ScummVM IRC Log - 12 Jul 2017

[Back to Index]

[00:00] --> omer_mor_ joined #scummvm.
[00:03] <-- omer_mor left irc: Ping timeout: 268 seconds
[00:25] --> girafe joined #scummvm.
[00:35] <-- SylvainTV left irc: Read error: Connection reset by peer
[00:55] <-- girafe left irc: Read error: Connection reset by peer
[01:27] <-- ComradeAnderson left irc: Ping timeout: 246 seconds
[01:28] --> ComradeAnderson joined #scummvm.
[01:31] <-- Korak left irc: Ping timeout: 268 seconds
[01:35] --> Joefish_ joined #scummvm.
[01:35] #scummvm: mode change '+v Joefish_' by ChanServ!ChanServ@services.
[01:35] <-- Joefish left irc: Ping timeout: 240 seconds
[01:44] --> Korak joined #scummvm.
[01:58] <-- Dominus left irc: Ping timeout: 260 seconds
[01:58] --> Dominus joined #scummvm.
[02:01] --> GitHub185 joined #scummvm.
[02:01] <GitHub185> [scummvm] dreammaster pushed 1 new commit to master: https://git.io/vQDDA
[02:01] <GitHub185> scummvm/master 51ebb01 Paul Gilbert: TITANIC: Fix Titania's nose remaining present after being picked up...
[02:01] GitHub185 (GitHub185@192.30.252.42) left #scummvm.
[02:07] --> GitHub122 joined #scummvm.
[02:07] <GitHub122> [scummvm] dreammaster pushed 1 new commit to master: https://git.io/vQDyk
[02:07] <GitHub122> scummvm/master 2bfe136 Paul Gilbert: TITANIC: Fix savegame names not showing in Continue dialog
[02:07] GitHub122 (GitHub122@192.30.252.41) left #scummvm.
[02:36] <-- dreammaster left irc:
[03:57] <-- Lightkey left irc: Ping timeout: 276 seconds
[04:09] --> Lightkey joined #scummvm.
[04:22] <-- omer_mor_ left irc: Ping timeout: 240 seconds
[04:25] --> omer_mor joined #scummvm.
[05:05] --> Strangerke_ joined #scummvm.
[05:07] <-- Strangerke left irc: Ping timeout: 240 seconds
[05:07] Nick change: Strangerke_ -> Strangerke
[05:28] <snover> hm. ida dosbox is not loving this phant1 executable (and/or i used the wrong shift delta when relocating the image). it looks like probably some IPC issue since nothing is CPU-bound but it still takes several minutes just to get the mouse cursor to draw in dosbox at game startup. i guess i will see tomorrow if i know enough to debug this.
[06:38] <-- balrog left irc: Ping timeout: 246 seconds
[06:51] --> balrog joined #scummvm.
[07:35] --> waltervn joined #scummvm.
[07:35] #scummvm: mode change '+o waltervn' by ChanServ!ChanServ@services.
[07:36] <waltervn> morning
[07:41] <wjp> snover: I seem to recall I got it working at least once before (for that fast-forward animation glitch), so hopefully we can figure it out. I vaguely recall some strangeness if I tried to break into the debugger too soon after startup
[07:42] <wjp> morning
[07:53] <-- LittleToonCat left irc: Remote host closed the connection
[08:19] --> Axy joined #scummvm.
[08:19] <-- Axy left irc: Changing host
[08:19] --> Axy joined #scummvm.
[08:33] <Simei> Morning
[08:34] <-- Axy left irc: Ping timeout: 260 seconds
[08:43] --> t0by joined #scummvm.
[08:43] #scummvm: mode change '+o t0by' by ChanServ!ChanServ@services.
[09:10] <-- t0by left irc: Ping timeout: 255 seconds
[09:20] --> t0by joined #scummvm.
[09:20] #scummvm: mode change '+o t0by' by ChanServ!ChanServ@services.
[09:36] --> ajax16384 joined #scummvm.
[09:36] #scummvm: mode change '+o ajax16384' by ChanServ!ChanServ@services.
[10:03] <t0by> Yawn
[10:03] <t0by> Morning
[10:10] --> criezy|Work joined #scummvm.
[10:10] #scummvm: mode change '+o criezy|Work' by ChanServ!ChanServ@services.
[10:42] --> _sev|work joined #scummvm.
[10:42] #scummvm: mode change '+o _sev|work' by ChanServ!ChanServ@services.
[11:00] <-- ajax16384 left irc: Quit: Leaving
[11:25] <-- _sev|work left irc: Quit: This computer has gone to sleep
[11:28] <-- waltervn left irc: Quit: Leaving
[11:29] --> _sev|work joined #scummvm.
[11:29] <-- _sev|work left irc: Changing host
[11:29] --> _sev|work joined #scummvm.
[11:29] #scummvm: mode change '+o _sev|work' by ChanServ!ChanServ@services.
[11:34] <-- t0by left irc: Read error: Connection reset by peer
[11:37] --> t0by joined #scummvm.
[11:37] #scummvm: mode change '+o t0by' by ChanServ!ChanServ@services.
[11:39] --> waltervn joined #scummvm.
[11:39] #scummvm: mode change '+o waltervn' by ChanServ!ChanServ@services.
[11:42] <-- t0by left irc: Remote host closed the connection
[11:43] --> t0by joined #scummvm.
[11:43] #scummvm: mode change '+o t0by' by ChanServ!ChanServ@services.
[11:54] --> GitHub152 joined #scummvm.
[11:54] <GitHub152> [scummvm] dreammaster pushed 1 new commit to master: https://git.io/vQygI
[11:54] <GitHub152> scummvm/master e5ab09d Paul Gilbert: TITANIC: Fix promenade fan noises when turned to full
[11:54] GitHub152 (GitHub152@192.30.252.40) left #scummvm.
[12:04] --> _sev|work_ joined #scummvm.
[12:04] #scummvm: mode change '+o _sev|work_' by ChanServ!ChanServ@services.
[12:07] <-- _sev|work left irc: Ping timeout: 260 seconds
[12:15] <Simei> Hi, everyone.
[12:16] <Simei> Is there an example how languages are handled with ConfMan and GUI ?
[12:17] <Simei> When there are several languages
[12:37] --> _sev|work__ joined #scummvm.
[12:38] --> Tomaz^W joined #scummvm.
[12:39] <-- Tomaz^W2 left irc: Ping timeout: 248 seconds
[12:40] --> ajax16384 joined #scummvm.
[12:40] #scummvm: mode change '+o ajax16384' by ChanServ!ChanServ@services.
[12:41] <-- Cruel` left irc: Ping timeout: 246 seconds
[12:41] <-- Mellified left irc: Ping timeout: 246 seconds
[12:41] <-- _sev|work_ left irc: Ping timeout: 246 seconds
[12:41] --> Cruel` joined #scummvm.
[12:43] <-- thesquib left irc: Ping timeout: 240 seconds
[12:46] --> thesquib joined #scummvm.
[12:57] <criezy|Work> Simei: you mean when a game is multi-lingual to let the user select which language to use?
[12:58] <Simei> yes!
[12:59] <criezy|Work> sword1 is doing that for subtitles
[12:59] <criezy|Work> Each localized version contains the speech in their language only but the subtitles for all languages, and we though it might be a good idea to let players use the subtitles language they want (which the original did not do)
[13:00] <Simei> The language is selected via GUI by user ?
[13:01] <criezy|Work> Yes
[13:01] <criezy|Work> And then use here: https://github.com/scummvm/scummvm/blob/master/engines/sword1/sword1.cpp#L129
[13:01] <criezy|Work> Thelanguage is selected in the Game Settings in the ScummVM launcher
[13:03] <Simei> Ok, I see. But how to add more languages to choose in gui ?
[13:08] <t0by> See if grepping for ADGameDescription in engines/ helps
[13:09] <criezy|Work> I am trying to remember, but that was a long time ago :P
[13:09] <criezy|Work> I think it is indeed related to the game description for the detection.
[13:09] <criezy|Work> Maybe using Common::UNK_LANG for the language field?
[13:14] <Simei> So, we need to choose from the whole list ? We can't set a certain range of languages to choose ?
[13:15] <Simei> ADGameDescription gives nothing :/
[13:18] <criezy|Work> I think so yes. From what I remember you get the whole list of lnguages, even if the game is onlay available in a few of those.
[13:19] <Simei> Ok, I see. Thanks!
[13:21] <t0by> Simei, your grep is broken ;)
[13:21] <-- _sev|work__ left irc: Quit: This computer has gone to sleep
[13:22] <t0by> make sure you use the -r switch
[13:22] <criezy|Work> Simei: I might be wrong.
[13:22] <criezy|Work> I don't have the source code here, so it is difficult to check
[13:22] <criezy|Work> Grep for checkGameGUIOptionLanguage
[13:22] <criezy|Work> to see what it does
[13:23] <criezy|Work> This is used here to populate the language popup https://github.com/scummvm/scummvm/blob/master/gui/editgamedialog.cpp#L155
[13:24] <t0by> criezy|Work, I haven't dealt *a lot* with localization, but once the ADGameDescription for game X version Y reports the correct language, shouldn't Simei be all set wrt GUI?
[13:26] <Simei> t0by, what does that mean?
[13:26] --> _sev|work__ joined #scummvm.
[13:33] <t0by> Which part?
[13:33] <t0by> I *suppose* you shouldn't have to deal with the GUI directly for what you have to do
[13:37] <Simei> There is a string to set for this checkGameGUIOptionLanguage
[13:37] <Simei> I look into the code to see which one it is
[13:39] <Simei> I think so
[13:45] --> Littleboy joined #scummvm.
[13:45] #scummvm: mode change '+o Littleboy' by ChanServ!ChanServ@services.
[13:55] --> ignalina joined #scummvm.
[13:59] <-- _sev|work__ left irc: Quit: This computer has gone to sleep
[14:01] --> _sev|work__ joined #scummvm.
[14:02] --> jamm joined #scummvm.
[14:37] --> Mellified joined #scummvm.
[15:01] <-- ignalina left irc: Quit: ignalina
[15:12] --> Strangerke_ joined #scummvm.
[15:14] <-- Strangerke left irc: Ping timeout: 248 seconds
[15:14] Nick change: Strangerke_ -> Strangerke
[15:26] --> Begasus joined #scummvm.
[15:39] <Simei> I think the list can be limited by adding languages using sth like dg.appendGUIOptions(getGameGUIOptionsDescriptionLanguage(x->language)); several times. But I didn't figure out other ways but to do it manually in the SludgeMetaEngine. What do you think of it?
[15:46] <-- Deledrius left irc: Quit: App.Exit
[16:00] <t0by> Simei, I'm sorry, *what* sort of behaviour are you trying to achieve?
[16:02] <-- Begasus left irc: Ping timeout: 255 seconds
[16:02] --> Deledrius joined #scummvm.
[16:03] <t0by> I don't understand your need to /limit/ (how?) a /list/ (which one?)
[16:03] --> Henke37 joined #scummvm.
[16:04] <-- TMM left irc: Quit: Ex-Chat
[16:05] --> LittleToonCat joined #scummvm.
[16:06] <Simei> I'm thinking about overriding sth in SludgeMetaEngine to add more informations in the GuiOption
[16:06] <t0by> Simei, I think engines/wintermute/detection_tables.cpp and detection.cpp are sufficiently clear an example
[16:07] <t0by> Simei, I understand, which information?
[16:07] <Simei> In sludge every game has a limited language list
[16:08] <t0by> I would suggest to mimic the approach used in WME
[16:08] <t0by> i.e. each different language pack is detected as a different game
[16:08] <Simei> For Verb Coin, there's only English and German
[16:08] <t0by> a second
[16:09] <t0by> let me have a look
[16:09] <t0by> s/different game/has a separate detection entry/
[16:09] <Simei> But how do I know which one it is
[16:09] <t0by> the gui then shows that in the form of a dropdown
[16:09] <t0by> a second, let me have a look at the code
[16:09] <Simei> without letting user choose?
[16:10] <Simei> https://github.com/scummvm/scummvm/blob/master/gui/editgamedialog.cpp#L155
[16:10] <Simei> It's just this one to check
[16:11] <t0by> I can't remember, wait a second
[16:12] <t0by> Oh, right, in Sludge there are *different languages* packed inside the same data file
[16:13] <t0by> or at least same dir
[16:13] --> Begasus joined #scummvm.
[16:14] <Simei> t0by: I see what you mean.
[16:14] <Simei> It will be better to do the same way as Wintermute
[16:14] <t0by> We do have a "Pick the game" dialog already iirc
[16:14] <Simei> Yes
[16:15] <t0by> Relevant logic should be in luancher.cpp
[16:15] <t0by> *launcher
[16:15] <t0by> You *should* be able to get away with that mechanism
[16:15] <t0by> I don't see an obvious way in which it could fail
[16:15] <t0by> unless Sludge allows you to switch languages mid-game
[16:16] <t0by> (But we might very well ignore that for the time being)
[16:17] <-- ajax16384 left irc: Read error: Connection reset by peer
[16:18] <t0by> Wintermute fails spectacularly at detection if I put several .dcp language packs in the same dir, though...
[16:19] <t0by> Let me see why
[16:19] <Simei> So I'll do it this way. Thanks!
[16:19] <Simei> OK, I'll do it so. Thanks!
[16:20] Last message repeated 2 time(s).
[16:20] <Simei> Sorry, there's a network problem :p
[16:21] <t0by> I'm thinking.
[16:27] <t0by> Simei, no, I fear you can't get away with using the "Pick the game" dialog.
[16:27] <snover> wjp: do you remember/have a note about the shift delta for phant1? is it 0x1af000?
[16:27] <t0by> I fear that would be a problem with the fallback tdetector, as it can return only *one* entry
[16:29] --> m_kiewitz joined #scummvm.
[16:29] #scummvm: mode change '+o m_kiewitz' by ChanServ!ChanServ@services.
[16:30] <wjp> snover: 0x1bf000, maybe?
[16:30] <wjp> but the relocation shouldn't impact freezes, only weirdness when breaking
[16:31] <t0by> wjp, snover: are there precedents for engines where a single data pack can support multiple languages?
[16:31] <Simei> t0by Sorry, why will there be a problem with fallback detection?
[16:32] <t0by> Simei, fallback detection returns only *one* entry, so I *guess* if you try fallback detection on a dir which cointains a game in *two* langagues, it will return only *one* ADGameDescription
[16:32] <snover> t0by: torin, kinda. if you have multiple ADGameDescription entries with the same file entries but different languages, scummvm will show a second dialogue asking you which language you want
[16:32] <t0by> (with *one* language)
[16:33] <snover> (same as with the platform)
[16:33] <t0by> snover, right, thanks, that's how I imagined it, but... what do you do about fallback detection in that case?
[16:33] <wjp> snover: are you getting freezes even just running it without actually breaking into the debugger?
[16:34] <snover> wjp: its not frozen, just unusably slow (slower than valgrind even), but yeah, it is happening without breaking.
[16:34] <-- jamm left irc: Ping timeout: 248 seconds
[16:35] <t0by> You're stick with UNK_LANG, I guess..
[16:35] <t0by> *stuck
[16:35] <t0by> hrm.
[16:35] <wjp> snover: did you increase cycles to a reasonable number? (20k+)
[16:35] <t0by> If there's precedent for that, I guess it's tolerable for Sludge too...
[16:36] <Simei> t0by, I don't think there's a problem for fallback detection
[16:36] <Simei> In this case we just use the GUI to select one
[16:37] <Simei> In sludge, every game has its own coding for languages
[16:37] <Simei> So I guess, it will be set to the default one
[16:38] <snover> wjp: ill give it a try, though when the game switches into protected mode dosbox automatically goes to 100% max cycles, and things run at normal speed so long as the debugger isnt hooked up
[16:38] <t0by> Simei, have a look at how ADGF_ADDENGLISH is used in sci
[16:38] <t0by> I *think* you will need something like that for Sludge as well
[16:39] <t0by> notice in particularadvancedDetector.cpp +126
[16:39] <t0by> *in particular advancedDetector.cpp
[16:39] <Simei> Ok. I note it. Sorry I'm not in front of the code. I'll have a look later.
[16:40] <snover> t0by: since the fallbackDetect method can only return one ADGameDescription you just gotta pick one language right now, i guess.
[16:40] <t0by> ^
[16:41] <t0by> Simei, that's exactly my problem
[16:41] <t0by> But well
[16:41] <t0by> There's precedent for that.
[16:42] <t0by> snover, don't you use the ADGF_ADDENGLISH flag for that?
[16:42] <t0by> Looking at the code it seems to me you use it for sneaking an extra language inside a game description
[16:44] <wjp> snover: oh, you may be using the dynamic core then?
[16:44] <wjp> should switch to "normal" core for debugging
[16:45] <t0by> Pfff.
[16:45] <Simei> But, in sludge, we can't do that for fall back detection any way. Because we can know which language is in an unknown game in advance because every sludge game has it's own language code
[16:46] <wjp> snover: speed is fine for me here in the dosbox debugger (core normal, cycles 20000)
[16:46] <Simei> They haven't set a standard for language yet.
[16:46] <t0by> Ouch.
[16:46] <wjp> snover: let me try to also get the ida link working again
[16:46] <t0by> Simei, I understand, sorry it took me a while to understand what you meant.
[16:46] <t0by> I'm a poor reader.
[16:46] <t0by> Simei: so yes, forget about fallback detection, let's rely on multiple ADGameDescription entries.
[16:47] <Simei> I haven't been very clear neither:)
[16:47] <Simei> Ok, I'll do it this way
[16:48] <t0by> Simei, if it's not standardized, in that case this makes it your responsibility to find a reasonable way to provide the engine with a (game, language) -> datafile matrix.
[16:49] <Simei> Ok
[16:54] <t0by> Simei, IIRC ConfMan can give you gameid *and* language
[16:55] <snover> wjp: hm, weird. it was using normal core already, but i guess something is messed up with the max cycles. fixed 20k cycles, performance is good.
[16:55] <t0by> You probably need to find a reasonable way to tie specific datafiles to detection entries.
[16:56] <t0by> I can't say I envy you...
[16:56] <t0by> snover: thanks for the SCI pointer
[17:00] <-- _sev|work__ left irc: Quit: This computer has gone to sleep
[17:05] <snover> its too hot in here, my computer is probably throttling too, which doesnt help anything.
[17:06] <wjp> (in the meantime, with some heavy hacked bitstream inlining, titanic in valgrind is about playable)
[17:09] <-- user9 left irc: Remote host closed the connection
[17:12] --> WooShell joined #scummvm.
[17:15] <m_kiewitz> wjp: wow
[17:15] <-- Begasus left irc: Ping timeout: 255 seconds
[17:17] <WooShell> meow =^.^=
[17:27] --> Begasus joined #scummvm.
[17:32] --> Farmboy0 joined #scummvm.
[17:32] <-- Farmboy0 left irc: Changing host
[17:32] --> Farmboy0 joined #scummvm.
[17:33] <-- criezy|Work left irc: Quit: Page closed
[17:35] --> Axy joined #scummvm.
[17:35] <-- Axy left irc: Changing host
[17:35] --> Axy joined #scummvm.
[17:36] <snover> wjp: so, with 1bf000, ida throws up some message about an instruction already existing and asking if i want to recreate it or something. i calculated a delta of 1b0000 by breaking in the dosbox debugger, looking at the code at eip, and then searching to find the location of the same code byte sequence in ida. is that a correct way to find the delta? do i also need to subtract the cseg start (in which case the delta becomes 1a0000, which
[17:36] <snover> is what it was in sq6?)
[17:38] <wjp> hm, that does sound reasonable
[17:39] <wjp> (and matching code is also what I do for protected mode games, yes)
[17:39] <wjp> (haven't really thought of a way to automatically detect what the loader did)
[17:45] <snover> sweet, thanks.
[17:47] --> girafe joined #scummvm.
[17:54] --> Begas_VBox joined #scummvm.
[17:55] <-- Begas_VBox left irc: Client Quit
[17:58] <-- Begasus left irc: Quit: Ex-Chat
[18:02] --> Begasus joined #scummvm.
[18:06] --> Begas_VBox joined #scummvm.
[18:07] <-- t0by left irc: Ping timeout: 268 seconds
[18:07] <-- Axy left irc: Ping timeout: 255 seconds
[18:16] --> TMM joined #scummvm.
[18:16] #scummvm: mode change '+o TMM' by ChanServ!ChanServ@services.
[18:43] <-- Begas_VBox left irc: Quit: Vision[0.9.8]: i've been blurred!
[18:44] <-- Begasus left irc: Quit: Ex-Chat
[18:44] --> ajax16384 joined #scummvm.
[18:44] #scummvm: mode change '+o ajax16384' by ChanServ!ChanServ@services.
[18:45] --> SylvainTV joined #scummvm.
[18:45] #scummvm: mode change '+o SylvainTV' by ChanServ!ChanServ@services.
[18:45] --> jamm joined #scummvm.
[18:48] --> t0by joined #scummvm.
[18:48] #scummvm: mode change '+o t0by' by ChanServ!ChanServ@services.
[18:53] <-- t0by left irc: Remote host closed the connection
[18:53] --> t0by joined #scummvm.
[18:53] #scummvm: mode change '+o t0by' by ChanServ!ChanServ@services.
[18:58] <-- t0by left irc: Ping timeout: 246 seconds
[18:59] <-- m_kiewitz left irc: Read error: Connection reset by peer
[19:02] --> t0by joined #scummvm.
[19:02] #scummvm: mode change '+o t0by' by ChanServ!ChanServ@services.
[19:10] --> m_kiewitz joined #scummvm.
[19:10] #scummvm: mode change '+o m_kiewitz' by ChanServ!ChanServ@services.
[19:17] <snover> wjp: the delta is actually 0x1b0000, at least for my phant1 executable with hash 36f1ce39db483894aeb4ec32644bbef7.
[19:21] --> K4T joined #scummvm.
[19:21] <-- K4T left irc: Client Quit
[19:44] <-- t0by left irc: Remote host closed the connection
[19:45] --> t0by joined #scummvm.
[19:45] #scummvm: mode change '+o t0by' by ChanServ!ChanServ@services.
[20:04] --> GitHub113 joined #scummvm.
[20:04] <GitHub113> [scummvm] bgK pushed 1 new commit to master: https://git.io/vQSot
[20:04] <GitHub113> scummvm/master c3bbae0 Bastien Bouclet: MOHAWK: Riven: Improve save load error handling...
[20:04] GitHub113 (GitHub113@192.30.252.42) left #scummvm.
[20:04] --> GitHub3 joined #scummvm.
[20:04] <GitHub3> [scummvm] criezy pushed 1 new commit to master: https://git.io/vQSom
[20:04] <GitHub3> scummvm/master 662009b Thierry Crozat: I18N: Update translations templates
[20:04] GitHub3 (GitHub3@192.30.252.34) left #scummvm.
[20:11] <-- _sev left irc: Quit: Leaving
[20:16] --> frankyboy_ joined #scummvm.
[20:19] Nick change: Joefish_ -> Joefish
[20:27] <wjp> snover: the delta, or the new "address of first segment" ?
[20:27] <snover> the delta. the address of cseg000 is at 0x1c0000 in dosbox
[20:28] <snover> -at
[20:31] --> _sev joined #scummvm.
[20:31] <-- _sev left irc: Changing host
[20:31] --> _sev joined #scummvm.
[20:31] #scummvm: mode change '+o _sev' by ChanServ!ChanServ@services.
[20:36] <-- ajax16384 left irc: Read error: Connection reset by peer
[20:45] --> _sev_ joined #scummvm.
[20:45] <-- _sev_ left irc: Changing host
[20:45] --> _sev_ joined #scummvm.
[20:45] #scummvm: mode change '+o _sev_' by ChanServ!ChanServ@services.
[20:47] <-- _sev left irc: Ping timeout: 240 seconds
[20:51] --> GitHub79 joined #scummvm.
[20:51] <GitHub79> [scummvm] criezy pushed 1 new commit to master: https://git.io/vQS1t
[20:51] <GitHub79> scummvm/master 34969d5 Thierry Crozat: SWORD1: Specify list of available languages in detection code...
[20:51] GitHub79 (GitHub79@192.30.252.40) left #scummvm.
[20:54] --> Strangerke_ joined #scummvm.
[20:56] <-- Strangerke left irc: Ping timeout: 248 seconds
[20:56] Nick change: Strangerke_ -> Strangerke
[20:57] <waltervn> I also ran into the "can only return one ADGameDescription" problem... I haven't really decided what to do with this in ADL yet. I may drop AD there and write custom code.
[21:00] --> DJWillis joined #scummvm.
[21:00] #scummvm: mode change '+o DJWillis' by ChanServ!ChanServ@services.
[21:00] <-- waltervn left irc: Quit: Leaving
[21:03] <wjp> is there a fundamental reason for it? Presumably easy to make it return a list?
[21:10] <snover> i cant think offhand of a reason not to return ADGameDescList instead, though i am uncertain about what the ownership rules are for that type
[21:16] --> Begasus joined #scummvm.
[21:21] --> Begas_VBox joined #scummvm.
[21:26] <snover> augh, ida causes the mouse to uncapture from dosbox every time a breakpoint is hit, even if the breakpoint is just a call to Message that returns 0 so it does not actually break
[21:27] <snover> since i am tracing celobj construction calls, whenever the mouse changes its view, like when hovering over a hotspot i lose the mouse
[21:31] <snover> ok, well, the good news is that the constructor call order and count are identical in scummvm and ssci. now lets see about those palette submissions&
[21:34] <-- heroux left irc: Ping timeout: 248 seconds
[21:37] <wjp> at some point I hacked dosbox to output certain registers and/or memory values whenever it would hit a breakpoint instead of actually breaking which got around this problem
[21:37] <wjp> (using the dosbox debugger instead of idados)
[21:38] <-- Henke37 left irc: Quit: ERR_SHUTDOWN
[21:38] <snover> that sounds like a reasonable workaround for this problem if i end up getting super frustrated
[21:39] <snover> i really must thank you again for idados, it really does make this debugging less painful
[21:42] <wjp> the main work was done by Yuv422, but I'm glad it's proving useful
[21:51] <-- Stormkeeper left irc: Quit: KVIrc 4.9.2 Aria http://www.kvirc.net/
[21:58] <-- Dark-Star left irc: Ping timeout: 255 seconds
[22:00] --> Stormkeeper joined #scummvm.
[22:03] <-- Begas_VBox left irc: Remote host closed the connection
[22:03] <-- Begasus left irc: Quit: Ex-Chat
[22:03] --> heroux joined #scummvm.
[22:28] <-- heroux left irc: Ping timeout: 246 seconds
[22:28] <-- Farmboy0 left irc: Remote host closed the connection
[22:29] --> heroux joined #scummvm.
[22:32] --> evil-t0by joined #scummvm.
[22:32] #scummvm: mode change '+o evil-t0by' by ChanServ!ChanServ@services.
[22:32] <-- t0by left irc: Remote host closed the connection
[22:32] --> Littleboy_ joined #scummvm.
[22:32] <-- Littleboy left irc: Disconnected by services
[22:32] #scummvm: mode change '+o Littleboy_' by ChanServ!ChanServ@services.
[22:34] <-- Dominus left irc: Ping timeout: 251 seconds
[22:35] <-- aquadran left irc: Remote host closed the connection
[22:35] --> aquadran joined #scummvm.
[22:35] #scummvm: mode change '+o aquadran' by ChanServ!ChanServ@services.
[22:37] --> Dominus joined #scummvm.
[22:39] --> DDR__ joined #scummvm.
[22:43] <-- WooShell left irc: Quit: If you understand or if you don't, if you believe or if you doubt - There's a universal justice, and the eyes of truth are always watching you.
[22:45] Nick change: DDR__ -> DDR
[22:47] <-- DDR left irc: Quit: DDR is not Dance Dance Revolution
[22:47] --> DDR joined #scummvm.
[22:56] --> dreammaster joined #scummvm.
[22:56] #scummvm: mode change '+o dreammaster' by ChanServ!ChanServ@services.
[23:00] <wjp> hi dreammaster
[23:00] <-- m_kiewitz left irc: Quit: technology isn't intrinsically good or evil. It's how it's used. Like the Death Ray.
[23:04] <dreammaster> Evening :)
[23:04] <wjp> FWIW, I think DMatrix makes more sense when its 4 vectors are columns rather than rows
[23:05] <wjp> you've reversed an impressive amount of geometry in star_control there, by the way
[23:05] <dreammaster> If you think so, I'll change it over, then.
[23:06] <dreammaster> Thanks. Lets just hope I never have to deal with something so complicated again. Everytime I'm ever tempted to look at reversing any 3d adventures, I'll point myself to Titanic and say.. never forget.. never again! :)
[23:07] --> Dark-Star joined #scummvm.
[23:07] <wjp> with that interpretation, DMatrix becomes a 3D affine transform
[23:08] <wjp> then DMatrix::fn4 is composition of transforms (so it could be called 'compose', or 'multiply' with some slight abuse of notation)
[23:08] <wjp> and DMatrix::fn1 is the inverse transform
[23:08] <dreammaster> It may be worth renaming the class as well then, to better underscore it's purpose. And differentiate it from FMatrix
[23:09] <wjp> applying the transform is DVector::fn1(DMatrix)
[23:12] <wjp> the rest of DVector does rotation stuff (rotations to align vectors with other vectors or axes and such), but still trying to wrap my head around how that is actually used in the rest of star_control :-)
[23:13] <dreammaster> Well, every little bit helps in the longrun, so even better names for the few functions will be helpful
[23:13] <dreammaster> Plus correcting the rows to cols
[23:18] <-- jamm left irc: Ping timeout: 248 seconds
[23:38] <-- evil-t0by left irc: Quit: Leaving
[23:41] --> GitHub146 joined #scummvm.
[23:41] <GitHub146> [scummvm] dreammaster pushed 2 new commits to master: https://git.io/vQ9UO
[23:41] <GitHub146> scummvm/master dc5645c Paul Gilbert: TITANIC: Change rows to cols in DMatrix, thanks to wjp
[23:41] <GitHub146> scummvm/master 256b275 Paul Gilbert: TITANIC: Rename DMatrix to DAffine, thanks to wjp
[23:41] GitHub146 (GitHub146@192.30.252.42) left #scummvm.
[23:48] <ScummBot_> Port build status changed with 256b2758: Failure: master-ios7, master-mingw-w32, master-openpandora, master-android_mips, master-android_x86, master-ios, master-wii, master-debian-x86, master-mingw-w64-cplusplus11, master-webos, master-osx_intel, master-psp2, master-gp2x, master-dingux, master-debian-x86-clang, master-gamecube, master-gp2xwiz, master-android_arm,
[23:48] <ScummBot_> master-mingw-w64, master-deb
[23:48] <ScummBot_> ian-x86_64
[23:50] <dreammaster> Damn. One sec..
[23:52] --> GitHub178 joined #scummvm.
[23:52] <GitHub178> [scummvm] dreammaster pushed 1 new commit to master: https://git.io/vQ9Tc
[23:52] <GitHub178> scummvm/master 906c6de Paul Gilbert: TITANIC: Fix daffine.h includes
[23:52] GitHub178 (GitHub178@192.30.252.45) left #scummvm.
[00:00] --- Thu Jul 13 2017