[00:03] snover, https://bugs.scummvm.org/ticket/9613 [00:03] reading backlog, hold please [00:03] snover, I think that after that is fixed lsl6hires is completable :) [00:04] Action: TMM listens to some elevator music [00:05] boy, thats one heck of a argument list [00:06] somewhat sparsely used apparently :) [00:06] something for explicit memory management it seems? [00:07] yeah it is for purging stuff from memory [00:07] but it has never worked the way lsl6hires is trying to use it here [00:08] it's the ending cutscenes, you have to wonder if this ever worked in scidhuv.exe either [00:08] well, it will free 0x2e4. and none of the rest :) [00:09] yeah, but if the original interpreter also didn't free the entire movie (I guess?) who would ever have noticed [00:09] not like they had valgrind back then [00:09] and after this the game exits [00:09] apparently someone did not know to design symmetric apis here [00:09] kLoad accepts multiple arguments [00:10] i mean, multiple resource numbers [00:10] kUnload only accepts one [00:10] in scummvm? or is this also like that in scidhuv.exe? [00:11] in the original game interpreter [00:11] sierra.exe/sier5.exe/sierraw.exe/sierrah.exe/scidhuv.exe/whatever they felt like calling it this release [00:11] qfg4 tries this too [00:12] SIERRA.EXE apparently [00:12] not that it matters :P [00:13] looks like these are the only two games that make bad calls [00:13] how did the original interpreter deal with faulty argument lists like that? [00:15] just used the first two arguments and ignore the rest [00:16] of course now i have to try to use logic to decide whether it is best to copy ssci and not unload all of them, or do what the game script author intended which might keep a memory leak from forming [00:16] ill probably just do what ssci does until it turns out to not be the right move. [00:18] hmm. [00:18] Last message repeated 1 time(s). [00:18] theres a workaround for this same room in non-hires LSL6 [00:18] in scummvm? [00:18] yes [00:19] yeah, exact same workaround even [00:19] makes me wonder if it was actually intended for hires. [00:19] except that it is in drawPic and not changeState? [00:19] oh. there are like 10s of these invalid calls in rm740 :) [00:20] 15 distinct invalid calls to kUnLoad [00:20] see line 790 of workarounds.cpp [00:20] yes, that is the one i was talking about [00:20] ah, I was looking at 736 before [00:21] if you look at all those lines it seems that the LSL6HIRES one was simply missed there and both lsl6 and lsl6hires have the same bug? [00:22] Im currently working on determining if that workaround was intended for hires all along or not [00:23] I really need to figure out a faster way of running blames on these files that have been arranged and rearranged so many times [00:23] ah you think that none of the workarounds are applicable to lsl6? [00:25] oh my god. i have gone back through like 8 commits now that are just clean up reorder move [00:25] change the workaround-types [00:26] adding room numbers to all workarounds [00:26] sheezus. finally. [00:26] its not wrong. so thats good. [00:27] snover, I can verify that adding the workaround exactly as it is for LSL6 to LSL6HIRES also allows the game to complete [00:28] so, there's that, at least :) [00:28] yeah [00:29] it seems that the 'SIERRA' background from the text box isn't cleared properly though [00:29] fairly minor thing, I'll make a separate bug for it [00:29] GitHub132 (~GitHub132@192.30.252.34) joined #scummvm. [00:29] [scummvm] csnover pushed 1 new commit to master: https://git.io/vP6Nu [00:29] scummvm/master 06d58c9 Colin Snover: SCI32: Fix crash in LSL6hires ending... [00:29] GitHub132 (GitHub132@192.30.252.34) left #scummvm. [00:29] that rendering glitch may be due to another bug i am working on [00:30] https://bugs.scummvm.org/ticket/9583 this one [00:30] ah yeah, sounds the same [00:31] I've just added it as a comment [00:31] btw: amazing work on getting these games to run so well, I also clicked around in SQ6 a bit and it all looks fantastic [00:32] thanks a lot :) [00:32] sq6 is like a middle finger from the past. at the very very end of the credits it uses a screen transition that is never used anywhere else in any other game [00:33] hahaha [00:34] just waiting for the entire credits of lsl6hires to finish now [00:37] is love for sail also supposed to work now? [00:37] no [00:38] the virtual machine needs upgrades for SCI3 to work [00:39] i am glad to see that the LSL6 hires remains gaudy up to the bitter end [00:39] it does [00:40] they committed [00:40] how many SCI3 games even exist? LSL7 and KQ8? [00:40] lsl7, rama, lighthouse, phantasmagoria 2 [00:41] and shivers 2, which is a sci3 game but doesnt use the VM [00:42] doesn't use the VM? [00:43] in sci3 sierra added a kernel abstraction layer so games could be written in c++ instead of sc [00:43] ohhh [00:43] all hardcoded logic? [00:43] yes [00:43] fun [00:52] snover, hahaha, as a final fuck you, the very last call the interpreter makes when quitting the ending is also invalid [00:52] kResCheck? [00:52] Attempt to free invalid entry 0037:0000 as bitmap! [00:53] sorry, not the same error [00:53] thats caused by loading games from scummvm gui [00:53] ah, ok, so I don't have to file that as a bug then? [00:53] the restore code is bypassed and something gets screwed up with ScrollWindow [00:53] it needs to be fixed eventually so you might as well open a ticket about it :) [00:54] ok [00:54] but this is because I hit 'load' from the scummvm gui for the save? [00:54] yes [00:55] if you have original save/load turn on and choose to load the game from inside the game, then that wont happen [00:55] ok, in that case lsl6hires is at least completely playable now [00:56] modulo some small bugs that may or may not be in the original too, I'll check [00:58] SylvainTV (~SylvainTV@LFbn-1-8392-241.w81-254.abo.wanadoo.fr) left irc: Read error: Connection reset by peer [01:23] Henke37b (~Henrik@81-227-16-59-no133.bredband.skanova.com) left irc: Quit: ERR_SHUTDOWN [01:29] sluicebox (~userid@cpe-45-48-227-201.socal.res.rr.com) left irc: Ping timeout: 260 seconds [01:35] demonimin (~demonimin@unaffiliated/demonimin) left irc: Ping timeout: 256 seconds [01:37] demonimin (~demonimin@unaffiliated/demonimin) joined #scummvm. [01:49] DominusExult (~dominus@194-166-156-125.adsl.highway.telekom.at) joined #scummvm. [01:49] DominusExult (~dominus@194-166-156-125.adsl.highway.telekom.at) left irc: Changing host [01:49] DominusExult (~dominus@unaffiliated/dominus) joined #scummvm. [01:50] Vampire0_ (~Vampire@jEdit/Vampire) joined #scummvm. [01:54] Dominus (~dominus@unaffiliated/dominus) left irc: Ping timeout: 248 seconds [01:54] Nick change: DominusExult -> Dominus [01:54] Vampire0 (~Vampire@jEdit/Vampire) left irc: Ping timeout: 268 seconds [01:58] GitHub39 (~GitHub39@192.30.252.46) joined #scummvm. [01:58] [scummvm] csnover pushed 1 new commit to master: https://git.io/vP6pU [01:58] scummvm/master 65d0c65 Colin Snover: SCI32: Fix crash clicking quit button during LSL6hires credits [01:58] GitHub39 (GitHub39@192.30.252.46) left #scummvm. [02:26] snover (~Adium@unaffiliated/snover) left irc: Quit: Leaving. [02:38] dreammaster (~dreammast@c-73-167-118-204.hsd1.ma.comcast.net) joined #scummvm. [02:38] #scummvm: mode change '+o dreammaster' by ChanServ!ChanServ@services. [04:03] Polynomial-C (~Poly-C@gentoo/developer/Polynomial-C) left irc: Ping timeout: 268 seconds [04:08] dreammaster (~dreammast@c-73-167-118-204.hsd1.ma.comcast.net) left irc: [04:35] Cheeseness (~cheesenes@ppp14-2-136-237.bras1.hba1.internode.on.net) joined #scummvm. [04:52] Littleboy (~littleboy@c-174-62-174-85.hsd1.ma.comcast.net) left irc: Read error: Connection reset by peer [05:08] LittleToonCat (~littlecat@sydnns0115w-156057019210.dhcp-dynamic.FibreOp.ns.bellaliant.net) left irc: Ping timeout: 260 seconds [05:30] LittleToonCat (~littlecat@sydnns0115w-047054148237.dhcp-dynamic.FibreOP.ns.bellaliant.net) joined #scummvm. [05:38] LittleToonCat (~littlecat@sydnns0115w-047054148237.dhcp-dynamic.FibreOP.ns.bellaliant.net) left irc: Remote host closed the connection [05:59] Polynomial-C (~Poly-C@gentoo/developer/Polynomial-C) joined #scummvm. [06:50] waltervn (~waltervn@82-171-142-149.ip.telfort.nl) joined #scummvm. [06:50] #scummvm: mode change '+o waltervn' by ChanServ!ChanServ@services. [06:59] m_kiewitz (~m_kiewitz@x4d03f3aa.dyn.telefonica.de) joined #scummvm. [06:59] m_kiewitz (~m_kiewitz@x4d03f3aa.dyn.telefonica.de) left irc: Changing host [06:59] m_kiewitz (~m_kiewitz@scummvm/undead/m-kiewitz) joined #scummvm. [06:59] #scummvm: mode change '+o m_kiewitz' by ChanServ!ChanServ@services. [07:42] m_kiewitz (~m_kiewitz@scummvm/undead/m-kiewitz) left irc: Read error: Connection reset by peer [07:45] m_kiewitz (~m_kiewitz@x4d03ee4c.dyn.telefonica.de) joined #scummvm. [07:45] m_kiewitz (~m_kiewitz@x4d03ee4c.dyn.telefonica.de) left irc: Changing host [07:45] m_kiewitz (~m_kiewitz@scummvm/undead/m-kiewitz) joined #scummvm. [07:45] #scummvm: mode change '+o m_kiewitz' by ChanServ!ChanServ@services. [08:34] WooShell (~Markus@ipbcc071f7.dynamic.kabel-deutschland.de) joined #scummvm. [08:35] good meowning =^.^= [08:47] WooShell: morn [08:55] rootfather (~rootfathe@unaffiliated/rootfather) joined #scummvm. [08:55] #scummvm: mode change '+o rootfather' by ChanServ!ChanServ@services. [08:55] g'day folks [09:13] SDL 2.0.5 achieved "prerelease" status \o/ [09:13] http://forums.libsdl.org/viewtopic.php?t=12145 [09:14] frankyboy_ (~franky@ppp109-252-159-76.pppoe.spdop.ru) joined #scummvm. [09:16] ajax16384 (~User@ip33.net130.n37.ru) joined #scummvm. [09:16] #scummvm: mode change '+o ajax16384' by ChanServ!ChanServ@services. [09:41] criezy I had to reinstall my building environment, tests for linear and anisotropic filtering will follow soon. [10:20] omer_mor (~Omer@46-117-132-33.bb.netvision.net.il) joined #scummvm. [10:22] omer_mor_ (~Omer@46-117-132-33.bb.netvision.net.il) left irc: Ping timeout: 244 seconds [10:33] frankyboy_ (~franky@ppp109-252-159-76.pppoe.spdop.ru) left irc: Quit: #E>6C O >B 20A [10:34] ny00123 (~ny00123@46-116-115-228.bb.netvision.net.il) joined #scummvm. [10:36] omer_mor_ (~Omer@46-117-132-33.bb.netvision.net.il) joined #scummvm. [10:38] omer_mor (~Omer@46-117-132-33.bb.netvision.net.il) left irc: Ping timeout: 244 seconds [11:26] Strangerke_ (~Strangerk@cable-85.28.84.13.coditel.net) joined #scummvm. [11:29] Strangerke (~Strangerk@cable-85.28.84.13.coditel.net) left irc: Ping timeout: 260 seconds [11:29] Nick change: Strangerke_ -> Strangerke [11:59] Harekiet (~harekiet@kittens.harekiet.com) left irc: Ping timeout: 260 seconds [12:00] SylvainTV (~SylvainTV@LFbn-1-8392-241.w81-254.abo.wanadoo.fr) joined #scummvm. [12:00] #scummvm: mode change '+o SylvainTV' by ChanServ!ChanServ@services. [12:06] Harekiet (~harekiet@kittens.harekiet.com) joined #scummvm. [12:31] Littleboy (~littleboy@c-174-62-174-85.hsd1.ma.comcast.net) joined #scummvm. [12:31] #scummvm: mode change '+o Littleboy' by ChanServ!ChanServ@services. [12:40] dreammaster (~dreammast@c-73-167-118-204.hsd1.ma.comcast.net) joined #scummvm. [12:40] #scummvm: mode change '+o dreammaster' by ChanServ!ChanServ@services. [12:48] Guest79952 (~tompsson@h-236-221.a199.priv.bahnhof.se) left irc: [12:50] GitHub158 (~GitHub158@192.30.252.45) joined #scummvm. [12:50] [scummvm] sev- pushed 1 new commit to master: https://git.io/vPiCs [12:50] scummvm/master 9db3a38 Eugene Sandulenko: PSP: Added release building script [12:50] GitHub158 (GitHub158@192.30.252.45) left #scummvm. [12:55] ajax16384 (~User@ip33.net130.n37.ru) left irc: Read error: Connection reset by peer [13:47] omer_mor (~Omer@46-117-132-33.bb.netvision.net.il) joined #scummvm. [13:49] omer_mor_ (~Omer@46-117-132-33.bb.netvision.net.il) left irc: Ping timeout: 265 seconds [13:58] wow, SDL2.0.4 is really messed up on Windows. [13:59] Fullscreen ScummVM, "Tab" to switch to another application, "Tab" again to switch back to ScummVM -> segfault in SDL2.dll. [14:00] "Tab" means Alt+Tab of course. [14:16] omer_mor_ (~Omer@46-117-132-33.bb.netvision.net.il) joined #scummvm. [14:18] omer_mor (~Omer@46-117-132-33.bb.netvision.net.il) left irc: Ping timeout: 260 seconds [14:27] rootfather (~rootfathe@unaffiliated/rootfather) left irc: [14:39] heroux (sandroco@gateway/shell/insomnia247/x-fdndxzhuntfplcxk) left irc: Ping timeout: 245 seconds [14:39] heroux (sandroco@gateway/shell/insomnia247/x-xmlrzavplqpskggg) joined #scummvm. [14:50] snover (~Adium@unaffiliated/snover) joined #scummvm. [14:50] #scummvm: mode change '+o snover' by ChanServ!ChanServ@services. [14:57] rootfather (~rootfathe@unaffiliated/rootfather) joined #scummvm. [14:57] #scummvm: mode change '+o rootfather' by ChanServ!ChanServ@services. [14:58] I tried the pre-release version of SDL 2.0.5, and thankfully, all known Windows issues are fixed. [14:58] But I don't think SDL 2.0.5 will be released on time for ScummVM 1.9.0 sadly. [15:15] LittleToonCat (~littlecat@sydnns0115w-047054148237.dhcp-dynamic.FibreOP.ns.bellaliant.net) joined #scummvm. [15:31] if it's that much of an improvement, ship it with the pre-release regardless? [15:35] or 2.0.3?! [16:05] snow_bckspc (~snow_bcks@ganon.dot-server.net) left irc: Ping timeout: 250 seconds [16:07] snow_bckspc (~snow_bcks@ganon.dot-server.net) joined #scummvm. [16:12] GitHub62 (~GitHub62@192.30.252.46) joined #scummvm. [16:12] [scummvm] csnover pushed 1 new commit to master: https://git.io/vPigl [16:12] scummvm/master abc4656 Colin Snover: SCI32: Fix workaround comment [16:12] GitHub62 (GitHub62@192.30.252.46) left #scummvm. [16:15] dreammaster (~dreammast@c-73-167-118-204.hsd1.ma.comcast.net) left irc: Ping timeout: 250 seconds [16:15] Lightkey @m_kiewitz I think both the pre-release and 2.0.3 are better options than 2.0.4... [16:48] m_kiewitz (~m_kiewitz@scummvm/undead/m-kiewitz) left irc: Ping timeout: 260 seconds [16:56] GitHub12 (~GitHub12@192.30.252.40) joined #scummvm. [16:56] [scummvm] rootfather opened pull request #849: CHEWY: Add detection entries for German versions (master...chewy-de-detection) https://git.io/vPiVc [16:56] GitHub12 (GitHub12@192.30.252.40) left #scummvm. [17:03] snover, I found a scummvm freeze in SQ6, it's near the ending when the blur filter gets reapplied to make it seem like the camera switches focus [17:03] snover, was that the thing you talked about yesterday? If not I have a save and I can make a bugreport [17:03] hm. i never had a problem with that during my playthrough [17:03] I'll make a bugreport then [17:04] ok [17:04] i am still chasing this weirdo rendering bug, somehow it was introduced when the palette manager for sci32 was separated from the sci16 one. so weird. [17:06] snover, https://bugs.scummvm.org/ticket/9614 [17:07] snover, I also saw some evidence of that priority bug in sq6, but only in 2 different scenes [17:13] snover, I'll try to get a gdb bt of the crash, maybe I can debug this myself [17:13] well, it's a freeze so, hopefully sigquit will still give me something useful [17:17] hum... maybe this is actually a problem in the GL renderer [17:17] http://paste.debian.net/880958/ trying again with the normal one [17:20] i can reproduce it with sdl2 [17:20] very odd [17:21] it is not actually stuck forever [17:21] snover, it really does get stuck forever with the OpenGL renderer [17:21] snover, just set the renderer to HQ2x and it doesn't happen [17:22] it is such a crappy effect too [17:22] I set the priority to 'normal' now and added the bt [17:29] Henke37 (~Henrik@81-227-16-59-no133.bredband.skanova.com) joined #scummvm. [17:35] m_kiewitz (~m_kiewitz@scummvm/undead/m-kiewitz) joined #scummvm. [17:35] #scummvm: mode change '+o m_kiewitz' by ChanServ!ChanServ@services. [17:35] what priority is that? [17:39] frankyboy_ (~franky@ppp46-138-41-38.pppoe.spdop.ru) joined #scummvm. [17:40] snover, oh, I thought for the ticket, sorry :) [17:41] I figured 'high' would be appropriate if it breaks a game, but if it's only a problem with the gl renderer it shouldn't be? I don't know... :P [17:42] i dont think it is a problem with just that renderer, as the transition is taking so long to complete without yielding that the window becomes unresponsive [17:42] GitHub83 (~GitHub83@192.30.252.45) joined #scummvm. [17:42] [scummvm] criezy closed pull request #847: Add graphics linear filtering feature (master...sdl-filtering) https://git.io/vPag4 [17:42] GitHub83 (GitHub83@192.30.252.45) left #scummvm. [17:42] GitHub45 (~GitHub45@192.30.252.41) joined #scummvm. [17:42] [scummvm] criezy pushed 11 new commits to master: https://git.io/vPioq [17:42] scummvm/master 30aae51 Thierry Crozat: OSYSTEM: Add kFeatureFilteringMode [17:42] scummvm/master 3e08c33 Thierry Crozat: GUI: Add checkbox and config option to enable/disable graphics filtering [17:42] scummvm/master c6ce1c8 Thierry Crozat: OPENGL: Add support for filtering feature... [17:42] GitHub45 (GitHub45@192.30.252.41) left #scummvm. [17:44] snover, you think? [17:44] I guess maybe [17:45] it's kind of weird it doesn't seem to be able to complete at all with gl though [17:45] but maybe I was just being impatient [17:50] borosky (~broosky@178235040163.dynamic-ww-05.vectranet.pl) left irc: Ping timeout: 250 seconds [17:51] GitHub127 (~GitHub127@192.30.252.42) joined #scummvm. [17:51] [scummvm] Strangerke closed pull request #849: CHEWY: Add detection entries for German versions (master...chewy-de-detection) https://git.io/vPiVc [17:51] GitHub127 (GitHub127@192.30.252.42) left #scummvm. [17:51] GitHub146 (~GitHub146@192.30.252.42) joined #scummvm. [17:51] [scummvm] Strangerke pushed 2 new commits to master: https://git.io/vPioV [17:51] scummvm/master 2abdea7 rootfather: CHEWY: Add detection entries for German versions [17:51] scummvm/master 7bd5ca5 Arnaud Boutonné: Merge pull request #849 from rootfather/chewy-de-detection... [17:51] GitHub146 (GitHub146@192.30.252.42) left #scummvm. [17:57] girafe (~girafe@LFbn-1-8015-136.w90-112.abo.wanadoo.fr) joined #scummvm. [18:09] kurtwr (~kurtwr@c-73-12-209-100.hsd1.ca.comcast.net) left irc: [18:14] kurtwr (~kurtwr@c-73-12-209-100.hsd1.ca.comcast.net) joined #scummvm. [18:15] cpasjuste_ (~cpasjuste@LFbn-1-5075-211.w90-104.abo.wanadoo.fr) joined #scummvm. [18:16] borosky (~broosky@178235040163.dynamic-ww-05.vectranet.pl) joined #scummvm. [18:21] Nick change: omer_mor_ -> omer_mor [18:21] hi [18:22] does anyone here know who's in charge of the scummvm demo server? [18:24] dreammaster (~dreammast@c-73-167-118-204.hsd1.ma.comcast.net) joined #scummvm. [18:24] #scummvm: mode change '+o dreammaster' by ChanServ!ChanServ@services. [18:28] frankyboy_ (~franky@ppp46-138-41-38.pppoe.spdop.ru) left irc: Quit: #E>6C O >B 20A [18:39] rootfather (~rootfathe@unaffiliated/rootfather) left irc: Ping timeout: 252 seconds [18:48] <_sev> omer_mor: what is 'scummvm demo server'? [18:49] _sev: https://www.scummvm.org/frs/demos/ [18:49] <_sev> okay, that is FRS, no idea why you called it a 'demo server' [18:49] <_sev> what's the problem? [18:50] demo hosting server :) [18:50] I asked for the valid police quest 2 demo to be added: http://forums.scummvm.org/viewtopic.php?p=83298#83298 [18:50] Anyone remind me how to how to suppress the console window when building in Visual Studio? [18:51] But apparently it was missed [18:51] <_sev> omer_mor: great. What is the proposed file description? [18:51] <_sev> omer_mor: and forums are hard to track, that's why it got missed. I'll add it now [18:51] It should replace https://www.scummvm.org/frs/demos/sci/pq2-dos-ni-demo-en-corrupt.zip [18:51] without the 'corrupt' suffix of course :) [18:51] <_sev> aha [18:52] Police Quest II: The Vengeance (DOS demo - Non-interactive) [18:53] Thanks _sev [18:53] what's FRS btw? [18:53] a dev? [18:56] <_sev> File Release System [18:57] GitHub14 (~GitHub14@192.30.252.45) joined #scummvm. [18:57] [scummvm-web] sev- pushed 1 new commit to master: https://git.io/vPiiM [18:57] scummvm-web/master 2693235 Eugene Sandulenko: WEB: PQ2 demo was provided [18:57] GitHub14 (GitHub14@192.30.252.45) left #scummvm. [18:57] <_sev> omer_mor: it's up [18:57] great - thanks! [18:57] <_sev> omer_mor: thanks a lot [18:57] I spent years looking for the demo... [18:58] Seems like the default suggestion, changing to SUBSYSTEM:WINDOWS, and the Entry Point to mainCRTStartup gives a linker error. And digging up old ScummVM Threads, an older historic version of the Compiling ScummVM instructions only says how to show the console, not how to hide it :P [18:59] <_sev> dreammaster: let me try to dig it too [18:59] I'm having two problems with master at the moment. 1) if I enable cloud support, I get a runtime PNG decoding error (I don't have libpng installed), and 2) if I disable cloud support, I get a linker error in gui/options.cpp [19:01] Thanks [19:02] <_sev> waltervn: what is the linker error in 2)? [19:03] sorry, not linker error, compile error [19:04] the first error is [19:04] ../gui/options.cpp: In member function 'virtual void GUI::GlobalOptionsDialog::close()': [19:04] ../gui/options.cpp:1540:37: error: '_selectedStorageIndex' was not declared in this scope [19:16] it looks like in options.h this is guarded by USE_CLOUD, and in options.cpp by USE_LIBCURL [19:16] Strangerke_ (~Strangerk@cable-85.28.84.13.coditel.net) joined #scummvm. [19:19] Strangerke (~Strangerk@cable-85.28.84.13.coditel.net) left irc: Ping timeout: 244 seconds [19:19] Nick change: Strangerke_ -> Strangerke [19:24] rootfather (~rootfathe@p57BBA988.dip0.t-ipconnect.de) joined #scummvm. [19:24] rootfather (~rootfathe@p57BBA988.dip0.t-ipconnect.de) left irc: Changing host [19:24] rootfather (~rootfathe@unaffiliated/rootfather) joined #scummvm. [19:24] #scummvm: mode change '+o rootfather' by ChanServ!ChanServ@services. [19:24] snover, I was able to complete SQ6 with the hq2x renderer just fine, just found one crash. I made a ticket for it. [19:39] #scummvm: mode change '+o Strangerke' by ChanServ!ChanServ@services. [19:53] ced117 (~ced117@opensuse/member/ced117) left irc: Quit: Quitte [20:02] Seems I figured out the problem. In OSystem_Win32, we manually have a config file option for the console, which is true by default. I was going frantic wondering why the console always opened, no matter what I did in the project files. It was only when I started tracing executing in Debug mode I realised that the application itself, not the runtime, was opening the console window :P [20:03] TMM: that is weird and concerning. [20:03] snover, the crash? or that it failed with the opengl backend? [20:03] the opengl/transition thing [20:03] yeah [20:05] snover, let me try again, it appears that the option I was using OpenGL (no filtering) no longer exists [20:07] there is a problem with that transition regardless because it screws up for a while here [20:07] anyone wants English Duckman? [20:08] GitHub72 (~GitHub72@192.30.252.42) joined #scummvm. [20:08] [scummvm] sev- pushed 1 new commit to branch-1-9: https://git.io/vPiDZ [20:08] scummvm/branch-1-9 cfe1bf4 Eugene Sandulenko: BUILD: This is 1.9.1pre [20:08] GitHub72 (GitHub72@192.30.252.42) left #scummvm. [20:08] snover, no ,it still happens [20:09] snover, no, it DOES eventually complete. it just takes a very ,very long time on opengl renderer [20:09] takes 2 or 3 minutes [20:09] you were right [20:09] snover: I guess you need to implement frame skipping? [20:10] m_kiewitz: maybe. also probably need to flush the event queue. [20:10] s/flush/pump [20:10] we had a similar problem with sci16 transitions [20:10] i had to implement dynamic frame skipping, but also frame delays as well [20:10] I wonder if it's because it tries to apply the effect in ever smaller rects [20:11] that would be a complete texture download/upload each time on gl, right? [20:16] snover, I've created one of those 'workarounds' for https://bugs.scummvm.org/ticket/9615 it seems to work. Should I just attach a patch to the ticket? [20:17] const SciWorkaroundEntry kNumCels_workarounds[] [20:17] one of those [20:28] waltervn (~waltervn@82-171-142-149.ip.telfort.nl) left irc: Quit: Leaving [20:30] GitHub2 (~GitHub2@192.30.252.41) joined #scummvm. [20:30] [scummvm] hpvb opened pull request #850: Add workaround for ego::lastCel. Fixes #9615 (master...fix-sq6-9615) https://git.io/vPiyy [20:30] GitHub2 (GitHub2@192.30.252.41) left #scummvm. [20:30] I made a PR [20:30] lol, as the github bot just said, sorry [20:30] Action: TMM slinks away [21:06] one small meta-comment: we add the engine or component as prefix to the commit messages; so "SCI32: Add workaround for ego::lastCel" in this case [21:06] oh, sorry, I can amend and force push? [21:07] yeah, that's fine [21:07] (github picks up on that) [21:07] wjp, game too? [21:07] SCI32: SQ6? [21:07] Just engine normally [21:07] ok [21:07] but adding SQ6 makes sense [21:08] "SCI32: Add workaround for SQ6 ego::lastCel" or so [21:08] The major point being that grepping the changelog is easy-ish [21:08] wjp, ok! [21:08] done [21:09] timofonic2 (~timofonic@129.125.20.95.dynamic.jazztel.es) joined #scummvm. [21:09] Idk how you treat sub-engines in ScummVM, presumably SCI32 is a sub-engine of SCI? [21:09] we talked about it and decided to go with SCI32: as prefix [21:09] How would HE-SCUMM games be prefixed? [21:09] I think it's the only case of this [21:09] up to SCUMM devs [21:10] in the SCI32 case there's just a difference between stable-SCI and unstable-SCI32 [21:10] I've only dealt with this in ResidualVM, where we have a mixture of GRIM:, EMI: and GRIM/EMI: depending on whether its shared code, or v1/v2 specific code [21:11] wjp, SQ6 is completeable btw :) some minor issues, but it is working. [21:11] great! [21:11] lsl6hires too [21:11] (thanks to snover :)) [21:12] he was fixing bugs as I was finding them, pretty cool [21:26] GitHub141 (~GitHub141@192.30.252.41) joined #scummvm. [21:26] [scummvm] sev- pushed 1 new commit to master: https://git.io/vPiHH [21:26] scummvm/master 5eb1da4 Eugene Sandulenko: FULLPIPE: Fix steps sounds [21:26] GitHub141 (GitHub141@192.30.252.41) left #scummvm. [21:27] wjp: can you see anything offhand in 4cfc387 that would cause drawing of cels that should be hidden by a higher plane? i mean, i *guess* this commit must have unmasked some other bad behaviour in the renderer since i have no idea how isolating palette32 would cause views to be drawn above higher-priority planes [21:28] ny00123 (~ny00123@46-116-115-228.bb.netvision.net.il) left irc: Quit: Leaving [21:28] (a really easy way to see this is to just start pq4 and then click on the dumpster) [21:41] dreammaster (~dreammast@c-73-167-118-204.hsd1.ma.comcast.net) left irc: [21:41] NuSuey (uid3556@gateway/web/irccloud.com/x-cmuhugxusqdlilhy) joined #scummvm. [21:51] WooShell (~Markus@ipbcc071f7.dynamic.kabel-deutschland.de) left irc: Quit: It's easy to laugh and it's easy to hate... it takes guts to be gentle and kind. [21:54] m_kiewitz (~m_kiewitz@scummvm/undead/m-kiewitz) left irc: Quit: technology isn't intrinsically good or evil. It's how it's used. Like the Death Ray. [21:57] snover: did bisecting point at that? [21:58] was just going to bed, but can have a look later [22:02] dreammaster (~dreammast@c-73-167-118-204.hsd1.ma.comcast.net) joined #scummvm. [22:02] #scummvm: mode change '+o dreammaster' by ChanServ!ChanServ@services. [22:08] ScottT (~ScottT@2401:a400:6100:f400:10fb:23cd:9d70:da80) joined #scummvm. [22:08] #scummvm: mode change '+o ScottT' by ChanServ!ChanServ@services. [22:09] ScottT1 (~ScottT@2401:a400:6100:f400:10fb:23cd:9d70:da80) left irc: Ping timeout: 245 seconds [22:21] Gentle (~tier@quassel/contributors/gentle) left irc: Ping timeout: 258 seconds [22:24] GitHub4 (~GitHub4@192.30.252.46) joined #scummvm. [22:24] [scummvm] sev- pushed 1 new commit to master: https://git.io/vPi52 [22:24] scummvm/master b9a7775 Eugene Sandulenko: DISTS: Add Win32 signature [22:24] GitHub4 (GitHub4@192.30.252.46) left #scummvm. [22:25] snover: I still get those artifacts at the commit before 4cfc387, by the way [22:26] Gentle (~tier@quassel/contributors/gentle) joined #scummvm. [22:29] GitHub40 (~GitHub40@192.30.252.45) joined #scummvm. [22:29] [scummvm] criezy pushed 2 new commits to master: https://git.io/vPidk [22:29] scummvm/master 873515a Thierry Crozat: GUI: Fix compilation when cloud is disabled but libcurl is not [22:29] scummvm/master a43692f Thierry Crozat: CONFIGURE: Disable libcurl and sdlnet when using --disable-cloud [22:29] GitHub40 (GitHub40@192.30.252.45) left #scummvm. [22:30] <_sev> criezy: interesting. and is it OK when sdl_net is enabled and cloud is not? [22:31] <_sev> criezy: and a43692f is an overkill. Some other features could use sdl_net [22:31] If you use both --disable-cloud and --enable-sdlnet, sdl_net will be disabled. [22:31] That is a temporary measure because currently the define is a real mess [22:32] <_sev> ok [22:32] There are compilation errors in other files than gui/options.cpp [22:32] <_sev> I see. I can also take a look into cleaning it up if you need help [22:33] Help is obviously always welcome ;) [22:33] <_sev> ok, I'll try to take a look tomorrow [22:33] I do think it makes more sense to disable features than libraries [22:34] wjp: yeah i bisected [22:37] 4cfc387 https://zetafleet.com/i/580400e94309e.png 7f53a26 https://zetafleet.com/i/5804010f451d5.png [22:38] 4cfc387^ looks exactly the same as that first image for me [22:39] well thats&interesting [22:39] slash frightening [22:39] do we still have sorting corner cases? [22:40] oh, valgrind complains [22:40] let's see [22:41] man. i do like ASan but it makes me overly confident that there are no memory problems when it doesnt do the same level of checks as valgrind. [22:42] oh, false alarm. Valgrind warning in that screen happened on old commit, but not on master [22:43] from what i could see looking at the plane lists it seems like plane priority isnt being respected when deciding which screen items to draw; the close-up is its own picture-type plane at higher priority than the main game plane [22:44] interesting [22:44] im compiling in linux now to see if i can reproduce that it is broken on 4cfc387^ [22:44] the valgrind warning was an uninitialized pictureId in Plane:setType [22:44] just in case it rings a bell [22:48] who called setType? [22:48] ==30684== at 0x4FA225: Sci::Plane::setType() (plane32.cpp:755) [22:48] ==30684== by 0x4F7F8C: Sci::Plane::Plane(Sci::reg_t) (plane32.cpp:86) [22:48] ==30684== by 0x4F093F: Sci::GfxFrameout::kernelAddPlane(Sci::reg_t) (frameout.cpp:378) [22:48] ==30684== by 0x4DCE1F: Sci::kAddPlane(Sci::EngineState*, int, Sci::reg_t*) (kgraphics32.cpp:94) [22:48] I _think_ these line numbers were from back in 7edc5cd891d903e8ae114238745a965c319d7fa3 [22:52] wjp: well, i can verify your results in linux. curious. [22:53] im not even going to attempt speculation on what bit of non-determinism is causing it to work properly for me on 4cfc387^ [22:56] looks like gfxcursor32 is doing some bad stuff & [22:56] not related. [22:56] I'm going to try sleep again; 'night :-) [22:56] ok. sleep well! [23:14] Henke37 (~Henrik@81-227-16-59-no133.bredband.skanova.com) left irc: Read error: Connection reset by peer [23:14] girafe (~girafe@LFbn-1-8015-136.w90-112.abo.wanadoo.fr) left irc: Read error: Connection reset by peer [23:15] Henke37 (~Henrik@81.227.16.59) joined #scummvm. [23:33] Henke37 (~Henrik@81.227.16.59) left irc: Quit: ERR_SHUTDOWN [23:56] SylvainTV (~SylvainTV@LFbn-1-8392-241.w81-254.abo.wanadoo.fr) left irc: Read error: Connection reset by peer [00:00] --- Mon Oct 17 2016