[00:26:10] <-- sbx has left IRC ("BRB")
[00:30:22] --> SB-X has joined #pentagram
[01:23:26] --> Enirya has joined #pentagram
[01:24:13] <Enirya> morning?
[01:24:50] <Colourless> morning :-)
[01:25:08] <Colourless> must be like 1am for you in nl
[01:25:30] <Enirya> more like 3.30. but I live on US time anyway :p
[01:25:48] <Enirya> was bored and decided to come in and say hi for no particular reason.
[01:26:30] * Enirya has been following pentagram since it was just a map viewer thingymabob. >_>
[01:26:54] <Colourless> :-)
[01:28:26] <Enirya> I love U8 for some reason while most people quite hated it.
[01:29:44] <Colourless> i quite liked it too
[01:30:11] <Enirya> it just had way better atmosphere than the other ultimas to me, and way better than any other game there was at the time.
[01:30:37] <Colourless> yeah i know what you mean.
[01:31:33] <Enirya> plus making that stupid guy who kept blowing you up for being a criminal take a walk into the water and get eaten never got old XD
[01:32:25] <Colourless> hehe
[01:33:13] <Enirya> anyway, just wanted to say keep up the great work ... and also, stop hogging all the coders for projects that need no art, us artists need our sparse few talented coders, dangit ;P
[01:33:26] <Colourless> hehe
[01:33:48] <Colourless> but it's easier to code when everything is already done :-)
[01:34:09] <Enirya> and its easier to make art when you don't need to do the code yourself as well ;_;
[01:35:09] <Colourless> starting a game from scratch is a lot lot lot harder than just recreating a game engine
[01:35:09] * Enirya tried that for years, one person devteams never make it too far ;P
[01:35:21] <Enirya> ... It is? O_o;
[01:36:25] <Enirya> but you're fumbling around guessing at things and constantly reverse engineering the original engine. instead of just coding something and then BAM, it's there.
[01:36:59] <Colourless> the reason it's harder is because you need all the content to use with the game engine
[01:37:35] <Colourless> when doing something like pentagram all the data is already there, we just need to make it work
[01:37:42] <Colourless> in particular things like all the usecode scripting
[01:37:52] <Colourless> that is a fairly substantial amount of code that we don't need to write
[01:38:07] <Enirya> hardly, you only need enough content to test the base functions ... then again I've never really tried too hard at a plot driven single player game, so I suppose in that scenario it might be more true.
[01:38:12] <Colourless> if we had to write all the usecode ourselves we'd probably never finish
[01:38:34] <Colourless> sort of thing if you were paid to do it, then you'd do it
[01:38:39] --> Kirben2 has joined #pentagram
[01:39:33] <Enirya> ah, I suppose you have a point there. I dunno, I just reason from the angle that the programmers program, the scripters script, the artists make art and etc. Which means the programmer's job wouldn't be significantly larger.
[01:39:56] <SB-X> especially true in U8 where usecode controls everything
[01:40:11] <Enirya> since you dont have to code any actual behaviour?
[01:40:18] <Enirya> well, little behaviour.
[01:40:56] <SB-X> especially true that it would take longer/be hard to finish
[01:41:19] <SB-X> but I don't know I just lurk here
[01:44:12] * Enirya theorizes that the job wouldnt necesarily be much longer/harder to finish, but that requires some form of scripting that can be done by people who are not the programmers.
[01:44:52] <Enirya> then again what do I know :p I once tried to understand the whole usecode thing and didnt get too far.
[01:55:56] <-- Kirben has left IRC (Read error: 110 (Connection timed out))
[02:14:22] <Enirya> hmm, well, time to skidaddle. buh-bye!
[02:14:27] <-- Enirya has left IRC ("<Kalciane> goddamnit, internet, i want my vanilla ice mp3. <Enirya> ... why? <Kalciane> it's funky fresh. <Enirya> actually i)
[02:21:05] <SB-X> actually i... ?
[02:29:01] <-- oaQ^ has left IRC (sterling.freenode.net irc.freenode.net)
[02:29:01] <-- Kirben2 has left IRC (sterling.freenode.net irc.freenode.net)
[02:29:01] <-- megawatt has left IRC (sterling.freenode.net irc.freenode.net)
[02:29:01] <-- SB-X has left IRC (sterling.freenode.net irc.freenode.net)
[02:29:01] <-- servus has left IRC (sterling.freenode.net irc.freenode.net)
[02:29:01] <-- paranoidd_away has left IRC (sterling.freenode.net irc.freenode.net)
[02:29:01] <-- wjp has left IRC (sterling.freenode.net irc.freenode.net)
[02:29:46] --> wjp has joined #pentagram
[02:29:46] --> paranoidd_away has joined #pentagram
[02:30:05] --> oaQ^ has joined #pentagram
[02:31:13] --> Kirben2 has joined #pentagram
[02:31:13] --> SB-X has joined #pentagram
[02:31:13] --> servus has joined #pentagram
[02:31:13] --> megawatt has joined #pentagram
[03:19:40] <-- Colourless has left IRC ("brb... hopefully")
[03:30:05] <SB-X> hmm, my identd returns the userid for sbx
[03:30:22] <SB-X> firstname.lastname@example.org
[04:07:39] --> freaks has joined #pentagram
[04:07:46] <freaks> hi there
[04:08:08] <freaks> i7ve just downloaded and install pentagram on my fedora 4 box,
[04:08:25] <freaks> since then i can't load game anymore
[04:08:27] <SB-X> hi
[04:08:39] <freaks> version mismatch ..
[04:08:43] --- paranoidd_away is now known as paranoidd
[04:08:47] <freaks> hi SB-X
[04:09:09] <SB-X> you couldn't load the games before installing pentagram could you?
[04:09:19] <freaks> i installed a previous version few weeks ago, the load/save were working nice
[04:09:30] <SB-X> oh
[04:09:35] <SB-X> yeah, gotta start a new game
[04:09:46] <freaks> no you get me wrong
[04:10:08] <freaks> load don't work at all, even if you start a new game ;)
[04:10:36] <SB-X> oh ok
[04:10:45] <SB-X> did you report it on sf bugtracker?
[04:10:45] <freaks> it say the save is version 212 and the game is version 0 ... the md5 keys looks identik tho
[04:10:51] <freaks> no
[04:10:58] <freaks> i don't know how to do it
[04:11:11] <SB-X> sf.net/projects/pentagram, Bugs
[04:11:14] <freaks> would you explain that to me ? so i could help
[04:11:17] * SB-X pokes wjp.
[04:11:22] <SB-X> or maybe he will wake up and help you
[04:11:28] <SB-X> i've got to go
[04:11:28] <SB-X> cya
[04:11:31] <-- SB-X has left IRC ("casts gate travel")
[04:12:14] <servus> Go here, freaks : http://sourceforge.net/tracker/?atid=471706&group_id=53819&func=browse
[04:30:54] <freaks> servus, ok, i created an sf.net account, and submited the bug report
[04:30:59] <freaks> thanks for the help
[04:31:39] <freaks> now i wonder how i can update my u8 to version 2.12
[04:32:34] <freaks> i have the game, i have the patch... i'm using linux, and wine can't do it, neither cedega ... the patch is an old dos mode prog
[04:32:46] <servus> DosBox?
[04:32:52] <freaks> hum :)
[04:32:52] <servus> Definitely DosBox.
[04:32:56] <freaks> excellent idea
[04:33:01] <freaks> :)
[04:33:21] --> Colourless has joined #Pentagram
[04:33:21] --- ChanServ gives channel operator status to Colourless
[04:54:55] <freaks> it worked nice i could apply the patch, but pentagram still refuse to load any saves
[04:55:09] <freaks> am i alone to experience this ?
[04:55:23] <freaks> i've just compiled the cvs..
[04:55:59] <servus> You actually have a save game worth... saving? : o)
[05:04:17] <Colourless> savegame format change sorry
[05:05:15] <Colourless> should hopefully be the last time we break savegames
[05:05:35] <Colourless> and sorry, there isn't anything we can do to make them work
[05:05:46] <freaks> no and no and no, you get me wrong here ,
[05:06:54] <Colourless> hmm
[05:07:02] <Colourless> so you saved, then loaded and it failed?
[05:07:26] <freaks> i got the cvs, i compiled the cvs, i installed the cvs, i STARTED A NEW GAME, i played 5minutes, i tryed to SAVE(result: ok) , then i continued to play, then i died !! then i tryed to load my save from less than 10 minutes ago: result FAIL !
[05:07:38] <freaks> Colourless, yes
[05:07:45] <freaks> exaclty :)
[05:07:46] <Colourless> odd...
[05:07:49] * servus plays satan's advocate and says "With an XML format, it'd be hard to break save games..."
[05:07:50] <freaks> yes
[05:08:23] <freaks> Colourless, it complain about the game being version 2.12 and the save being version 0
[05:08:44] <freaks> (the md5 keys looks ok, but the game still refuse to load)
[05:08:46] <servus> Maybe you could fix that in a hex editor *grin*
[05:09:24] <freaks> the possibilities of success are low ;)
[05:09:45] <Colourless> should even need to use a hex editor
[05:09:55] <Colourless> savegames are just zip files
[05:09:59] <Colourless> *shouldn't
[05:10:07] <freaks> i dunno yet of a good one, in linux world
[05:10:24] <freaks> (hex ed, i meant)
[05:10:37] <freaks> Colourless, really? then what should i do ?
[05:11:02] <Colourless> i don't konw what the problem is
[05:11:22] <freaks> "load saved game fail because of version mismatch"
[05:11:42] <freaks> you want a screenshot ?
[05:11:46] <freaks> wait..
[05:11:52] <Colourless> no
[05:14:50] <Colourless> savegames working fine here
[05:16:09] <servus> Maybe you did not commit a file.
[05:16:19] <Colourless> wjp made the changes
[05:16:20] <servus> Or maybe freaks did not do a make clean before compiling?
[05:16:30] <Colourless> just updated cvs then and compiled
[05:16:30] <servus> Well, wjp has been known to forget files in commits : o)
[05:17:01] <Colourless> as i just said, i updated, then compiled
[05:17:04] <servus> freaks: What sort of system are you on?
[05:18:30] <servus> Endian problem, maybe?
[05:22:30] <freaks> here it is:
[05:22:44] <freaks> http://keisangi.free.fr/u8load.png
[05:22:48] <freaks> http://keisangi.free.fr/u8save.png
[05:23:41] <freaks> i'm using a p3 at 500mhz (yea i know poor me) running linux fedora core 4
[05:25:44] <freaks> when i accessed the cvs i used "pentagram" as the module name here:
[05:26:51] <freaks> cvs -z3 -d:pserver:email@example.com:/cvsroot/pentagram co -P modulename
[05:27:06] <Colourless> i think there is a bug in the saving code...
[05:27:29] <freaks> yup
[05:28:01] <Colourless> lets see...
[05:28:26] <freaks> cool
[05:29:36] <Colourless> no.. it saves correctly
[05:30:19] <freaks> you mean it work for you ?
[05:30:35] <freaks> did you compiled the last cvs ? in that short amount of time ?
[05:31:08] <Colourless> takes me a couple minutes to compile
[05:31:40] <freaks> you should have a better proc than mine then ;)
[05:31:59] <freaks> also, why "make uninstall" insn't implemented ;)
[05:32:00] <freaks> ?
[05:32:09] <Colourless> Athlon 1000 but i don't use gcc
[05:32:20] <freaks> what do you use ??
[05:32:23] * freaks curious
[05:32:49] <Colourless> MSVC.Net
[05:33:57] <Colourless> ah ha. found the problem
[05:35:09] <Colourless> line 198 of GameInfo.cpp change version to GameInfo::version (or this->version)
[05:37:31] <Colourless> fix is going to use this->version so you should update for file to that if you don't want to get a cvs conflict
[05:37:54] <freaks> a microsoft ddeep shit proprietary software from hell that will taint and chain your soul for ever in the dark side of redmond?
[05:38:08] <freaks> you should stop using micro$oft
[05:38:16] <freaks> sorry i couldn't resist
[05:39:01] <freaks> i plan to get the whole cvs and recompile it completly, it it ok ?
[05:39:05] <freaks> is
[05:39:21] <Colourless> cvs update times are usually slow
[05:39:26] <Colourless> so you may not be able to get the fix for some hours
[05:39:38] <freaks> what should i do then?
[05:40:06] <freaks> aah
[05:40:08] <freaks> lol
[05:40:21] <freaks> w8 i'm fixing my eyes better
[05:40:22] <servus> Could you not log on to the SourceForge SSH shell, do an update there, then rsync over?
[05:42:09] <Colourless> anonymous cvs has a big delay
[05:42:31] <freaks> version = std::strtol(parts.c_str(), 0, 0);
[05:42:34] <Colourless> well it used to. supposedly it was going to be shortened i think or something
[05:42:36] <freaks> here the line 198
[05:42:41] <Colourless> make that
[05:42:47] <Colourless> this->version = std::strtol(parts.c_str(), 0, 0);
[05:43:04] <freaks> of pentagram/games/gameinfo.cpp
[05:43:15] <Colourless> yes
[05:43:34] <Colourless> and just run make
[05:43:42] <freaks> so i just have to add "this->" ?
[05:43:43] <Colourless> you wont need to make clean
[05:43:47] <Colourless> yes just add this->
[05:43:55] <freaks> allright :)
[05:43:57] <freaks> cool
[05:44:02] <freaks> i test
[05:45:01] <freaks> even make clean isn't necessary ? hum nice
[05:45:54] <Colourless> nope
[05:46:08] <Colourless> only really need make clean when .h files are modded
[05:50:42] <freaks> hummm
[05:50:48] <freaks> nice info
[05:51:10] <freaks> well, it worked \(^o^)/
[05:51:26] <freaks> i can save and load again :)
[05:51:30] <freaks> THANKS
[05:53:53] <freaks> a lot even
[06:29:55] --- Kirben2 is now known as Kirben
[06:29:56] --- ChanServ gives channel operator status to Kirben
[07:15:09] --- freaks is now known as freaks^afk
[07:21:25] --> sbx has joined #pentagram
[09:03:05] <wjp> eek; sorry about that
[09:05:18] <wjp> I should really test things like that better :/
[09:06:57] <Colourless> let me guess, you tested on a u8 version with modified usecode (like me)
[09:08:12] <wjp> yup...
[09:08:24] <wjp> and that always sets version to 0
[09:09:20] <wjp> maybe I should set version to 666 or something when the autodetect failed :-)
[09:09:39] <Colourless> i approve of that
[09:10:11] <wjp> or maybe 999 :-)
[09:10:33] <wjp> before we get sued for corrupting children or something ;-)
[09:10:43] <Colourless> it's called pentagram...
[09:46:25] <wjp> one fairly 'big' thing we still need to do is the automatic adjusting of the Avatar's movement when it's blocked
[09:47:14] <wjp> and a related thing is figuring out why Arcadion has fallen through the floor in one of the bug reports
[09:49:09] <wjp> I wonder at which level the movement adjustments should be done
[09:49:13] <wjp> maybe at several
[09:49:45] <wjp> sometimes large steps are converted into smaller steps, and sometimes movement is shifted to the side a bit
[09:50:01] <wjp> the second bit _could_ go into AnimationTracker
[09:50:20] <wjp> not sure if it should be enabled for all animations of all actors, though
[09:50:48] <Colourless> i actually have very little idea how pentagrams animation system is setup
[09:52:06] <wjp> two classes take care of things, mainly: AnimationTracker and ActorAnimProcess
[09:52:26] <wjp> AnimationTracker can perform an animation, but doesn't necessarily move an Actor
[09:52:57] <wjp> (so it can also be used to test if an animation can run without being blocked, or if an attack animation would hit something)
[09:53:36] <wjp> ActorAnimProcess actually moves an Actor with the AnimationTracker telling it what to do exactly
[09:55:04] <Colourless> well for avatar stepping/running i think that something as simple as retrying a blocked animation shifted left/right by a few units (maybe repeated a few times say 2,4 and 8 units either side or maybe even bigger) to see if they can be made and if so do it
[09:55:23] <Colourless> in exult it was easy, just did test for 1 unit
[09:56:04] <wjp> stepping up/down is currently handled in AnimationTracker
[09:56:48] <Colourless> hmm it is actually the exact same sort of thing as running up stairs really
[09:56:53] <Colourless> just sideways
[09:56:55] <Colourless> i didn't think of that
[09:58:23] <wjp> the more I think about it, the more I think the minor sideway shifts should go into AnimationTracker
[09:58:40] <wjp> would also allow the Pathfinder to use them for NPCs which have to go through doors, for example
[09:58:57] * Colourless nods
[09:59:05] <Colourless> that sounds like it would save processing
[10:00:51] <wjp> I'm not sure if it should be enabled for all animations, or just for the 'walk-type' ones
[10:10:51] <Colourless> i am going go guess only ones that are on ground
[11:17:38] --- sbx is now known as SBX|AFK
[11:48:18] <wjp> yeah, I guess it wouldn't hurt for non-movement animations anyway
[11:48:53] <wjp> so when an on-ground animation frame is blocked or unsupported, try shifting sideways a small amount
[11:49:17] <wjp> then there's the replacement of an animation by a completely different one in AvatarMoverProcess
[11:49:40] <wjp> if you can't take a normal step, try in a different direction or a smaller step
[11:50:28] <Colourless> yep sounds good
[11:51:17] <wjp> I wonder if NPCs falling through the floor could be a fastarea problem
[11:51:27] <Colourless> it's possible
[11:52:18] <wjp> fastarea consists of whole glob-sized chunks, right?
[11:52:28] <wjp> (i.e., no partial ones)
[11:52:31] <Colourless> yep
[12:17:32] <Colourless> wjp i've noticed a slight problem in the Gump destructor. It calls virtual function Close() which is kind of bad since because of the was destruction is done, it will only ever call Gump::close
[12:17:58] <Colourless> of course i'm sort of stumped with how to fix the problem
[12:18:11] <Colourless> not that i've seen a problem yet
[12:19:02] <Colourless> we could probably put all the 'closing' code into the destructor if needed i'm going to guess to get rid of the gotcha that could happwn
[12:19:06] <Colourless> *happen
[12:22:55] <wjp> hm
[12:23:28] <wjp> aren't gumps always closed before they're destructed?
[12:24:23] <Colourless> i have no idea
[12:24:58] <wjp> except maybe maybe resetting engine
[12:25:19] <wjp> s/maybe maybe/maybe when/
[12:26:44] <wjp> the usual procedure should be that we call Gump::Close() when we want to get rid of it and then the parent deletes it
[12:27:14] <Colourless> yes
[12:28:46] <wjp> I'd say just get rid of the Close call
[12:28:52] <wjp> and hope everything still works :-)
[12:29:47] <wjp> it _should_ be the case that the only place where it will make a difference is the shutdown/reset procedure
[12:30:05] <wjp> and there it shouldn't matter that notifyprocesses aren't notified
[12:30:49] <Colourless> yeah i'm going to guess that it should be fine
[12:38:45] <Colourless> i guess it will mean following proper precedure with gumps like we do with items (i.e. call Gump::close() like we call Item::destroy())
[12:39:00] * wjp nods
[12:39:40] <wjp> we should already be doing that in most places, I think
[12:39:52] <wjp> (I hope :-) )
[12:52:59] <Colourless> bed time for me
[12:53:02] <Colourless> night
[12:53:13] <wjp> good night
[12:53:15] <-- Colourless has left IRC ("casts improved invisibility")
[13:18:03] <-- Kirben has left IRC ("System Meltdown")
[15:58:26] <-- freaks^afk has left IRC ("Leaving")
[17:56:22] --- SBX|AFK is now known as sbx
[17:56:25] --- sbx is now known as SB-X
[20:19:49] <-- paranoidd has left IRC (".")
[23:21:57] --> Colourless has joined #Pentagram
[23:21:57] --- ChanServ gives channel operator status to Colourless
[23:30:29] <-- watt has left IRC (Read error: 60 (Operation timed out))
[23:47:12] --> Kirben has joined #pentagram
[23:47:12] --- ChanServ gives channel operator status to Kirben
[23:57:21] <-- Kirben has left IRC (Read error: 104 (Connection reset by peer))