[Back to Index]

[00:03] <-- m_kiewitz left irc: Quit: technology isn't intrinsically good or evil. It's how it's used. Like the Death Ray.
[00:25] <-- klusark left irc: Ping timeout: 250 seconds
[00:38] --> klusark joined #scummvm.
[01:01] <-- Henke37 left irc: Quit: ERR_SHUTDOWN
[01:04] <-- WinterGrascph left irc: Ping timeout: 260 seconds
[01:30] --> DominusExult joined #scummvm.
[01:34] <-- Dominus left irc: Ping timeout: 244 seconds
[01:34] Nick change: DominusExult -> Dominus
[01:38] --> blitter joined #scummvm.
[02:23] <-- Strangerke left irc: Ping timeout: 264 seconds
[02:33] <-- bouncy left irc: Ping timeout: 272 seconds
[02:45] --> bouncy joined #scummvm.
[03:28] --> jammm joined #scummvm.
[03:28] <-- jammmm left irc: Read error: Connection reset by peer
[03:35] <-- ST left irc: Read error: Connection reset by peer
[03:47] <-- snover left irc: Quit: Leaving.
[04:01] --> Poly-C joined #scummvm.
[04:03] <-- Cruel` left irc: Ping timeout: 240 seconds
[04:04] <-- Polynomial-C left irc: Ping timeout: 244 seconds
[05:03] --> Mellified_Man_ joined #scummvm.
[05:04] --> balrog_ joined #scummvm.
[05:04] snow_bckspc (~snow_bcks@ganon.dot-server.net) got netsplit.
[05:04] Mellified_Man (~Mellified@eddie.mellified.com) got netsplit.
[05:04] VonGuardian (sid17526@gateway/web/irccloud.com/x-tutjzpvqyiknsqti) got netsplit.
[05:04] balrog (~balrog@unaffiliated/balrog) got netsplit.
[05:06] Nick change: balrog_ -> balrog
[05:06] Possible future nick collision: balrog
[05:15] VonGuardian (sid17526@gateway/web/irccloud.com/x-tutjzpvqyiknsqti) got lost in the net-split.
[05:15] Mellified_Man (~Mellified@eddie.mellified.com) got lost in the net-split.
[05:15] snow_bckspc (~snow_bcks@ganon.dot-server.net) got lost in the net-split.
[06:00] --> ny00123 joined #scummvm.
[06:29] --> uruk-hai joined #scummvm.
[06:29] #scummvm: mode change '+o uruk-hai' by ChanServ!ChanServ@services.
[06:36] --> frankyboy_ joined #scummvm.

[07:17] <-- uruk-hai left irc: Quit: Leaving
[07:18] --> Begasus joined #scummvm.
[07:25] --> uruk-hai joined #scummvm.
[07:25] #scummvm: mode change '+o uruk-hai' by ChanServ!ChanServ@services.
[07:25] --> waltervn joined #scummvm.
[07:26] #scummvm: mode change '+o waltervn' by ChanServ!ChanServ@services.
[07:26] <waltervn> morning
[07:30] --> m_kiewitz joined #scummvm.
[07:30] #scummvm: mode change '+o m_kiewitz' by ChanServ!ChanServ@services.
[07:31] <Begasus> morning
[07:51] <-- ced117 left irc: Quit: Quitte
[07:54] --> ced117 joined #scummvm.
[08:10] <-- jammm left irc: Quit: Leaving
[08:10] --> jamm joined #scummvm.
[08:15] --> salty-horse joined #scummvm.
[08:15] #scummvm: mode change '+o salty-horse' by ChanServ!ChanServ@services.
[08:19] <salty-horse> need help from a scumm hacker: where are the definitions of where to move the characters when clicking on an on-screen object? is it in a "walk to" script?
[08:29] <salty-horse> hmm if I run several instances of ScummVM, and add a new game in one of them, close that window, then close the others, will my new game be in the final config file or not? :)
[08:32] <salty-horse> the scumm "objects" debug command should print object names. why doesn't it? will that overflow the ascii table? if so, truncate the object names
[08:44] <salty-horse> inconsistent scumm function name: getObjOrActorName vs getObjectOrActorXY
[08:46] <-- jamm left irc: Read error: Connection reset by peer
[08:47] --> jamm joined #scummvm.
[08:48] <-- _sev left irc: Quit: Leaving
[08:48] --> _sev joined #scummvm.
[08:48] <-- _sev left irc: Changing host
[08:48] --> _sev joined #scummvm.
[08:48] #scummvm: mode change '+o _sev' by ChanServ!ChanServ@services.
[09:01] --> Henke37 joined #scummvm.
[09:02] <salty-horse> should the debug output try to fit the old gui resolution or should I not worry about that? (scumm's "actors" table already overflows it)
[09:05] <-- Begasus left irc: Quit: Vision[0.9.7-H-20140108]: i've been blurred!
[09:09] <salty-horse> should I commit this? https://i.imgur.com/JV1gzyn.png
[09:10] <m_kiewitz> nice
[09:11] <waltervn> salty-horse: is that some kind of table-printing functionality for common?
[09:12] <salty-horse> waltervn, no, I just added a name field to the current debug tables in scumm's code
[09:13] <salty-horse> (the string is truncated if the name is too long)
[09:13] <waltervn> so this table printing is just custom scumm code?
[09:13] <wjp> is truncating really what you want in debugging output?
[09:14] <wjp> (of course I have no idea what the goal of the field is)
[09:17] <-- ced117 left irc: Quit: Quitte
[09:17] --> GitHub16 joined #scummvm.
[09:17] <GitHub16> [scummvm] Tkachov opened pull request #774: Add ScrollContainerWidget (master...the-container-box-pr2) https://git.io/vKvE7
[09:17] GitHub16 (GitHub16@192.30.252.45) left #scummvm.
[09:17] <salty-horse> wjp, I just want to print the name and keep the table from being messed up :) the name is useful so you know what the objects refer to
[09:17] <salty-horse> waltervn, yes. it's just printfs with "---------------"
[09:19] <rootfather> good morning peeps
[09:20] <rootfather> oh, github is counting the correct numbers of commits again
[09:20] <rootfather> no stupid 10.000+ anymore
[09:20] <-- frankyboy_ left irc: Read error: Connection reset by peer
[09:21] --> frankyboy_ joined #scummvm.
[09:23] --> GitHub106 joined #scummvm.
[09:23] <GitHub106> [scummvm] salty-horse pushed 1 new commit to master: https://git.io/vKvuJ
[09:23] <GitHub106> scummvm/master cfeae98 Ori Avtalion: SCUMM: Add actor/object names to debug output
[09:23] GitHub106 (GitHub106@192.30.252.45) left #scummvm.
[09:23] <-- Lightkey left irc: Ping timeout: 272 seconds
[09:26] <-- frankyboy_ left irc: Read error: Connection reset by peer
[09:27] Action: waltervn summons Strangerke
[09:27] --> frankyboy_ joined #scummvm.
[09:28] <-- frankyboy_ left irc: Remote host closed the connection
[09:36] <-- uruk-hai left irc: Quit: http://www.kiwiirc.com/ - A hand crafted IRC client
[09:36] --> Lightkey joined #scummvm.
[09:38] --> t0by joined #scummvm.
[09:38] #scummvm: mode change '+v t0by' by ChanServ!ChanServ@services.
[09:38] <t0by> PEOPLE
[09:38] Last message repeated 1 time(s).
[09:38] <t0by> http://superadventuresingaming.blogspot.it/2011/10/buffy-vampire-slayer-gbc.html
[09:38] <t0by> I want this.
[09:44] <m_kiewitz> https://www.youtube.com/watch?v=Lct6x-XqWrw
[09:53] <t0by> m_kiewitz: https://www.youtube.com/watch?v=fYpwEYFRrEQ
[09:54] <t0by> I always had the lingering suspicion it was a deliberate hommage
[10:18] --> Strangerke joined #scummvm.
[10:18] #scummvm: mode change '+o Strangerke' by ChanServ!ChanServ@services.
[10:18] <Strangerke> hi guys
[10:19] <rootfather> hi Strangerke
[10:19] <wanwan> hi
[10:20] <rootfather> hi wanwan
[10:20] <rootfather> btw, any news regarding the progress of Quux?
[10:22] --> uruk-hai joined #scummvm.
[10:22] #scummvm: mode change '+o uruk-hai' by ChanServ!ChanServ@services.
[10:23] <salty-horse> what's the status of gabriel knight 1 support?
[10:24] <wjp> rootfather: quux? why do you ask?
[10:24] <wanwan> rootfather: doing code cleanup, but the progress is quite slow
[10:24] <wjp> oh, I guess that name has multiple meanings
[10:26] <rootfather> wjp yeah wanwan used "Quux" as placeholder and I forgot the real name of the game :D
[10:28] <wjp> salty-horse: in how much detail do you want to know the status?
[10:29] <salty-horse> wjp, why isn't it "officially" supported? any big missing features?
[10:29] --> WinterGrascph joined #scummvm.
[10:32] <wjp> labelling any SCI2+ games officially supported isn't being discussed yet
[10:33] <wjp> but it's getting closer I suppose
[10:34] <wjp> I really wouldn't want to remove the freedom of breaking things yet by saying some games are "supported"
[10:35] <salty-horse> wjp, so GK1 is completable?
[10:35] <m_kiewitz> wjp: i still wonder why all sorts of speed issues do not happen in ScummVM even when I remove the current throttler. Well, it also seems even then the games run at quite a poor frame rate and at the same time ScummVM doesn't use up lots of CPU
[10:35] <m_kiewitz> no idea why that is
[10:35] <-- TMM left irc: Remote host closed the connection
[10:35] <m_kiewitz> i would like to properly patch out all of those bugs...
[10:36] <m_kiewitz> i think back then the timer bugs in gk1 happened also in ScummVM, that's why I added patches
[10:36] <wjp> salty-horse: I don't know if anyone has tried GK1 recently
[10:36] <m_kiewitz> i will gladly try a playthrough :P
[10:37] <salty-horse> someone mentioned it online, so I hacked a patch that removes gk1's "two spaces after a period" in the subtitles.
[10:37] <m_kiewitz> two spaces after a period? is than an issue from the original game? did they make the texts that way intentionally?
[10:40] --> WooShell joined #scummvm.
[10:42] <wjp> the original has that
[10:42] <wjp> it's not an uncommon typographical convention as far as I know (at least historically)
[10:46] <salty-horse> m_kiewitz, https://twitter.com/saltyhorse/status/743340964001964032
[10:47] <salty-horse> any objection to making it a configurable setting? :)
[10:48] <m_kiewitz> im sure some people will complain about it :P
[10:48] <m_kiewitz> but no problem, how did you do it? patching text resources?
[10:52] <salty-horse> I modified GfxText32::drawText to look behind and not draw a second consecutive space. (Not bullet-proof, since I didn't check for a period before the two spaces, and the game might use spaces may be used to align text)
[10:52] <wjp> I really don't think that's worth it
[10:55] <rootfather> Strangerke have you thought about the "snapshot-builds" I proposed a few days ago?
[10:55] <rootfather> do you think it is worth discussing on the mailing list?
[10:58] <wjp> I can't follow your reasoning at all
[10:59] <wjp> (reading back)
[10:59] <wjp> feel free to bring it up, but I can't see what the specific concept of weekly snapshots would improve
[11:02] <wjp> but it's great to think and discuss about how we can make things clearer in any case
[11:11] --> SylvainTV joined #scummvm.
[11:11] #scummvm: mode change '+o SylvainTV' by ChanServ!ChanServ@services.
[11:14] <WooShell> good meowning =^.^=
[11:37] <-- uruk-hai left irc: Quit: This computer has gone to sleep
[11:45] <-- ced117_ left irc: Ping timeout: 244 seconds
[11:45] --> uruk-hai joined #scummvm.
[11:45] #scummvm: mode change '+o uruk-hai' by ChanServ!ChanServ@services.
[11:46] --> ced117_ joined #scummvm.
[12:01] <-- Strangerke left irc: Ping timeout: 272 seconds
[12:15] --> criezy joined #scummvm.
[12:15] #scummvm: mode change '+o criezy' by ChanServ!ChanServ@services.
[12:26] #scummvm: mode change '+v WinterGrascph' by ChanServ!ChanServ@services.

[12:32] <salty-horse> all we care about here is Moonbase Commander :P
[12:37] Action: t0by Wololololololo
[12:38] <t0by> Oh look, all of a sudden it's full of AOE fans in here.
[12:38] <t0by> (or was it "unemalemanemadus"?)
[12:38] <t0by> (no, that was Warcraft II, Google tells me)
[12:44] <WooShell> WOLOLO!
[12:44] Action: rootfather Changes color
[12:44] Action: rootfather Rogan?
[12:53] <WooShell> hm. Van Helsing looks nice, but the reviews are abysmal..
[12:55] --> TMM joined #scummvm.
[12:57] <WooShell> and somehow I don't get this marketing approach of "similar items"
[12:57] <WooShell> "you have played minecraft and craft the world, so you really must like terraria and underworld and bla bla bla"
[12:58] <WooShell> NO, stupid Steam.. i've got that block-pushing genre covered. show me something i don't have a dozen versions of, yet..
[13:00] <rootfather> Same thing with Amazon
[13:00] <rootfather> "You purchased a new chair recently. Look, we have 100 more chairs in stock!"
[13:02] --> snover joined #scummvm.
[13:02] #scummvm: mode change '+o snover' by ChanServ!ChanServ@services.
[13:10] --> snow_bckspc joined #scummvm.
[13:17] <m_kiewitz> rootfather: sometimes amazon even shows me products that I already purchased
[13:39] --> ST joined #scummvm.
[13:39] #scummvm: mode change '+o ST' by ChanServ!ChanServ@services.
[13:41] <snover> salty-horse: it is probable that someone, somewhere, in a sci game script used spaces in a row to modify text alignment, because thats just how sierra rolled. also, some change like that one you made to drawText would probably break kEditText if a user inserted two spaces in a row
[13:42] <snover> as much as i love good typography i dont think that scummvm should really be in the business of trying to fix that particular thing, since its become a stylistic choice
[13:42] <salty-horse> snover, oh, I haven't considered user input :)
[13:43] <snover> (when i was in grade school they taught us to use two spaces between sentences; it wasnt until i did research on my own many years later that i learned this convention only existed because of monospacing typewriters)
[13:43] <salty-horse> it's mostly a hack to see how it would look :)
[13:43] <m_kiewitz> that's why i think text resource changes would be a better approach, even that could cause issues somewhere of course
[13:44] <salty-horse> m_kiewitz, I can "protect" it to only apply to text drawn in the "subtitles" area of the screen, when there's a conversation.
[13:44] <snover> im not sure&what is the problem that is being solved by doing this?
[13:46] <snover> salty-horse: ^
[13:47] <salty-horse> snover, it's not a problem. just addressing a nitpick :)
[13:47] <snover> a nitpick that sentences have two spaces between them?
[13:48] <salty-horse> yes. it's not something I strongly think should be officially supported. I was just asking to see if there's interest
[13:48] <snover> in my mind, changing that in the sci engine would be like nitpicking that they used american english instead of commonwealth english and changing all the words that are different in CwE
[13:50] <salty-horse> Yeah, it's not an obvious mistake like the typo in Loom.
[13:52] <salty-horse> Come to think of it, there's no "preserve typos" flag :)
[13:52] --> ced117 joined #scummvm.
[13:53] <m_kiewitz> and also no "preserve game bugs" flag :P
[13:53] <wjp> despite the words being similar, "typo" and "typographic style" are really not on the same level
[13:53] <m_kiewitz> i was actually asked about exactly that. it seems people like crashing games
[13:53] <salty-horse> should the Loom typo fix be documented here? https://www.scummvm.org/compatibility/DEV/loom/
[13:54] <snover> not any more than every bug fix is documented there, i would think.
[13:54] <salty-horse> m_kiewitz, it's important for game preservation to allow for the game's original behavior.
[13:55] <salty-horse> snover, I really do think every fixed bug should be documented as well :)
[13:55] <snover> they are documented, but how many places do you need it to be documented?
[13:55] <snover> i mean at least in sci engine, workarounds and script patches are all (or mostly all) commented on
[13:55] <salty-horse> snover, accessible to laypeople, not in code.
[13:56] <snover> not sure why laypeople would care about this.
[13:57] <m_kiewitz> then I guess we should implement "Oops" / "You tried something that we didn't think of. Try taking a different approach to the situation. SCI version 0.000.685" message box and then close down ScummVM :P
[13:58] <snover> i personally feel that the approach to bug fixing is good the way it is right now.
[13:59] <m_kiewitz> i documented a few AGI bugs on the wiki, doing that now for SCI would be quite a bit of work
[13:59] <t0by> "preserve lame puzzles" flag
[14:00] <t0by> "[x] Including the cat-moustache-ID puzzle (EXPERT ONLY)"
[14:01] <m_kiewitz> [x] include the goat 'puzzle' (ONLY IN CASE YOU WANT TO GO CRAZY)"
[14:04] <t0by> "[x] Enable support for pretty much any King's Quest game (m_kiewitz, is that you?)"
[14:04] <t0by> :P
[14:06] <m_kiewitz> [x] Enable support for special keyboard shortcut, that will use everything in the inventory on everything in the current room (that's how you solve most puzzles anyway, so why not use this shortcut)
[14:06] <salty-horse> > use all on all
[14:08] <t0by> https://www.youtube.com/watch?v=8PIJcwsAvNg#t=10m02s
[14:08] <t0by> lest we forget
[14:09] <t0by> arf
[14:09] <t0by> I think I have... forgotten how the goat puzzle was supposed to work
[14:10] <t0by> A much better goat puzzle: http://www.cafepress.co.uk/+cartoon-goat+puzzles
[14:12] <m_kiewitz> t0by: that's the guy that played apple IIgs games too, right? I remember that voice from some let's plays that I watched when playing through AGI games
[14:20] <snover> m_kiewitz: do you know offhand why keyboard events in sci might get dropped if you type too quickly?
[14:20] <snover> i just remembered i wanted to ask that.
[14:21] <m_kiewitz> they are?
[14:21] <t0by> m_kiewitz, perhaps?
[14:22] <m_kiewitz> when using the original interpreter too?
[14:22] <snover> nope, just scummvm
[14:22] <m_kiewitz> i never experienced that, but maybe i simply do not type fast enough :P
[14:22] <snover> its kEditText; i havent debugged yet to see if there was some dumb thing i did that would cause this to happen
[14:22] <snover> its just something on my to-do list.
[14:22] <m_kiewitz> sci32 + sci16, or just sci32?
[14:23] <snover> good question. let me try sci16 real quick&
[14:24] <snover> looks like its my problem :)
[14:26] <m_kiewitz> or well maybe VM for sci32
[14:26] <m_kiewitz> at least in theory it could also be some tiny difference in event handling
[14:28] <snover> kEditText runs its own event loop and has to do a speed throttle
[14:29] <m_kiewitz> oh well then yeah
[14:29] <m_kiewitz> weird that sierra did that though. in sci16 it works differently
[14:29] <m_kiewitz> so you can never enter text while the game is actually running in sci32?
[14:30] <snover> the kernel call doesnt return back to the VM until the text field is unfocused by clicking away from it or hitting one of the unfocus keys
[14:31] <m_kiewitz> really weird
[14:34] <wjp> in the main scummvm they can/could get eaten by tooltips stealing focus
[14:34] <wjp> not sure if that's fixed yet
[14:34] <wjp> s/scummvm/scummvm GUI/
[14:35] <wjp> but I suspect that's not really relevant here
[14:35] <snover> indeed.
[14:38] <-- criezy left irc: Quit: criezy
[14:43] --> criezy joined #scummvm.
[14:43] #scummvm: mode change '+o criezy' by ChanServ!ChanServ@services.
[14:49] <m_kiewitz> wow, qfg4
[14:49] <m_kiewitz> https://www.gog.com/forum/quest_for_glory_series/qfg4_cant_restore_game
[14:49] <m_kiewitz> it seems whatever is happening, it's quite broken atm
[14:50] <m_kiewitz> (not ScummVM, but playing in dosbox using the original interpreter)
[14:50] <snover> ive seen similar problems with the savegame engine in phant1
[14:50] <snover> i dont know what it is doing, but obviously something gets corrupted and then it get stuck in a 100% cpu VM loop
[14:50] <snover> the two might be unrelated, but it sure is annoying.
[14:51] <m_kiewitz> in ScummVM too?
[14:51] <snover> yeah
[14:51] <m_kiewitz> that should get figured out and patched
[14:51] <snover> it might actually be *only* in scummvm, i dont know yet
[14:51] <m_kiewitz> never played sci32 games except for Gabriel Knight 1
[14:51] <m_kiewitz> those people are surely playing in dosbox
[14:52] <m_kiewitz> so it has to be the scripts/interpreter/whatever
[14:52] <m_kiewitz> gk1 worked fine, still gk1 is probably the least buggy sci32 game so that makes sense
[14:52] <m_kiewitz> qfg4 is quite buggy
[14:53] <m_kiewitz> can you enter debugger when it's at 100% CPU?
[14:53] <snover> i think so. at the worst i can enter the xcode debugger
[14:53] <snover> might wait for wjp to finish doing what hes doing with savegame code before looking into it, it might be something that is solved just by fixing the virtual slots thing
[14:54] <snover> there is some special code for phant1 in scummvm save code
[14:54] <m_kiewitz> still then it would work when using the original interpreter
[14:54] <m_kiewitz> and it seems it doesn't
[14:56] --> Strangerke joined #scummvm.
[14:56] <wjp> (fixing which virtual slots thing?)
[14:58] <wjp> (as an aside: that word 'virtual' is really over-used in kfile.cpp... virtual slots, virtual handles, virtual index files...)
[14:58] <snover> its virtually endless
[15:00] <wjp> but so far the only virtual slots I'm looking at are the Torin autosave slots
[15:00] <snover> PHANTASMAGORIA_SAVEGAME_INDEX
[15:00] <snover> that thing
[15:01] <wjp> I'm rewriting at least the backend of that
[15:01] <snover> hopefully once thats done, whatever corruption that causes phant1 to get stuck in a loop after entering a name ceases to be a thing
[15:02] <snover> though it could also be a problem with kString, or just a totally unrelated problem
[15:02] <snover> i guess well figure it out eventually :)
[15:04] <wjp> the current implementation is likely writing uninitialized memory to file; no idea what happens with that when it's read back
[15:05] <wjp> but I'm trying to avoid looking at the current implementation too much
[15:05] <snover> that doesnt sound too good& :)
[15:06] <snover> anyway, im going to keep working on VMDs. another thing where i am impressed they work at all right now, the opcode for closing the vmd file was being interpreted as the opcode for playing the vmd
[15:09] <snover> new question for the group: do i bother to implement those awful blacklines, or just always do pixel double?
[15:14] <waltervn> IMHO we should implement the blacklines
[15:16] <snover> ok
[15:20] <snover> implement wins i guess :)
[15:29] <waltervn> \o/
[15:35] <-- t0by left irc: Ping timeout: 246 seconds
[15:36] --> t0by joined #scummvm.
[15:36] #scummvm: mode change '+v t0by' by ChanServ!ChanServ@services.
[15:38] <t0by> > virtually endless
[15:38] <t0by> well played, snover
[15:38] <t0by> well played
[15:38] <snover> thanks t0by!
[15:48] <-- t0by left irc: Ping timeout: 246 seconds
[15:59] <-- criezy left irc: Quit: criezy
[16:00] --> criezy joined #scummvm.
[16:00] #scummvm: mode change '+o criezy' by ChanServ!ChanServ@services.
[16:00] <-- criezy left irc: Client Quit
[16:17] --> plipfishy joined #scummvm.
[16:24] --> Cruel` joined #scummvm.
[16:46] <-- jamm left irc: Read error: Connection reset by peer
[16:47] --> jamm joined #scummvm.
[16:50] plipfishy (whore@apostlegayrape.space) left #scummvm ("WeeChat 1.5-dev").
[16:58] --> Littleboy joined #scummvm.
[16:58] #scummvm: mode change '+o Littleboy' by ChanServ!ChanServ@services.
[17:11] <-- uruk-hai left irc: Quit: Leaving
[17:28] <-- Tkachov left irc: Remote host closed the connection
[17:32] --> Tkachov joined #scummvm.
[17:32] #scummvm: mode change '+v Tkachov' by ChanServ!ChanServ@services.
[17:34] <wjp> ok, there are definitely bugs in the current phantasmagoria save file handling
[17:34] <wjp> and my rewrite replaces them by a new and exciting different set of bugs
[17:35] <snover> ha :)
[17:35] <snover> if we just keep piling the bugs up high enough, it just might work&!
[17:39] <wjp> oh, that was less exciting than it appeared
[17:40] <wjp> it seems fairly happy now
[17:43] <snover> sweet!
[17:43] <wjp> so in theory this should also work for the SWAT FileIO problem
[17:44] <wjp> which was actually the reason I started looking at this
[17:45] <snover> awesome. that is the most important game, after all :)
[17:46] <wjp> clearly
[17:48] <wjp> of course I don't seem to actually have it installed
[17:48] <snover> that is shocking.
[17:49] <wjp> I know, right
[17:49] <snover> i dont know how much of the pqswat trouble at the beginning was caused by save issues and how much is caused by incomplete kRobot. it gets stuck on a kRobot one it gets to the second part of the title screen
[17:58] <wjp> I get to the main menu, and the first screen after "new carreer", and then it hits a signature error on kAddToFront
[17:59] <wjp> which is much further than without these changes it seems
[18:00] <snover> indeed, yes. i cant even skip the title screen to get anywhere useful
[18:04] <m_kiewitz> wjp: yeah, SWAT is so awesome. It's the most important SCI game ever made :P
[18:06] <snover> sometimes, i feel like scaling issues are going to haunt me until the end of time&
[18:19] <salty-horse> SWAT is a SCI game?
[18:19] <salty-horse> (only police quest game I own and have played)
[18:20] <m_kiewitz> the horrible first one, yes
[18:20] <m_kiewitz> SWAT4 is great and uses afaik the Half-Life engine
[18:20] <m_kiewitz> and you should play the others
[18:21] <m_kiewitz> especially Police Quest 1+2
[18:25] <salty-horse> I don't like games where you can die frequently or get stuck without you knowing it
[18:26] <salty-horse> I played PQ1 up to the part where you get in the car without it exploding
[18:28] <m_kiewitz> lol
[18:28] <m_kiewitz> In PQ2+3 I think you can't die that often.
[18:28] <snover> sierra learned their lesson eventually and most sci32 games have a try again button on the death screen
[18:28] <m_kiewitz> and PQ3 got a Jan Hammer soundtrack
[18:29] <snover> (or maybe all of them, i dont know)
[18:29] <m_kiewitz> In Gabriel Knight 1 you also can't die that often, but i think it's game over when you die
[18:30] <m_kiewitz> at least if I remember it right, haven't really played it for ages now
[18:35] --> frankyboy_ joined #scummvm.
[18:36] <wjp> hurray, 6 files changed, 146 insertions(+), 268 deletions(-)
[18:36] <wjp> more features, less code :-)
[18:36] <snover> 👍
[18:37] <m_kiewitz> uh oh
[18:37] <snover> The framerate on these VMDs is so low I cant tell if the throttling is bad or if the video is just that choppy naturally
[18:38] <snover> m_kiewitz: whats wrong?
[18:39] <wjp> (oh, and one less use of the word virtual)
[18:39] <m_kiewitz> the more changes, the more new bugs
[18:40] <m_kiewitz> well not always of course
[18:40] <snover> https://zetafleet.com/i/57780aad3d2fb.png hmm. not quite right yet.
[18:41] <m_kiewitz> if the videos were only available at a higher resolution :/
[18:41] <wjp> m_kiewitz: it's mainly replacing VirtualIndexFile, which for some reason re-implements most of the Common::Read/WriteStream infrastructure
[18:42] <m_kiewitz> then it shouldn't matter that much, that one was used by sci32 only, right?
[18:42] <wjp> yes
[18:42] <snover> generally speaking i think i can say that the new sci32 stuff reduces the number of bugs, most of the time, except for when i accidentally break the build :)
[18:42] <wjp> sci32 only
[18:42] <snover> so, er, some of the time.
[18:42] <m_kiewitz> my AGI rewrite didn't create that many issues as well, still...
[18:43] <m_kiewitz> and well the new sci32 stuff is of course way better than the old, because the old was guess work
[18:44] <m_kiewitz> that's basically writing new code, the old one was basically only useful to get a feel about how things somewhat work and was inaccurate as hell
[18:50] <snover> wjp: i think you can that PR #741 too
[18:50] <snover> land that*
[18:54] <snover> m_kiewitz: as far as the video goes& in the future, it is an area where upgrading the display to hi-colour and adding some bilinear filter would really boost the quality&
[18:55] <m_kiewitz> VMDs are hi-color and not 256 colors only?
[18:55] <snover> no
[18:56] <snover> theyre 256, but an interpolation would need more than that to look good
[18:58] <snover> hmmmmmm.
[18:58] <m_kiewitz> are the situations, where VMDs need to blend in with the background?
[18:58] <m_kiewitz> *there
[18:59] <waltervn> wjp did some tests with hi-colour for the screen transitions didn't he?
[18:59] <snover> its possible, games can ask the VMD kernel call to yield back to the VM
[18:59] <snover> waltervn: i remember something like that, now that you mention it.
[18:59] <snover> and the frames are drawn into a bitmap screenitem
[18:59] <snover> so they can be given priority and placement
[19:00] <snover> however, i am not sure if this was ever actually used or not, there are a ton of seemingly unused subops
[19:02] <snover> only 12 out of 32 subops are ever used (and actually, it might be only 9i still dont know which vm opcodes leave things on the stack, so may have read script disassembly a little bit wrong)
[19:04] <-- m_kiewitz left irc: Quit: technology isn't intrinsically good or evil. It's how it's used. Like the Death Ray.
[19:06] --> m_kiewitz joined #scummvm.
[19:06] <-- m_kiewitz left irc: Changing host
[19:06] --> m_kiewitz joined #scummvm.
[19:06] #scummvm: mode change '+o m_kiewitz' by ChanServ!ChanServ@services.
[19:09] <snover> oh, phantasmagoria, you and your weirdo screen size&
[19:10] <snover> i just&dont&understand.
[19:57] --> GitHub10 joined #scummvm.
[19:57] <GitHub10> [scummvm] wjp opened pull request #775: SCI32: Improve FileIO read-write support (master...sci32_readwrite) https://git.io/vKvFr
[19:57] GitHub10 (GitHub10@192.30.252.40) left #scummvm.
[20:00] <-- rootfather left irc: Read error: Connection reset by peer
[20:02] <snover> wjp: are you on case-sensitive filesystem
[20:02] <snover> ?
[20:02] <wjp> yes
[20:03] <wjp> am I doing stupid things?
[20:03] <snover> no i was just curious about the lowercase swat.dat
[20:03] <snover> i thought the game made it uppercase
[20:04] <snover> id have to look at wrapFilename to see if that is a problem i guess
[20:06] <-- frankyboy_ left irc: Read error: Connection reset by peer
[20:07] --> frankyboy_ joined #scummvm.
[20:07] <wjp> the kernel function gets lowercase swat.dat here
[20:07] <snover> interesting. alright.
[20:14] <-- frankyboy_ left irc: Read error: Connection reset by peer
[20:14] --> frankyboy_ joined #scummvm.
[20:21] <-- frankyboy_ left irc: Read error: Connection reset by peer
[20:22] --> frankyboy_ joined #scummvm.
[20:24] --> snover1 joined #scummvm.
[20:24] #scummvm: mode change '+o snover1' by ChanServ!ChanServ@services.
[20:24] <-- snover left irc: Disconnected by services
[20:24] Nick change: snover1 -> snover
[20:24] <snover> alright. thanks internet connection.
[20:24] <snover> https://zetafleet.com/i/5778230bcdd66.jpg alllllmost.
[20:30] --> go2m joined #scummvm.
[20:32] <-- frankyboy_ left irc: Read error: Connection reset by peer
[20:33] <-- goom left irc: Ping timeout: 276 seconds
[20:33] --> frankyboy_ joined #scummvm.
[20:38] <-- frankyboy_ left irc: Read error: Connection reset by peer
[20:40] --> frankyboy_ joined #scummvm.
[20:46] <-- frankyboy_ left irc: Read error: Connection reset by peer
[20:47] --> frankyboy_ joined #scummvm.
[20:48] <-- ny00123 left irc: Quit: Leaving
[20:51] <-- frankyboy_ left irc: Read error: Connection reset by peer
[20:52] --> frankyboy_ joined #scummvm.
[20:57] --> GitHub4 joined #scummvm.
[20:57] <GitHub4> [scummvm] wjp pushed 1 new commit to master: https://git.io/vKvAz
[20:57] <GitHub4> scummvm/master 79b0309 Willem Jan Palenstijn: SCI32: Fix dropping events in kEditText
[20:57] GitHub4 (GitHub4@192.30.252.40) left #scummvm.
[20:57] <wjp> snover: ^
[20:58] <snover> aww. thanks!
[21:00] <snover> i seem to keep picking calls to implement that have their own event loops&
[21:01] <-- frankyboy_ left irc: Read error: Connection reset by peer
[21:02] --> frankyboy_ joined #scummvm.
[21:02] --> NuSuey joined #scummvm.
[21:02] <snover> also i think there is something wrong with the altify function in the event system
[21:02] <snover> when i try using debug commands they often activate from the wrong keys
[21:02] <snover> or wont activate using the right keys
[21:03] <snover> hot rectangles are wedged into the event system, so if i am the one that implements that ill probably look into it then.
[21:06] <-- frankyboy_ left irc: Read error: Connection reset by peer
[21:07] --> frankyboy_ joined #scummvm.
[21:10] <-- frankyboy_ left irc: Read error: Connection reset by peer
[21:17] <-- salty-horse left irc: Remote host closed the connection
[21:22] --> frankyboy_ joined #scummvm.
[21:23] <m_kiewitz> snover: im quite sure that alitfy is correct for sci16
[21:23] <m_kiewitz> but maybe something got changed in sci32
[21:24] <m_kiewitz> it was a nightmare to get all of that right, especially because for example linux + windows do not give us the same keys in those cases
[21:25] <m_kiewitz> and then there was afaik also some change between sci0 and sci1/1.1
[21:26] <snover> could be a macos problem too perhaps?
[21:32] <m_kiewitz> im not sure if another dev uses MacOS
[21:32] <m_kiewitz> in theory yes, well a SDL problem
[21:32] <m_kiewitz> SDL is totally broken, it doesn't return the same keys on all platforms
[21:33] <m_kiewitz> there are all sorts of special cases
[21:33] <m_kiewitz> you could for example try to run quest for glory 1 sci0
[21:33] <m_kiewitz> and then check if all the hotkeys work
[21:33] <snover> is SDL2 better?
[21:33] <m_kiewitz> if they do, then it's a sci32 problem. if they don't, then it may be a MacOS issue
[21:33] <snover> good suggestion :) i will try qfg1
[21:33] <m_kiewitz> haven't used that yet
[21:34] <m_kiewitz> i should have take notes on all hotkeys, i think you could also try larry 1 sci to check if Alt-X works
[21:34] <m_kiewitz> *taken
[21:34] <m_kiewitz> ah wait, it's Ctrl-Alt X for larry 1 sci
[21:35] <m_kiewitz> but that's also something to check
[21:35] <m_kiewitz> Ctrl-Alt X skips age check right at the start, well it shoul
[21:35] <m_kiewitz> should
[21:38] <m_kiewitz> oh right, and at least original AGI also accepted all sorts of special characters, but didn't support them correctly
[21:38] <snover> ah, sweet. intro videos in shivers now arent messing up the screen :]
[21:38] <m_kiewitz> so it showed bad characters or didn't show those at all. I added filtering to text input in ScummVM AGI, so that those characters are filtered and not used at all
[21:38] <m_kiewitz> except for the Russian translations, that use such special characters (and those work)
[21:39] <snover> i'm trying qfg1 now, then i need to go for a while
[21:40] <snover> m_kiewitz: what is a hotkey in qfg1 what i can try?
[21:40] <m_kiewitz> i think there also were changes between sci0 and sci1, which means you should also try qfg1 vga
[21:40] <m_kiewitz> Strg-I should show inventory
[21:41] <m_kiewitz> ah qfg1 sci0 uses lots of Ctrl-hotkeys
[21:41] <m_kiewitz> let me think, which game used Alt-hotkeys....
[21:43] <m_kiewitz> well you could in any case try larry 1 AGI
[21:43] <m_kiewitz> and try to skip age check
[21:43] <m_kiewitz> i think larry 1 AGI used Alt-X
[21:43] <m_kiewitz> i really should have documented all of that when i tried everyything out
[21:43] <snover> seems i did not copy my larry collection so ill have to try that one later
[21:44] <snover> ctrl+whatever seems to be ok for qfg1old
[21:49] <snover> the skip check in larry1vga works
[21:49] <snover> though i also think x and i were the two keys that *did* work already
[21:49] <snover> anyway. who knows. its not that important right now. :)
[21:50] <m_kiewitz> which one don't work?
[21:50] <snover> another good question
[21:52] <snover> in GK1CD, alt+i opens an unlabelled box with a text input, alt+q shows cast information (should be alt+s), alt+x gives help about the commands (should be alt+h)
[21:52] <snover> most of the reset of them do not work
[21:52] <snover> (i dont expect all of them to work, of course :))
[21:53] <snover> alt+comma says "on" and then "off"
[21:53] <snover> so theyre just like&all remapped all over the place weirdly
[21:54] <m_kiewitz> ah it seems Alt-keys are used for built in game debug functions
[21:54] <snover> yes, and i would love to get the second page of inventory items for sq6 to show up :)
[21:54] <m_kiewitz> ah right
[21:55] <m_kiewitz> do you have king's quest 4?
[21:55] <snover> yes
[21:55] <-- frankyboy_ left irc: Quit: #E>6C O >B 20A
[21:56] <snover> anyway i really need to go, just let me know if i am supposed to try something or whatever and i can do it when i get back :) thanks
[21:57] <m_kiewitz> hmm not sure if kq4 sci even has the debug mode
[21:57] <m_kiewitz> you got police quest 2 too?
[21:58] <m_kiewitz> ah im stupid
[21:58] <m_kiewitz> you can of course use qfg1 sci0
[21:58] <m_kiewitz> enter "razzle dazzle root beer" as command
[21:58] <m_kiewitz> that will enable debug mode
[21:59] <m_kiewitz> Alt-E - shows ego coordinates
[21:59] <m_kiewitz> Alt-P - switches to priority screen
[21:59] <m_kiewitz> Alt-C - switches to control screen
[22:00] <m_kiewitz> ah i just see, Alt-V doesn't seem to work, that one should switch to visual screen (display screen)
[22:00] <m_kiewitz> no idea if the key event doesn't work or if the debug command isn't implemented correctly
[22:01] <m_kiewitz> Alt-S also doesn't work, although it seems ScummVM grabs that even to save screenshots :( whoever implemented that one, urgh
[22:01] <m_kiewitz> Alt-T to teleport, works for me
[22:02] <m_kiewitz> Alt-M shows memory
[22:02] <m_kiewitz> Alt-B shows money + changes money
[22:02] <m_kiewitz> Alt-X should set all stats to 80
[22:03] <m_kiewitz> Alt-K changes skill points
[22:03] <m_kiewitz> everything works for me except Alt-S + Alt-V
[22:04] <m_kiewitz> the debug command is also not the same for all versions, razzle dazzle root beer works for my version dated 11.10.1989
[22:04] <m_kiewitz> there was another one, but i cant recall it atm
[22:05] --> MrCuddle joined #scummvm.
[22:05] <-- Cheeseness left irc: Ping timeout: 244 seconds
[22:06] MrCuddle (Simpa@h-88-94.a230.priv.bahnhof.se) left #scummvm.
[22:10] <m_kiewitz> ah nvm, Alt-V actually works. debug mode simply shuts down when you exit a room
[22:11] <m_kiewitz> So only Alt-S doesn't work and that's probably because of the screenshot functionality, which should probably get switched to another shortcut maybe Ctrl-Alt-S
[22:12] <m_kiewitz> wjp: what do you think about screenshot functionality?
[22:27] <-- wanwan left irc: Ping timeout: 264 seconds
[22:30] <-- WinterGrascph left irc: Ping timeout: 260 seconds
[23:09] <snover> m_kiewitz: imback, reading logs&
[23:11] <snover> m_kiewitz: in qfg1 sci0, alt+e works, nothing else works
[23:11] <snover> well, alt+q works too
[23:12] <snover> alt+i gives me id number of the object? so that might be inventory and working
[23:12] <snover> alt+. gives me ego coordinates too
[23:12] <m_kiewitz> wait, nothing else works, but then all sorts of things do work?!
[23:12] --> _sev_ joined #scummvm.
[23:12] <-- _sev_ left irc: Changing host
[23:12] --> _sev_ joined #scummvm.
[23:12] #scummvm: mode change '+o _sev_' by ChanServ!ChanServ@services.
[23:12] <m_kiewitz> which ones don't work?
[23:12] <snover> the ones you mentioned dont work
[23:13] <snover> alt+p no, alt+c no, alt+t no, alt+m no, alt+b no, alt+k no
[23:13] <snover> alt+x no
[23:13] <-- _sev left irc: Ping timeout: 258 seconds
[23:13] <m_kiewitz> omg, so I guess SDL causes those
[23:14] <m_kiewitz> i think someone tried all of those on Linux for me (well Linux running on x86)
[23:14] <m_kiewitz> alt-v/alt-c works?
[23:14] <snover> nope
[23:15] <m_kiewitz> urgh
[23:15] <m_kiewitz> can you please check what you get right at the start of the keyboard event from ScummVM?
[23:16] <m_kiewitz> i really wonder why a few work for you it makes no sense
[23:16] <m_kiewitz> if we could only dump SDL for good
[23:16] <m_kiewitz> it's so broken
[23:16] <snover> someone is going to murder me if i dont go afk for a few minutes. i will let you know as soon as i get back. sorry for the inconvenience
[23:16] <m_kiewitz> kk
[23:30] <snover> so&lets see here.
[23:33] <snover> m_kiewitz: ev.key.keysym.scancode = SDL_SCANCODE_LALT ev.key.keysym.mod = 0, then ev.key.keysym.scancode SDL_SCANCODE_X, ev.key.keysym.mod = 256
[23:33] <m_kiewitz> on which keypress?
[23:33] <snover> alt+x
[23:33] <m_kiewitz> Alt-x?
[23:35] <snover> then, event.kbd.keycode = KEYCODE_x and event.kbd.flags = 2
[23:35] <snover> so i see nothing wrong in what SDL says
[23:37] <m_kiewitz> .ascii is?
[23:37] <snover> 8776
[23:37] <m_kiewitz> o_O
[23:37] <m_kiewitz> WTF
[23:38] <m_kiewitz> it should be 120
[23:38] <m_kiewitz> what is SDL doing...
[23:38] <snover> SDL doesnt do that, mapKey does that&
[23:39] <m_kiewitz> mapKey?
[23:39] <m_kiewitz> in the backend?
[23:39] <snover> yes
[23:39] <snover> SDL keysym.sym is 120
[23:39] <snover> so we can stop blaming sdl now ;)
[23:39] <m_kiewitz> so the backend is doing it? why? who implemented that?
[23:40] <m_kiewitz> and why was it implemented for some Alt- keys only?!
[23:40] <snover> // FIXME: This is all a bit fragile because in mapKey we derive the ascii [&]
[23:40] <snover> fixme, indeed.
[23:41] <m_kiewitz> ARGH
[23:41] <m_kiewitz> who why
[23:41] <m_kiewitz> and why is it MacOS only?
[23:42] <m_kiewitz> ah, you are in fact using SDL2 then it seems
[23:42] <snover> yes, SDL2 here
[23:42] <m_kiewitz> so i guess SDL2 support is broken then?!
[23:42] <m_kiewitz> do we use SDL2 officially atm? i hope not
[23:44] <m_kiewitz> SDL: Add experimental support for SDL2.
[23:44] <m_kiewitz> experimental indeed :P
[23:44] <snover> it seems to work better than SDL1 in every other way
[23:45] <m_kiewitz> well keyboard input is a bit important :P
[23:45] <snover> eh
[23:45] <snover> questionable ;)
[23:45] <m_kiewitz> so i guess you should also try all the other keys, like all Ctrl- keys, Alt- keys and Ctrl-Alt-keys and so on
[23:46] <m_kiewitz> i wonder why none of that was tested
[23:48] <m_kiewitz> but wait, mapKey does this?
[23:48] <snover> just because something is broken in one case doesnt mean it wasnt tested :)
[23:48] <m_kiewitz> is there another mapKey? because the one that is active for me doesn't seem to be able to produce that mess
[23:49] <-- WooShell left irc: Quit: Zu gotdy od mpy nrmy stpimf. Zu drvpmf zrsmd aogy jrt iq pt viy jrt yp yjr htpimf.
[23:49] <m_kiewitz> hmm, i guess i should look into switching to SDL2 and grabbing some SDL2 library
[23:50] <snover> the ascii value is 120 when there is no modifier and 8776 when there is an alt modifier
[23:50] <m_kiewitz> 120 is 'x'
[23:51] <m_kiewitz> and it should be like that all the time
[23:53] <snover> convUTF8ToUTF32 is doing the right thing, the value in 'n' is H
[23:54] <snover> because on mac os, alt+x = H
[23:55] <snover> so everything is doing the right thing here, the character *is* 8776
[23:56] <m_kiewitz> can you switch to SDL1 just to check what happens when using the older one?
[23:56] <snover> uh. perhaps. lets see.
[23:56] <snover> one moment.
[23:56] <snover> compiling.,
[23:56] <m_kiewitz> .ascii should be "x", that's what it should be. That's always the nightmare, because SDL doesn't return the same on all platforms, which it simply should
[23:57] <snover> no, its being retrieved from a text input API
[23:57] <snover> the text being input when you hold alt and press X in mac os is H
[23:58] <snover> if you want the scancode of the key, use the scancode value
[23:58] <snover> thats still valid
[23:58] <snover> scancode SDL_SCANCODE_X, mod 256.
[23:58] <snover> ScummVM event.keycode KEYCODE_x, flags 2
[23:59] <snover> use of 'ascii' property in this case seems to be incorrect use of the event api.
[00:00] --- Sun Jul 3 2016