Ping timeout: 248 seconds]
20170722 10:10:58-!- vultraz_iOS changed the topic of #wesnoth-dev to: 1.14 not delayed for accelerated_rendering | Wesnoth Developers Channel | >>> Want to help? Go here: http://r.wesnoth.org/t42911 (and thanks!) <<< | Discord Server: https://discord.gg/tSmJS2E | Logs: http://irclogs.wesnoth.org | Bug tracker: http://bugs.wesnoth.org
20170722 10:16:45-!- Kwandulin [~Kwandulin@p200300760F12B3DCAC8159F81658B184.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20170722 10:24:58-!- JyrkiVesterinen [~JyrkiVest@85-23-197-3.bb.dnainternet.fi] has quit [Quit: .]
20170722 10:36:53-!- timotei_ [~timotei@wesnoth/developer/timotei] has joined #wesnoth-dev
20170722 10:53:50-!- Kwandulin [~Kwandulin@p200300760F12B3DC9C1561EE063619EB.dip0.t-ipconnect.de] has joined #wesnoth-dev
20170722 11:01:00-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20170722 11:03:57-!- horrowind [~Thunderbi@p2003008E6C0B2623964452FFFE0220ED.dip0.t-ipconnect.de] has joined #wesnoth-dev
20170722 11:06:27-!- mjs-de [~mjs-de@x4db6df19.dyn.telefonica.de] has quit [Remote host closed the connection]
20170722 11:30:59-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection]
20170722 11:31:05-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev
20170722 11:56:44-!- horrowind [~Thunderbi@p2003008E6C0B2623964452FFFE0220ED.dip0.t-ipconnect.de] has quit [Quit: horrowind]
20170722 12:06:00-!- JyrkiVesterinen [~JyrkiVest@85-23-197-3.bb.dnainternet.fi] has joined #wesnoth-dev
20170722 12:20:01-!- Appleman1234 [~quassel@124x38x163x22.ap124.ftth.ucom.ne.jp] has quit [Ping timeout: 255 seconds]
20170722 12:22:01-!- Appleman1234 [~quassel@124x38x163x22.ap124.ftth.ucom.ne.jp] has joined #wesnoth-dev
20170722 12:49:39-!- irker644 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20170722 12:49:39< irker644> wesnoth: Jyrki Vesterinen wesnoth:master 4e4d7b527790 / / (11 files in 3 dirs): Rewrite the FPS cap implementation https://github.com/wesnoth/wesnoth/commit/4e4d7b527790a6bcb083dccf4cfb22cd8d5e3a92
20170722 12:52:57-!- Kwandulin [~Kwandulin@p200300760F12B3DC9C1561EE063619EB.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20170722 12:58:25 * zookeeper blinks at commit description
20170722 13:01:35< JyrkiVesterinen> :P
20170722 13:20:23< zookeeper> JyrkiVesterinen, was that something that profiling led you to?
20170722 13:23:13< JyrkiVesterinen> Attempting to profile, yes.
20170722 13:23:33< JyrkiVesterinen> The very first thing the profiler summary shows is a CPU usage graph.
20170722 13:24:10< JyrkiVesterinen> And I immediately noticed "wait... the game is not using a full core".
20170722 13:24:44< JyrkiVesterinen> Which led me to conclude that the game must have been throttling itself, instead of running out of available CPU power.
20170722 13:27:13< zookeeper> righty
20170722 13:28:13< zookeeper> sounds great, i'll try it this evening and see if it makes things perform better for me as well
20170722 13:34:23-!- gfgtdf [~chatzilla@x4e3683d8.dyn.telefonica.de] has joined #wesnoth-dev
20170722 13:34:39< gfgtdf> JyrkiVesterinen: so now the game needs more cpu usage?
20170722 13:35:04< JyrkiVesterinen> Yes, obviously. It's drawing more frames than before.
20170722 13:35:12< zookeeper> sounds more like "will actually max out a single core if it needs to"
20170722 13:35:31< gfgtdf> JyrkiVesterinen: and why did you rempove the --max-fps  switch ?
20170722 13:36:04< JyrkiVesterinen> TimeBeginPeriod() is a bit bad, as it increases the power usage of Windows itself. But it's still better than busy looping, which would be the other way to achieve better sleep granularity.
20170722 13:36:43< JyrkiVesterinen> I removed the --max-fps switch because it was the easiest way to make the target FPS dynamic (dependent on actual refresh rate, instead of hardcoded to 50).
20170722 13:37:17< JyrkiVesterinen> And it's not that important anyway when the target FPS automatically matches the refresh rate.
20170722 13:37:48< gfgtdf> well before max-fps youcl be used to reduce wensoths cou usage i'd think
20170722 13:38:15< gfgtdf> could* cpu*
20170722 13:38:30< JyrkiVesterinen> Hmm. Maybe I need to add it back later, then.
20170722 13:39:58< gfgtdf> hmm maybe i just need to buy a new pc ...
20170722 13:48:18-!- travis-ci [~travis-ci@ec2-54-197-147-29.compute-1.amazonaws.com] has joined #wesnoth-dev
20170722 13:48:19< travis-ci> wesnoth/wesnoth#14460 (master - 4e4d7b5 : Jyrki Vesterinen): The build was broken.
20170722 13:48:19< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/256361964
20170722 13:48:19-!- travis-ci [~travis-ci@ec2-54-197-147-29.compute-1.amazonaws.com] has left #wesnoth-dev []
20170722 14:07:37-!- Alkenrinnstet [~alkenrinn@42.61.217.253] has joined #wesnoth-dev
20170722 14:22:24-!- Bonobo [~Bonobo@61.68.206.22] has quit [Ping timeout: 260 seconds]
20170722 14:25:11-!- moongazer [~moongazer@59.95.137.122] has joined #wesnoth-dev
20170722 14:30:09-!- Kwandulin [~Kwandulin@p200300760F12B3DC9C1561EE063619EB.dip0.t-ipconnect.de] has joined #wesnoth-dev
20170722 14:43:55-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection]
20170722 14:44:01-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev
20170722 15:18:29-!- vultraz_iOS [uid24821@wesnoth/developer/vultraz] has quit [Quit: Connection closed for inactivity]
20170722 15:31:32-!- mjs-de [~mjs-de@x4db6df19.dyn.telefonica.de] has joined #wesnoth-dev
20170722 15:37:33< gfgtdf> smowhow all text attachements to github give me "SignatureDoesNotMatch" errors, is there a way to work argound those?
20170722 15:43:36-!- moongazer [~moongazer@59.95.137.122] has quit [Ping timeout: 260 seconds]
20170722 15:49:00-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has joined #wesnoth-dev
20170722 15:50:06-!- irker644 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20170722 15:51:32< gfgtdf> celticminstrel: do you whtehe rit possible to workaround the "SignatureDoesNotMatch" error on gthub when trying to see attachements?
20170722 15:51:55< celticminstrel> Huh?
20170722 15:53:23< gfgtdf> celticminstrel: when i try to open attachemnts, for exmaple in https://github.com/wesnoth/wesnoth/issues/1320 the fiel in post#2 i always get a "SignatureDoesNotMatch" error
20170722 15:55:01< celticminstrel> That's weird, because I could download it easily.
20170722 15:55:38< gfgtdf> celticminstrel: display_event_error.cfg ? just now ?
20170722 15:56:01< celticminstrel> Yup.
20170722 15:56:05< APic> k
20170722 15:58:11< gfgtdf> hmm
20170722 15:59:21< gfgtdf> ok it seems the 'noscipt blocked xss attempt' is related
20170722 15:59:34< gfgtdf> now i have to figure out how to add an exception
20170722 16:01:58-!- gfgtdf [~chatzilla@x4e3683d8.dyn.telefonica.de] has quit [Quit: ChatZilla 0.9.93 [Firefox 54.0.1/20170628075643]]
20170722 16:02:38-!- gfgtdf [~chatzilla@x4e3683d8.dyn.telefonica.de] has joined #wesnoth-dev
20170722 16:04:10< Ravana_> should https://github.com/wesnoth/wesnoth/issues/1690 be closed?
20170722 16:05:41< gfgtdf> yes i think it should
20170722 16:26:06< celticminstrel> Definitely.
20170722 16:54:07-!- Kwandulin [~Kwandulin@p200300760F12B3DC9C1561EE063619EB.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20170722 17:01:27-!- moongazer [~moongazer@59.95.137.227] has joined #wesnoth-dev
20170722 17:03:48-!- Kwandulin [~Kwandulin@p200300760F12B3DC3912D4DFBB6C1D1B.dip0.t-ipconnect.de] has joined #wesnoth-dev
20170722 17:22:35< gfgtdf> any opinion on making a 1.13.9 release soon ?
20170722 17:40:44< pydsigner> I think we should finish checking JyrkiVesterinen's FPS fix first at least
20170722 17:41:00< pydsigner> I need to check zookeeper's change with ANL yet still as well
20170722 17:48:15< gfgtdf> pydsigner: i think he just added force_lock_settings=yes  to it
20170722 17:50:33< gfgtdf> pydsigner: if you want to work on a new land after 1.13.9 you could maybe add a difficulcy option (slider or dropdownbutton), similar to what 2p dark forcast has. Someone complained  that it was too easy right now.
20170722 17:51:02< pydsigner> Too easy ey? Haha that's funny
20170722 17:51:08< pydsigner> I'm such a scrub
20170722 17:53:29-!- DeFender1031 [~DeFender1@217.132.16.231] has joined #wesnoth-dev
20170722 17:54:54< DeFender1031> Greetings, my fellow weslanders.
20170722 18:05:04< DeFender1031> "Which led me to conclude that the game must have been throttling itself, instead of running out of available CPU power." Great. So now we'll hsave performance issues AND melt our CPUs. :P
20170722 18:06:22< JyrkiVesterinen> Hey. When you have a problem "the game runs too slowly", and you notice that it doesn't peg a CPU core, then the first think you should do is MAKE it peg a core. ;)
20170722 18:07:06< DeFender1031> JyrkiVesterinen, I know. I'm just being facetiously pessimistic.
20170722 18:07:47< DeFender1031> One question I do have though is wether removing the frame cap will change the behavior of [print] and [message] duration= attributes.
20170722 18:08:11< DeFender1031> (On the other hand, measuring the duration in frames is asinine and should probably not be the behavior.)
20170722 18:08:45< DeFender1031> (But it is currently, at least according to the docs, and thus, I ask.)
20170722 18:10:22< DeFender1031> JyrkiVesterinen, ^^^
20170722 18:11:43< JyrkiVesterinen> The cap is not removed. It's just rewritten.
20170722 18:12:01< JyrkiVesterinen> (But I changed the default from 50 FPS to the screen refresh rate, usually 60 FPS.)
20170722 18:12:37< JyrkiVesterinen> Which means that durations measured in frames are indeed different than before, and even system-dependent.
20170722 18:17:16< celticminstrel> FPS fix?
20170722 18:20:03< JyrkiVesterinen> https://github.com/wesnoth/wesnoth/commit/4e4d7b527790a6bcb083dccf4cfb22cd8d5e3a92
20170722 18:20:23< JyrkiVesterinen> The performance problems in master were caused by a poorly implemented FPS cap.
20170722 18:21:27< DeFender1031> JyrkiVesterinen, which is why having the timing based on frames is asinine.
20170722 18:21:40< JyrkiVesterinen> Agreed.
20170722 18:22:15< DeFender1031> though, at this point, replacing it to mean "seconds" would be saner and close enough to the original behavior to not cause issues...
20170722 18:22:26< DeFender1031> (I'd also suggest changing the default to 60)
20170722 18:22:46< celticminstrel> I would've assumed that [delay] was specified in either seconds or milliseconds.
20170722 18:22:52< DeFender1031> (For the same raisin.)
20170722 18:22:57< DeFender1031> reason*
20170722 18:23:15< celticminstrel> If it's seconds you need to treat it as a floating-point, of course.
20170722 18:23:22< JyrkiVesterinen> I see I'm not the only one here who sometimes mixes up homophones. :P
20170722 18:24:16< DeFender1031> oh wait... not seconds
20170722 18:24:20< DeFender1031> I'm an idiot.
20170722 18:24:31< DeFender1031> It'd be 60th of a second.
20170722 18:24:35< DeFender1031> Brainfart.
20170722 18:24:54< JyrkiVesterinen> 50th of a second because the old cap was 50 FPS.
20170722 18:25:10< DeFender1031> hmm
20170722 18:25:17< DeFender1031> yeah, that's not really a sane interval either.
20170722 18:25:51< DeFender1031> were there not things dependent on it, I'd say it shuold be ms like everything else
20170722 18:28:19< celticminstrel> Has anyone actually checked what [delay] did before?
20170722 18:28:28< celticminstrel> Or is this all speculative?
20170722 18:28:36< DeFender1031> No one mentioned delay.
20170722 18:28:45< DeFender1031> We're talking about [message] and [duration]
20170722 18:28:48< DeFender1031> er
20170722 18:28:50< celticminstrel> Oh, print and message.
20170722 18:28:52< DeFender1031> [print]
20170722 18:28:58< DeFender1031> yeah.
20170722 18:29:03< celticminstrel> I assumed those used the same implementation as delay though...
20170722 18:29:05< DeFender1031> and those are clearly documented.
20170722 18:29:11< DeFender1031> not in the slightes.
20170722 18:29:24< celticminstrel> ...why the heck not?
20170722 18:29:29< DeFender1031> Why should anything in wesnoth use the same implementation as anything else?
20170722 18:29:35< DeFender1031> That would just be madness!
20170722 18:29:37< celticminstrel> For sanity?
20170722 18:29:45< DeFender1031> Sarcasm, dude.
20170722 18:30:00< DeFender1031> But no, delay uses ms, message and print use frames.
20170722 18:33:24< celticminstrel> Ultimately that should be changed then... but that's easier said than done...
20170722 18:36:23-!- Kwandulin [~Kwandulin@p200300760F12B3DC3912D4DFBB6C1D1B.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20170722 18:42:39-!- Kwandulin [~Kwandulin@p200300760F12B3DCC05ADA6B0C068540.dip0.t-ipconnect.de] has joined #wesnoth-dev
20170722 18:43:13< shadowm> JyrkiVesterinen: https://pastebin.com/raw/KTgVsuNQ[3~
20170722 18:43:19< shadowm> https://pastebin.com/raw/KTgVsuNQ  rather
20170722 18:44:50< JyrkiVesterinen> Right, I already know.
20170722 18:45:09< JyrkiVesterinen> As gfgtdf asked me to being --max-fps back, I think I'll just do that.
20170722 18:48:24< DeFender1031> celticminstrel, the question is really how to do it sanely in a way that both makes sense and doesn't cause all sorts of backwards-compatibility frustration... the sanest approach would be to deprecate the "duration" attribute for removal (level 3 in the doc I threw together) in favor of a "duration_ms" tag which works as expected, while at the same time switching "duration=" over to mean "50th of a second" rather than "frame". Then once the 
20170722 18:48:26< DeFender1031> deprecation period has passed, switch "duration" to just mean milliseconds, and wrap the "_ms" version to it.
20170722 18:48:51< DeFender1031> (Also, the deprecation message could mention that it'll change to mean ms in the next version...)
20170722 18:49:03< celticminstrel> What's this about a doc you threw together?
20170722 18:49:13< DeFender1031> you don't remember?
20170722 18:49:24< DeFender1031> oh... you weren't involved in that conversation, were you?
20170722 18:49:28< celticminstrel> Was I not?
20170722 18:49:33< DeFender1031> https://wiki.wesnoth.org/CompatibilityStandards
20170722 18:50:00< celticminstrel> Oh, I think I was involved in the conversation before you threw the doc together, though.
20170722 18:50:18< DeFender1031> Well, whatever.
20170722 18:51:17< DeFender1031> That's what came out of that conversationthat happened after one too many arguments about it between zookeeps and vult on the topic.
20170722 18:51:55< DeFender1031> (AFAIK, it still has yet to be fully read through by anyone here other than me...)
20170722 18:52:24< celticminstrel> Well, I'm reading it now.
20170722 18:52:27< DeFender1031> (But that's fine. The necessary technical requirements for the different levels is not in place yet...)
20170722 18:52:34< celticminstrel> I would've read it earlier if I'd known it was done.
20170722 18:53:11< DeFender1031> (I'll probably get to it myself in about a month once I have some more free time if no one else does in the meantime.)
20170722 18:53:34< celticminstrel> What do you mean by necessary technical requirements?
20170722 18:53:45< DeFender1031> (And I hope to God that my workload does actually lighten then...)
20170722 18:54:35< DeFender1031> Well, the idea would be that there would be deprecation functions in all the necessary places (wml, lua, c++) for each level, and an option when turning on debug mode to set what level of notice you want to see.
20170722 18:55:06< celticminstrel> We probably need something in the WML preprocessor too.
20170722 18:55:17< celticminstrel> In order to deprecate macros that aren't ActionWML.
20170722 18:55:18< DeFender1031> That way, content devs can work on updating their use of features that are slated for imminent removal before updating the ones that just happen to be sorta ugly.
20170722 18:55:42< DeFender1031> well, a WML tag that says "this whatever is deprecated" could be used in deprecated macros.
20170722 18:55:59-!- crimson_penguin [~crimson_p@ec2.happyspork.com] has joined #wesnoth-dev
20170722 18:55:59-!- crimson_penguin [~crimson_p@ec2.happyspork.com] has quit [Changing host]
20170722 18:55:59-!- crimson_penguin [~crimson_p@wesnoth/developer/crimsonpenguin] has joined #wesnoth-dev
20170722 18:56:04< DeFender1031> Ah, that's what you mean.
20170722 18:56:15< DeFender1031> Meh, that tag could be implemented such that it'd work anyway.
20170722 18:56:18< DeFender1031> anywhere*
20170722 18:56:42< celticminstrel> What, are you proposing the WML parser should detect the tag and print a message?
20170722 18:57:03< DeFender1031> That's one way to do it, yes.
20170722 18:57:14-!- Jetrel_bot [~Jetrel@ec2.happyspork.com] has joined #wesnoth-dev
20170722 18:57:18< DeFender1031> That's not the only way to do it, though.
20170722 18:58:18< celticminstrel> Ooh, replacing removed interfaces with an error stub for one release doesn't sound like a bad idea...
20170722 18:58:19-!- DDR [~david@ec2.happyspork.com] has joined #wesnoth-dev
20170722 18:59:18< DeFender1031> Thank you. I thought so.
20170722 19:00:08< celticminstrel> It'd be nice if we could retroactively apply these rules for everything changed in 1.13, but that might be too much of a chore...
20170722 19:00:48< DeFender1031> It'd certainly be tedious...
20170722 19:01:08< DeFender1031> But it would be ideal.
20170722 19:01:09< celticminstrel> (Also, some of the framework for it isn't in place yet, I guess.)
20170722 19:01:11< celticminstrel> Yeah.
20170722 19:01:41< celticminstrel> My thought for deprecating macros was some means to mark the macro as deprecated (perhaps a #deprecated directive) which causes the preprocessor to print a message whenever the macro is used.
20170722 19:01:42< DeFender1031> The tedium would probably be more in excavating commit logs to figure out all the things that were removed.
20170722 19:01:51< celticminstrel> Yeah.3
20170722 19:02:11< celticminstrel> And of course, things only added after 1.12 probably don't need to be maintained.
20170722 19:02:15< DeFender1031> celticminstrel, that's... a smarter approach than mine.
20170722 19:05:19< celticminstrel> I think the put_unit parameter order should probably be level 2...
20170722 19:05:28< DeFender1031> huh?
20170722 19:05:40< celticminstrel> The order of parameters to wesnoth.put_unit was changed.
20170722 19:06:00< celticminstrel> On the other hand, backstab shold probably be level 1...
20170722 19:06:09< celticminstrel> As in the backstab key.
20170722 19:06:17< DeFender1031> ah
20170722 19:06:34< celticminstrel> Though I'm not sure on that...
20170722 19:07:14< DeFender1031> Well, look, the beauty is that anything which someone thinks is a 1 that turns out to be a 2 can have its level raised as such wehn necessary.
20170722 19:07:36< DeFender1031> just like 2s will semi-regularly be raised to 3s, and so on.
20170722 19:07:51< DeFender1031> And 3s by definition end up as 4s.
20170722 19:10:13< DeFender1031> This deprecation policy is actually probably the ideal for pretty much any piece of software, IMHHO (double-H intentional, for "hardly", of course :P).
20170722 19:10:52< celticminstrel> ...hardly honest? :S
20170722 19:10:59< gfgtdf> does [message] duration currenly even do anything? I couldn't find it in the implemtation
20170722 19:11:07< celticminstrel> Good question.
20170722 19:11:32< celticminstrel> But if it doesn't it's likely because something was missed in the porting to Lua, so it should be added back.
20170722 19:14:05< DeFender1031> celticminstrel, the H usually means "humble"...
20170722 19:14:12< DeFender1031> hence, "hardly humble"
20170722 19:14:26< gfgtdf> celticminstrel: i couldn'T find it in the 1.13 implementation eigher https://github.com/wesnoth/wesnoth/blob/1.12/src/game_events/action_wml.cpp
20170722 19:14:35< DeFender1031> gfgtdf, supposedly, it forces a minimum duration for the messate to remain onscreen.
20170722 19:14:38< celticminstrel> Really? I've always thought of it as "honest".
20170722 19:14:43< celticminstrel> But I guess that works too.
20170722 19:15:44< DeFender1031> celticminstrel, google results all say "humbl", except for urban dictionary which has "honest" in parens as an alternate meaning.
20170722 19:15:51< DeFender1031> humble*
20170722 19:16:57< gfgtdf> DeFender1031: ye i know
20170722 19:17:03< DeFender1031> JyrkiVesterinen, just finished compiling latest master. The perforrmance is definitely improved, but still pretty laggy in a few places, and clearly still worse than in 1.12
20170722 19:17:12< DeFender1031> Still, it's a marked improvement.
20170722 19:18:00< JyrkiVesterinen> I think remaining problems shouldn't be too hard to fix either.
20170722 19:19:56< gfgtdf> celticminstrel: i now think [message] duration= does nothign since 1.5.9 when [message] was ported to gui2
20170722 19:20:39< celticminstrel> So let's reimplement it then?
20170722 19:21:10< DeFender1031> Still have the one on [print] though, no?
20170722 19:21:24< gfgtdf> celticminstrel: but not with 'frames'
20170722 19:21:30< celticminstrel> Of course.
20170722 19:21:43< DeFender1031> btw, if you're gonna do stuff wth message timing code, I have a wish feature which is setting a max duration for a message as well.
20170722 19:21:57< DeFender1031> (so that you can do things like time an action sequence to music.)
20170722 19:22:09< celticminstrel> Do you mean a min duration?
20170722 19:22:12< DeFender1031> no, max.
20170722 19:22:20< DeFender1031> this duration tag IS min duration already
20170722 19:22:33< celticminstrel> How can you time a dialogue sequence to music if the player can dismiss it at any time?
20170722 19:22:37< DeFender1031> meaning, I want a tag that makes the message auto-dismiss at a certain time.
20170722 19:22:55< celticminstrel> If you want to time a dialogue sequence to music, then what you need is a way to disallow user dismissal.
20170722 19:23:04< pydsigner> ^
20170722 19:23:07< DeFender1031> truth is, this feature would be better as a tag that makes the duration automatic.
20170722 19:23:43< DeFender1031> i was thinking set the duration of min and max to the same value, so the user can't dismiss it early, and it'll auto-dismiss itself
20170722 19:24:01< pydsigner> I'm opposed to this.
20170722 19:24:04< DeFender1031> but you're right, an "auto_dismiss=yes" tag might be better.
20170722 19:24:10< DeFender1031> pydsigner, why>?
20170722 19:24:14< celticminstrel> Which is pydsigner opposed to exactly?
20170722 19:24:30< pydsigner> Locking message dismissal times
20170722 19:24:40< DeFender1031> pydsigner, why?
20170722 19:24:42< celticminstrel> So all of it?
20170722 19:24:45< pydsigner> 1) it completely breaks UX
20170722 19:25:17< pydsigner> Do I click to get past the message? Do I not click?
20170722 19:25:34< pydsigner> What if I step away from my computer, am I going to miss dialog?
20170722 19:25:38< DeFender1031> no it doesn't, any more than not being able to pause animations does. If you're doing an action sequence, you expect the user to either be watching or to pause the game by going to the menu.
20170722 19:26:03< pydsigner> DeFender1031: which would then break your music synchronization.
20170722 19:26:14< DeFender1031> there can be some visual difference between a timed message and not.
20170722 19:26:53< DeFender1031> pydsigner, 1. that's true, but it'd be at least possible to time it in cases where the user doesn't pause and 2. not if pausing were to eventually also pause the music.
20170722 19:27:03< pydsigner> 2) This would suck for testing. Like right now for FPS stuff, I'm continually holding down space to skip past all the opening dialog.
20170722 19:27:19< celticminstrel> Uh, that's what escape is for. :P
20170722 19:27:56< DeFender1031> and if the messages were also accellerated?
20170722 19:28:22< DeFender1031> Ideally, I'd want the music and sfx to accellerate with the accelleration as well.
20170722 19:28:29< DeFender1031> but whatever.
20170722 19:28:33< celticminstrel> That's asking a bit much...
20170722 19:28:36< pydsigner> You can't accelerate music
20170722 19:28:43< celticminstrel> Accelerating music is difficult.
20170722 19:28:48< pydsigner> Without breaking it
20170722 19:28:52< celticminstrel> The naive approach changes the pitch.
20170722 19:29:01< celticminstrel> So what you need to do is basically skip half the samples.
20170722 19:29:09< celticminstrel> Much like scaling an image down.
20170722 19:29:09< DeFender1031> Basically, I'm into making action sequences and want more tools to make them feel fuller.
20170722 19:29:41< celticminstrel> Which of course degrades the quality.
20170722 19:29:46< DeFender1031> pydsigner, yes, you can. celticminstrel yes, and that would be the expected behavior while accellerating such a sequence.
20170722 19:29:58< celticminstrel> Because that's inevitable when discarding data.
20170722 19:30:10< pydsigner> And after a certain point in acceleration, you can't really pull it off anymore.
20170722 19:30:31< DeFender1031> fine, so ignore my idea for now. maybe sometime in the future when the code is cleaner and lends itself more easily to tightly-knit action sequences, I'll bring it up again.
20170722 19:30:37< pydsigner> People making action scenes intended to be played at 3x would have their random bg music completely destroyed
20170722 19:30:46-!- Kwandulin [~Kwandulin@p200300760F12B3DCC05ADA6B0C068540.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20170722 19:30:54< pydsigner> Basically what we'd need is a cutscene mode
20170722 19:30:56< DeFender1031> pydsigner, action sequences would always be intended to be played at 1x.
20170722 19:31:09< DeFender1031> If you're rushing through it, you should expect to miss things.
20170722 19:31:25< pydsigner> DeFender1031: there are add-ons that say to use a certain multiplier for cutscenes
20170722 19:31:30< pydsigner> Because units move so slowly.
20170722 19:31:40-!- irker990 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20170722 19:31:40< irker990> wesnoth: gfgtdf wesnoth:gfgtdf_fix1473 ac608e49c21f / src/pathfind/pathfind.cpp: attempt to fix #1473: cannot step on hexes with high movecosts https://github.com/wesnoth/wesnoth/commit/ac608e49c21feaa7f13bdf98e1666020afbe5724
20170722 19:32:03< DeFender1031> yeah, well, that's a hack and I would also like a feature that allows setting the speed of various stuff during cutscenes.
20170722 19:32:20< pydsigner> Hide the in-game UI, so that the player knows there's something going on
20170722 19:32:27< pydsigner> And then have everything be timed
20170722 19:32:36< celticminstrel> gfgtdf: Please don't merge that unless the attempt is confirmed. Otherwise it'll autoclose the issue.
20170722 19:32:40< pydsigner> With Esc skipping the whole thing.
20170722 19:32:45< pydsigner> That's a massive project though.
20170722 19:32:52< DeFender1031> so that you could do [move_unit] / duration_per_hex=150
20170722 19:33:03< celticminstrel> Ooooh.
20170722 19:33:05< pydsigner> DeFender1031: I'm picturing it as
20170722 19:33:12< pydsigner> [cutscene]
20170722 19:33:23< gfgtdf> i cannot confrm it before i merge it... . But i can manualyl repoen the issue if it didn't fix it.
20170722 19:33:31< pydsigner>  [element] duration=200ms [move_unit]
20170722 19:33:31< celticminstrel> :|
20170722 19:33:41< DeFender1031> pydsigner, a special [cutscene] tag that gives more control and times stuff and whatnot would also work, yeah.
20170722 19:34:12< celticminstrel> So this would also ignore the acceleration settings?
20170722 19:34:48< DeFender1031> pydsigner, essentially, I have grand ideas about being able to do some real action to enhance stories with cinematics, and I'm hoping people listen to my ideas to take baby steps in the direction that makes the system as a whole friendlier toward proper cinematics.
20170722 19:35:28< DeFender1031> celticminstrel, I would imagine that it would have its own accelleration settings which holding shift or running with accelleration on would then even further accellerate, perhaps.
20170722 19:35:49< DeFender1031> Though, again, ideally each move_unit would be independantly timable.
20170722 19:36:17< DeFender1031> so you can do things like have one unit walk in slowly and another run away screaming
20170722 19:36:34< celticminstrel> Right, that's what prompted my "ooooh".
20170722 19:36:53< pydsigner> I think you'd just want to have a separate cutscene speed control
20170722 19:36:56< celticminstrel> I guess if it was skipped with escape you'd still need to play it out at max acceleration though...
20170722 19:37:13< celticminstrel> Maybe max acceleration plus not showing moves.
20170722 19:37:28< pydsigner> You'd basically skip all delays
20170722 19:37:34< pydsigner> duration=0ms on everything
20170722 19:37:40< pydsigner> So you just end up with the final state
20170722 19:37:58< celticminstrel> Basically disable all animations, I guess.
20170722 19:38:14< pydsigner> DeFender1031: I think you'd have to allow that perfectly timed music could be hurt
20170722 19:38:27< pydsigner> Because I don't think SDL has music acceleration capabitilies.
20170722 19:38:32< DeFender1031> that'd be nice. I currently have a hack in my cutscene of where in the scene to start so you don't have to run through the whole thing.
20170722 19:38:37< pydsigner> "capabitilies" lol
20170722 19:39:04< celticminstrel> SDL doesn't really have music capabilities at all?
20170722 19:39:28< celticminstrel> Not sure what SDL_Mixer has; I think there's some stuff related to altering audo, though.
20170722 19:39:50< DeFender1031> pydsigner, it'd work if accelleration skipped a scene entirely, orif someone stuck to running at normal speed though.
20170722 19:40:21< DeFender1031> accelleartion could also silence the music, and the music could seek to where it would have been up to when accelleration stops...
20170722 19:40:26< DeFender1031> there's a few workarounds.
20170722 19:40:36< DeFender1031> obviously, none of this is going to happen any time soon
20170722 19:40:48< DeFender1031> but it's an interesting idea to think about.
20170722 19:41:03< pydsigner> DeFender1031: silenced music is what I was imagining actually
20170722 19:41:09< pydsigner> So we're on the same page there at least
20170722 19:41:53< pydsigner> I know that you can start music with SDL >mixer< [happy now?] from an offset
20170722 19:42:01< pydsigner> So you could catch up afterwards
20170722 19:42:15< pydsigner> Since everything is exactly metered out in the CS frames.
20170722 19:42:31< pydsigner> The GUI stuff is already possible at least
20170722 19:43:00< pydsigner> You'd just tell Wesnoth to use a theme like what's in IftU/AtS
20170722 19:43:41< pydsigner> But the game field would accept 0 input and you'd have to somehow figure out all the x duration stuff
20170722 19:44:03< celticminstrel> pydsigner: And that theme is mainlined now, too.
20170722 19:44:07< pydsigner> Ah ok
20170722 19:44:12< pydsigner> Even simpler
20170722 19:44:28< pydsigner> Would simultaneous actions be allowed?
20170722 19:44:50< pydsigner> [cutscene] [simultaneous]
20170722 19:44:56< celticminstrel> Ideally?
20170722 19:46:02< pydsigner> [element] [delay 100] [move_unit 400] [/element] [element]
20170722 19:46:15< pydsigner> Oops, extra element there at the end
20170722 19:46:37< pydsigner> [element] [move_unit 300] [/element]
20170722 19:46:42< pydsigner> [/simultaneous]
20170722 19:47:13< DeFender1031> pydsigner, not as such, I think. Not at the moment.
20170722 19:47:35< DeFender1031> Though, simultaneous movement of units (or at least in move_uits_fake) would be very nice.
20170722 19:48:08< DeFender1031> (My cutscene has a point where a bunch of people ideally ought to all be shuffling out at once, but instead it moves one at a time, because wesnoth.)
20170722 19:52:43< zookeeper> DeFender1031, you should never make messages auto-dismiss after a given time. it's a turn-based game and you're supposed to be able to take your time, so breaking that principle in a scenario just so you can time a cutscene to music or whatever is... folly.
20170722 19:52:44< DeFender1031> Also, ideally, the [cutscene] tag would fade out the interface bars when it starts, and fade them back in when it ends.
20170722 19:53:23< DeFender1031> zookeeper, again, if it's a cutscene, then it's not running in the same kind of turn-based mode, and you're expected to watch in real-time.
20170722 19:53:34< DeFender1031> zookeeper, I agree with you for messages in general, though.
20170722 19:53:35< pydsigner> zookeeper: that was my point, hence the suggestion of a delimited cutscene
20170722 19:53:56< zookeeper> DeFender1031, not if what you have on your screen is the kind of message that you know is not a real-time thing
20170722 19:54:19< JyrkiVesterinen> I have seen turn-based games where cutscenes usually wait for me to proceed, but not always.
20170722 19:54:22< DeFender1031> zookeeper, as I said, there would be some visual indicator that it was a different kind of message.
20170722 19:54:26< JyrkiVesterinen> So it isn't unprecedented.
20170722 19:54:38< DeFender1031> It could even be a completely different styl.
20170722 19:54:42< DeFender1031> style*
20170722 19:54:55< zookeeper> floating text? :>
20170722 19:55:03< pydsigner> You don't want to break the aesthetic of the game
20170722 19:55:10< pydsigner> zookeeper: floating dialogues :P
20170722 19:56:04< DeFender1031> Yeah, I'd suggest it be something a little more stylized and... portrait-having... than floating text, but a little less solid-looking than the regular message screen.
20170722 19:56:36< DeFender1031> zookeeper, also, if this were implemented after my navigable messages idea, then you could always go back to read the stuff that happened in the cutscene as well...
20170722 19:57:01< DeFender1031> zookeeper, and if it were after your unlimited undo slider, then really there's no concern over people missing it.
20170722 19:57:19< zookeeper> do you really need/want that for multiple messages from different speakers? or would it be good enough to be able to make a single message's text appear in timed bursts?
20170722 19:57:21< zookeeper> kinda like...
20170722 19:57:22< zookeeper> this
20170722 19:57:27< DeFender1031> "oh, I missed that message..." *slightly rewinds*
20170722 19:57:37< zookeeper> *music sync breaks*
20170722 19:57:45< pydsigner> DeFender1031: what, with a record scratch for the music?
20170722 19:57:50< zookeeper> ha
20170722 19:57:52< DeFender1031> zookeeper, I'm not sure I know what you're asking.
20170722 19:58:10< DeFender1031> pydsigner, yeah, that can be an easter egg. :P
20170722 19:58:46< pydsigner> DeFender1031: I realize you're kinda throwing a wishlist at the wall
20170722 19:59:14< pydsigner> But a lot of this stuff feels like it's asking a lot
20170722 20:00:23< DeFender1031> pydsigner, at this point, I'm just discussing hypothetical super-awesome-what-if-we-coulds.
20170722 20:00:59< DeFender1031> The serious suggestions were things that would be much smaller steps toward potentially this kind of thing.
20170722 20:01:11< DeFender1031> But in no way do I expect it to happen any time soon.
20170722 20:01:44< zookeeper> i do think it's fundamentally a pipedream to be able to really sync cutscenes to music
20170722 20:01:51< DeFender1031> Basically, I'm saying "it'd be cool if it were fully-featured enough to do this, so why not do small things that might inch us closer, even if we never fully get there."
20170722 20:02:14< DeFender1031> zookeeper, I agree with you that there's a high likelihood of that.
20170722 20:02:32< DeFender1031> Still, no reason not to do simple things which make it a little easier to make cutscenes...
20170722 20:02:53< pydsigner> Truly syncing music with cutscenes means you're sitting down with a composer and producing a soundtrack specifically for your campaign
20170722 20:03:24< DeFender1031> ...
20170722 20:03:30< APic> …
20170722 20:03:36< pydsigner> …...
20170722 20:03:39< DeFender1031> There's a step before that, I think.
20170722 20:03:55< zookeeper> i'm not really sure what i think of acceleration in cutscenes. on one hand, acceleration is for making the game animations run faster/slower (well... slower is used for animation testing only, i'd imagine), so maybe it shouldn't affect cutscenes.
20170722 20:04:08< DeFender1031> There's plenty of great music, especially in the UMC packs, which could have a cutscene reasonably timed to them without seeming forced.
20170722 20:04:16< pydsigner> DeFender1031: what, cueing music at specific times?
20170722 20:05:03< pydsigner> If you're just cueing tracks, you can do most of that without fixed timings at all.
20170722 20:05:06< DeFender1031> pydsigner, there's a step in between cueing music at specific times (currently doable) and having a composer specifically write a piece times to a cutscene.
20170722 20:05:20< DeFender1031> which is timing your cutscene to existing music.
20170722 20:05:57< zookeeper> if i wanted to sync music to a cutscene right now, i'd cut the music track into pieces, make the pieces loopable, and then change the music within the cutscene... after i'd have managed to persuade someone to implement changing of music via crossfade.
20170722 20:06:01< pydsigner> >Wesnoth 2.0 because a music video creation platform
20170722 20:06:47< zookeeper> lack of ability to smoothly fade from one track to another is IMO the #1 obstacle in varying music in cutscenes or even mid-scenario in general.
20170722 20:06:57< pydsigner> zookeeper: true
20170722 20:07:10< pydsigner> Again, here we're bumping into SDL limitations
20170722 20:08:41< pydsigner> We'd have to switch to PortAudio or something.
20170722 20:09:03< DeFender1031> The ability to crossfade WOULD be a good first step.
20170722 20:09:17< DeFender1031> btw, there is a way to sort of hack a crossfade.
20170722 20:09:34< zookeeper> pydsigner, SDL only supports one "music" channel?
20170722 20:09:44< DeFender1031> cut the overlap and play it as a [sound] instead.
20170722 20:09:48< zookeeper> yeah
20170722 20:09:54< DeFender1031> but that's very icky
20170722 20:10:03< zookeeper> is it? sounds relatively simple
20170722 20:10:03< DeFender1031> ad if the volumes are set at different levels...
20170722 20:10:20< JyrkiVesterinen> zookeeper: Yes, SDL_Mixer supports only one music channel.
20170722 20:10:33< DeFender1031> zookeeper, icky in terms of it's not The Right Way To Do Things (TM).
20170722 20:12:56< DeFender1031> there also likely would be a noticeable seam in the music, and the thing I said about the volume levels.
20170722 20:13:13< DeFender1031> (would be interesting to test, though...)
20170722 20:13:15< zookeeper> stop the current music track, cut out the crossfade duration's worth of music-to-come, play that on a regular sound channel with the right volume and a fade-out. doesn't sound like codethulhu to me in principle. dunno if the seam would be a problem.
20170722 20:14:21< JyrkiVesterinen> Still sounds extremely hacky. It would be better to just use OpenAL or something instead of SDL_Mixer.
20170722 20:14:42< DeFender1031> "the right volume" isn't doable, users can adjust the music volume and sound volume separately IIRC.
20170722 20:15:14< pydsigner> DeFender1031: I think SDL allows adjusting sound volume on a per-channel basis
20170722 20:15:55< DeFender1031> pydsigner, how does that help?
20170722 20:16:50< pydsigner> The in-Wesnoth sound volume adjustments are even granular between sound types
20170722 20:17:12< pydsigner> So you could have a hidden music crossfade channel set to whatever music volume is set to
20170722 20:17:20< pydsigner> But it's still awful and shouldn't be done.
20170722 20:18:43< DeFender1031> Oh, you're talking about if the crossfade is implemented internally... When I came up with the idea just now I was thinking about it as a WML hack.
20170722 20:18:55< DeFender1031> Internally it could be controlled a lot more precisely, yeah.
20170722 20:19:01< DeFender1031> Still a terrible way to do it.
20170722 20:31:10-!- celmin [~celticmin@unaffiliated/celticminstrel] has joined #wesnoth-dev
20170722 20:34:01-!- travis-ci [~travis-ci@ec2-23-22-251-232.compute-1.amazonaws.com] has joined #wesnoth-dev
20170722 20:34:02< travis-ci> wesnoth/wesnoth#14461 (gfgtdf_fix1473 - ac608e4 : gfgtdf): The build failed.
20170722 20:34:02< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/256436503
20170722 20:34:02-!- travis-ci [~travis-ci@ec2-23-22-251-232.compute-1.amazonaws.com] has left #wesnoth-dev []
20170722 20:36:34< irker990> wesnoth: Jyrki Vesterinen wesnoth:master dc5e8b36ea93 / / (7 files in 3 dirs): Restore --max-fps https://github.com/wesnoth/wesnoth/commit/dc5e8b36ea93d764f672352fd2779fa25dd3057e
20170722 20:42:10-!- moongazer [~moongazer@59.95.137.227] has quit [Quit: Leaving]
20170722 21:10:55-!- JyrkiVesterinen [~JyrkiVest@85-23-197-3.bb.dnainternet.fi] has quit [Quit: .]
20170722 21:20:44< gfgtdf> i'm unable to open the command bix in current master
20170722 21:20:56< gfgtdf> does anyone have an idea why that is ?
20170722 21:23:21< gfgtdf> also can't see the map in a mp game (when i use 2 clients, client 2 has a black map, the sidebar still shows information about the currently mouseover hex though)
20170722 21:24:56< gfgtdf> oh the map gets normal as soon as its player2s turn
20170722 21:35:55-!- vultraz_iOS [uid24821@wesnoth/developer/vultraz] has joined #wesnoth-dev
20170722 21:39:54-!- Duthlet [~Duthlet@dslb-188-101-088-243.188.101.pools.vodafone-ip.de] has quit [Quit: leaving]
20170722 21:41:31< irker990> wesnoth: gfgtdf wesnoth:master 8a7a3a41ff22 / src/game_state.cpp: attempt to fix map beeing black in mp games #1852 https://github.com/wesnoth/wesnoth/commit/8a7a3a41ff22cea4e0ca482a5e599fee015914f2
20170722 21:45:20< gfgtdf> hmm i'm actually sure i tested mp games multiple times on 1.13.8 how did i not notice that bug before
20170722 21:46:13-!- vultraz_iOS changed the topic of #wesnoth-dev to: 1.13.9 scheduled for sometime in August| Wesnoth Developers Channel | >>> Want to help? Go here: http://r.wesnoth.org/t42911 (and thanks!) <<< | Discord Server: https://discord.gg/tSmJS2E | Logs: http://irclogs.wesnoth.org | Bug tracker: http://bugs.wesnoth.org
20170722 21:46:59< shadowm> I don't know about performance, but an -O3 build is still around 32% CPU idle.
20170722 21:47:16< shadowm> (In-game.)
20170722 21:47:41< shadowm> Compared to 5% with Wesnoth 1.12.
20170722 21:48:13< vultraz_iOS> on master?
20170722 21:49:59< shadowm> Yes.
20170722 21:50:58-!- gfgtdf [~chatzilla@x4e3683d8.dyn.telefonica.de] has quit [Quit: ChatZilla 0.9.93 [Firefox 54.0.1/20170628075643]]
20170722 22:05:54< irker990> wesnoth: Charles Dang wesnoth:accelerated_rendering b09d965679f4 / src/display.hpp: Display: exit render_scaled_to_zoom early if passed texture is null https://github.com/wesnoth/wesnoth/commit/b09d965679f42710507d8f7da8eaa81e02e272b1
20170722 22:05:57< irker990> wesnoth: Charles Dang wesnoth:accelerated_rendering 319e5c1e62d4 / src/ (display.cpp display.hpp): Display: avoid temp vectors when rendering partial fog/shround images on hex https://github.com/wesnoth/wesnoth/commit/319e5c1e62d4cc4155e49b74d7018a1a8920aca5
20170722 22:06:00< irker990> wesnoth: Charles Dang wesnoth:accelerated_rendering b95f56405c95 / src/halo.cpp: Halos: fixed incorrect drawing origin when at non-default zoom https://github.com/wesnoth/wesnoth/commit/b95f56405c95dc17af61e3a2e2e318a420f21d4f
20170722 22:06:03< irker990> wesnoth: Charles Dang wesnoth:accelerated_rendering c144222b9afd / src/ (gui/core/canvas.cpp sdl/render_utils.hpp sdl/window.cpp): Moved SDL_SetRenderDrawColor wrapper to utils file and added a second overload https://github.com/wesnoth/wesnoth/commit/c144222b9afd2d4235dfd18b6f807d785801c08b
20170722 22:06:06< irker990> wesnoth: Charles Dang wesnoth:accelerated_rendering 1bc69ce3a3a3 / src/gui/core/ (canvas.cpp canvas.hpp): GUI2/Canvas: optimized cached texture management https://github.com/wesnoth/wesnoth/commit/1bc69ce3a3a3e666a0798189ab3a8a0d884a5968
20170722 22:06:29< vultraz_iOS> wedge009: 
20170722 22:06:42< vultraz_iOS> please test a_r again to see if you still get graphics glitches
20170722 22:06:55< vultraz_iOS> on whatever driver ( I think it was D3d for you) that you got them
20170722 22:09:41< zookeeper> eh eh. i can't easily reset back to an earlier commit _without_ destroying all local changes (and thus requiring stashing)? that's inconvenient.
20170722 22:10:03< vultraz_iOS> of course you can do that
20170722 22:10:12< celmin> Um.
20170722 22:10:38< vultraz_iOS> you use TGit, IIRC, so you want the "Mixed" reset option
20170722 22:10:56< celmin> Pretty sure you can't reset without stashing.
20170722 22:11:19< celmin> Maybe that "Mixed" reset option automatically stashes and unstashes for you.
20170722 22:11:26< vultraz_iOS> dunno
20170722 22:11:31< zookeeper> "leave working tree untouched, reset index" <- means it doesn't actually change the files.
20170722 22:11:50< celmin> If you reset without stashing, if it's a hard reset you'll lose everything, and if it's not a hard reset, everything that happened after the commit you reset to would be considered local changes.
20170722 22:12:00< zookeeper> whether i'm interpreting the description correctly is of course another question
20170722 22:12:17< vultraz_iOS> you will retain the changes of the commits past the ones you reset as local
20170722 22:12:23< vultraz_iOS> so if you have commits A and B
20170722 22:12:26< vultraz_iOS> and reset to A
20170722 22:12:30< vultraz_iOS> B's changes will be local changes
20170722 22:12:44< celmin> "leave working tree untouched, reset index" means that everything you staged for commit is no longer staged for commit, but no files are changed on disk.
20170722 22:13:04< celmin> Since the index is what you have staged for commit.
20170722 22:13:06< zookeeper> what's staging?
20170722 22:13:18< zookeeper> i'm not kidding. i've never looked it up.
20170722 22:13:28< celmin> "git add" is staging.
20170722 22:13:40< celmin> So is "git rm".
20170722 22:16:28< zookeeper> ok, great. mixed reset didn't seem to mess up anything.
20170722 22:19:35< zookeeper> ...it just requires me to then manually revert the local changes it produces
20170722 22:23:32< celmin> Like I said.
20170722 22:23:42< celmin> It would've probably been easier to stash,
20170722 22:24:36< zookeeper> i avoid stashing because it causes a vast flood of "file changed, reload?" pop-ups in notepad++ that i have to click through
20170722 22:25:09< zookeeper> (and then it loses my bookmarks on those files too, if i actually reload)
20170722 22:26:29< pydsigner> Use a different editor?
20170722 22:28:23< zookeeper> yeah, because switching editors whenever your current one has a slight inconvenience (if this even counts as the editor's fault) is a good idea?
20170722 22:29:18< APic> Emacs > *
20170722 22:30:09< zookeeper> no comment
20170722 22:30:21< celmin> Ew emacs.
20170722 22:36:22-!- travis-ci [~travis-ci@ec2-54-197-147-29.compute-1.amazonaws.com] has joined #wesnoth-dev
20170722 22:36:23< travis-ci> wesnoth/wesnoth#14464 (master - dc5e8b3 : Jyrki Vesterinen): The build is still failing.
20170722 22:36:23< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/256448508
20170722 22:36:23-!- travis-ci [~travis-ci@ec2-54-197-147-29.compute-1.amazonaws.com] has left #wesnoth-dev []
20170722 23:16:57-!- Bonobo [~Bonobo@61.68.206.22] has joined #wesnoth-dev
20170722 23:17:42-!- travis-ci [~travis-ci@ec2-23-22-251-232.compute-1.amazonaws.com] has joined #wesnoth-dev
20170722 23:17:43< travis-ci> wesnoth/wesnoth#14465 (master - 8a7a3a4 : gfgtdf): The build is still failing.
20170722 23:17:43< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/256460543
20170722 23:17:43-!- travis-ci [~travis-ci@ec2-23-22-251-232.compute-1.amazonaws.com] has left #wesnoth-dev []
20170722 23:20:52-!- deep-book-gk_ [~1wm_su@5.62.43.15] has joined #wesnoth-dev
20170722 23:21:25-!- deep-book-gk_ [~1wm_su@5.62.43.15] has left #wesnoth-dev []
20170722 23:28:53< irker990> wesnoth: Sofartin wesnoth:accelerated_rendering 0e77976ae879 / projectfiles/Xcode/Wesnoth.xcodeproj/project.pbxproj: Updated Xcode project for macOS targets https://github.com/wesnoth/wesnoth/commit/0e77976ae879852a321e35836525426b0feaf738
20170722 23:29:05< celmin> …who's that?
20170722 23:29:22< vultraz_iOS> a PR
20170722 23:29:29< celmin> Oh.
20170722 23:29:56< celmin> Oh yeah, mattsc, can you update the XCode project on master?
20170722 23:30:07< celmin> This branch commit reminds me that it's missing things even on master.
20170722 23:30:24< celmin> (Which is definitely going to cause conflicts when you next rebase, Vultraz.)
20170722 23:30:46< vultraz_iOS> FML ;_;
20170722 23:30:52< celmin> ?
20170722 23:31:02< vultraz_iOS> i hate conflicts
20170722 23:31:17< celmin> vultraz_iOS: Just taking the version of the project on your side of the merge should suffice, I think.
20170722 23:31:28< vultraz_iOS> i see
20170722 23:31:30< vultraz_iOS> alright
20170722 23:32:03< celmin> And if it doesn't I'm sure someone else will fix it. >_>
20170722 23:40:09< mattsc> celmin: I’ll try
20170722 23:43:29-!- mjs-de [~mjs-de@x4db6df19.dyn.telefonica.de] has quit [Remote host closed the connection]
20170722 23:47:04-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection]
--- Log closed Sun Jul 23 00:00:49 2017