[Back to Index]

  
[00:06] <grogbot> <IlDucci> And while I kinda see a difference in the understand_graphics document, I can't seem to find something that makes both names different
[00:06] <grogbot> <IlDucci> Could you please help me out explain this, or is this a case of "could be rewritten"?
[00:07] <-- Smedles left irc: Ping timeout: 268 seconds
[00:08] --> Smedles joined #scummvm.
[00:42] --> gsi_ joined #scummvm.
[00:45] <-- gsi left irc: Ping timeout: 245 seconds
[01:08] <Scummette> [scummvm] criezy pushed 1 new commits to master: https://git.io/JGAYl
[01:08] <Scummette> scummvm/master 8559354 criezy: DOC: Update number of stretch modes
[01:12] <grogbot> <criezy> I really struggled to find good names for those, so suggestions are welcome.
[01:21] <grogbot> <criezy> Pixel-perfect results in big "pixels" that have exactly the expected aspect ratio, for example 4:3, and scale each pixel in the horizontal direction by an integral factor (but vertically it may be non-integral, causing some blurring). In practice that means that original pixels do not necessarily all have the same size in the result image. For example when scaling 320x200 to 1280x960 (x4 plus aspect ratio correction) 4 lines of pixels out of
[01:21] <grogbot> 5 are duplicated 5 times, and 1 line out of 5 is duplicated 4 times. So 1/5th of the big pixels are smaller that the rest.
[01:24] <grogbot> <criezy> On the other hand the even pixel mode scales each pixel by an integral factor both horizontally and vertically but at the cost of changing the aspect ratio. With the example above it would produce an image at 1280x1000 instead of 1280x960, but that means that each big pixel is exactly 4x an original pixel horizontally and 5x vertically. So all the big "pixel" in the resulting image have the same size.
[01:25] <grogbot> <criezy> This only make a difference when aspect ratio correction is used. For games that do not need aspect ratio correction (for example games whose original resolution is 640x480), or when the aspect ratio correction is disabled, the two produce the exact same result.
[01:51] --> DominusExult joined #scummvm.
[01:52] <-- Dominus left irc: Ping timeout: 268 seconds
[01:52] Nick change: DominusExult -> Dominus
[02:57] <-- Smedles left irc: Quit: http://quassel-irc.org - Chat comfortably. Anywhere.
[03:01] --> Smedles joined #scummvm.
[03:29] LePhilousophe[m] <-- (~lephilous@scummvm/LePhilousophe) left irc: Remote host closed the connection
[03:30] LePhilousophe[m] --> (~lephilous@scummvm/LePhilousophe) joined #scummvm.
[03:30] #scummvm: mode change '+o LePhilousophe[m]' by ChanServ!ChanServ@services.libera.chat
[04:14] --> Sylvain joined #scummvm.
[04:17] <-- SylvainTV left irc: Ping timeout: 264 seconds
[04:31] --> zokum_ joined #scummvm.
[04:35] <-- niska left irc: Ping timeout: 264 seconds
[04:35] <-- zokum left irc: Ping timeout: 264 seconds
[04:35] <-- bgK left irc: Ping timeout: 264 seconds
[04:35] --> niska joined #scummvm.
[04:35] Nick change: zokum_ -> zokum
[04:35] --> bgK joined #scummvm.
[04:35] #scummvm: mode change '+o bgK' by ChanServ!ChanServ@services.libera.chat
[05:52] <Scummette> [scummvm] ScummVM-Translations pushed 3 new commits to master: https://git.io/JGpkK
[05:52] <Scummette> scummvm/master 1d7e05b lotharsm: I18N: Update translation (German)
[05:52] <Scummette> scummvm/master d06bb88 goodoldgeorge: I18N: Update translation (Hungarian)
[05:52] <Scummette> scummvm/master d07cf1b marcosoutsider: I18N: Update translation (Portuguese (Brazil))
[06:29] <-- Strangerke left irc: Changing host
[06:29] --> Strangerke joined #scummvm.
[06:29] #scummvm: mode change '+o Strangerke' by ChanServ!ChanServ@services.libera.chat
[06:29] <Strangerke> hi guys
[06:33] LePhilousophe[m] (~lephilous@scummvm/LePhilousophe) got netsplit.
[06:33] eriktorbjorn (~d91tan@scummvm/eriktorbjorn) got netsplit.
[06:34] ids1024 (~ids1024@iandouglasscott.com) got netsplit.
[06:34] borosky (~borosky@89-64-74-154.dynamic.chello.pl) got netsplit.
[06:34] djsrv (~djsrv@165.22.183.89) got netsplit.
[06:34] Lightkey (~Lightkey@p4fd7cbbf.dip0.t-ipconnect.de) got netsplit.
[06:34] LePhilousophe[m] (~lephilous@scummvm/LePhilousophe) returned to #scummvm.
[06:34] eriktorbjorn (~d91tan@scummvm/eriktorbjorn) returned to #scummvm.
[06:34] ids1024 (~ids1024@iandouglasscott.com) returned to #scummvm.
[06:34] borosky (~borosky@89-64-74-154.dynamic.chello.pl) returned to #scummvm.
[06:34] djsrv (~djsrv@165.22.183.89) returned to #scummvm.
[06:34] Lightkey (~Lightkey@p4fd7cbbf.dip0.t-ipconnect.de) returned to #scummvm.
[06:34] #scummvm: mode change '+ooo LePhilousophe[m] eriktorbjorn djsrv' by osmium.libera.chat
[06:56] <grogbot> <OzzieMonkey> Hey guys, so I've encountered a pretty serious bug with ScummVM after updating Windows 10. When running in OpenGL mode, the graphics go all weird like the image I've attached of Sam & Max Hit the Road. This happens with every game. I used to be able to run in this mode with no issues and it's my preferred setting to get that crisp pixel look.
[06:56] <grogbot> <OzzieMonkey> https://cdn.discordapp.com/attachments/581224061091446795/850992201898262538/ScummVM_bug.png
[07:00] <-- Strangerke left irc: Read error: Connection reset by peer
[07:09] <grogbot> <antoniou79> I cannot reproduce this with a local build (Windows 10), latest code. Which build are you testing with? Also could you share the relevant sections from your scummvm.ini file (the scummvm section and samnmax one)
[07:27] <grogbot> <antoniou79> Also cannot reproduce with the buildbot dev build, nor the snapshot from the Downloads page.
[07:31] <grogbot> <rootfather> Same here, no issues on my end. Maybe a driver update sneaked in breaking openGL for you, @OzzieMonkey?
[07:34] --> Strangerke joined #scummvm.
[07:34] #scummvm: mode change '+o Strangerke' by ChanServ!ChanServ@services.libera.chat
[07:34] <grogbot> <antoniou79> Hm, I see I have a pendn
[07:37] <grogbot> <antoniou79> Othewise I'm on 21H1 build 19043.985
[07:37] <Scummette> [scummvm] Strangerke pushed 3 new commits to master: https://git.io/JGp58
[07:37] <Scummette> scummvm/master 5ecae97 Strangerke: TRECISION: Fix warning in MSVC
[07:37] <Scummette> scummvm/master 5d27d3a Strangerke: TRECISION: Pass a pointer instead of SDText in drawText
[07:37] <Scummette> scummvm/master 61dc5d3 Strangerke: TRECISION: Change the order of initialization of GraphicsManager to please MSVC, fix missing initialization
[07:37] <grogbot> <rootfather> I'm on 19042.1023 for reference
[07:38] <grogbot> <rootfather> Good luck.
[07:38] --> Strangerke_ joined #scummvm.
[07:38] <-- Strangerke_ left irc: Changing host
[07:38] --> Strangerke_ joined #scummvm.
[07:38] #scummvm: mode change '+o Strangerke_' by ChanServ!ChanServ@services.libera.chat
[07:40] <grogbot> <OzzieMonkey> Daaaamn, I have the same version. No idea what the issue is then cause that's literally the only thing I've changed about my PC and everything worked fine before. I even fiddled with my graphics card settings and no dice.
[07:40] <-- Strangerke left irc: Ping timeout: 268 seconds
[07:40] <grogbot> <OzzieMonkey> It's definitely 21H1?
[07:41] <grogbot> <OzzieMonkey> Actually, I wonder if it's the new AMD drivers
[07:42] <grogbot> <rootfather> Yeah, I only use the AMD drivers provided via Windows Update, I never used their driver packages
[07:45] <grogbot> <OzzieMonkey> Right ok. Fow now I'm ok with using 3x to get a good level of pixelated goodness. I might try older drivers at some point and see if that does the trick
[07:47] <-- Strangerke_ left irc: Ping timeout: 264 seconds
[07:48] <grogbot> <antoniou79> I'll try and test on an old laptop next, which historically seemed to have issues (it doesn't have AMD card, but it does have one of those old, arguably crappy Intel HD 2000 ones)
[07:49] <grogbot> <antoniou79> @rootfather is the Windows snapshot from the Downloads page, a release build? (or at least built with --enable-optimizations)?
[07:50] <grogbot> <rootfather> The Windows snapshots are --enable-optionizations --enable-all-engines
[07:50] <grogbot> <antoniou79> Someone on the forum mentions crackles with MT-32 emulation and I wonder if it could be due to a debug build: https://forums.scummvm.org/viewtopic.php?f=2&t=16301
[07:50] <grogbot> <antoniou79> nice. I don't know which build they are running, but maybe it's not an optimized one
[07:51] <grogbot> <rootfather> I heard about the crackles a while ago, but I can't really replicate. I did extensive tests using the same toolchain when I updated munt the last time
[07:51] <grogbot> <rootfather> :/
[07:52] <grogbot> <rootfather> And looking at the 32bit build, maybe their hardware is too weak, IIRC munt is pretty CPU intense
[07:53] <grogbot> <antoniou79> It could be low specs hardware. I haven't checked with my RPi for a long time -- it's usually where I'd get issues with MT-32 emulation. On my main PC I don't get crackles or glitches
[07:54] <grogbot> <antoniou79> In the post they mention that MUNT on its own runs fine though
[07:54] <grogbot> <rootfather> I think it only affects older RPis. I tried MT-32 on my Pi 400 when I built the AppImage and I was suprised to find out that at least the 4/400 is powerful enough for flawless MT32 emulation
[07:54] <grogbot> <antoniou79> I have a 3B
[07:55] <grogbot> <rootfather> Funny thing is that the Pi 4 is powerful enough while my Intel Core i5 laptop CPU sometimes has issues with MT32 😄
[07:55] <grogbot> <antoniou79> lol. Oh, yeah, my laptop struggles with the debug builds on MT32 emulation too. (Loom intro is a good quick thing to test).
[08:26] <-- TMM_ left irc: Quit: https://quassel-irc.org - Chat comfortably. Anywhere.
[08:26] --> TMM_ joined #scummvm.
[08:54] --> SylvainTV__ joined #scummvm.
[08:57] <-- Sylvain left irc: Ping timeout: 268 seconds
[09:47] <grogbot> <IlDucci> Taking a page from RetroArch, what would you think about "Pure integer scaling"?
[09:48] <grogbot> <IlDucci> Another options: Pixel-perfect (match aspect ratio) for "Pixel-perfect" and "Pixel-perfect" for "Even pixel scaling"
[09:49] <grogbot> <antoniou79> Should these also get a tooltip for a quick further explanation -- or is this not possible for dropdown items?
[10:26] <-- borosky left irc:
[10:33] --> borosky joined #scummvm.
[11:37] --> athrxx joined #scummvm.
[11:41] <-- athrxx left irc: Client Quit
[11:43] --> athrxx joined #scummvm.
[11:49] <grogbot> <_athrxx> I'd say, "Pure integer scaling" is completely non-descriptive to >99% of users. Something like "Pixel-perfect (with aspect ratio correction)" and "Pixel-perfect (w/o aspect ratio correction)" might be better to understand. And should also be straightforwart to translate...
[11:51] <grogbot> <_athrxx> I agree that the current wording "pixel-perfect" and "even pixels" isn't very helpful
[12:34] <eriktorbjorn> Is there anyone around who has the DVD version of Zork Grand Inquisitor, and is running the Windows version of ScummVM? I'm wondering if the hack I made to get the cutscene volume right (for me) is actually breaking the volume for others: https://bugs.scummvm.org/ticket/11181
[12:40] <grogbot> <IlDucci> That sounds indeed pretty much descriptive, if it fits in lower-res ports
[13:08] <-- TMM_ left irc: Quit: https://quassel-irc.org - Chat comfortably. Anywhere.
[13:08] --> TMM_ joined #scummvm.
[13:09] <grogbot> <criezy> eriktorbjorn: I can at least confirm that with the GOG version the volume sounds similar for he low resolution and the high resolution videos on macOS currently, and I don't hear any clipping.
[13:11] <-- athrxx left irc: Quit: Leaving
[13:11] <grogbot> <criezy> Also that bug seems to be about the low resolution videos since he mentions that the high resolution videos cause a crash for him. So the volume boosting for the high resolution videos should be irrelevant (unless it is also somehow applied on the low resolution ones)?
[13:14] <grogbot> <criezy> "Pixel-perfect (w/o aspect ratio correction)" might be misleading though. At least I would understand it as meaning that no aspect ratio correction is applied (and thus that the he aspect ratio correction option is ignored). But it still does an approximative aspect ratio correction.
[13:16] <grogbot> <_athrxx> @criezy Okay, how about "almost pixel-perfect" and "even more pixel-perfect"? 😛
[13:18] <grogbot> <_athrxx> Or "Pixel-perfect with aspect ratio correction" vs "Pixel-perfect with a bit less aspect ratio correction"?
[13:20] <grogbot> <_athrxx> "Even Pixels" isn't that bad actually, if you know what it is supposed to mean 😉
[13:23] <grogbot> <_athrxx> "Pixel-perfect (uneven pixels)" vs "Pixel-perfect (even pixels)"?
[13:35] <grogbot> <rootfather> I wonder if the "old" Pixel-perfect method can be even called "Pixel-perfect" anymore
[13:36] <grogbot> <rootfather> Since we now have the true pixel perfect scaling
[13:36] <grogbot> <rootfather> So what about "Integer Scaling" for the "old" pixel-perfect scaler and "Pixel perfect" for the new pixel perfect secaler?
[13:50] <grogbot> <criezy> "Pixel-perfect (uneven pixels)" and "Pixel-perfect (even pixels)" sound at least better than the current names I think (even if not perfect 😛 ).
[14:20] --> Sylvain joined #scummvm.
[14:24] <-- SylvainTV__ left irc: Ping timeout: 272 seconds
[14:28] <Scummette> [scummvm] athrxx pushed 5 new commits to master: https://git.io/JZeNC
[14:28] <Scummette> scummvm/master c57243a athrxx: SCUMM: (ZAK) - fix bug #12278 ("Zak's positioning in the intro")
[14:28] <Scummette> scummvm/master 74ed6d9 athrxx: SCUMM: (SCUMM1-3) - (very minor) fix for walk dir interpolation
[14:28] <Scummette> scummvm/master f5ed4d5 athrxx: SCUMM: (SCUMM1-3) - fix actor facing for diagonal walks
[14:47] <-- ced117 left irc: Changing host
[14:47] --> ced117 joined #scummvm.
[15:03] <grogbot> <BeleG> From a user's perspective, I don't think you can get away with referring to two different modes as both being "pixel-perfect". At least my poor lay mind struggles understanding what it means!
[15:04] <Scummette> [scummvm] athrxx pushed 1 new commits to master: https://git.io/JZvLE
[15:04] <Scummette> scummvm/master 1f7c52a athrxx: SCUMM: whitespace
[15:06] <grogbot> <_athrxx> yes, uneven kind of implies imperfect
[15:12] --> ny00123 joined #scummvm.
[15:19] <Dark-Star> hm, even I, who understands the technicalities behind scaling a bit, wouldn't know which option to choose. does "uneven" mean "non-square" ? orsomething else? also, is it for correcting games FROM uneven pixels TO square pixels? Or will the result be that the pixels I see are uneven (i.e. non-square? or that every n-th line will be a bit smaller?)
[15:25] <Dark-Star> how about "emulate original aspect ratio" vs. "stretch display horizontally (ignore aspect ratio)" or something like that? This would make it clearer that the first option would provide the result closest to the original, and it would make it clear what happens when aspect correction is off (horizontal stretching)
[15:26] <grogbot> <_athrxx> No, it actually means that the pixels will differ in size. If I understand it correctly (not guaranteed 😅 ) there will be two pixel sizes, some pixels will be scales to the one size, some to the other, in order to reach the desired number of lines
[15:27] <grogbot> <rootfather> "Perfect scaling" and "More perfect scaling". That's it. 😛
[15:28] <grogbot> <BeleG> At the end of the day, as long as the modes are well documented in the wiki it probably doesn't matter much either way, but considering that most users will not be checking the wiki I'd try to use something as descriptive as possible (which doesn't seem like an easy task!)
[15:28] <Dark-Star> maybe add some sample images to the GUI to make it clear what happens? exaggerating the effects a bit? because, let's be honest here, configuring ScummVM's video settings is not exactly trivial :-D
[15:29] <grogbot> <_athrxx> The problem with these modes is that you won't see much of a difference on a thumbnail
[15:29] <Dark-Star> yes that's why it should be exaggerated/zoomed in
[15:31] <grogbot> <_athrxx> Unfortunately, some mode make more sense for certain games while other modes are better for other games. So even having a sane default setting isn't that simple...
[15:31] <grogbot> <criezy> We could start by adding some images to https://docs.scummvm.org/en/latest/advanced_topics/understand_graphics.html#stretch-modes
[15:32] <grogbot> <antoniou79> I ran the GOG (DVD) version of Zork GI, with ScummVM on Windows. The crash issue with the high quality videos has been fixed (I think it was the past week). The volume of the video sounds ok to me.
[15:32] <grogbot> <antoniou79> But, I cannot control it from ScummVM
[15:32] <grogbot> <antoniou79> The volume controls for Music, Speech, SFX do not affect it at all. You'd expect that one would
[15:34] <Dark-Star> @criezy true, I had to read the explanations multiple times and I'm still not sure what the difference between "pixel-perfect" and "even pixels" is :-D
[15:35] <Dark-Star> pixel perfect does aspect correction first and scales by a single factor, while even pixels scales by different factors to implicitly get aspect correction that way?
[15:38] <Dark-Star> isn't "pixel-perfect scaling with aspect correction" always inferior to "even pixels scaling with aspect correction" due to the artifacts of the aspect correction?
[15:39] <-- _sev left irc: Quit: This computer has gone to sleep
[15:40] --> _sev joined #scummvm.
[15:40] #scummvm: mode change '+o _sev' by ChanServ!ChanServ@services.libera.chat
[16:12] <Scummette> [scummvm] sev- closed pull request #3044: DIRECTOR: allow setting of video casts to castNum (master...director-set-cast-movies) https://git.io/JGXXS
[16:12] <Scummette> [scummvm] sev- pushed 1 new commits to master: https://git.io/JZv6f
[16:12] <Scummette> scummvm/master a7f16ff rvanlaar: DIRECTOR: allow setting of video casts to castNum
[16:14] <grogbot> <ghoost82> I've tested my German version, but it was not deteced. After adding it to the detection table I was able to boot it up. I played it for about 10 minutes and it seems to working fine. {"bklynlgo.bmp", 0, "1dfb703349a46f8ec183de107992b7f5", 33118}, {"pvteye.ex_", 0, "5ca171c4e8d804c7277887277d049f03", 600576},
[16:16] <grogbot> <gu3> Thanks a lot!
[16:17] <grogbot> <gu3> btw, can you check if inside your game there is a demo folder? I know the spanish version includes the demo, but the in the french is missing..
[16:22] <grogbot> <ghoost82> there is a German demo in a folder called PVTDEMO ``` {"pvtdemo.ex", 0, "17156cbac7d14b08f4e351ac0e16a889", 599040}, {"bklynlgo.bmp", 0, "1dfb703349a46f8ec183de107992b7f5", 33118}, ```
[16:23] <grogbot> <gu3> Great, thanks again!
[16:27] <grogbot> <ghoost82> No problem, it is always a pleassure for me to support the SCUMMVM team
[16:36] <grogbot> <gu3> make
[18:28] <-- ny00123 left irc: Quit: Leaving
[18:31] --> ny00123 joined #scummvm.
[18:51] --> chraist joined #scummvm.
[18:52] <-- chraist left irc: Client Quit
[19:00] --> ny00123_ joined #scummvm.
[19:02] <-- ny00123 left irc: Ping timeout: 245 seconds
[19:11] --> frankyboy_ joined #scummvm.
[19:19] --> DominusExult joined #scummvm.
[19:20] <-- Dominus left irc: Ping timeout: 244 seconds
[19:20] Nick change: DominusExult -> Dominus
[19:31] --> SupSuper joined #scummvm.
[19:31] #scummvm: mode change '+o SupSuper' by ChanServ!ChanServ@services.libera.chat
[19:40] <-- SupSuper left irc: Ping timeout: 268 seconds
[19:46] <-- frankyboy_ left irc: Quit: ...
[19:48] --> frankyboy_ joined #scummvm.
[20:13] --> SupSuper joined #scummvm.
[20:13] #scummvm: mode change '+o SupSuper' by ChanServ!ChanServ@services.libera.chat
[21:39] <-- SupSuper left irc: Quit: Rip
[21:41] <-- ny00123_ left irc: Quit: Leaving
[22:22] <Scummette> [scummvm] ScummVM-Translations pushed 3 new commits to master: https://git.io/JZUe7
[22:22] <Scummette> scummvm/master e2d307e Timpii: I18N: Update translation (Finnish)
[22:22] <Scummette> scummvm/master dda96bc : I18N: Update translation (French)
[22:22] <Scummette> scummvm/master 578c68a antoniou79: I18N: Update translation (Greek)
[22:26] <Scummette> [scummvm] sev- pushed 4 new commits to master: https://git.io/JZUvd
[22:26] <Scummette> scummvm/master 238127f sev-: SCUMM HE: Added support for French Socks game
[22:26] <Scummette> scummvm/master d7f82b8 sev-: ASYLUM: Added override keywords
[22:26] <Scummette> scummvm/master 275cad9 sev-: LILLIPUT: Added missing reference to debug flags
[22:26] <Scummette> [scummvm] ScummVM-Translations pushed 1 new commits to master: https://git.io/JZUvA
[22:26] <Scummette> scummvm/master 0d44072 : I18N: Update translation files
[22:34] <Scummette> [scummvm] bluegr closed pull request #3041: COMMON: fixed invalid read in zlib read stream (master...pr/event-recorder-valgrind) https://git.io/JGKzk
[22:34] <Scummette> [scummvm] bluegr pushed 2 new commits to master: https://git.io/JZUTt
[22:34] <Scummette> scummvm/master b259e16 mgerhardy: COMMON: fixed invalid read in zlib read stream
[22:34] <Scummette> scummvm/master 6772dab mgerhardy: COMMON: replaced NULL with nullptr in wrapCompressedReadStream
[23:51] <Scummette> [scummvm] ScummVM-Translations pushed 1 new commits to master: https://git.io/JZUzk
[23:51] <Scummette> scummvm/master d3cacb7 sev-: I18N: Update translation (Russian)
[23:51] <Scummette> [scummvm] ScummVM-Translations pushed 1 new commits to master: https://git.io/JZUzt
[00:00] --- Mon Jun 7 2021