ScummVM IRC Log - 01 Sep 2010

[Back to Index]

[00:42] <-- Robin_Watts left irc: Quit: Pop!
[00:44] <-- GoGi left irc: Quit: Leaving
[00:50] <-- D0SFreak left irc: Quit: If I'm not back in 5 minutes....just wait longer!
[01:13] <-- Sho_ left irc: Remote host closed the connection
[01:37] --> D0SFreak joined #scummvm.
[01:41] <-- Ceriand|work left irc: Quit: Leaving.
[01:43] <-- waltervn left irc: Quit: ChatZilla 0.9.86-rdmsoft [XULRunner 1.9.2/20100222070821]
[01:51] <-- Javacat left irc: Quit: zzzz
[01:52] <-- jackster left irc: Quit: jackster
[01:55] --> Ceriand|desktop joined #scummvm.
[01:56] --> pablog joined #scummvm.
[02:09] <-- SylvainTV left irc: Ping timeout: 255 seconds
[02:27] --> Vampire0 joined #scummvm.
[02:29] <-- Vampire0_ left irc: Ping timeout: 240 seconds
[03:47] <-- nego left irc: Quit: Lost terminal
[03:47] --> nego joined #scummvm.
[03:56] <-- D0SFreak left irc: Ping timeout: 265 seconds
[03:58] --> peres joined #scummvm.
[03:58] #scummvm: mode change '+o peres' by ChanServ!ChanServ@services.
[03:58] <peres> hello
[03:59] <digitall> Morning...
[04:17] <-- clone2727 left irc: Quit: later
[04:17] digitall (digitall@cpc1-hitc2-0-0-cust861.9-2.cable.virginmedia.com) left #scummvm.
[05:03] <-- FreeTHX left irc: Ping timeout: 240 seconds
[05:09] <acks> yay, the latest sam&max is out :D (i know the tell tale games releases are unrelated to scummvm, but this is the only channel i'm in that is even slightly interested in the genre of games >.> )
[05:11] <acks> kind of stinks that the extras DVD is only free if you bought straight from telltale, I got it on PSN >.>
[05:44] --> Hkz joined #scummvm.
[05:44] #scummvm: mode change '+o Hkz' by ChanServ!ChanServ@services.
[06:03] --> m_kiewitz joined #scummvm.
[06:03] #scummvm: mode change '+o m_kiewitz' by ChanServ!ChanServ@services.
[06:04] <-- Newa left irc: Ping timeout: 245 seconds
[06:20] <-- Hkz left irc: Quit: Leaving
[06:40] <-- pablog left irc: Quit: Ex-Chat
[06:56] <-- Strangerke left irc: Ping timeout: 240 seconds
[06:58] <Q_u_a_r_K> Ooooh, season final
[06:58] <Q_u_a_r_K> of Sam & Max
[07:16] --> FreeTHX joined #scummvm.
[07:20] <Lightkey> lol
[07:20] <Lightkey> you guys are LATE
[07:21] <Lightkey> *makesnoteDrMcDoyfoureyes*
[07:35] --> ajax16384 joined #scummvm.
[07:35] #scummvm: mode change '+o ajax16384' by ChanServ!ChanServ@services.
[07:48] <-- DJW|Home left irc: Ping timeout: 265 seconds
[08:06] --> William__ joined #scummvm.
[08:06] <-- WilliamC left irc: Disconnected by services
[08:06] Nick change: William__ -> WilliamC
[08:08] --> James|GlideM joined #scummvm.
[08:08] #scummvm: mode change '+v James|GlideM' by ChanServ!ChanServ@services.
[08:09] --> dreammaster joined #scummvm.
[08:09] #scummvm: mode change '+o dreammaster' by ChanServ!ChanServ@services.
[08:13] --> William__ joined #scummvm.
[08:13] --> ny00123 joined #scummvm.
[08:13] <-- WilliamC left irc: Disconnected by services
[08:13] Nick change: William__ -> WilliamC
[08:27] --> Strangerke joined #scummvm.
[08:27] #scummvm: mode change '+o Strangerke' by ChanServ!ChanServ@services.
[08:27] --> SylvainTV joined #scummvm.
[08:27] <Strangerke> hi guys
[08:30] <fuzzie> morning
[08:31] <dreammaster> G'day
[08:36] --> digitall joined #scummvm.
[08:36] #scummvm: mode change '+o digitall' by ChanServ!ChanServ@services.
[08:52] <-- dreammaster left irc: Ping timeout: 260 seconds
[09:14] <-- Kerbox left irc: Read error: Connection reset by peer
[09:16] --> Kerbox joined #scummvm.
[09:53] --> Hkz joined #scummvm.
[09:53] #scummvm: mode change '+o Hkz' by ChanServ!ChanServ@services.
[10:09] --> dreammaster joined #scummvm.
[10:09] #scummvm: mode change '+o dreammaster' by ChanServ!ChanServ@services.
[10:22] <-- raorn left irc: Read error: Connection reset by peer
[10:23] --> raorn joined #scummvm.
[10:27] --> Robin_Watts joined #scummvm.
[10:27] #scummvm: mode change '+o Robin_Watts' by ChanServ!ChanServ@services.
[10:35] <-- Kerbox left irc: Read error: Connection reset by peer
[10:36] --> lunavorax joined #scummvm.
[10:38] --> Kerbox joined #scummvm.
[10:46] <CIA-11> m_kiewitz * r52479 /scummvm/trunk/engines/sci/sci.cpp:
[10:46] <CIA-11> SCI: disable dialog replacement for jones
[10:46] <CIA-11> fixes odd crash right at the start, although you can't save/restore in jones o_O (fixes bug #3057080)
[10:49] <-- Zaarin left irc: Ping timeout: 240 seconds
[10:50] --> Newa joined #scummvm.
[10:58] --> royan joined #scummvm.
[11:16] --> DJWillis joined #scummvm.
[11:16] #scummvm: mode change '+o DJWillis' by ChanServ!ChanServ@services.
[11:34] --> jackster joined #scummvm.
[11:37] <-- _sev left irc: Ping timeout: 240 seconds
[11:41] <-- FreeTHX left irc: Ping timeout: 272 seconds
[11:55] --> _sev joined #scummvm.
[11:55] #scummvm: mode change '+o _sev' by ChanServ!ChanServ@services.
[11:55] --> Bluddy|work joined #scummvm.
[12:01] <-- dreammaster left irc:
[12:04] --> D0SFreak joined #scummvm.
[12:07] <Strangerke> Ha! Someone tested Fascination! :)
[12:11] <-- jackster left irc: Quit: jackster
[12:12] <LordHoto> hehe
[12:13] --> bushwakko joined #scummvm.
[12:24] <-- Littleboy left irc:
[12:27] <-- James|GlideM left irc: Ping timeout: 265 seconds
[12:37] <wjp> the subtleties of strict aliasing rules give me a headache
[12:57] <Bluddy|work> wjp: it's not that complicated. To quote the gcc manual: an object of one type is assumed never to reside at the same address as an object of a diferent type,unless the types are almost the same. For example, an unsigned int can alias an int, but not a void* or a double.
[12:57] <Bluddy|work> wjp: unions are an exception though
[13:05] --> Lubomyr joined #scummvm.
[13:06] <dhewg> ohboi, those READ_ WRITE_ funcs are ugly
[13:15] --> Littleboy joined #scummvm.
[13:19] <Bluddy|work> wjp: you committing or am i?
[13:29] <wjp> ah, sorry, I wasn't on irc for a while :-)
[13:30] <wjp> go for it :-)
[13:31] <wjp> I agree the basic idea behind the strict aliasing isn't that complicated. It gets annoying when casting between structs and non-structs
[13:32] <wjp> it seems that with two different Unaligned32 { uint32 val; } structs, gcc thinks they don't alias. But their val members do both alias with int*'s. So this aliassing thing would seem to be non-transitive. Which confuses me.
[13:35] <-- raorn left irc: Read error: No route to host
[13:36] <-- Lubomyr left irc: Ping timeout: 265 seconds
[13:39] <CIA-11> Bluddy * r52480 /scummvm/trunk/common/endian.h:
[13:39] <CIA-11> COMMON: changed read/write endian function to use __may_alias__ attribute
[13:39] <CIA-11> This is a better solution for the gcc aliasing problem that happens when
[13:39] <CIA-11> aliasing a struct onto something else. What happens is that the compiler assumes
[13:39] <CIA-11> no aliasing can happen when -O2 and -O3 are activated, and a call to
[13:39] <CIA-11> READ_UINT32() followed by WRITE_UINT32() and another READ_UINT32() will be
[13:39] <CIA-11> optimized to return the original read value instead of re-reading.
[13:42] <dhewg> and this fixes bs2/psp when removing -fno-strict-aliasing again?
[13:42] <Bluddy|work> dhewg: it does
[13:43] <dhewg> i understand the problem, but that gcc4+NEEDS_ALIGNMENT code for those funcs is ugly. isnt there anything better we can use?
[13:44] <Bluddy|work> wjp: from gcc's docs, it seems like gcc takes no responsibility once strict-aliasing is enabled, if you take a pointer, cast it and dereference.
[13:44] --> raorn joined #scummvm.
[13:44] <-- WilliamC left irc: Ping timeout: 240 seconds
[13:46] <Bluddy|work> dhewg: i tried another way of doing it (recommended to me by a ncie guy from gcc) and it doesn't work. you can see it in my email. there seems to be no other way of producing the desired ASM code
[13:46] --> clone2727 joined #scummvm.
[13:46] #scummvm: mode change '+o clone2727' by ChanServ!ChanServ@services.
[13:47] <wjp> is your conversation with the gcc guy on a public mailing list, by the way?
[13:47] <wjp> but I still see no reason why having a single Unaligned struct outside of the two functions shouldn't work. As far as I can tell it abides by all the rules
[13:48] <Bluddy|work> wjp: it's a bug report on gcc bugzilla. number 45462
[13:48] <wjp> (but I do definitely agree that using __may_alias__ is the safest solution)
[13:49] --> Nevyn4 joined #scummvm.
[13:49] <Bluddy|work> wjp: you're not instantiating the struct, you're casting the pointer to the struct type, so why should it matter where it's defined?
[13:49] <dhewg> there's no other way to produce good code for it? no gcc buildin or whatever?
[13:50] <wjp> it matters if you're casting a pointer to the same struct type twice, or to two different struct types
[13:51] <wjp> by the way, did this unaligned_uint32 thing cause crashes in gcc or in scummvm?
[13:51] <Bluddy|work> dhewg: not as far as i could tell. another guy on gcc suggested memcpy, since gcc is supposed to optimize memcpy (making it inline for small copies). but i checked and at least on psp it doesn't do it.
[13:52] <Bluddy|work> wjp: in scummvm
[13:52] <wjp> bus errors?
[13:52] <Bluddy|work> yes
[13:53] <LordHoto> Bluddy|work: I think gcc only optimized memcpy when it's properly declared as builtin and no "custom" memcpy is inserted somewhere
[13:53] <LordHoto> optimizes even
[13:54] <Bluddy|work> LordHoto: if you're referring to wrap-memcpy, i removed it for the test
[13:54] <CIA-11> Bluddy * r52481 /scummvm/trunk/ (backends/platform/psp/Makefile configure):
[13:54] <CIA-11> PSP: removed strict-aliasing
[13:54] <CIA-11> It's no longer needed after adding __may_alias__ flag in endian.h.
[13:55] <LordHoto> Bluddy|work: did you check for memcpy to be a builtin?
[13:55] <LordHoto> Bluddy|work: maybe your toolchain is setup not to have that
[13:55] <LordHoto> anyway I am not sure right now anymore
[13:56] <Bluddy|work> how is memcpy made builtin?
[13:56] --> WilliamC joined #scummvm.
[13:56] peres (peres@unaffiliated/peres) left #scummvm ("Leaving").

[13:57] <LordHoto> Bluddy|work: this has some info about what should be builtin by default: http://gcc.gnu.org/onlinedocs/gcc-4.1.2/gcc/Other-Builtins.html
[14:26] <CIA-11> mthreepwood * r52482 /scummvm/trunk/engines/mohawk/ (6 files):
[14:26] <CIA-11> MOHAWK: Implement xbookclick
[14:26] <CIA-11> It is now possible to trap Gehn in the trap book. Side note: Riven is now completable from the beginning provided you know the D'ni number system already (and the marble puzzle is just hacked to always work).
[14:36] --> Mataniko2 joined #scummvm.
[14:49] <-- WilliamC left irc: Ping timeout: 264 seconds
[14:51] <-- Kirben left irc: Ping timeout: 252 seconds
[14:54] <-- Hkz left irc: Quit: Leaving
[14:55] <-- ajax16384 left irc: Read error: Connection timed out
[14:56] --> ajax16384 joined #scummvm.
[14:56] #scummvm: mode change '+o ajax16384' by ChanServ!ChanServ@services.
[15:00] <Bluddy|work> LordHoto: great link. I checked the PSP toolchain and indeed it uses -fno-builtin
[15:01] <Bluddy|work> LordHoto: since it uses newlib
[15:01] <LordHoto> Bluddy|work: probably because it uses newlin instead of glibc
[15:01] <LordHoto> Bluddy|work: yeah :-)
[15:02] <LordHoto> Bluddy|work: I guess newlib doesn't have a memcpy with the __builtin_ prefix, thus gcc doesn't bother about optimizing it
[15:04] <Bluddy|work> LordHoto: for some reason they add the flag in the PSP toolchain. Not sure why, but I'm not going to mess with it. But since I override memcpy anyway with --wrap, maybe I can make my version __builtin_
[15:05] <LordHoto> Bluddy|work: yeah I guess it's safer not to mess with it for all the functions out there :-)
[15:08] --> James|GlideM joined #scummvm.
[15:08] #scummvm: mode change '+v James|GlideM' by ChanServ!ChanServ@services.
[15:13] <-- clone2727 left irc: Quit: bbl
[15:14] <Bluddy|work> question: where are the defines mentioned in endian.h from? For example, in one place we have __mips__ and in another _MIPS_ARCH_MIPS32R2. __psp__ is defined with lower case (ie. it isn't activated -- I'm going to change it). Are the other mips defines standard?
[15:17] <Bluddy|work> OK actually I see __mips__ is defined by the PSP's gcc so I guess it's standard
[15:17] <logix> __psp__ is defined by pspsdk's psp-cpp
[15:18] <dhewg> yeah, its part of the psp toolchain itself
[15:18] <Bluddy|work> ok thanks.
[15:18] <logix> and also by devkitpro's psp-cpp
[15:18] <dhewg> echo | gcc -E -dM -x c - | sort
[15:19] <Bluddy|work> excellent thanks for that
[15:22] <Bluddy|work> wow i sure learned a lot thanks to this bug
[15:34] --> Zaarin joined #scummvm.
[15:52] <-- Ceriand|desktop left irc: Read error: Connection reset by peer
[15:53] <-- bushwakko left irc: Quit: bushwakko
[16:00] --> timofonic joined #scummvm.
[16:01] <timofonic> Hello
[16:01] <Zaarin> hi
[16:01] <timofonic> How's all? I see quite good, new engines coming and GSoC being a success. Congratulations :)
[16:02] <timofonic> And translation support, that's very good!
[16:06] --> Ceriand|work joined #scummvm.
[16:11] --> MetalSnake joined #scummvm.
[16:21] <-- Strangerke left irc: Ping timeout: 252 seconds
[16:28] --> Strangerke joined #scummvm.
[16:28] #scummvm: mode change '+o Strangerke' by ChanServ!ChanServ@services.
[16:34] <-- lubyou left irc: Ping timeout: 258 seconds
[16:44] <CIA-11> sev * r52483 /scummvm/trunk/ (10 files in 2 dirs): i18n: Update Russian translation
[16:47] --> lubyou joined #scummvm.
[16:47] <-- lubyou left irc: Changing host
[16:47] --> lubyou joined #scummvm.
[16:57] --> Hkz joined #scummvm.
[16:58] #scummvm: mode change '+o Hkz' by ChanServ!ChanServ@services.
[16:58] <Hkz> hi
[16:58] <Strangerke> hi Hkz
[17:02] --> Sho_ joined #scummvm.
[17:09] --> jackster joined #scummvm.
[17:23] <-- Strangerke left irc: Quit: Bye!
[17:24] --> ST2 joined #scummvm.
[17:24] <-- ST left irc: Disconnected by services
[17:24] Nick change: ST2 -> ST
[17:24] #scummvm: mode change '+o ST' by ChanServ!ChanServ@services.
[17:27] <-- jackster left irc: Quit: jackster
[17:31] --> WooShell joined #scummvm.
[17:33] <WooShell> nabbend
[17:35] <-- Kerbox left irc: Ping timeout: 252 seconds
[17:35] --> Kerbox joined #scummvm.
[17:38] [vEX] --> (~vex@c-83-233-205-109.cust.bredband2.com) joined #scummvm.
[17:38] #scummvm: mode change '+o [vEX]' by ChanServ!ChanServ@services.
[17:40] --> Javacat joined #scummvm.
[18:01] --> WilliamC joined #scummvm.
[18:24] <-- ny00123 left irc: Quit: Leaving
[18:28] --> ny00123 joined #scummvm.
[18:36] <DrMcCoy> o_O: http://www.youtube.com/watch?v=LVwmtwZLG88
[18:38] <logix> DrMcCoy: see also http://www.boingboing.net/2010/08/20/explain-this-photo.html and http://www.boingboing.net/2010/08/31/more-cool-rolling-sh.html
[18:39] <logix> (and comments)
[18:39] <DrMcCoy> "This looks shopped. I can tell from some of the pixels and from seeing quite a few shops in my time"
[18:39] <DrMcCoy> Ah, explains everythign :P
[18:42] <DrMcCoy> But yeah, I know about the rolling shutter :P
[18:53] --> Strangerke joined #scummvm.
[18:53] #scummvm: mode change '+o Strangerke' by ChanServ!ChanServ@services.
[18:56] --> salty-horse joined #scummvm.
[18:56] #scummvm: mode change '+o salty-horse' by ChanServ!ChanServ@services.
[18:56] <-- Kerbox left irc: Read error: Connection reset by peer
[18:57] --> Kerbox joined #scummvm.
[19:15] <-- Kerbox left irc: Read error: Connection reset by peer
[19:16] --> Kerbox joined #scummvm.
[19:20] --> jackster joined #scummvm.
[19:20] <-- jackster left irc: Client Quit
[20:14] <-- Sho_ left irc: Remote host closed the connection
[20:18] <CIA-11> m_kiewitz * r52484 /scummvm/trunk/engines/sci/ (7 files in 2 dirs):
[20:18] <CIA-11> SCI: set master volume correctly
[20:18] <CIA-11> and merge it together with global volume, fixes bug #3053104)
[20:25] <-- James|GlideM left irc: Read error: Connection reset by peer
[20:34] <m_kiewitz> wtf
[20:34] <m_kiewitz> im just trying to pay 20.50pounds via paypal
[20:34] <m_kiewitz> and paypal wants to get 32EUR (!) from my account
[20:34] <m_kiewitz> nice try :P
[20:35] <m_kiewitz> according to yahoo its currently around 24EUR
[20:35] <m_kiewitz> or did i miss some huge financial crash
[20:45] <logix> sounds like amazon
[20:45] <logix> if you order from amazon.co.uk, don't let them calculate the exchange rates
[20:46] <logix> they'll charge you like 1.5 eur/gbp
[20:46] <m_kiewitz> i switched to creditcard, they now took 20.50pounds :P
[20:46] <digitall> Much better..
[20:46] <logix> (actually it's not quite that bad, but they do charge a lot more than if you just pay them in gbp)
[20:47] <m_kiewitz> ohhh
[20:47] <m_kiewitz> http://content.usatoday.com/communities/gamehunters/post/2010/08/telltale-games-times-back-to-the-future-project/1
[20:47] <m_kiewitz> Thanks to USA Today, we now know the following: Marty McFly and Doc Brown (voiced by Christopher Lloyd) are in it; Bob Gale, the screenwriter for the three films, is working with Telltale on the plot; there are five episodes planned; and the DeLorean, Hill Valley, and 1985 will be in the games.
[20:47] <m_kiewitz> yay
[20:47] <logix> actually it might have been that bad... i could imagine it being 20% higher than the official exchange rate, so with todays rates that would be close to 1.50 eur/gbp
[20:47] <logix> m_kiewitz: yep, read that too :)
[20:48] <m_kiewitz> my creditcard company will take a little share, but its still much cheaper that way
[20:48] <logix> yep
[20:48] <digitall> The odd thing is doing stuff GBP to EUR.. they screw us as well..
[20:48] <-- Vaikungfu left irc: Ping timeout: 272 seconds
[20:49] <m_kiewitz> back to the future trilogy in 34 seconds http://www.youtube.com/watch?v=iBAgxrRj69o :P
[20:50] --> Vaikungfu joined #scummvm.
[20:50] <-- D0SFreak left irc: Quit: If I'm not back in 5 minutes....just wait longer!
[20:53] --> kreegee joined #scummvm.
[21:02] [md5] --> (~asd-@unaffiliated/md5/x-729473) joined #scummvm.
[21:02] #scummvm: mode change '+o [md5]' by ChanServ!ChanServ@services.
[21:04] <-- lunavorax left irc: Read error: Connection reset by peer
[21:04] <-- Ge0rG left irc: Read error: Operation timed out
[21:05] [vEX] <-- (~vex@c-83-233-205-109.cust.bredband2.com) left irc: Quit: http://niechift.com
[21:06] <DrMcCoy> m_kiewitz: :D
[21:07] --> Ge0rG joined #scummvm.
[21:17] --> waltervn joined #scummvm.
[21:17] #scummvm: mode change '+o waltervn' by ChanServ!ChanServ@services.
[21:23] <kreegee> hm, what library does scummvm use for video decoding? (looks like its not ffdshow...)
[21:25] <fuzzie> i don't think it uses a library, see graphics/video/
[21:26] <kreegee> oh
[21:27] <kreegee> thx
[21:27] <kreegee> just checked it out
[21:27] <DrMcCoy> kreegee: But some of our video decoders are (heavily) based on ffmpeg code
[21:27] <DrMcCoy> Maybe even most :P
[21:28] <kreegee> just opened the cinepak.cpp and saw that
[21:28] <kreegee> was just wondering why riven/myst lacks codecs, since all missing codecs have been reverseengineered years ago
[21:29] <kreegee> now i know
[21:29] <fuzzie> well, reverse engineered doesn't mean accurately implemented in any case, apparently
[21:29] <kreegee> (and its obviously nothing easy enough to fidle with during a first night...)
[21:29] <kreegee> fuzzie, i guess most codecs in ffdshow are pretty good implementations
[21:30] --> spookypeanut joined #scummvm.
[21:30] #scummvm: mode change '+o spookypeanut' by ChanServ!ChanServ@services.
[21:38] <-- ny00123 left irc: Quit: Leaving
[21:40] --> D0SFreak joined #scummvm.
[21:42] <-- ajax16384 left irc: Quit: i am quit
[21:51] <-- ST left irc: Ping timeout: 258 seconds
[21:52] --> ST joined #scummvm.
[21:52] #scummvm: mode change '+o ST' by ChanServ!ChanServ@services.
[22:05] --> jackster joined #scummvm.
[22:10] <-- Mataniko2 left irc: Ping timeout: 265 seconds
[22:12] <-- salty-horse left irc: Quit: Leaving
[22:18] --> bushwakko joined #scummvm.
[22:18] <-- bushwakko left irc: Client Quit
[22:21] --> Sho_ joined #scummvm.
[22:22] <-- Sho_ left irc: Changing host
[22:22] --> Sho_ joined #scummvm.
[22:27] <-- MetalSnake left irc: Quit: zzZZzz
[22:29] <-- spookypeanut left irc: Ping timeout: 265 seconds
[22:30] --> spookypeanut joined #scummvm.
[22:30] #scummvm: mode change '+o spookypeanut' by ChanServ!ChanServ@services.
[22:35] <m_kiewitz> HAHAHA
[22:35] <m_kiewitz> http://www.youtube.com/watch?v=VfqCcATPlGQ
[22:35] <m_kiewitz> laura bow 2 in 5 seconds
[22:35] <-- WooShell left irc:
[22:39] <kreegee> lol
[22:39] <kreegee> hm
[22:39] <kreegee> somebody here involved with the mohawk stuff?
[22:43] <[md5]> kreegee: what are you looking for?
[22:43] <kreegee> it doesn't recognize my games correctly
[22:43] <kreegee> just fiddling around at the moment
[22:44] <kreegee> i got this strange trilogy dvd
[22:44] <[md5]> are you talking about Myst or Riven?
[22:44] <kreegee> riven
[22:44] <digitall> Ahem..
[22:44] <kreegee> when i choose the dvd root, it finds riven and myst on it
[22:44] <kreegee> but both entries then start myst
[22:45] Action: [md5] points an accusing finger at digitall
[22:45] <kreegee> if i choose the myst2 folder, it adds riven, but can't fint the tsplit
[22:45] <[md5]> er
[22:45] <[md5]> don't add files straight from the DVD
[22:45] <digitall> kreegee : The Advanced Detector in ScummVM has issues when two games are in the same directory..
[22:45] <kreegee> i i choose the All folder (the other with a a_Data-file), it adds riven, but doesn't fine the extras.mhk
[22:45] <kreegee> ok
[22:45] <[md5]> put them in separate folders
[22:45] <digitall> Yep basically, copy the data and seperate them
[22:47] <kreegee> ok
[22:47] <kreegee> easier said than done
[22:47] <kreegee> this dvd is a mess
[22:52] <digitall> Two secs..
[22:53] <kreegee> it's ok
[22:54] <kreegee> just found the datafiles page in the wiki
[22:55] <-- Hkz left irc: Read error: Operation timed out
[22:58] <digitall> Ah.. was just going to refer you there.
[22:58] <-- spookypeanut left irc: Ping timeout: 265 seconds
[22:59] <kreegee> Could not load stack tspit!
[22:59] <kreegee> well
[23:00] <kreegee> grep tells me that tspit is in the a_Data file
[23:00] <kreegee> which is there
[23:00] <kreegee> strange
[23:04] <-- Kerbox left irc: Read error: Connection reset by peer
[23:06] --> Kerbox joined #scummvm.
[23:09] <CIA-11> strangerke * r52485 /scummvm/trunk/ (5 files in 2 dirs): HUGO: Fix labyrinth bug in Hugo 2 for Windows
[23:13] <timofonic> !seen criezy
[23:13] <echelog-1> timofonic: criezy was last seen in #scummvm 1 day, 2 hours, 19 minutes, and 38 seconds ago: <criezy> I was actually going to make the same remark ;)
[23:13] <CIA-11> strangerke * r52486 /scummvm/trunk/dists/engine-data/hugo.dat: HUGO: Updated Hugo.dat with the fix for labyrinth bug
[23:17] <digitall> kreegee : a_data.mhk , not a_Data
[23:17] <kreegee> lets try that
[23:17] <-- m_kiewitz left irc: Quit: bye
[23:17] <kreegee> osx is case insensitive though
[23:18] <digitall> I doubt that is totally true as it has a Unix base.. but clone2727, the main Riven developer is the Mac Fiend...
[23:18] <logix> actually per default hfs+ is only case preserving
[23:18] <kreegee> well
[23:18] Action: digitall summons clone2727
[23:18] <logix> oh, you said *in*sensitive
[23:18] <kreegee> it preserves cases
[23:19] <kreegee> but you can access files insensitive
[23:19] <kreegee> i thought
[23:22] <kreegee> well, renaming didn't change a thing
[23:22] [md5] <-- (~asd-@unaffiliated/md5/x-729473) left irc: Quit: nn
[23:25] <digitall> Hmm
[23:26] <digitall> Could you do a "ls -al" and "md5sum -b *" on the files?
[23:26] <digitall> and send the results to me via DCC?
[23:27] <digitall> kreegee: Are you OK with that?
[23:29] <kreegee> one sec
[23:29] <kreegee> on the files i extracted or on the hole dvd?
[23:29] <kreegee> *whole
[23:30] <kreegee> i added some debug stuff to the changeStackTo-function, and apparently it loads a_Data.mhk
[23:33] <digitall> One thing.. You may need to run the installer to get extras.mhk
[23:33] <kreegee> i got the extras.mhk
[23:33] <kreegee> http://kreegee.pastebin.com/yqFMpY8S
[23:33] <digitall> Though I doubt that is linked..
[23:33] <kreegee> these are the extracted files
[23:33] <kreegee> now for the hole dvd
[23:33] <digitall> OK Two secs.. Will compare to mine
[23:34] <digitall> I don't have all the versions of Riven though..
[23:34] <kreegee> i'm not sure if these are all files, some might be missing, but both a_data.mhk and b_data.mhk reference tspit
[23:34] <kreegee> (only a_data.mhk gets loaded though)
[23:35] <digitall> You may have a unknown version of some kind.. What does the detector say when you add the game? No md5sums complaints?
[23:35] <kreegee> nope
[23:35] <digitall> Hmm..
[23:36] <digitall> Two secs..
[23:37] --> criezy joined #scummvm.
[23:37] #scummvm: mode change '+o criezy' by ChanServ!ChanServ@services.
[23:37] <criezy> Hi
[23:38] <digitall> kreegee: You appear to be missing quite a number of files of *_data.mhk
[23:38] --> clone2727 joined #scummvm.
[23:38] #scummvm: mode change '+o clone2727' by ChanServ!ChanServ@services.
[23:38] <digitall> i.e. b2_data.mhk, g_data.mhk etc.
[23:38] <digitall> CLONE!
[23:38] <clone2727> Yes, I see there was some conundrum involving Mohawk going on :P
[23:38] <kreegee> ok
[23:39] <digitall> If you can't find them on the DVD, either you have very odd version.. or they are packed in the installers setup cabinets.
[23:39] <digitall> You may need to use unshield or cabextract to extract them... or run the installer in Windows.
[23:39] <clone2727> The DVD does not need the installer to be run
[23:40] <digitall> kreegee: For clone2727, please describe the exact version you have...
[23:40] <kreegee> found some more files
[23:40] <kreegee> german trilogy
[23:40] <digitall> The name of the collection...
[23:40] <kreegee> this stupid dvd has the same files in different folders
[23:40] <clone2727> kreegee: The sound files? Or the data ones?
[23:40] <kreegee> data
[23:40] <kreegee> a_data for example is in two directories
[23:40] <kreegee> exactly the same file
[23:41] <digitall> Then just take one :-)
[23:41] <kreegee> copying the missing stuff now
[23:41] <kreegee> sorry for probably just wasting your time ;)
[23:41] <digitall> Not a waste of time.. I'm watching Inspector Morse...
[23:41] <kreegee> The following directories allow Riven to be played directly
[23:41] <kreegee> // from the DVD.
[23:41] <kreegee> (from riven.cpp)
[23:41] <kreegee> ;)
[23:42] <kreegee> oh
[23:42] <kreegee> lets try something
[23:42] --> FreeTHX joined #scummvm.
[23:50] <clone2727> kreegee: You're still getting "[22:59] <kreegee> Could not load stack tspit! " ?
[23:50] <clone2727> kreegee: You need t_Data1.mhk, t_Data2.mhk, and t_Sounds.mhk for tspit
[23:51] <clone2727> The reference to "tspit" in the a_Data.mhk file is how aspit knows to get to tspit (via the NAME resource)
[23:51] <kreegee> ok
[23:51] <kreegee> this dvd only has t_Data, no t_data1 or 2
[23:52] <clone2727> That's fine
[23:52] <kreegee> anyway
[23:52] <kreegee> its working now ;)
[23:53] <clone2727> kreegee: As for your codec question, Riven doesn't need any more. Only Myst Masterpiece Edition does (SVQ1, SVQ3, QDM2). QDM2 is almost in working condition
[23:56] <kreegee> ok, thx
[23:56] <kreegee> as said, just playing around and looking if i come across a bug which i might fix myself (and implementing a codec is probably not a part of that set)
[23:58] Action: digitall starts twitching at mention of the codec of doom...
[23:58] <clone2727> kreegee: One thing that you could do is implement a more complete version of the movie display matrix within QuickTime
[23:59] <clone2727> (qt_player.cpp:588 and 684)
[00:00] --- Thu Sep 2 2010