#pentagram@irc.freenode.net logs for 25 Jun 2003 (GMT)

Archive Today Yesterday Tomorrow
Pentagram homepage


[01:04:41] --> Kirben has joined #pentagram
[01:04:41] --- ChanServ gives channel operator status to Kirben
[05:02:16] --> Cashman has joined #pentagram
[05:02:19] <-- Cashman has left IRC (Client Quit)
[07:41:31] --> Colourless has joined #Pentagram
[07:41:32] --- ChanServ gives channel operator status to Colourless
[07:41:56] <Colourless> inlined virtual methods. The how and why
[07:42:09] <Colourless> it's all pretty simple. You inline, when you call a specific implementation
[07:42:27] <Colourless> for instance by doing SomeClass::TheInlinedVirtualFunciton()
[07:42:52] <Colourless> inlining a pure virutal function... as fingolfin says, cute :-)
[07:43:05] <-- Colourless has left IRC (Client Quit)
[07:43:58] --> Cashman has joined #pentagram
[07:48:28] <-- Cashman has left IRC (Client Quit)
[08:24:38] --> Cashman has joined #pentagram
[08:40:54] <-- Cashman has left IRC ()
[08:57:33] --> Zxcvb_ has joined #pentagram
[08:58:00] <Zxcvb_> anyone alive here?
[08:59:42] <DarkeZzz> Somewhat. *grin*
[09:00:11] <Zxcvb_> any idea why the trick where you get executed by beren but live and therefore can't be executed again works?
[09:00:29] <Zxcvb_> basically, you drink a purple and black potion, then steal from rhian the gem dealer
[09:00:33] <DarkeZzz> You'll have to ask someone else, I've never played u8. *grin*
[09:01:18] <Zxcvb_> as beren is summoned for the execution you start a conversation and time it so you are at a place where you select what to say next when beren blows you to bits
[09:01:37] <Zxcvb_> you will then survive the execution, but you must leave the screen (area, zone) before it realizes you are dead
[09:01:39] <DarkeZzz> What's a purple and a black potion for?
[09:01:49] <Zxcvb_> purple is invulnerable, and black is invisible
[09:02:12] <Zxcvb_> but you must be in conversation, and at a place where you choose what to say next just as beren destroys you
[09:02:29] <Zxcvb_> I am guessing it has to do with the fact that ultima 8 uses cooperative multithreading in usecode
[09:02:30] <DarkeZzz> Heh. Sounds like a race condition.
[09:02:46] <Zxcvb_> DarkeZzz: cause by cooperative multithreading instead of preemptive?
[09:03:21] <Zxcvb_> DarkeZzz: perhaps the death/execution thread can't get enough time to set the avatar_death flag to yes?
[09:04:00] <Zxcvb_> DarkeZzz: all I know is that you must have the invisible and immune to damage flags set, and you must be in conversation at a conversation menu at just the right time
[09:04:51] <DarkeZzz> I don't think it'd make a difference. I think it's probably caused by the fact you've have a process_exclude set on the avatar's shape, so that no other processes can act upon it. The process can't remove your invlunerability, (which it doesn't notice), beren attacks and thinks he's killed you (presumably it's a 'insta-kill' attack in the middle of usecode), them something weird happens.
[09:05:17] <Zxcvb_> DarkeZzz: but why is invisibility and immunity to damage required?
[09:05:50] <DarkeZzz> Immunity to damage would be required so that Devon can't reduce your hp to less then 0 with his attack.
[09:05:58] <Zxcvb_> DarkeZzz: beren
[09:06:01] <DarkeZzz> Have you tried it without the invisibility?
[09:06:05] <DarkeZzz> Yeah, him, sorry. *grin*
[09:06:07] <Zxcvb_> DarkeZzz: yeah, doesn't work
[09:06:16] <DarkeZzz> Curious.
[09:06:20] <Zxcvb_> DarkeZzz: you knew devon, yet you haven't played it
[09:06:36] <Zxcvb_> DarkeZzz: perhaps invisibility blocks the immunity flag from being removed
[09:06:51] <Zxcvb_> DarkeZzz: you must go invisible _after_ becoming immune to damage
[09:07:33] <Zxcvb_> DarkeZzz: perhaps the game designers thought of the immunity to damage idea and put in special code to handle it, but it only removes the last flag set if in conversation
[09:09:42] <Zxcvb_> who is often here who works on ultima 8 usecode?
[09:09:47] <DarkeZzz> I don't think so. Don't really know though. I'll glance at u8's usecode.
[09:10:28] <Zxcvb_> you have a decompiler that works with ultima 8?
[09:10:53] <DarkeZzz> Only a disassembler atm. The decompiler just barely works.
[09:11:48] <Zxcvb_> how are you supposed to play ultima 7 on the zaurus anyway?
[09:12:05] <Zxcvb_> it seems that most of the time is spent avoiding fights because of the difficulty in controlling the avatar
[09:12:34] <DarkeZzz> I dunno. Never even seen a Zaraus except in pictures. *grin*
[09:14:05] <DarkeZzz> Ok, it looks like Beren yells that he'll destroy you, targets the avatar, then puts him in stasis...
[09:15:06] <Zxcvb_> DarkeZzz: no, he destroys you with an insta-death attack
[09:15:21] <Zxcvb_> DarkeZzz: with graphics of little wriggling bits
[09:16:39] <Zxcvb_> www.it-he.org/ultima8.htm
[09:16:43] <Zxcvb_> click on "magna carta"
[09:17:44] * DarkeZzz eeews at Beren's usecode. It's a mess. He's really not sure what's happening where in it.
[09:18:36] <Zxcvb_> so is that mess why the trick works?
[09:20:51] <Zxcvb_> could he actually put you into stasis, make you invisible, dispaly the wiggling graphics and reduce hp to 0?
[09:21:05] <Zxcvb_> apparently the invis is so you will be visible after the event
[09:21:36] <DarkeZzz> *nod* The invisibility would be so it would wear off and return you to the appropriate shape so you were visible again, I would guess.
[09:22:16] <Zxcvb_> and the conversation is to prevent the stasis?
[09:22:50] <Zxcvb_> that seems like an ugly way to do the execution
[09:23:52] <DarkeZzz> Hmm... yeah, the conversation 'should' have prevented the status, depending upon how they did things.
[09:25:06] <Zxcvb_> so basically the conversation prevents the stasis or freezing, the invincibility stops the hit point loss, and the invisibility restores your shap
[09:25:15] --> wjp has joined #pentagram
[09:25:15] --- ChanServ gives channel operator status to wjp
[09:25:20] <Zxcvb_> shape, that is
[09:25:32] <wjp> invisibility doesn't make you change shape
[09:25:37] <wjp> it just sets the invisibility flag
[09:25:50] <wjp> I seem to remember that the 'Blow Up!' cheat also makes you invisible
[09:26:08] <wjp> so it's a reasonable assumption that the invisibility potion is to make you visible again afterward
[09:26:14] <Zxcvb_> wjp: I was wondering why invisibility was needed to help survive beren's execution
[09:26:27] <wjp> Darke: process exclude isn't on a shape; it's on an <item num, process type> pair
[09:26:29] <Zxcvb_> wjp: and why the conversation and invulnerability potions were so important
[09:26:54] <wjp> it's possible they accidentally made two different processes use the same process type, I guess
[09:26:55] <Zxcvb_> that is, how does that trick work?
[09:27:24] <wjp> once we have Beren blowing you up working in pentagram we can take a look :-)
[09:27:38] <wjp> (or maybe it does actually already work?)
[09:28:11] <Zxcvb_> I would have guessed the invisiblity and immunity to damage flags, combined with the conversation don't give the check_beren_death thread any time to run
[09:28:16] <wjp> btw, I've been meaning to ask... what exactly happens when you drink an invisibility potion?
[09:28:46] <Zxcvb_> you turn translucent as if you were made of glass, and monsters ignore you
[09:28:48] <wjp> time to run isn't an issue, I would guess; while you're talking things have plenty of time to run
[09:29:06] <wjp> does the rest of the rest of the screen change in any way like in U7?
[09:29:25] <Zxcvb_> that is only if the conversation and other code give up time to the beren stuff
[09:29:33] <wjp> they do
[09:29:38] <Zxcvb_> no, it's more like ultima 9
[09:29:41] <wjp> if they don't the screen won't even get updated
[09:30:01] <wjp> conversation doesn't do anything while waiting for an answer
[09:30:14] <wjp> it is suspended while waiting for you to reply
[09:30:16] <Zxcvb_> apparently beren's usecode is a real mess
[09:30:34] <wjp> just means we'd need to spend more time figuring it out :-)
[09:31:06] <Zxcvb_> how well is pentagram coming along anyway?
[09:31:13] <wjp> very well :-)
[09:32:55] <Zxcvb_> I noticed exult took a while
[09:33:10] <Zxcvb_> is it that the last 5% of compatibility and fixes takes a good 80% of the time?
[09:33:19] <wjp> probably :-)
[09:33:30] <wjp> but we're not in the last 5% yet with pentagram :-)
[09:33:50] <Zxcvb_> so many ultima emulators being worked on
[09:42:41] <-- DarkeZzz has left IRC (kornbluth.freenode.net irc.freenode.net)
[09:43:20] --> DarkeZzz has joined #pentagram
[09:49:57] <-- Zxcvb_ has left IRC ("ircII EPIC4-1.0.1 -- Are we there yet?")
[09:55:58] <-- wjp has left #pentagram ("bbl")
[10:48:55] --> Fingolfin has joined #pentagram
[10:48:56] --- ChanServ gives channel operator status to Fingolfin
[11:22:58] --> Cashman has joined #pentagram
[11:23:05] <Cashman> hi
[11:26:07] <Fingolfin> lo
[11:51:23] <-- Cashman has left IRC ()
[13:20:37] --> Colourless has joined #Pentagram
[13:20:42] --- ChanServ gives channel operator status to Colourless
[13:21:09] <Colourless> hi
[13:22:17] <Fingolfin> hi here too
[13:27:26] <Colourless> Fingolfin: i sent an email asking why you changed ButtonWidget.cpp
[13:28:26] <Colourless> hmm, now looking at the function i think i can see why, BUT, that was the incorrect fix
[13:30:39] <Fingolfin> what change did I make?
[13:30:40] * Fingolfin looks
[13:31:09] <Colourless> you commented out a line- Gump *ret = SimpleTextWidget::OnMouseDown(button,mx,my);
[13:31:10] <Colourless> +// Gump *ret = SimpleTextWidget::OnMouseDown(button,mx,my);
[13:31:24] <Fingolfin> yeah I see your mail now
[13:31:48] <Colourless> thinking about it, there 'should' have been the line if (ret) return ret;
[13:31:54] <Fingolfin> I didn't check OnMouseDown for side effects I guess. but essentially, remove "Gump *ret" and I am happy - it's not used, thus causing a warning
[13:31:58] <Fingolfin> ah :-)
[13:33:21] <Fingolfin> will you fix it, or want me to do it?
[13:42:37] <-- Colourless has left IRC ("bbl")
[14:17:11] <-- Kirben has left IRC ("System Meltdown")
[14:30:54] --> Colourless has joined #Pentagram
[14:30:54] --- ChanServ gives channel operator status to Colourless
[14:40:41] <Colourless> thx Fingolfin
[14:49:53] <Fingolfin> sorry for breaking it in the first place, should have looked more carefully at the code, it's a pretty obvious mistake in retrospect :-)
[15:32:17] <Colourless> got to go
[15:32:20] <-- Colourless has left IRC ("casts invisibility")
[15:42:34] --> wjp has joined #pentagram
[15:42:34] --- ChanServ gives channel operator status to wjp
[16:25:47] --- Fingolfin is now known as Fingolfin|away
[17:22:27] <wjp> savegames... I'll go ramble a bit :-)
[17:22:38] <wjp> we need to save processes and items, mainly
[17:22:52] <wjp> s/items/Objects/, to be more exact
[17:23:40] <wjp> with objects, we need to make sure any container-contents links are preserved
[17:24:04] <wjp> hm, dinner; will continue later :-)
[17:46:36] <wjp> ...continuing...
[17:47:03] <wjp> probably by changing Containers to use objids instead of Item*s
[17:48:50] <wjp> I think we should version savegames somehow, to prevent incompatibility problems
[17:50:08] <wjp> processes: we should probably store processes in the order they're on the Kernel's process list
[17:50:36] <wjp> (just to prevent weird savegame-induced sync issues)
[17:51:04] <wjp> (or at least store that order somewhere; doesn't necessarily have to be the order in which the processes are stored, obviously)
[17:53:13] <wjp> it would be nice to structure the save/load mechanism in such a way that a subclass can call its base class' save/load function
[17:53:30] <wjp> (and pretty much required :-) )
[18:01:59] <wjp> savegame structure...
[18:02:12] <wjp> we could go with something file-based like the original and U7
[18:02:19] <wjp> or something more block bases
[18:02:23] <wjp> s/bases/based/
[18:03:31] <wjp> (of course, at some level they're quite the same...)
[18:03:47] <wjp> filenames are just a way of identifying blocks, I guess :-)
[18:03:54] --- Fingolfin|away is now known as Fingolfin
[18:06:16] <wjp> wb :-)
[18:06:27] <wjp> please say something so it at least _looks_ like I'm not talking to myself ;-)
[18:07:33] <wjp> oh... just thought up a pretty convincing reason why we can't use objids for Container contents...
[18:07:50] <wjp> containers+contents don't have objids at all times :-)
[18:07:55] <wjp> (oops :-) )
[18:08:25] <Fingolfin> yo willem
[18:08:27] <Fingolfin> hm
[18:08:29] <wjp> because we need to store non-loaded maps too
[18:08:52] <wjp> not a big problem, though
[18:09:08] <Fingolfin> making a versioned save game system can be tricky, but is something I'd absolutley recommend to everybody. Ideally one where you can still open older save game versions in newer version of the engine, of course :-)
[18:09:22] <wjp> just need to think things through a bit to prevent having two separate systems for loaded/non-loaded items
[18:09:42] <wjp> won't that convert some loading functions into huge if statements? :-)
[18:09:58] <Fingolfin> not if you do it right
[18:10:40] <Fingolfin> I doN't claim we did it fully right in ScummVM, but we did the closest to right we could do while still staying compatible with the "old" save game system (where each savegame revision invalidated all old savegames).
[18:10:49] <Fingolfin> still situation may be different here again
[18:11:01] <wjp> we don't have an old save game system :-)
[18:11:12] <Fingolfin> the most annoying issues (besides versioning) are object references / pointeres of course
[18:11:25] <wjp> we don't really have any of those
[18:11:40] <Fingolfin> are there any cases where an object (in the C++ sense; I.e. an instance of a class) is being referred to in more than one place?
[18:11:41] <wjp> most object pointers are ids internally
[18:11:56] <Fingolfin> id's is the best of course
[18:11:58] <wjp> (since they have to be accessible in usecode anyway, in a non-pointer format)
[18:12:01] <Fingolfin> cause it's trivial to serialize those
[18:12:33] <wjp> the only place I can think of where we don't use ids for objects is with Containers/contents, like I mentioned just when you returned
[18:13:03] <wjp> but that's fairly simple to do too, for example by storing contents directly after a container followed by an end-of-contents marker
[18:14:19] <Fingolfin> yup that would be exactly what I propose
[18:14:33] <Fingolfin> but it only works well if the objects inside the container are not pointed at in some other place
[18:14:47] <Fingolfin> hence my question whether it's possible that an object is pointed to in more than one place :-)
[18:14:53] <wjp> should be only by id
[18:15:19] <Fingolfin> if that's not a case, then reading/writing a container simply would write the container desc, then a "numberOfChildren" int, then simply write out all the children. that would even work with nested children
[18:15:55] <Fingolfin> and when you read stuff in, it registeres itself with the object manager (or wahtever it's called in pentagram... still haven't read up on the source :-/)
[18:16:15] <wjp> kernel :-)
[18:16:42] <Fingolfin> what was the URL to the doxygen stuff again?
[18:17:00] <wjp> http://www.math.leidenuniv.nl/~wpalenst/pentagram
[18:17:14] <wjp> I'm sure you'll be able to find it from there ;-)
[18:17:17] * Fingolfin bookmarks it this time
[18:17:40] <wjp> I've also been keeping cvs snaps of some projects (follow the cvs link from there)
[18:18:07] <wjp> but you're probably a member of most of those projects :-)
[18:18:24] <Fingolfin> I think for both pentagram & scummvm doxygen, it might be a good idea to insert a text on the front page which gives people some pointers where to start... like "execution flow starts here and here" (with links) and stuff like that
[18:18:34] <Fingolfin> heh
[18:20:10] <wjp> yeah, might be a good idea
[18:20:38] <wjp> some kind of high-level code overview
[18:21:08] <wjp> "graphics things are in the graphics/ dir; world stuff is in the world/ dir; ", etc... ;-)
[18:24:02] <wjp> the main thing to look at first is kernel/GUIApp.cpp, I guess
[18:24:12] <wjp> it mainly runs things
[18:24:36] <wjp> world/CurrentMap.cpp handles map issues
[18:24:52] <wjp> world/Item.cpp does item-related things (surprise, no? ;-) )
[18:25:17] <wjp> filesys/FileSystem.h for an overview of file services
[18:25:24] <wjp> (virtual paths... :-) )
[18:25:45] <wjp> graphics/*RenderSurface.* for low-level graphics
[18:25:56] <wjp> world/ItemSorter.cpp for, well, sorting items (for rendering)
[18:26:12] <wjp> usecode/UCMachine.cpp handles usecode
[18:26:54] <wjp> conf/ is Exult's config stuff, with multi-file support added
[18:30:26] <wjp> where do I look in scummvm to get an idea of its savegame system?
[18:34:52] <Fingolfin> in scumm/saveload.cpp
[18:35:16] <Fingolfin> mind you: don't copy that. I'd do many things differently if I had to implement the system
[18:35:16] <Fingolfin> =)
[18:35:20] * wjp sighs... it's nearly impossible to get through to anoncvs these last few days :-(
[18:35:22] <Fingolfin> but some things aren't that bad
[18:35:25] <Fingolfin> :-(
[18:37:07] <wjp> hm, all the data to save is in saveload.cpp?
[18:37:33] <wjp> I'd probably prefer to move more to the classes that need to be saved
[18:50:48] <wjp> definitely some useful ideas in there
[18:51:53] <wjp> oh, I see not all save data is in saveload.cpp :-)
[18:53:38] <wjp> wrt. OFFS/SIZE: forty-two? :-)
[18:58:44] <Fingolfin> not all save data: some is in actor.cpp / imuse*.cpp... ideally i'd spread out the data over the classes it belongs into, but that code was written to stay compatible to the old format, so my hands are pretty tied
[18:58:52] <Fingolfin> and yeah, 42 was of course the only possible value =)
[18:59:14] <wjp> yeah, I noticed afterwards
[19:04:48] * wjp converts some 'accidental' internal pointers to IDs :-)
[19:18:13] <wjp> I've been wondering... the SF activity rating is a percentile, right?
[19:19:03] <wjp> if we're number 272 with 95.39%, does that mean there are about 5900 active projects?
[19:19:31] <wjp> (out of over 10 times that number total projects)
[19:22:52] <-- DarkeZzz has left IRC (kornbluth.freenode.net irc.freenode.net)
[19:23:38] --> DarkeZzz has joined #pentagram
[19:36:31] --> Dark-Star has joined #pentagram
[19:37:04] <wjp> hi
[19:49:37] <-- Dark-Star has left IRC (kornbluth.freenode.net irc.freenode.net)
[19:49:37] <-- wjp has left IRC (kornbluth.freenode.net irc.freenode.net)
[19:49:37] <-- Fingolfin has left IRC (kornbluth.freenode.net irc.freenode.net)
[19:49:37] <-- DarkeZzz has left IRC (kornbluth.freenode.net irc.freenode.net)
[19:49:57] --> wjp has joined #pentagram
[19:49:57] --> Dark-Star has joined #pentagram
[19:49:57] --> DarkeZzz has joined #pentagram
[19:49:57] --> Fingolfin has joined #pentagram
[20:14:36] <Fingolfin> wjp: I think you shouldn't try to put *any* senseful meaning on the SF.net stats :-)
[20:47:41] <wjp> you have a point there :-)
[21:20:08] --> Dominus has joined #pentagram
[21:20:19] <wjp> hi
[21:20:40] <Dominus> hi
[21:20:49] <Dominus> first: I need to report my absence for the coming two weeks
[21:21:12] <Dominus> I'll be in italy and most likely totally away from any kind of web access
[21:22:00] <Dominus> or better: I'll not bring my laptop with me as at the place we'll be staying members of my family tend to 'loose' things
[21:22:10] <wjp> uh, ah
[21:22:15] <Dominus> my grandpa 'lost' his mercedes there once...
[21:22:27] <wjp> ugh
[21:23:20] <Dominus> exactly. I won't try my luck with my precious laptop
[21:23:59] <Dominus> my grandparents own a little place there and it has been robed (as in everything that was not bolted to the wall) got taken away
[21:25:26] <Dominus> about savegames: will they be displayed like in the original?
[21:25:42] <wjp> I haven't made up my mind about the interface yet
[21:25:53] <wjp> I kind of dislike the original's interface, though
[21:26:19] <Dominus> hmm, let me quickly check :-)
[21:26:32] <Fingolfin> yeah, it's not that great... but seriously, the interface is probably less of a concern than the savegame code itself :-)
[21:26:50] <wjp> yeah, which is why I haven't thought about it much, yet :-)
[21:27:06] <wjp> obviously we'll need more savegames than the original supported
[21:27:40] <wjp> having infinitely many savegames in Exult has kind of got me thinking we need a better way to order them than linearly
[21:27:49] <wjp> having 'chapters' of savegames in the diary might be nice
[21:34:01] <wjp> hmm.. the list of things to save is quite lengthy :-)
[21:34:01] <-- Dominus has left IRC (Read error: 104 (Connection reset by peer))
[21:41:11] --> Dominus has joined #pentagram
[21:41:19] <wjp> wb
[21:41:43] <Dominus> *cough*
[21:41:45] <Dominus> hi again
[21:42:02] <Dominus> somehow my keyboard stopped working and I was trapped within u8 and could not end the game
[21:42:56] <Dominus> something to remember once the interface is somewhat working - some way to quit without a mouse
[21:43:38] <wjp> yeah, I'd like that too
[21:44:05] <Dominus> I was quite suprised there is no way in the original
[21:44:16] <wjp> hmm... we do have a couple of pointer lying around
[21:44:17] <wjp> s/pointer/pointers/
[21:44:46] <wjp> some no problem, some problem :-)
[21:45:11] <wjp> just need to replace those pointers with objids, I guess
[21:45:53] <Dominus> looking at the original's savegame display I'm conviced that it is rather boring...
[21:46:10] <Dominus> I'd like something more like Exult's way with screenshots...
[21:46:24] <wjp> We need a miniature screenshot, snapshot of the avatar's paperdoll, ... :-)
[21:47:07] <wjp> with our gump architecture they should be quite easy to do... just get the gumps to paint to a temporary RenderSurface
[21:47:34] <Dominus> why of the avatar's paperdoll?
[21:47:43] <wjp> why not? :-)
[21:47:59] <wjp> makes things less 'boring'? ;-)
[21:49:03] <Dominus> and a snapshot of the Avatar's whereabouts on the worldmap...
[21:49:08] <Dominus> if there were one
[21:49:12] <wjp> heh :-)
[21:49:26] <wjp> sure, draw me a world map and I'll implement it ;-)
[21:49:38] <wjp> (someday, maybe ;-) )
[21:49:44] <Dominus> I just need to scan it :-)
[21:49:56] <wjp> ah, but that one's useless
[21:50:20] <Dominus> yeah, I always wondered why they included that one
[21:50:35] <wjp> I wonder how well the maps would fit together at the edges
[21:50:40] <wjp> (the in-game maps, that is)
[21:50:48] <Dominus> hmm
[21:50:51] <wjp> well, it wouldn't be an ultima without a cloth map ;-)
[21:51:09] <Dominus> he he, even if it is totally useless
[21:51:25] <Dominus> they made the game that much harder to play with no useful map
[21:52:11] <wjp> I've been really lost several times
[21:52:22] <wjp> just because I missed a turn in the catacombs :-)
[21:52:38] <wjp> "WHERE IS Argentrock Isle?!?!" :-)
[21:52:54] <Dominus> another thing to look forward to: high screen res to get a bit more orientated
[21:53:00] <wjp> I must've tried every boat in the game to see if I could row it :-)
[21:53:08] <Dominus> he he
[21:53:24] <wjp> (days later: "Hey, what's this door for?" ;-) )
[21:53:39] <Dominus> hehe
[21:53:53] <wjp> the good old days :-)
[21:54:31] <Dominus> I recently gave The Dig another round with scummvm
[21:54:46] <Dominus> and was lost at exactly the same point as years before
[21:54:47] <wjp> never played that
[21:55:06] <Dominus> nice adventure...
[21:55:34] * wjp hmms about the savegame interface
[21:55:58] <wjp> I don't know
[21:56:07] <wjp> having to scroll through a diary is too slow for my taste
[21:56:11] <Dominus> and then I downloaded a new snapshot and the game was broken and I had overwritten my savegames already and couldn'T use the older version with it anymore...
[21:56:23] * Dominus nods
[21:56:45] <wjp> lots of dig-related changes to scummvm recently, IIRC
[21:56:58] <wjp> from font rendering to getting stuck in rooms
[21:57:07] <Dominus> how about putting the exult interface in there
[21:57:10] <Dominus> joking
[21:57:24] <wjp> I'm considering it :-)
[21:57:24] <Dominus> have to look again when I'm back
[21:58:13] <Dominus> but the diary is just to cumbersome
[21:58:38] <wjp> how about saving/loading from the console? :-)
[21:58:41] <Dominus> I want a list of my savegames and when I choose one I want to get some infos about it
[21:58:58] <Dominus> he he
[21:59:01] <wjp> "# save 37"
[21:59:07] <wjp> "# info 42"
[21:59:18] <wjp> :-)
[22:00:04] <Dominus> that sounds nice and easy, will you handle the user support for that ? :-)
[22:01:28] <wjp> no, that's your job, remember ;-P
[22:01:36] * wjp runs and hides
[22:01:41] * wjp ducks, too ;-)
[22:01:58] * Dominus throws the faq and misses of course
[22:02:19] * Dominus throws the documentation as wjp 'unducks'
[22:02:23] <wjp> you're supposed to say you hit, after which I'm supposed to come with a way not to get hit ;-)
[22:02:28] <wjp> ah, better :-)
[22:02:49] * wjp gets squarely hit by the documentation while we was trying to make witty remarks
[22:02:54] <wjp> s/we/he/
[22:03:03] <Dominus> :-)
[22:03:37] <wjp> (good thing we don't have that much documentation for pentagram yet :-) )
[22:04:17] <Dominus> how will you save in pentagram? directly to file and gamedat (or sth similar)?
[22:04:30] <Dominus> and quicksave to gamedat or also to a file?
[22:05:24] <wjp> gamedat is history
[22:05:40] <Dominus> I remember you talking about that
[22:05:49] <Dominus> what will replace it?
[22:05:57] <Dominus> just save to file?
[22:06:05] <wjp> RAM :-)
[22:06:20] <Dominus> :-)
[22:06:27] <wjp> quicksave and normal saves will go straight to a file from RAM
[22:06:45] <wjp> quicksave should be exactly the same as a normal save
[22:06:49] <wjp> (much less confusing)
[22:07:07] * wjp remembers a lot of "which file is the quicksave?" questions
[22:07:10] <Dominus> yes (but with a distinct filename?)
[22:07:41] <wjp> probably
[22:07:51] <wjp> maybe just index 0 or something
[22:07:51] <wjp> we'll see
[22:08:14] <Dominus> index 0 sounds fine
[22:36:39] <-- wjp has left IRC ("Zzzz...")
[23:31:35] <-- Dark-Star has left IRC ()
[23:36:05] <-- Dominus has left IRC ("enough for now")