ScummVM IRC Log - 12 Nov 2017

[Back to Index]

[00:02] --> Lightkey joined #scummvm.
[00:29] --> Strangerke joined #scummvm.
[00:47] #scummvm: mode change '+o Strangerke' by ChanServ!ChanServ@services.
[01:06] <-- girafe left irc: Read error: Connection reset by peer
[01:09] --> TMM joined #scummvm.
[01:09] <-- TMM left irc: Changing host
[01:09] --> TMM joined #scummvm.
[01:09] #scummvm: mode change '+o TMM' by ChanServ!ChanServ@services.
[01:10] <snover> TMM: good evening!
[01:25] <dreammaster> Commit bot seems to have died again..
[01:29] --> Strangerke_ joined #scummvm.
[01:29] <-- Strangerke left irc: Ping timeout: 268 seconds
[01:29] Nick change: Strangerke_ -> Strangerke
[01:31] <snover> its understandable, github is a ragtag startup with few staff to keep things running
[01:33] <dreammaster> Well, luckily it's not a major issue, so long as the underlying commits & pulls still work
[01:39] <ScummBot> Port build status changed with 4a76e6fb: Failure: master-amigaos4
[01:45] <-- TMM left irc: Quit: Ex-Chat
[01:47] --> TMM joined #scummvm.
[01:47] <-- TMM left irc: Changing host
[01:47] --> TMM joined #scummvm.
[01:47] #scummvm: mode change '+o TMM' by ChanServ!ChanServ@services.
[02:03] <-- waltervn left irc: Quit: Leaving
[02:07] <snover> dreammaster: did you see the amigaos failure?
[02:08] <dreammaster> Yes, I've got a fix pending, but I figured I'd commit in the remainder of the giveTake method at the same time
[02:08] <snover> no worries. just checking due to some& unchecked ocd
[02:16] --> GitHub38 joined #scummvm.
[02:16] <GitHub38> [scummvm] csnover opened pull request #1053: DREAMWEB: Use accurate memory reclamation for Ex transfers (master...dreamweb-mem) https://git.io/vFwgm
[02:16] GitHub38 (GitHub38@192.30.252.34) left #scummvm.
[02:17] <-- ccawley2011 left irc: Quit: Page closed
[02:24] --> GitHub119 joined #scummvm.
[02:24] <GitHub119> [scummvm] csnover pushed 1 new commit to master: https://git.io/vFwgE
[02:24] <GitHub119> scummvm/master e6228f3 Colin Snover: SCI32: Fix integer overflow calculating audio duration...
[02:24] GitHub119 (GitHub119@192.30.252.45) left #scummvm.
[02:24] <snover> oh hi githubbot.
[02:27] <Deledrius> GithubBot takes long coffee breaks.
[02:30] --> GitHub16 joined #scummvm.
[02:30] <GitHub16> [scummvm] dreammaster pushed 1 new commit to master: https://git.io/vFw2T
[02:30] <GitHub16> scummvm/master c068dd8 Paul Gilbert: XEEN: Added more giveTake
[02:30] GitHub16 (GitHub16@192.30.252.35) left #scummvm.
[02:44] --> DJW|Home joined #scummvm.
[02:44] #scummvm: mode change '+o DJW|Home' by ChanServ!ChanServ@services.
[02:44] <-- dreammaster left irc:
[02:48] <-- DJWillis left irc: Ping timeout: 258 seconds
[02:48] --> GitHub99 joined #scummvm.
[02:48] <GitHub99> [scummvm] csnover pushed 1 new commit to master: https://git.io/vFw27
[02:48] <GitHub99> scummvm/master 0d8afad Colin Snover: GRAPHICS: Fix buffer overflow in drawBevelSquareAlg...
[02:48] GitHub99 (GitHub99@192.30.252.42) left #scummvm.
[02:59] <-- Dominus left irc: Ping timeout: 260 seconds
[03:00] --> Dominus joined #scummvm.
[03:04] --> Begasus joined #scummvm.
[03:06] --> Begas_VBox joined #scummvm.
[03:24] --> GitHub90 joined #scummvm.
[03:24] <GitHub90> [scummvm] csnover opened pull request #1054: SCUMM: Fix race condition in MOD player (master...scumm-mod-mutex) https://git.io/vFwaS
[03:24] GitHub90 (GitHub90@192.30.252.41) left #scummvm.
[03:25] <snover> weirdly, scummvm is a lot less crashy in xcode 9.1 with sdl 2.0.7
[03:26] <snover> almost like 2.0.6 has some bugs in it&
[03:44] <-- SylvainTV left irc: Read error: Connection reset by peer
[04:36] --> nast joined #scummvm.
[04:41] <Lightkey> https://www.humblebundle.com/day-of-the-devs-2017 last day to get literally Everything
[05:20] --> koyaanisqatsi` joined #scummvm.
[05:20] <koyaanisqatsi`> is broken sword director's cut supported in some way?
[05:25] <Lightkey> only in the sense that the original is included if you buy it on GOG.com
[05:26] <-- ignalina left irc: Quit: ignalina
[05:28] <koyaanisqatsi`> so director's cut is a whole new engine?
[05:29] <Lightkey> jupp, ask joostp all about it
[05:30] <koyaanisqatsi`> that's a shame!
[05:43] <-- nast left irc: Quit: Leaving
[05:44] <koyaanisqatsi`> and it's windows only
[05:45] <Lightkey> no it isn't? joostp did the Linux port
[05:47] <koyaanisqatsi`> i thought just the regular version was linux
[05:47] <koyaanisqatsi`> linux often gets screwed that way on gog
[06:22] <-- Begas_VBox left irc: Quit: Vision[0.10.2]: i've been blurred!
[06:31] --> Begas_VBox joined #scummvm.
[07:07] <-- koyaanisqatsi` left irc: Quit: 🐕💩
[07:54] --> Henke37 joined #scummvm.
[08:14] --> jamm joined #scummvm.
[08:40] <-- _sev left irc: Quit: This computer has gone to sleep
[08:42] --> _sev joined #scummvm.
[08:42] #scummvm: mode change '+o _sev' by ChanServ!ChanServ@services.
[08:55] <rootfather> I wonder if there is a plugin for Trac that would post chagnes to the issue tracker to IRC
[08:55] <rootfather> pretty much like the GitHub bot
[09:00] <-- _sev left irc: Quit: This computer has gone to sleep
[09:02] --> _sev joined #scummvm.
[09:02] #scummvm: mode change '+o _sev' by ChanServ!ChanServ@services.
[09:03] <-- _sev left irc: Client Quit
[09:04] --> _sev joined #scummvm.
[09:04] #scummvm: mode change '+o _sev' by ChanServ!ChanServ@services.
[09:08] <-- LittleToonCat left irc: Remote host closed the connection
[09:23] --> frankyboy_ joined #scummvm.
[09:36] --> ajax16384 joined #scummvm.
[09:36] #scummvm: mode change '+o ajax16384' by ChanServ!ChanServ@services.
[09:37] --> GitHub97 joined #scummvm.
[09:37] <GitHub97> [scummvm] bgK pushed 1 new commit to master: https://git.io/vFw1R
[09:37] <GitHub97> scummvm/master ce3d5eb Bastien Bouclet: GRAPHICS: Fix buffer overflow in drawBevelSquareAlgClip...
[09:37] GitHub97 (GitHub97@192.30.252.35) left #scummvm.
[09:43] <-- kurtwr left irc:
[09:46] <-- antlarr2 left irc: Ping timeout: 268 seconds
[09:47] --> kurtwr joined #scummvm.
[10:23] --> waltervn joined #scummvm.
[10:23] #scummvm: mode change '+o waltervn' by ChanServ!ChanServ@services.
[10:24] <waltervn> morning
[10:33] <bgK> BTW, the VC++ compiler is available as a conventient zip file on nuget now, I was wondering about using it to create a buildbot worker running it through wine
[10:33] <bgK> nevermind
[11:04] --> dreammaster joined #scummvm.
[11:04] #scummvm: mode change '+o dreammaster' by ChanServ!ChanServ@services.
[11:04] --> GitHub79 joined #scummvm.
[11:04] <GitHub79> [scummvm] dreammaster pushed 1 new commit to master: https://git.io/vFwSK
[11:04] <GitHub79> scummvm/master da87097 Paul Gilbert: XEEN: Fix compiler warnings & AmigaOS errors
[11:04] GitHub79 (GitHub79@192.30.252.42) left #scummvm.
[11:25] <ScummBot> Port build status changed with da870978: Success: master-amigaos4
[11:27] --> antlarr joined #scummvm.
[11:43] <-- rootfather left irc:
[11:49] --> rootfather joined #scummvm.
[11:49] <-- rootfather left irc: Changing host
[11:49] --> rootfather joined #scummvm.
[11:49] #scummvm: mode change '+o rootfather' by ChanServ!ChanServ@services.
[12:04] <-- TMM left irc: Quit: Ex-Chat
[12:10] <-- jamm left irc: Ping timeout: 240 seconds
[12:14] <-- Strangerke left irc: Ping timeout: 240 seconds
[12:21] <-- dreammaster left irc:
[12:23] --> Strangerke joined #scummvm.
[12:40] --> mortalius_ joined #scummvm.
[12:40] <-- mortalius left irc: Ping timeout: 240 seconds
[12:47] --> SylvainTV joined #scummvm.
[12:47] #scummvm: mode change '+o SylvainTV' by ChanServ!ChanServ@services.
[12:50] --> criezy joined #scummvm.
[12:50] #scummvm: mode change '+o criezy' by ChanServ!ChanServ@services.
[13:17] --> jamm joined #scummvm.
[13:31] <logix> you *just* missed him
[13:31] <rootfather> yeah I see
[13:32] <rootfather> I wonder if he is even still involved with scummvm...
[13:36] <logix> he didn't fall entirely off the face of the planet 1.5 years ago, he made a small commit in august: https://github.com/scummvm/scummvm/commit/2fe893
[13:40] <criezy> And he has been commenting on windows-related pull request on GitHub from time to time as well.
[13:40] <rootfather> that's good to know
[13:53] --> vasc joined #scummvm.
[13:54] <vasc> hey folks. just as a curiosity which tools do you guys use/recommend for analyzing game data files? i've had enough with 'od' and 'dd' and 'printf "%x"'
[14:03] <logix> hexdump?
[14:05] <vasc> looks similar to od
[14:05] <logix> for a more serious answer - when looking at unknown data files, I also do quite some printf "%x", other than that really hexdump (and xxd), but perhaps python is closer to what you're actually looking for
[14:06] <logix> I usually end up writing some scripts that do something like "read this and that, dump the guess/known fields of a data structure with comments, dump the raw unknown fields" etc.
[14:06] <vasc> hmm. that's more or less what i'm doing right now.
[14:08] <logix> ok, my main real point was that I automate the known stuff in python scripts
[14:08] <vasc> i see
[14:17] <vasc> python makes sense because of the interpreter.
[14:17] <vasc> you can open a file and print it out and go back and forward inside the interpreter
[14:19] <criezy> I have seen people recommend Synalize It!. That's only for mac but there might be similar software on other platforms.
[14:20] <logix> I wouldn't be surprised if there's some tool that at least automates *some* things - like finds common markers at 32 bit aligned addresses, detects small 32 bit numbers (indicating sizes of objects) etc.
[14:21] <logix> I'm not sure if it would help all *that* much because I assume the more it finds the more false positives you'd get
[14:22] <logix> an those are probably more of a hassle than unknown data, at least once you reach a certain not too high threshold
[14:28] <logix> criezy: https://hexinator.com/faq/ - this seems to share code with synalize it
[14:28] <logix> or perhaps "hexinator" is the name they chose for the linux/windows ports, I'm not quite sure
[14:35] <vasc> i use this a lot:
[14:35] <vasc> https://www.scadacore.com/tools/programming-calculators/online-hex-converter/
[14:35] <vasc> oh that looks nice
[14:35] <criezy> logix: Indeed, it seems to mostly be the same application ported to Linux and Windows. With a different name.
[14:51] --> nast joined #scummvm.
[14:52] --> GitHub196 joined #scummvm.
[14:52] <GitHub196> [scummvm-web] rootfather pushed 2 new commits to master: https://git.io/vFwxv
[14:52] <GitHub196> scummvm-web/master b946f08 rootfather: WEB/DE: Add note that Revolution Software gave us the permission to distribute the Sword1/2 cutscene packs
[14:52] <GitHub196> scummvm-web/master f97f1d0 rootfather: WEB/DE: Improve translation for daily builds to match references
[14:52] GitHub196 (GitHub196@192.30.252.42) left #scummvm.
[14:57] <-- rsn8887 left irc: Quit: EliteBNC - http://elitebnc.org (Auto-Removal: idle account/not being used)
[15:18] <logix> vasc: heh, is that SCADA as in stuxnet?
[15:19] <logix> hm, from a quick glance on the wikipedia page SCADA seems to be more like an industry standard and not a specific product/implementation
[15:20] <logix> so ok, not really related (ISTR stuxnet targeting Siemens devices)
[15:35] <-- nast left irc: Quit: Leaving
[15:41] <snover> Synalize It! is pretty great.
[16:08] <rootfather> isn't today the day the release cycle starts?
[16:10] <-- jamm left irc: Ping timeout: 258 seconds
[16:11] <-- _sev left irc: Quit: This computer has gone to sleep
[16:12] <Lightkey> https://twitter.com/AdventureTreff/status/929727233454993408
[16:13] <Lightkey> today is AdventureX
[16:15] --> _sev joined #scummvm.
[16:15] <-- _sev left irc: Changing host
[16:15] --> _sev joined #scummvm.
[16:15] #scummvm: mode change '+o _sev' by ChanServ!ChanServ@services.
[16:16] <-- _sev left irc: Client Quit
[16:18] --> _sev joined #scummvm.
[16:18] #scummvm: mode change '+o _sev' by ChanServ!ChanServ@services.
[16:58] <-- _sev left irc: Quit: This computer has gone to sleep
[16:59] --> _sev joined #scummvm.
[16:59] <-- _sev left irc: Changing host
[16:59] --> _sev joined #scummvm.
[16:59] #scummvm: mode change '+o _sev' by ChanServ!ChanServ@services.
[17:17] --> LittleToonCat joined #scummvm.
[17:23] --> nast joined #scummvm.
[17:32] <-- macdude22 left irc: Quit: Textual IRC Client: www.textualapp.com
[17:35] <-- _sev left irc: Quit: This computer has gone to sleep
[17:36] --> macdude22 joined #scummvm.
[17:37] --> _sev joined #scummvm.
[17:37] <-- _sev left irc: Changing host
[17:37] --> _sev joined #scummvm.
[17:37] #scummvm: mode change '+o _sev' by ChanServ!ChanServ@services.
[17:45] <-- _sev left irc: Quit: This computer has gone to sleep
[17:46] --> _sev joined #scummvm.
[17:46] <-- _sev left irc: Changing host
[17:46] --> _sev joined #scummvm.
[17:46] #scummvm: mode change '+o _sev' by ChanServ!ChanServ@services.
[17:49] <rootfather> snover I'm going to set up a VM with windows 7 to check if the SDL2 image wrapping bug is only visible in Windows 10
[17:49] <snover> rootfather: sounds good!
[17:49] <rootfather> I now have validated it on 3 systems...
[17:50] <snover> three systems with different video cards?
[17:50] <rootfather> one AMD and 2x Intel
[17:50] <rootfather> I don't have a nvidia card :(
[17:50] <rootfather> so technically 2x AMD and 2x Intel if you count the system I first discovered this issue
[17:55] <-- _sev left irc: Quit: This computer has gone to sleep
[17:56] --> _sev joined #scummvm.
[17:56] #scummvm: mode change '+o _sev' by ChanServ!ChanServ@services.
[17:56] <-- rootfather left irc:
[17:59] <-- nast left irc: Quit: Leaving
[17:59] <-- vasc left irc: Remote host closed the connection
[18:07] --> ny00123 joined #scummvm.
[18:08] <criezy> I am wondering if the solution here might be to document that users can set the SDL_RENDER_DRIVER environment variable to specify one, and in particular "opengl" or "direct3d11" which both seem to fix this issue.
[18:12] <-- Strangerke left irc: Ping timeout: 240 seconds
[18:12] <criezy> Although we could specify one in the source code as well since the environment variable would take priority anyway if defined, so that would not prevent users from using a different one.
[18:21] --> Strangerke joined #scummvm.
[18:25] --> rootfather joined #scummvm.
[18:25] <-- rootfather left irc: Changing host
[18:25] --> rootfather joined #scummvm.
[18:25] #scummvm: mode change '+o rootfather' by ChanServ!ChanServ@services.
[18:25] <rootfather> yep, neither an issue with the driver nor the windows version.
[18:25] <rootfather> now to Linux
[18:25] <rootfather> :D
[18:26] <criezy> From your test it looks like this is an issue with the directx render driver in SDL2, since either using directx11 or opengl fixes it.
[18:29] <rootfather> yep
[18:29] <rootfather> just to make sure :P
[18:32] <rootfather> when I initially opened this bugreport I searched for upstream bug reports but couldn't find proper ones
[18:33] <rootfather> and looking at the number of applications and games that use SDL I think such issues would have been surfaced earlier...
[18:33] <rootfather> especially on a huge platform like win32
[18:33] <rootfather> (huge as in "number of installations"
[18:33] <rootfather> )
[18:40] <snover> according to some googling the D3D11 renderer is not supported in Windows 7
[18:40] <snover> maybe this has improved since 2.0.4/2015
[18:41] <snover> i guess it uses some 11.1 APIs that werent backported to Windows 7.
[18:42] <criezy> I am not sure it is supported on any system. Last time I lokked (when rootfather opened that bug report) it was still marked as beta in the source code.
[18:42] <snover> so then what does setting direct3d11 to do? change it to software rendering?
[18:42] <snover> -to
[18:43] <criezy> No, it is present, so it does use it.
[18:43] <criezy> Or when it is not usable, it falls back to the directx renderer, which is the one with the issue.
[18:43] <snover> ok. i see.
[18:44] <criezy> software rendering is the last resort fallback when nothing else is available/
[18:47] <Lightkey> rootfather: I think it's fair to say it's huge in every aspect
[18:53] --> _sev|work joined #scummvm.
[18:53] #scummvm: mode change '+o _sev|work' by ChanServ!ChanServ@services.
[18:54] <rootfather> criezy, do you remember why we haven't decided to use dx11 when I reported the issue?
[18:55] <-- _sev|work left irc: Client Quit
[18:55] --> _sev|work joined #scummvm.
[18:55] <-- _sev|work left irc: Changing host
[18:55] --> _sev|work joined #scummvm.
[18:55] #scummvm: mode change '+o _sev|work' by ChanServ!ChanServ@services.
[18:55] <-- _sev|work left irc: Client Quit
[18:57] <criezy> rootfather: I am not sure we should a renderer ourselves. I would hope that SDL provides by default the best one, although I don't know why the default on Windows is still direct3d. Which is why I wrote 45 minutes ago that the solution might be to document that users can set the SDL_RENDER_DRIVER environment variable to specify one if the default don't work for them.
[19:00] <rootfather> criezy where have you written that? I must have missed that somehow oO
[19:01] <criezy> Here, on IRC.
[19:01] <criezy> At 18:10
[19:02] <criezy> For the log it's actually 18:08. It looks like it does not have exactly the same time as my computer.
[19:02] <rootfather> ah I wasn't online at that time
[19:05] <logix> my timestamp agrees with you, just that I would have specified it in CET, not UTC :)
[19:05] --> digitall joined #scummvm.
[19:05] #scummvm: mode change '+o digitall' by ChanServ!ChanServ@services.
[19:05] <criezy> I specified it in GMT, not UTC :P
[19:06] <logix> then shame on you for not saying so :P
[19:06] <-- ajax16384 left irc: Read error: Connection reset by peer
[19:06] <snover> sorry, i only use swatch time
[19:08] <rootfather> criezy, setting an environment variable doesn't sound like a very good out-of-the-box experience sadly :(
[19:08] <logix> I guess it's around 0.86 beats or so right now then
[19:08] <snover> rootfather: the problem is that one line of pixels at the left thats a little brighter?
[19:09] <rootfather> nope
[19:09] <logix> oh wait, a day is 1000 beats
[19:09] <rootfather> snover, let me try to explain
[19:09] <logix> then around 840 or so I guess
[19:09] <rootfather> when you look at the image, the very first pixel line of the image is "wrapping around" to the very last pixel line
[19:10] <rootfather> and the pixel column on the very right is "duplicated" or wrapped to the very left pixel line
[19:10] <digitall> Evening...
[19:10] <snover> rootfather: ah, ok. i got you now.
[19:10] <rootfather> so as if the whole image is shifted one pixel line to the top and to the right and then wrapping around
[19:10] <digitall> Was just wondering if we can fix a couple of minor issues in code before the release freeze...
[19:11] <rootfather> snover somewhat hard to explain with my english skills
[19:11] <logix> ha, 841, https://www.swatch.com/en_us/internet-time/
[19:11] <Lightkey> snover is hard to explain in any language
[19:11] <digitall> The Sword2.5 engine has an embedded Lua interpreter to deal with it's scripts... It is still throwing about 10 GCC warnings associated with casts.
[19:11] <digitall> engines/sword25/util/lua/ldblib.cpp: In function void hookf(lua_State*, lua_Debug*):
[19:11] <digitall> engines/sword25/util/lua/ldblib.cpp:213:37: warning: cast from type const char* to type void* casts away qualifiers [-Wcast-qual]
[19:12] <digitall> lua_pushlightuserdata(L, (void *)&KEY_HOOK);
[19:12] <rootfather> what the heck are those swatch beat things
[19:12] <digitall> and lines 249 and 254 of that file.
[19:12] <snover> digitall: is that lua code modified from upstream at all?
[19:13] <snover> or i guess maybe another way of asking this question is, are those qual-casts in the original lua code?
[19:13] <digitall> snover: I think so... It has a fair amount of fixes associated with PLUTO which was used for savegames IIRC... the upstream had major portability hell.
[19:14] <snover> ok, then i dont see any issue with fixing the casts to stop dropping const
[19:15] <digitall> Except that I don't think they are const in usage... at least I am not sure without unpicking all this code for the Lua interpreter.
[19:17] <digitall> Hmm... I think the sticking point with the sentinel const is lua_pushlightuserdata(L, sentinel); in loadlib.cpp ... if that is only passing to read.... then it should be fine.
[19:19] <digitall> https://github.com/scummvm/scummvm/blob/master/engines/sword25/util/lua/loadlib.cpp#L161
[19:20] <snover> since lua is actually C this is always going to be a bit of a mess
[19:21] <digitall> LUA_API void *lua_touserdata (lua_State *L, int idx) ..
[19:21] <digitall> Yes, the pushlightuserdata and lua_touserdata return / take void pointers... not const void or otherwise... so a bit of careful casting is needed to cue the compiler not to whinge.
[19:22] <rootfather> hm... the current windows buildbot is still using sdl 1.2.15
[19:22] <snover> rootfather: yes, one of the many reasons why i am working on fixing it :)
[19:22] <snover> i can cobble together a 2.0.x build for you from the new worker if youd like
[19:22] <snover> (i think it will be 2.0.5 since that is what debian comes with)
[19:23] <rootfather> snover no need to do that, I compile everything by myself :P
[19:24] <snover> im glad for kirbens builds, sdl 1.x is a hot mess these days
[19:25] <snover> i mean, it always was, but it has so many problems with modern graphics hardware.
[19:25] <snover> rootfather: https://bugzilla.libsdl.org/show_bug.cgi?id=1878
[19:26] <snover> this is your bug.
[19:27] <snover> its only had a patch for 4.5 years, no big deal :)
[19:27] <-- Begasus left irc: Ping timeout: 240 seconds
[19:28] <rootfather> wow thanks snover
[19:28] <rootfather> so they simply ignored this?!
[19:29] <criezy> They likely forgot about it. We do that sometimes too. :P
[19:29] <rootfather> :D
[19:29] <snover> i thought i had an account on their bug tracker but apparently i dont have a bug on their bug tracker
[19:29] <snover> er
[19:29] <snover> BRAIN. WORDS.
[19:29] <snover> an account*
[19:39] --> nast joined #scummvm.
[19:39] --> Begasus joined #scummvm.
[19:42] <-- nast left irc: Client Quit
[19:45] <-- _sev left irc: Quit: This computer has gone to sleep
[19:45] --> _sev joined #scummvm.
[19:45] <-- _sev left irc: Changing host
[19:45] --> _sev joined #scummvm.
[19:45] #scummvm: mode change '+o _sev' by ChanServ!ChanServ@services.
[19:47] <rootfather> I don't have an account their too
[19:47] <rootfather> snover should I report it or do you want to do that?
[19:48] <snover> well, i seem to have quite a backlog ahead of me of crashes and memory leaks, so i will graciously accept your help here to take ownership and get this addressed upstream :)
[19:50] <snover> https://zetafleet.com/i/5a08a61cb9ed9.png from 30 seconds of running the game&
[20:00] --> nast joined #scummvm.
[20:04] <snover> im also not so sure that a 5-year-old texture clamping issue supersedes 25 crashes :)
[20:05] <rootfather> :P
[20:07] <rootfather> yeah I'll lover the priority now that we know it's an upstream bug
[20:07] <-- _sev left irc: Quit: This computer has gone to sleep
[20:12] <snover> now pick a crash and fix it ;)
[20:15] <rootfather> when you teach me C++ first... :P
[20:15] <snover> so demanding!
[20:18] <digitall> snover: Quite a few of the crashes you have identified are tricky to investigate as they require games or game language variants which are rarer...
[20:18] <digitall> bug 6548 need a French version of Gregory... etc.
[20:19] <digitall> 7085, 6272, 6025 is doubly bad as it need an older OSX machine and Bud Tucker which is a pretty rare game to find.
[20:21] --> dreammaster joined #scummvm.
[20:21] #scummvm: mode change '+o dreammaster' by ChanServ!ChanServ@services.
[20:21] <snover> well, i looked for the word crash and this is what came up. luck of the draw :)
[20:21] <snover> kyrandia one has a patch and is on gog
[20:21] <digitall> True... Just that some are probably real, but hard to investigate as they need a developer to have that game. I do have Tucker (but only the German version).
[20:22] <digitall> Yes, the others I didn't mention are either not dependent on a specific game or the games are relatively easy to find on GOG or EBay... and quite a few of the devs will have copies in their collections
[20:23] <snover> the Neverhood RU one looks like it could be just data corruption on the OPs side
[20:23] <snover> oh, i probably need to pay more attention and scrub these links to illegal downloads from the description&
[20:24] <digitall> Possibly. Talk to _sev about Russian variants. They are often slightly odd in terms of data modification on the legit copies...
[20:25] <snover> in any case i dont really expect all of these to be solved/solvable, and i am actually surprised at how many of these games arent available for purchase
[20:25] <snover> at GOG
[20:25] <digitall> Quite a few localised versions have "damaged" datafiles on the original media i.e. the code work to localise them was not done very well... or was not sanctioned by the original game developers.
[20:26] <digitall> Yes, we do support quite a few obscure games... and I hope that more will get into GOG as it will allow these issues to get fixed.
[20:27] <digitall> Anyway, just explaining that some of the languishing bugs are probably real... but may be a PITA to investigate with getting a copy of the exact game variant together with a developer who has time and motivation... so some do languish.
[20:27] <digitall> "The Long Tail" of bug reports. We do eventually fix them... but some do take quite a long time to do that.
[20:32] <digitall> snover: I really can't see how to fix those warnings on Sword25's lua. ...
[20:36] #scummvm: mode change '+o Strangerke' by ChanServ!ChanServ@services.
[20:36] <digitall> Strangerke: Evening.
[20:36] --> _sev joined #scummvm.
[20:36] #scummvm: mode change '+o _sev' by ChanServ!ChanServ@services.
[20:36] <Strangerke> hi digitall :)
[20:41] <digitall> Ah think I might have a fix... It is not a nice fix, but it does work I think
[20:43] <snover> digitall: if you are not sure then make a PR for feedback
[20:43] <-- frankyboy_ left irc: Remote host closed the connection
[20:44] <snover> im going to be in and out for a bit here.
[20:45] <digitall> snover: I know... though for a 2 line fix, a PR is just massive overkill.
[20:45] <digitall> I think publish and be damned. If someone wants to come up with a better fix to Sword25's Lua, be my guest.
[20:47] <-- _sev left irc: Quit: This computer has gone to sleep
[20:51] --> _sev joined #scummvm.
[20:51] <-- _sev left irc: Changing host
[20:51] --> _sev joined #scummvm.
[20:51] #scummvm: mode change '+o _sev' by ChanServ!ChanServ@services.
[20:52] --> GitHub51 joined #scummvm.
[20:52] <GitHub51> [scummvm] digitall pushed 1 new commit to master: https://git.io/vFrZ1
[20:52] <GitHub51> scummvm/master 5199f9d D G Turner: SWORD25: Suppress GCC Cast Qualification Warnings....
[20:52] GitHub51 (GitHub51@192.30.252.35) left #scummvm.
[20:55] <rootfather> speaking of warnings...
[20:55] <rootfather> in general, would it make sense to open a bug report and collect *all* the GCC warnings I experience during a compilation run?
[20:55] <digitall> If this is about the ones in the prince engine...
[20:56] <rootfather> I haven't checked yet where those warnings appear
[20:56] <digitall> rootfather: Yes, that might be good.
[20:56] <rootfather> one bug report for each engine or "all in one"?
[20:56] <digitall> We try not to do "all in one" bugs... they never get closed.
[20:57] <rootfather> okay
[20:57] <digitall> We can do some now if you get just minor ones. Mostly my build with GCC-5.4.0 is clean apart from issues in prince and those issues in Sword25
[20:58] <rootfather> I'm on GCC 7.x
[20:59] <rootfather> will take some time... current PC is slow as hell
[20:59] <digitall> Ah, I did suspect that might be more strict by default and will have more -Wall turned on by defauly
[21:00] <rootfather> yep there are some warnings due to -Wall-
[21:00] <rootfather> I think in a few version GCC will complain about *everything*
[21:00] <digitall> I meant more that -Wall doesn't actually turn on Warnings All ... despite the name.
[21:00] <digitall> https://gcc.gnu.org/onlinedocs/gcc-7.1.0/gcc/Warning-Options.html
[21:01] <digitall> You need -Wextra and even then there are more... GCC turns more on by default with each version... so these tend to turn up
[21:02] <rootfather> there are already some warnings with the default settings
[21:02] <digitall> Yes, it is mainly that we want to stop to start with :)
[21:02] <rootfather> :D
[21:03] <rootfather> I wonder how important those warnings are in general... indicators for severe issues or more indicators for "bad coding style"?
[21:04] <digitall> A bit of both... but hard to spot the dangerous ones if there is tons of "noise" by default... hence trying to remove the minor ones to see the more worrying ones.
[21:07] <-- nast left irc: Quit: Leaving
[21:07] <snover> digitall: even a one line PR can be worth it if you are not sure if it is the right thing to do.
[21:08] <wjp> Yeah, I agree. Not overkill
[21:09] <digitall> Maybe... and yet PR1043 is still open.
[21:10] <wjp> Ah, scumm variants
[21:10] <snover> committing the wrong thing is worse than having a PR that stays open for a bit.
[21:11] <snover> i dont like that these things remain open for so long either, fwiw.
[21:11] <snover> i dont know enough about all the engines to make an educated review, unfortunately.
[21:14] <wjp> The scumm variant code is quite subtle and error-prone, and not straightforward to test :-(
[21:23] --> rootfather|2nd joined #scummvm.
[21:23] <-- rootfather left irc: Ping timeout: 240 seconds
[21:24] Nick change: rootfather|2nd -> rootfather
[21:24] <-- rootfather left irc: Changing host
[21:24] --> rootfather joined #scummvm.
[21:24] #scummvm: mode change '+o rootfather' by ChanServ!ChanServ@services.
[21:30] <-- rootfather left irc:
[21:32] --> rootfather joined #scummvm.
[21:32] #scummvm: mode change '+o rootfather' by ChanServ!ChanServ@services.
[21:40] <logix> rootfather: https://en.wikipedia.org/wiki/Swatch_Internet_Time (sorry for the delay)
[22:02] <-- Begas_VBox left irc: Quit: Vision[0.10.2]: i've been blurred!
[22:02] --> Farmboy0 joined #scummvm.
[22:03] <-- Begasus left irc: Quit: Ex-Chat
[22:03] <-- Farmboy0 left irc: Changing host
[22:03] --> Farmboy0 joined #scummvm.
[22:03] digitall (digitall@unaffiliated/digitall) left #scummvm ("ttfn").
[22:06] <-- Henke37 left irc: Quit: ERR_SHUTDOWN
[22:12] --> _sev|work joined #scummvm.
[22:12] <-- _sev|work left irc: Changing host
[22:12] --> _sev|work joined #scummvm.
[22:12] #scummvm: mode change '+o _sev|work' by ChanServ!ChanServ@services.
[22:27] <-- _sev|work left irc: Quit: Leaving
[22:29] <-- _sev left irc: Quit: This computer has gone to sleep
[22:32] --> girafe joined #scummvm.
[22:38] --> Strangerke_ joined #scummvm.
[22:40] <-- Strangerke left irc: Ping timeout: 240 seconds
[22:40] Nick change: Strangerke_ -> Strangerke
[22:43] --> _sev joined #scummvm.
[22:43] <-- _sev left irc: Changing host
[22:43] --> _sev joined #scummvm.
[22:43] #scummvm: mode change '+o _sev' by ChanServ!ChanServ@services.
[22:43] --> GitHub113 joined #scummvm.
[22:43] <GitHub113> [scummvm] criezy pushed 1 new commit to master: https://git.io/vFr4J
[22:43] <GitHub113> scummvm/master 086b6e7 Thierry Crozat: I18N: Update translation (French)...
[22:43] GitHub113 (GitHub113@192.30.252.42) left #scummvm.
[22:46] --> TMM joined #scummvm.
[22:46] <-- TMM left irc: Changing host
[22:46] --> TMM joined #scummvm.
[22:46] #scummvm: mode change '+o TMM' by ChanServ!ChanServ@services.
[22:48] --> vasc joined #scummvm.
[22:59] --> ccawley2011 joined #scummvm.
[23:01] <-- waltervn left irc: Quit: Leaving
[23:30] <-- ny00123 left irc: Quit: Leaving
[23:36] <-- vasc left irc: Quit: Leaving
[23:39] <-- TMM left irc: Remote host closed the connection
[23:46] <-- Lightkey left irc: Ping timeout: 240 seconds
[23:47] --> GitHub130 joined #scummvm.
[23:47] <GitHub130> [scummvm] dreammaster pushed 1 new commit to master: https://git.io/vFrRh
[23:47] <GitHub130> scummvm/master 4abcbcb Paul Gilbert: XEEN: Added remainder of giveTake
[23:47] GitHub130 (GitHub130@192.30.252.42) left #scummvm.
[23:55] --> TMM joined #scummvm.
[23:55] #scummvm: mode change '+o TMM' by ChanServ!ChanServ@services.
[00:00] --- Mon Nov 13 2017