ScummVM IRC Log - 04 Dec 2017

[Back to Index]

[00:01] <-- girafe left irc: Read error: Connection reset by peer
[00:10] <-- waltervn left irc: Quit: Leaving
[00:38] <-- ccawley2011 left irc: Quit: Page closed
[01:36] <snover> back now.
[01:39] <-- ignalina left irc: Quit: ignalina
[01:44] --> Vanfanel joined #scummvm.
[01:47] <Vanfanel> snover: I just built scummvm from GIT and I'm still getting wrong cursor movement on the dreamweb engine. Did you give up on fixing it?
[01:53] <snover> Vanfanel: i fixed one terrible bug, then i think i said i had some higher priority things i had to deal with (because& i still do, release and vm2 running out of disk space and whatnot). i am not actually 100% sure what exactly is happening, though i am *pretty* sure it is a rounding/truncation problem, though thats not a very satisfyingly specific answer. i thought it would be fixed by not changing the coordinate of the opposing a
[01:53] <snover> xis and it was still happening after i did that and i didnt have time to continue to investigate.
[01:55] <snover> i am pretty sure the root of the issue is that physical pixels can be interpolations of two virtual pixels because of the non-square PAR (or because of any non-integral scaling, really) and because the warp moves to the top it ends up truncate rounding down to the next highest pixel and thats why it ends up bumping up a row
[01:56] <snover> so maybe moving into the centre of the virtual pixel instead would fix the problem
[01:57] <Vanfanel> snover: I am not in a hurry, don't worry, I just wanted to know if some of the recent pushes were supposed to fix it. Just wanted to report it was still present.
[02:07] <snover> i would like to fix it at some point because it is annoying and happens everywhere.
[02:07] <snover> or, well, everywhere where there is some interpolating going on.
[02:22] <-- Dominus left irc: Ping timeout: 260 seconds
[02:23] --> Dominus joined #scummvm.
[02:32] <-- Vanfanel left irc: Ping timeout: 255 seconds
[02:38] --> travis-ci joined #scummvm.
[02:38] <travis-ci> scummvm/scummvm#4991 (master - a5bc891 : Colin Snover): The build passed.
[02:38] <travis-ci> Change view : https://github.com/scummvm/scummvm/compare/ef33d8a2fb98...a5bc89102e79
[02:38] <travis-ci> Build details : https://travis-ci.org/scummvm/scummvm/builds/311097102
[02:38] travis-ci (travis-ci@ec2-54-166-73-11.compute-1.amazonaws.com) left #scummvm.
[02:54] <-- SylvainTV left irc: Read error: Connection reset by peer
[03:11] <-- dreammaster left irc:
[03:35] --> macdude22 joined #scummvm.
[04:05] --> GitHub48 joined #scummvm.
[04:05] <GitHub48> [scummvm] criezy pushed 1 new commit to master: https://git.io/vbOF9
[04:05] <GitHub48> scummvm/master 71ff55e Thierry Crozat: I18N: Regenerate translations data file
[04:05] GitHub48 (GitHub48@192.30.252.35) left #scummvm.
[04:47] <snover> wjp: did you ever try just using the g++-arm-linux-gnueabihf for rpi?
[04:48] <snover> instead of the custom image with the outdated linaro one
[04:49] <snover> g++-arm-linux-gnueabihf is newish, so it would be unsurprising to me that the old compiler existed because g++-arm-linux-gnueabihf didnt already
[04:50] <snover> and The Internet suggests that g++-arm-linux-gnueabihf will just work.
[05:38] <-- balrog left irc: Ping timeout: 252 seconds
[05:46] --> balrog joined #scummvm.
[05:53] --> Strangerke_ joined #scummvm.
[05:55] <-- Strangerke left irc: Ping timeout: 248 seconds
[05:55] Nick change: Strangerke_ -> Strangerke
[06:22] <-- _marc` left irc: Ping timeout: 255 seconds
[06:34] --> _marc` joined #scummvm.
[07:36] --> rootfather joined #scummvm.
[07:37] <-- rootfather left irc: Changing host
[07:37] --> rootfather joined #scummvm.
[07:37] #scummvm: mode change '+o rootfather' by ChanServ!ChanServ@services.
[08:07] <wjp> snover: no, hadn't tried yet
[08:20] <-- TMM left irc: Quit: Ex-Chat
[08:34] <-- LittleToonCat left irc: Remote host closed the connection
[08:37] --> Strangerke|work joined #scummvm.
[08:37] <Strangerke|work> hi guys
[08:37] <madmoose> hi Strangerke|work
[08:51] --> ignalina joined #scummvm.
[09:40] --> TMM joined #scummvm.
[09:40] <-- TMM left irc: Changing host
[09:40] --> TMM joined #scummvm.
[09:40] #scummvm: mode change '+o TMM' by ChanServ!ChanServ@services.
[09:54] <-- rootfather left irc:
[10:14] <-- user9 left irc: Ping timeout: 246 seconds
[10:46] --> rootfather joined #scummvm.
[10:49] <-- ignalina left irc: Quit: ignalina
[10:56] --> ccawley2011 joined #scummvm.
[11:11] --> ignalina joined #scummvm.
[11:42] --> jam__ joined #scummvm.
[12:01] <-- ccawley2011 left irc: Ping timeout: 260 seconds
[12:14] --> dreammaster joined #scummvm.
[12:14] #scummvm: mode change '+o dreammaster' by ChanServ!ChanServ@services.
[12:20] --> ccawley2011 joined #scummvm.
[12:29] <-- dreammaster left irc:
[12:43] --> user9 joined #scummvm.
[12:47] <-- ST left irc: Ping timeout: 276 seconds
[12:49] --> ST joined #scummvm.
[12:49] #scummvm: mode change '+o ST' by ChanServ!ChanServ@services.
[13:17] <-- Tomaz^ left irc: Ping timeout: 248 seconds
[13:18] <-- ccawley2011 left irc: Ping timeout: 260 seconds
[13:57] --> Tomaz^ joined #scummvm.
[13:59] --> Littleboy joined #scummvm.
[13:59] #scummvm: mode change '+o Littleboy' by ChanServ!ChanServ@services.
[14:10] --> ccawley2011 joined #scummvm.
[14:26] --> dimo joined #scummvm.
[14:28] <dimo> When installing scummvm 1.9.0 from the download page of scummvm (distribution ubuntu 16.04) and run scummvm I see version 1.7.0
[14:29] <dimo> shoulnd't it be 1.9.0 on the menu screen? or did I do something wrong during the installation
[14:30] <logix> my first wild guess is that you have both, 1.7.0 and 1.9.0 installed - did you have 1.7.0 installed before all this?
[14:34] --> waltervn joined #scummvm.
[14:34] #scummvm: mode change '+o waltervn' by ChanServ!ChanServ@services.
[14:39] <dimo> yes, I had it installed via the ubuntu software centre, and removed it also from the ubuntu software centre
[14:41] <-- Littleboy left irc: Read error: Connection reset by peer
[14:43] <logix> I suppose that should have uninstalled it - how are you starting it right now, do you have a menuentry or something or are you starting it from a terminal?
[14:48] <logix> also do these two things: 1) open a terminal and type "scummvm -v", 2) open a terminal and type "dpkg -l scummvm\*"
[14:48] <dimo> I can see the scummvm icon on the menu
[14:48] <logix> (both commands without the quotation marks)
[14:48] <dimo> also from the terminal running scummvm
[14:49] <dimo> ScummVM 1.7.0 (Mar 29 2016 03:04:41)
[14:49] <dimo> thats what I get ....
[14:50] <dimo> Desired=Unknown/Install/Remove/Purge/Hold | Status=Not/Inst/Conf-files/Unpacked/halF-conf/Half-inst/trig-aWait/Trig-pend |/ Err?=(none)/Reinst-required (Status,Err: uppercase=bad) ||/ Name Version Architecture Description +++-==============-============-============-================================= ii scummvm 1.7.0+dfsg-2 amd64 engine for several graphical adve ii scummvm-data 1.7.0+dfsg-2 all
[14:50] <logix> all that sounds as if you don't have scummvm 1.9.0 installed after all
[14:51] <logix> what did you do to install it? did you click on the link, download it, click on the local file in a file manager and get asked if you want to install it or something like that?
[14:51] <dimo> ok so I am going on the binary distributions page, and I download the corresponding distribution for my system which is ubuntu 16.04 64bit
[14:57] <dimo> when I clicked on the distribution binary I chose to open with software install (default)
[14:58] <dimo> then a software centre window opens and under details I can see the version number 1.9.0
[14:59] <dimo> once I press install, the version number changes to 1.7.0 (!)
[15:01] <logix> I don't know ubuntu very well and even less so its graphical tools - what I *GUESS* is happening is that software center opens the package (the file you just downloaded), extracts the name of the package (i.e., the name of the software to be installed), and then goes through its own database to determine how to install that
[15:01] <logix> then it proceeds *not* by installing the contents of the file you just downloaded and opened, but instead by downloading scummvm from its master package repository
[15:02] <logix> as if you had opened software center, searched for "scummvm" there etc.
[15:03] <logix> I suggest you 1) remove scummvm in software center (don't delete the file you just downloaded, keep that!), then open a terminal in the directory containing the file and install it via "sudo dpkg -i scummvm-1.9.0_ubuntu_x86.deb" (you will probably have to adjust that file name, I'm just guessing here)
[15:04] <logix> holy inconsistent enumeration, batman - "open a terminal" is list item "2)" and "install it via ..." is list item "3)" :)
[15:06] <-- waltervn left irc: Quit: Leaving
[15:12] <dimo> thanks a lot!
[15:12] <dimo> from the terminal it works
[15:12] <dimo> now the 1.7.0 has been updated to 1.9.0 ! fially
[15:12] <dimo> finally!
[15:15] <logix> yay, awesome!
[15:15] <logix> I don't know what went wrong there, that sounds like pretty strange behavior
[15:16] <logix> but anyway, you have the updated version, that's all that matters
[15:20] <dimo> true! thanks for helping out, can't wait to play some of the newly supported games
[15:25] <-- dimo left irc: Quit: Page closed
[15:37] <-- jam__ left irc: Ping timeout: 258 seconds
[16:11] <rsn8887> csnover: Whenever you have time, could you check my modified PR for the new Vita-Worker? https://github.com/csnover/scummvm-docker/pull/21 I think it might be good now. Thanks!
[16:12] <rsn8887> csnover: I realized that the necessary changes to get a non-crashing binary were minimal. My first approach was to mimic exactly what I did to install the SDK on the old buildbot. But thanks to your comments, I realize that the Docker approach is vastly different.
[16:13] <snover> rsn8887: yes, i will look at it shortly. have you sent those patches upstream?
[16:14] <rsn8887> csnover: I only updated the PR on csnover/scummvm-buildbot. Where else shall I sent them?
[16:15] <snover> to the repo used by vitasdk, https://github.com/xerpi/SDL-Vita
[16:16] <snover> or if it is some SDL bug, then back to SDL itself
[16:18] <rsn8887> snover: I will try, but that will take a while. Xerpi has been outdated for a long time. Everybody was using Cpasjuste's or my repo, which are now identical....
[16:30] <snover> if it is so, then maybe vitasdk shouldnt be using that repo.
[16:41] <-- Strangerke left irc: Ping timeout: 248 seconds
[16:55] <-- Cheeseness left irc: Ping timeout: 276 seconds
[17:00] <-- ccawley2011 left irc: Ping timeout: 260 seconds
[17:00] --> Strangerke joined #scummvm.
[17:01] --> ccawley2011 joined #scummvm.
[17:02] --> Henke37 joined #scummvm.
[17:06] <-- Strangerke left irc: Ping timeout: 248 seconds
[17:07] --> ajax16384 joined #scummvm.
[17:07] #scummvm: mode change '+o ajax16384' by ChanServ!ChanServ@services.
[17:14] --> mslith12 joined #scummvm.
[17:16] --> Strangerke joined #scummvm.
[17:28] --> girafe joined #scummvm.
[17:33] <-- TMM left irc: Quit: Ex-Chat
[17:44] <-- mslith12 left irc: Quit: Page closed
[17:44] <-- Strangerke left irc: Ping timeout: 250 seconds
[17:44] --> ny00123 joined #scummvm.
[17:44] --> Strangerke joined #scummvm.
[17:45] --> LittleToonCat joined #scummvm.
[18:09] --> waltervn joined #scummvm.
[18:09] #scummvm: mode change '+o waltervn' by ChanServ!ChanServ@services.
[18:11] <-- ccawley2011 left irc: Ping timeout: 260 seconds
[18:14] --> Farmboy0 joined #scummvm.
[18:14] <-- Farmboy0 left irc: Changing host
[18:14] --> Farmboy0 joined #scummvm.
[18:21] --> TMM joined #scummvm.
[18:21] <-- TMM left irc: Changing host
[18:21] --> TMM joined #scummvm.
[18:21] #scummvm: mode change '+o TMM' by ChanServ!ChanServ@services.
[18:30] --> salty-horse joined #scummvm.
[18:30] <-- salty-horse left irc: Changing host
[18:30] --> salty-horse joined #scummvm.
[18:30] #scummvm: mode change '+o salty-horse' by ChanServ!ChanServ@services.
[18:33] --> criezy joined #scummvm.
[18:33] #scummvm: mode change '+o criezy' by ChanServ!ChanServ@services.
[18:34] <criezy> Hi guys
[18:34] <wjp> hello
[18:34] <snover> hi criezy.
[18:36] <eriktorbjorn> Is it just me, or does using Ctrl-Alt-A in windowed mode to toggle aspect ratio correction no longer change the size of the window?

[18:40] <-- girafe left irc: Read error: Connection reset by peer
[18:41] <eriktorbjorn> Ok. Seems a bit odd to me that starting, say, Monkey Island 1 with -g2x --aspect-ratio gives me a 640x480 pixel window, but starting it with -g2x and then pressing Ctrl-Alt-A gives me a 640x400 pixel window, but I assume there's a reason.
[18:43] <-- TMM left irc: Remote host closed the connection
[18:43] <snover> since it is possible to manually resize the window now, the window size only gets reset when engines start/end or the explicit resizing key combos are used. i suppose changing the AR could resize the window in just the y-dimension proportionally, though i am not sure thats any better than keeping the window the same size so it just appears that the x-dimension gets squashed
[18:44] <snover> (previously the window would get reset any time there was any mode change, which was bad for games like dreamweb which changed resolutions, and bad for games in SCI32 which switch pixel modes to render videos with linear interpolation)
[18:45] <eriktorbjorn> I don't know either. It just surprised me.
[18:45] <eriktorbjorn> I thought maybe I had activated the OpenGL renderer by mistake at first. :-)
[18:46] <snover> i guess it is a good sign if you cant tell, since that means we are actually finally achieving some level of feature parity
[18:47] <eriktorbjorn> Don't read too much into it. I never used the OpenGL renderer other than to see that it worked. :-)
[18:47] <snover> can someone tell me the historical reason why there is a null backend and why the buildbot builds the null backend?
[18:50] <logix> snover: wild guess - to allow hacking on systems without any kind of graphics system installed (headless systems, no X11 libs, no sdl, ...)?
[18:50] <snover> logix: what kind of system would that be?
[18:50] <snover> i guess i can speculate as well as anyone and i am looking for the non-speculative answer :)
[18:51] <snover> https://zetafleet.com/i/5a25990a91cfe.png finally!&
[18:52] <logix> snover: I only install that stuff if I need it - usually that means "if it's a desktop system of some kind"
[18:53] <logix> snover: but yes, all pure speculation, I don't know... and that only speculates about the "existence" part of your question, I don't know why buildbot would build that
[18:53] --> SylvainTV joined #scummvm.
[18:53] #scummvm: mode change '+o SylvainTV' by ChanServ!ChanServ@services.
[18:56] --> TMM joined #scummvm.
[18:56] <-- TMM left irc: Changing host
[18:56] --> TMM joined #scummvm.
[18:56] #scummvm: mode change '+o TMM' by ChanServ!ChanServ@services.
[18:57] <logix> snover: for many tools some kind of null-bucket is there to make sure that the tool handles certain input correctly - feed it valid/invalid data and it should either return 0 or an error (or such), but that doesn't make that much sense with scummvm (yet?)
[18:58] <wjp> the null backend dates back to 2002
[19:52] --> girafe joined #scummvm.
[20:10] flesk[m] --> (fleskmatri@gateway/shell/matrix.org/x-uvjljslikvrzrxkh) joined #scummvm.
[20:10] --> Strangerke_ joined #scummvm.
[20:13] <-- Strangerke left irc: Ping timeout: 248 seconds
[20:13] Nick change: Strangerke_ -> Strangerke
[20:21] Nick change: flesk[m] -> flesk_
[20:32] --> Cheeseness joined #scummvm.
[20:57] <-- eriktorbjorn left irc: Ping timeout: 255 seconds
[20:58] <-- ignalina left irc: Quit: ignalina
[20:58] <-- ajax16384 left irc: Read error: Connection reset by peer
[21:06] --> ccawley2011 joined #scummvm.
[21:15] #scummvm: mode change '+o Strangerke' by ChanServ!ChanServ@services.
[21:36] --> ignalina joined #scummvm.
[21:43] Nick change: rootfather -> rootfather|afk
[22:16] <-- rootfather|afk left irc: Read error: Connection reset by peer
[22:17] <-- waltervn left irc: Quit: Leaving
[22:17] <-- TMM left irc: Remote host closed the connection
[22:28] <snover> bgKa: have you seen this with the PS3 toolchain with c++11 enabled? https://gist.github.com/csnover/efca6775596bc99e2a005d7426a506a2 i have only done a very minimal amount of research, though i did find this in some altivec header, not sure how responsible it is and need to run for a bit: /* You are allowed to undef these for C++ compatibility. */ [&] #define bool __bool
[22:29] <-- salty-horse left irc: Quit: Leaving
[22:31] --> TMM joined #scummvm.
[22:31] #scummvm: mode change '+o TMM' by ChanServ!ChanServ@services.
[22:38] <wjp> __vector(4) __bool int ? huh
[22:39] <-- Tomaz^ left irc: Ping timeout: 260 seconds
[22:41] <wjp> some gcc bug reports suggest this may need -std=gnu++11 instead of c++11
[22:41] <wjp> (e.g., https://gcc.gnu.org/bugzilla/show_bug.cgi?id=78263 )
[22:43] <-- Henke37 left irc: Quit: ERR_SHUTDOWN
[22:54] <-- criezy left irc: Quit: criezy
[22:55] --> criezy joined #scummvm.
[22:55] #scummvm: mode change '+o criezy' by ChanServ!ChanServ@services.
[22:55] <-- criezy left irc: Client Quit
[23:03] <-- Lightkey left irc: Ping timeout: 255 seconds
[23:16] --> Lightkey joined #scummvm.
[23:20] <-- ny00123 left irc: Quit: Leaving
[23:26] <-- Farmboy0 left irc: Remote host closed the connection
[23:36] --> dreammaster joined #scummvm.
[23:36] #scummvm: mode change '+o dreammaster' by ChanServ!ChanServ@services.
[23:49] <snover> wjp: good sleuthing! there are a couple of platforms where gnu++11 seems to be called for, that is just a remarkably spectacular example of a failure.
[23:50] <Lightkey> sleuthing sounds like something I might be good at
[23:51] <Lightkey> ah, no, I was thinking of slothing
[23:52] <snover> one is a set of strdup calls, which probably shouldnt have been there anyway, so i got rid of those using Common::String. the other one is a missing vsnprintf, which is really a platform issue since afaik that C99 API should exist in a conforming C++11 environment. im a bit torn on how to fix it since the true solution is to use variadic templates. in the meantime i will probably just use gnu++11 or -U__STRICT_ANSI__ to deal with
[23:52] <snover> it.
[00:00] --- Tue Dec 5 2017