ScummVM IRC Log - 06 Jul 2021

[Back to Index]

[01:19] <-- Lightkey left irc: Ping timeout: 256 seconds
[01:32] --> Lightkey joined #scummvm.
[01:50] --> DominusExult joined #scummvm.
[01:53] <-- Dominus left irc: Ping timeout: 265 seconds
[01:53] Nick change: DominusExult -> Dominus
[02:42] <Scummette> [scummvm] dreammaster pushed 1 new commits to master: https://git.io/JcP4l
[02:42] <Scummette> scummvm/master 55d9f90 dreammaster: AGS: Added TPF demo detection entry
[02:51] <Scummette> [scummvm] djsrv pushed 4 new commits to master: https://git.io/JcPRC
[03:08] <Scummette> scummvm/master b3b55e5 djsrv: GRAPHICS: MACGUI: Add parent to MacFontRun
[03:08] <Scummette> scummvm/master 831c014 djsrv: DIRECTOR: Make FXmp available to MacText
[03:08] <Scummette> scummvm/master 3fda461 djsrv: GRAPHICS: MACGUI: Handle FontXPlatformMap in MacFontManager::getFont
[03:08] <Scummette> [scummvm] dreammaster pushed 1 new commits to master: https://git.io/JcPz7
[03:08] <Scummette> scummvm/master f8ac707 dreammaster: AGS: Add more explicit error message when trying to run pre-2.5 games
[03:42] --> gsi_ joined #scummvm.
[03:45] <-- gsi left irc: Ping timeout: 265 seconds
[06:40] <-- balrog left irc: Ping timeout: 268 seconds
[06:52] --> balrog joined #scummvm.
[07:03] <eriktorbjorn> Well, at least I got a reply from GOG support yesterday about Maniac Mansion. Unfortunately, the guy seemed confused about the problem, because he asked me if I had difficulties progressing forward with my game. So now I've attached two savegames, explaining that the problem is that their cracked version makes it *too easy* to progress. Let's hope he gets it this time.
[07:05] <grogbot> <Kebabounet> Ah and the later versions miss at least a joke: in the screen where you select your characters, their description is supposed to appear in two lines, with a pause between them. Thats because the first line is supposed to be an impressive description, while the second line actually makes the description a bit more deceptive
[07:07] <grogbot> <Kebabounet> "Award winning photographer&& for the college newspaper." Etc
[07:22] <eriktorbjorn> Kebabounet: I don't see that pause in either DOS version, even in DOSBox.
[08:22] <grogbot> <Kebabounet> Ah sorry I dont remember if it was a pause, or just a new line that went missing, or maybe something just appearing in the C64 version
[08:22] <grogbot> <Kebabounet> I do remember an online guide mentioning that but that was years ago
[08:56] <-- reset left irc: Quit: reset
[09:04] <Scummette> [scummvm] eriktorbjorn pushed 1 new commits to master: https://git.io/JcXib
[09:04] <Scummette> scummvm/master ce52a33 eriktorbjorn: STARK: Avoid adding \0 to a String
[09:05] <eriktorbjorn> Kebabounet: Judging by https://www.youtube.com/watch?v=9WVF0o9Lq2Q the C64 version had only one line of text at the top, where the DOS version has two.
[09:08] <eriktorbjorn> The GOG support guy said he would "pass on my findings to our Product Team for further analysis, that said I am unable to guarantee any solutions here at this time".
[09:08] <eriktorbjorn> I'm a bit irked that he asked me if he was correct that "it does not prevent you from finishing the game in any way" and that this "simplifies the game and you aren't happy that it isn't faithful to the original". I would have thought that was obvious by now.
[09:10] <eriktorbjorn> But I did say that yes, this was correct, and asked if he would be satisfied if he bought some fps game only to find out that it had some third-party crack that turned him invulnerable for some of the trickier parts. After all, this too would not prevent him from finishing the game. Just make it simpler.
[09:11] --> reset joined #scummvm.
[09:35] <grogbot> <antoniou79> If the issue is with the game data (and not ScummVM) wouldn't it be an easy fix? Just provide the original files. And maybe the booklet with the codes if required.
[09:35] <grogbot> <antoniou79> I mean an easy fix for GOG (or Disney)
[09:39] <eriktorbjorn> You'd think so. But it seems to depend a lot on which support person answers.
[09:46] <eriktorbjorn> And a quick look at Lucasfilm Game's twitter would suggest that their first seven or eight priorities now are Star Wars. (It doesn't help that when they post something that isn't Star Wars, you can find replies like "We want Star Wars games not this stupid 16 bit game".)
[09:46] <eriktorbjorn> Odd, because as I recall it the original Lucasfilm Games made hardly any Star Wars games.
[09:47] <grogbot> <madmoose> eriktorbjorn: Right, because the rights to make Star Wars games had already been sold.
[09:48] <grogbot> <madmoose> Merchandising!
[09:59] <eriktorbjorn> Oh for cryin' out loud... "Thank you for the clarification and I quite understand. Unfortunately, as it is, the version of the game is powered by ScummVM and this is the version available at our store. That said, I will make sure to pass on your feedback ..."
[10:31] <grogbot> <madmoose> "You say you understand but every other thing you say betrays that statement."
[10:44] <-- TMM_ left irc: Quit: https://quassel-irc.org - Chat comfortably. Anywhere.
[10:44] --> TMM_ joined #scummvm.
[10:44] <eriktorbjorn> I'll try that next time. This time I politely pointed out that "IT - IS - NOT - A - SCUMMVM - PROBLEM!!!", why I think I've earned the benefit of doubt in this case, how I've been through this before, how we used to get ScummVM bug reports about this problem long before they were selling the game, how it reflects badly on both them and us, etc.
[11:16] --> Maturion joined #scummvm.
[11:29] <-- Maturion left irc: Ping timeout: 240 seconds
[12:14] <-- reset left irc: Quit: reset
[13:01] <grogbot> <thoth> Is this the same as the version that got distributed with the newer Maniac releases? Because that would at least explain how this happened...
[13:02] --> ajax16384 joined #scummvm.
[13:06] <eriktorbjorn> I've only seen it in the GOG release of Maniac Mansion, but I wouldn't be surprised if the Steam release is the same. Only the enhanced version is affected. The even-lower-resolution original version appears to be identical to what I got with my old Day of the Tentacle CD, and is not cracked. (When playing in DOSBox, it disables the copy protection by having the security door open. ScummVM does the same thing.)
[13:10] <eriktorbjorn> The Maniac Mansion in the remastered version of Day of the Tentacle seems fine too, but I don't know if there's any way to extract that to run in ScummVM.
[13:55] --> CatButts joined #scummvm.
[13:58] <CatButts> are there any other projects out there like MUNT for the MT-32(or any other LA synth, for that matter)?
[13:58] <CatButts> besides MUNT
[13:59] <Scummette> [scummvm] sev- pushed 6 new commits to master: https://git.io/Jc1EN
[13:59] <Scummette> scummvm/master abaf4c3 sev-: SAGA2: Fix initialization of SpellTarget. CID 1457848, CID 1457849
[13:59] <Scummette> scummvm/master d4c8cfb sev-: SAGA2: Properly set actor assignment on loading. CID 1457852, CID 1457891
[13:59] <Scummette> scummvm/master aa12af7 sev-: SAGA2: Fix Band initialization. CID 1457853, CID 1457980
[13:59] <Deledrius> eriktorbjorn, sounds like he's just following a script, unfortunately :/
[14:01] <grogbot> <timofonic> I have no idea. But it would be amazing to have something similar for Yamaha, Casio, Ensoniq
[14:01] <CatButts> for Ensoniq, there was a VST, I think
[14:01] <CatButts> closed source
[14:03] <grogbot> <thoth> Maniac in DoTT Remastered can be extracted the same way that vanilla DoTT can be extracted from DoTT Remastered (the file format has been pretty exhaustively documented at this point -- google is your friend). It's literally just Maniac Mansion DOS v2 (so, enhanced). Seemingly unmodified, but I haven't verified it.
[14:10] --> ccawley2011 joined #scummvm.
[14:10] #scummvm: mode change '+o ccawley2011' by ChanServ!ChanServ@services.libera.chat
[14:16] <CatButts> ah yeah, this thing https://www.youtube.com/watch?v=AaGUgNP40bI
[14:16] <CatButts> for the Ensoniq
[14:23] <Scummette> [scummvm] sev- pushed 2 new commits to master: https://git.io/Jc1a3
[14:23] <Scummette> scummvm/master f032f05 sev-: SAGA2: Replace rand() with _random
[14:23] <Scummette> scummvm/master 7cbd245 sev-: SAGA2: Replaced rest of rand() calls
[14:45] <Scummette> [scummvm] sev- pushed 6 new commits to master: https://git.io/Jc1oA
[14:54] <Scummette> scummvm/master c9c77e4 sev-: SAGA2: Fixed side effect in assertion. CID 1457988
[14:54] <Scummette> scummvm/master 20a3dd4 sev-: SAGA2: Cleanup hResMgr
[14:54] <Scummette> scummvm/master b429407 sev-: SAGA2: Use safer string copy function
[14:54] <Scummette> [scummvm] sev- pushed 3 new commits to master: https://git.io/Jc16X
[14:54] <Scummette> scummvm/master 6e8aed5 sev-: SAGA2: Fix potential null dereferencing. CID 1457864
[14:54] <Scummette> scummvm/master 827496b sev-: SAGA2: Fix potential NULL dereference. CID 1458002
[14:54] <Scummette> scummvm/master 79003be sev-: SAGA2: Fix new[]/delete mismatch
[15:08] --> Maturion joined #scummvm.
[15:12] <Scummette> [scummvm] a-yyg pushed 3 new commits to master: https://git.io/Jc11k
[15:12] <Scummette> scummvm/master 6c857c3 a-yyg: SAGA2: Add more debug output for saveload
[15:12] <Scummette> scummvm/master 2c7416f a-yyg: SAGA2: Implement Timers and Calender saveloading
[15:12] <Scummette> scummvm/master a7650b1 a-yyg: SAGA2: Implement Worlds loading
[15:37] --> athrxx joined #scummvm.
[15:47] <Scummette> [scummvm] sev- pushed 3 new commits to master: https://git.io/Jc1Hi
[15:53] sandvich[m] --> (~sandvichm@2001:470:69fc:105::51fc) joined #scummvm.
[15:54] <sandvich[m]> Hi everyone,
[15:54] <sandvich[m]> I just wanted to check if ScummVM has any versions (built from source included) that can run the longest journey?
[15:55] <sandvich[m]> I see that there's no release that supports it, but I figured maybe a specific commit is still worth looking at?
[15:55] <grogbot> <SupSuper> since residualvm was merged, the latest daily should do it
[15:55] <sandvich[m]> okay, I'll have a look, thanks :
[15:55] <sandvich[m]> * okay, I'll have a look, thanks :D
[15:57] <sandvich[m]> ah, looks like no RPM/snap/flatpak version in the daily builds
[15:58] <sandvich[m]> if I just try to build the main branch on github, that should work, right?
[15:58] <grogbot> <SupSuper> yeah
[15:58] <sandvich[m]> perfect :)
[15:58] <grogbot> <SupSuper> there's compiling instructions on the wiki if you need
[15:58] <sandvich[m]> thanks
[16:18] <sandvich[m]> I wasn't expecting the hardest step to be cloning! Still only at 40% xD
[16:19] <sandvich[m]> I will be patient though :)
[16:23] --> ny00123 joined #scummvm.
[16:35] <eriktorbjorn> thoth: You're right, I was able to extract the .lfl files using the Double Fine Explorer. Those files are identical to the Maniac Mansion I have on floppies (that I can no longer read, alas) in the "Classic Adventures From LucasArts" collection. In the GOG version, 00.lfl and 43.lfl are different. (I still don't know why 00.lfl is different, but 43.lfl appears to be the script handling keypads in the game.)
[16:36] <grogbot> <sev> sandvich[m]: You may build only the stark engine. ./configure --disable-all-engines --enable-engine=stark
[16:37] <grogbot> <sev> and also, if you ran configure without parameters, then stark engine is not being built
[16:37] <sandvich[m]> useful, I'll do that. Thanks :)
[16:37] <grogbot> <sev> and I hope you're running with -j <cpus+2>
[16:37] <sandvich[m]> I still haven't finished the clone :p
[16:37] <sandvich[m]> I did see that in the wiki, so yeah, I'll do that ^^
[16:37] <grogbot> <sev> yes, our repo grew a lot
[16:39] <sandvich[m]> finally finished, starting the compile now ^^
[16:42] <sandvich[m]> hmm
[16:42] <sandvich[m]> ```
[16:42] <sandvich[m]> WARNING: Stark failed to instantiate engine: Game data not found (target 'tlj-win', path '<game install dir>')!
[16:43] <sandvich[m]> ah, looks like it's just down to me missing something in the copy
[16:43] <sandvich[m]> it works :D
[16:44] <sandvich[m]> other than the large initial clone, that was pretty painless. I was expecting a lot worse
[16:48] <eriktorbjorn> Well, there are some newly introduced bugs lately, but hopefully nothing that affects that game engine. :-)
[16:48] <sandvich[m]> worst case, I can always jump back a few merges
[16:49] <sandvich[m]> as I say, the build was basically nothing. Took a couple of minutes and worked first time. It was just the clone that was painful ^^
[16:50] <eriktorbjorn> The SCI engine seems to be the one that's hit the hardest. Most SCI games I've tried today crashed either during the intro, or very early in the game.
[16:51] <sandvich[m]> thanks for the help @SupSuper, _sev and eriktorbjorn
[16:53] <LePhilousophe[m]> sandvich: you can use git clone --depth 1 if you don't want to pull the full history (or download the source tarball at github)
[16:53] <sandvich[m]> I probably should have. Oh well :)
[16:53] <sandvich[m]> I just wasn't expecting it to be so massive
[17:11] <eriktorbjorn> These are the SCI games that currently crash for me, by the way: https://bugs.scummvm.org/ticket/12721#comment:3
[17:17] <sandvich[m]> works beautifully, btw :D
[17:18] <athrxx> eriktorbjorn That rect commit should be reverted. We haven't even recovered from the Common::String changes. Do we really want to have another source of bugs for the rest of the year?
[17:24] --> SylvainTV__ joined #scummvm.
[17:27] <-- Sylvain left irc: Ping timeout: 246 seconds
[17:39] <grogbot> <DreamMaster> Are there any performance implications to having a templated based rect and point class? That way we could have Rect/Point class being standard, but also have a Rect16/Point16 for engines like scumm that are doing dodgy stuff that rely on the 16-bit size
[18:02] <-- Maturion left irc: Ping timeout: 252 seconds
[18:16] <Scummette> scummvm/master 42c888f sev-: SAGA2: Fix out-of-bounds read.. CID 1457984
[18:16] <Scummette> scummvm/master 055a316 sev-: SAGA2: Fix allocation size. CID 1457903
[18:16] <Scummette> scummvm/master c019397 sev-: SAGA2: Fix object copying. CID 1458027
[18:16] <Scummette> [scummvm] neuromancer pushed 1 new commits to master: https://git.io/JcMOs
[18:16] <Scummette> scummvm/master 223c398 neuromancer: PRIVATE: disable pause menu when police bust is enabled
[18:27] <grogbot> <sev> @DreamMaster no, no implications. the only thing templates make worse is compile time (neglible in this case) and code size, which is also expected
[19:07] --> reset joined #scummvm.
[19:23] --> Maturion joined #scummvm.
[19:40] <grogbot> <DreamMaster> Alright. Then tonight I'll modify Rect and Point to be a base template BaseRect and BasePoint tonight, which will have a derived typedef BaseRect<int32> Rect; typedef BaseRect<int16> Rect16, and likewise for Point. Then there'll be an option, for a scumm dev to a do a global search & replace of Common::Rect with Common::Rect16 rather than look into individual bugs.
[19:41] <grogbot> <DreamMaster> It also means I'll be able to simplify some of the classes in Titanic, since now I can have a derived floating-point rect & point class for the starmap without having to create my own from scratch
[19:49] <grogbot> <DreamMaster> Probably useful in tSage for that stupid Sunflower field minigame in Ringworld. But I'm loathe to touch that monstrosity again, even for a brief testing after making a change 😛
[19:50] <grogbot> <DreamMaster> Actually, maybe not define a Rect16/Point16 in Common to avoid naming confusion for autocompletion. Engines like scumm can just as easily define their own typedef for Rect16 if needed
[19:54] <grogbot> <athrxx> You're aware that SCI is even more broken? And what about the other engines? Maybe it would be better to just make the int16 variant the default.
[20:05] <Scummette> [scummvm] sev- pushed 8 new commits to master: https://git.io/JcM0W
[20:05] <Scummette> scummvm/master 967f745 sev-: SAGA2: Initialize button.cpp classes
[20:05] <Scummette> scummvm/master 2223717 sev-: SAGA2: Fix potential null dereference. CID 1457869
[20:05] <Scummette> scummvm/master 55a9867 sev-: SAGA2: Move globalContainerList to Saga2Engine
[20:10] <grogbot> <DreamMaster> Actually, now that you mention it, I realise that makes sense. I was still in the mindset that the Surface class was 32-bit, so that the Rect class it used would need to be 32-bits as well. But now that it's signed 16-bits rather than 32, then a 16-bit rect class can still be the default across the board, but the base template will allow engines that specifically want it to do so
[20:14] <grogbot> <DreamMaster> It'll also make it unlikely for other existing engines to have similar issues with 32-bit rects/poitns as well
[20:15] <grogbot> <athrxx> The thing is: many engines are basically abandoned and don't see much testing. There aren't that much active devs who're just waiting to clean up their engines 😉
[20:17] <grogbot> <athrxx> Even for SCUMM, if you take look in the credits section: it's basically left to @sev and erictorbjorn. Kirben is still mentioned as active, since he never said that he wants to be retired.
[20:22] <grogbot> <SupSuper> i can see a templated rect being useful for future engines. though i think we already have float rects from the residualvm math stuff
[20:22] <grogbot> <SupSuper> mostly i'm dreading the usual int vs int32 issues 😛
[20:24] <grogbot> <sev> @DreamMaster what about leaving Rect alone, but introducing Rect32 ?
[20:24] <grogbot> <sev> otherwise you will potentially affect tons of engines
[20:25] <grogbot> <sev> leaving along = template-wise, BaseRect<uint16> is Rect, and BaseRect<int32> is Rect32
[20:25] <grogbot> <sev> then you switch AGS to that
[20:30] <grogbot> <DreamMaster> I think we're in agreement. Though BaseRect<int16> will be Rect, to match what Rect used to be, and hopefully avoid issues in all the existing engines, not just scumm, and AGS, Titanic, and others that need it can do their own local typedef of BaseRect<int32> to be Rect32, Rec, FRect, or whatever
[20:43] <Scummette> [scummvm] a-yyg pushed 1 new commits to master: https://git.io/JcMzf
[20:43] <Scummette> scummvm/master a931df7 a-yyg: SAGA2: Un-stub openBook/openParchment
[20:44] <Scummette> [scummvm] djsrv pushed 4 new commits to master: https://git.io/JcMzt
[20:44] <Scummette> scummvm/master e9d9b9f djsrv: DIRECTOR: Revert FXmp handling
[20:44] <Scummette> scummvm/master a2848f5 djsrv: DIRECTOR: Handle FXmp mappings in loadFontMapV4
[20:44] <Scummette> scummvm/master f64e941 djsrv: DIRECTOR: Add comment about ignored char mapings
[20:59] <-- ny00123 left irc: Quit: Leaving
[21:01] <-- Harekiet left irc: Ping timeout: 256 seconds
[21:12] <-- speachy left irc: Quit: WeeChat 3.0.1
[21:12] --> speachy joined #scummvm.
[21:14] <-- Maturion left irc: Ping timeout: 240 seconds
[21:18] <-- speachy left irc: Changing host
[21:18] --> speachy joined #scummvm.
[21:35] <-- TMM_ left irc: Quit: https://quassel-irc.org - Chat comfortably. Anywhere.
[21:35] --> TMM_ joined #scummvm.
[21:40] <-- athrxx left irc: Quit: Leaving
[21:46] <-- ccawley2011 left irc: Quit: Leaving
[21:56] <-- ajax16384 left irc: Read error: Connection reset by peer
[22:02] --> Harekiet joined #scummvm.
[23:49] <Scummette> [scummvm] sev- pushed 1 new commits to master: https://git.io/JcMQY
[23:49] <Scummette> scummvm/master 1ff8ead sev-: SAGA2: Init Saga2Engine class members. CID 1457993
[00:00] --- Wed Jul 7 2021