#exult@irc.freenode.net logs for 10 Aug 2002 (GMT)

Archive Today Yesterday Tomorrow
Exult homepage

[00:16:23] <-- Fingolfin has left IRC ("good night")
[01:22:35] --> ShadwChsr has joined #exult
[02:00:37] --- Darke|zzZ is now known as Darke
[04:32:42] --> Kirben2 has joined #exult
[04:32:42] <-- Kirben has left IRC (Read error: 104 (Connection reset by peer))
[06:25:29] <-- ShadwChsr has left IRC ()
[06:31:15] --> ShadwChsr has joined #Exult
[08:26:48] <-- ShadwChsr has left IRC (Read error: 104 (Connection reset by peer))
[08:54:00] --- ChanServ gives channel operator status to Darke
[10:24:34] --> Fingolfin has joined #exult
[10:25:07] <Fingolfin> yp
[10:25:08] --- ChanServ gives channel operator status to Fingolfin
[10:25:37] --> wjp has joined #exult
[10:25:38] --- ChanServ gives channel operator status to wjp
[10:25:40] <wjp> hi
[10:25:48] <Fingolfin> hi
[10:25:57] <Fingolfin> wow, only one minute difference between us, willem :-)
[10:26:06] <wjp> :-)
[10:30:24] <Darke> Hi all. *grin* Nice timing.
[12:36:22] <-- Kirben2 has left IRC (capek.openprojects.net irc.openprojects.net)
[12:36:22] <-- Fingolfin has left IRC (capek.openprojects.net irc.openprojects.net)
[12:36:22] <-- Darke has left IRC (capek.openprojects.net irc.openprojects.net)
[12:36:49] --> Kirben2 has joined #exult
[13:00:51] --> Darke has joined #exult
[13:01:26] --- ChanServ gives channel operator status to Darke
[13:14:29] --> Fingolfin has joined #exult
[13:15:11] * Darke bows. Hi Fingolfin.
[13:15:26] * Darke hmms... and must sleep. Night all. *grin*
[13:15:42] --- Darke is now known as Darke|zzZ
[13:21:20] <-- Kirben2 has left IRC ("System Meltdown")
[13:45:08] <-- Fingolfin has left IRC (capek.openprojects.net irc.openprojects.net)
[13:52:37] --> Fingolfin has joined #exult
[14:10:22] <-- Fingolfin has left IRC (capek.openprojects.net irc.openprojects.net)
[14:10:24] --> Fingolfin has joined #exult
[14:16:48] --> Colourless has joined #Exult
[14:16:48] --- ChanServ gives channel operator status to Colourless
[14:17:03] <Colourless> hi
[14:20:14] <wjp> hi
[14:25:54] <Fingolfin> hi ryan
[14:28:26] <-- Fingolfin has left IRC ("gotta get some work done")
[16:16:14] <wjp> hm, I think I'll go do something else for a while
[16:16:20] <wjp> I'll bbl
[16:16:26] <-- wjp has left IRC ()
[16:49:59] --> ShadwChsr has joined #Exult
[16:50:03] --> Fingolfin has joined #exult
[17:09:46] <Colourless> hi ShadwChsr, wb Fingolfin
[17:19:37] <Fingolfin> yo
[17:32:59] <ShadwChsr> hi
[17:50:03] <-- Fingolfin has left IRC ("gotta get some work done")
[18:38:28] <ShadwChsr> Anyone here? :)
[19:08:02] --> SB-X has joined #exult
[19:08:06] <SB-X> hello
[19:08:32] <SB-X> ShadwChsr
[19:08:36] <SB-X> have any new screenshots? :)
[19:09:00] <ShadwChsr> Hey! :)
[19:09:47] <ShadwChsr> I'm reworking alot of the way maps are structured
[19:10:03] <SB-X> btw, i never made heads or tails out of the source, because i havnt had a change to look at the pascal tutorial yet
[19:10:22] <SB-X> but i still have it
[19:10:30] <SB-X> are the maps going to be like U8 now?
[19:11:19] <ShadwChsr> Yeah, diamond in shape instead of square
[19:11:33] <ShadwChsr> that allows my X&Y axis to stay constant throughout the entire game
[19:11:55] <ShadwChsr> Only thing I'm stuck on right now is pixel->tile coordinate conversion... its 80% done, just can't get the X coordinates right :P
[19:12:23] <ShadwChsr> ... unless you have ideas? :)
[19:12:41] <SB-X> picture is interesting, thought you were already doing it diamond shape?
[19:13:15] <ShadwChsr> Nope, maps were square.. axis was 'normal' ;) Of course, everything went to hell when I tried to do collision detection
[19:13:37] <ShadwChsr> so with colourless' knowledge & info I recoded it ;)
[19:13:53] <SB-X> but in the old screens, it looks.. rotated
[19:13:54] <ShadwChsr> *almost* done, just gotta get that pixel->tile conversion working... its being..crabby :P
[19:14:05] <SB-X> same isometric style, so i assumed the tiles were diamond on a diamond grid
[19:14:18] <ShadwChsr> See the blue diamond representing the entire (small) map?
[19:14:37] <ShadwChsr> in the old system that was more-or-less square instead of a diamond - with 'jagged' edges
[19:14:42] <SB-X> i see
[19:15:12] <ShadwChsr> Y+1 would move down, in the new system Y+1 moves down and left :)
[19:15:19] <SB-X> what made you think of doing that?
[19:15:33] <SB-X> sounds like it would be difficult from the start
[19:15:50] <ShadwChsr> A) its easier to 'splice' ajoining maps together
[19:16:17] <ShadwChsr> B) collision detection is already defined with that 'rotated' axis. If it wasn't rotated, i'd need to define verticies, much more difficult (and slower)
[19:16:31] <ShadwChsr> C) Makes the code alot simpler, with the exception of pixel->tile coordinate conversion :P
[19:16:44] <ShadwChsr> It actually speeds up the game too :)
[19:16:45] <SB-X> hmm
[19:16:51] <SB-X> by "down" you mean down on the screen right?
[19:16:55] <ShadwChsr> yeah
[19:17:08] <SB-X> down in the world would be... down still :)
[19:17:42] <ShadwChsr> Problem is, when a user 'clicks' on the screen, thats in pixel coordinates. I need to rotate those coordinates 45 degrees to get them on the correct axis
[19:17:52] <ShadwChsr> then from there do some division to determine the tile that the mouse is over
[19:18:19] <SB-X> thats what you mean by pixel->tile conversion? im not very familiar with these concepts
[19:18:20] <ShadwChsr> when I move the mouse down the screen, the selected tile is under the mouse
[19:18:38] <ShadwChsr> if i move the mouse left or right, the 'error' increases the farther I get from the center of the map :P
[19:18:53] <ShadwChsr> Well, for example, I know what pixel the mouse is over
[19:19:05] <ShadwChsr> I need to take those pixel (XY) coordinates and figure out what tile the mouse is over
[19:19:06] <ShadwChsr> :)
[19:19:43] <SB-X> maybe pentagram mapdisp already does it?
[19:20:12] <ShadwChsr> probably, although it would do it differently since u8 used a different concept of 'tiles'
[19:20:24] <ShadwChsr> The tile grid was very small and tiles overlaped grid cells
[19:21:35] <SB-X> Colourless would know this instantly :)
[19:21:38] <SB-X> i have to think about it
[19:21:53] <ShadwChsr> Yeah I know ;)
[19:22:47] <SB-X> it took a day for me to figure out how to do it with a non-isometric engine
[19:23:03] <SB-X> but it turned out to be so simple that i think it shouldnt have taken me that long
[19:26:21] <SB-X> do you have the offset of the top-left most tile on the screen?
[19:27:08] <ShadwChsr> there is no top left most tile
[19:27:10] <ShadwChsr> its a diamond
[19:27:10] <ShadwChsr> :)
[19:27:17] <SB-X> i mean top
[19:27:21] <ShadwChsr> yep
[19:27:43] <ShadwChsr> my code looks VERY close now - just a little bit of 'creeping' error, but not much ;)
[19:28:10] <-- Colourless has left IRC ("i am so tired")
[19:28:18] <SB-X> :|
[19:29:51] <SB-X> whats the error you still have? im still working on the basic function
[19:30:07] <SB-X> creeping = the error that gets bigger the further away the cursor is ?
[19:30:13] <SB-X> that is due to dividing
[19:30:17] <ShadwChsr> woooo I may have got it...
[19:30:22] <ShadwChsr> yeah and it is due to dividing
[19:30:27] <ShadwChsr> I was 0.8 off ;)
[19:30:36] <ShadwChsr> One pixel made all the difference
[19:30:40] <SB-X> cool
[19:31:05] <SB-X> do you have a tile rotator?
[19:31:56] <ShadwChsr> Sigh.. still doin it
[19:31:57] <ShadwChsr> rotator?
[19:32:14] <SB-X> i wonder if you have worked on rotating images before
[19:32:24] <SB-X> because i am curious about how fast it can be done
[19:32:32] <ShadwChsr> ahh no
[19:33:00] <SB-X> i will have to make a test program sometime
[19:33:07] <SB-X> what is the error you have now?
[19:39:09] <ShadwChsr> Well its
[19:39:14] <ShadwChsr> *very* close, just not perfect
[19:39:27] <ShadwChsr> its like 4 pixels off when you get to tile 20
[19:39:40] <SB-X> do you have a demo?
[19:39:45] <SB-X> i have a delphi compiler now
[19:39:49] <SB-X> but im not sure how to use it
[19:40:08] <SB-X> kylix
[19:40:10] <ShadwChsr> :)
[19:41:13] <SB-X> if you send me src code and i get it to compile, you can tell me where calculation is and i can try to experiment
[19:42:37] <ShadwChsr> just a sec :)
[19:46:18] <SB-X> whats Forms.dcu
[19:46:19] <SB-X> ?
[19:47:50] <ShadwChsr> SWEET
[19:47:51] <ShadwChsr> it works!
[19:47:57] <SB-X> Game.dpr:uses UGRFFiles_Custom in 'UGRFFiles_Custom.pas';
[19:48:00] <ShadwChsr> .dcu is a delphi compiled unit
[19:48:01] <SB-X> what did you have to change?
[19:48:10] <ShadwChsr> a .pas files converts into a .dcu before they are linked into the .exe
[19:48:19] <ShadwChsr> err not convert i mean compile ;)
[19:48:36] <SB-X> i was just trying to compile the old source i had, i dont think i tried before
[19:48:52] <ShadwChsr> It worked cause I converted all my hard-coded constants into proper functions.. I must have made a typo somewhere :)
[19:49:18] <ShadwChsr> Sky's the limit now.... Now I can do some serious coding :)
[19:49:31] <SB-X> i love and hate that feeling
[19:49:34] <SB-X> but anyway
[19:49:39] <SB-X> can you send the source now? :)
[19:49:57] <ShadwChsr> Well, go to www.sourceforge.net/projects/lupercalia
[19:50:01] <ShadwChsr> download the latest source posted there
[19:50:02] <SB-X> ok
[19:50:25] <ShadwChsr> My current source is being reworked with the coordinate systems so it wont work properly.. yet
[19:51:18] <SB-X> where is C++ code?
[19:52:39] <SB-X> -downloading Lupercalia-Client-Src-
[19:53:01] <ShadwChsr> There is none
[19:54:10] <SB-X> sf says Languages: C++, Delphi/Kylix
[19:54:40] <ShadwChsr> Server will be in C++
[19:55:02] <ShadwChsr> Right now I'm writing the client graphics engine. Most other projects seem to only focus on the server so I decided to do the opposite
[19:56:35] <SB-X> i am currently working on the protocol for my project
[19:57:17] <SB-X> colourless gave me an idea the other day that caused me to slighty change the message format today
[19:58:23] <SB-X> and then I kept working on the message front-end, and reworking it, until i settled on "engine_message((void*)command, &arg1, &arg2, ...);"
[19:58:46] <SB-X> now ive got to try and get the server to auth the client
[19:59:35] <ShadwChsr> btw where are you from?
[19:59:43] <SB-X> so im still working on it like you
[20:00:13] <SB-X> OK, USA
[20:00:23] <ShadwChsr> BC, Canada here
[20:00:34] <ShadwChsr> so our time zomes are simmilar.. not on other sides of the world
[20:00:36] --> wjp has joined #exult
[20:00:36] --- ChanServ gives channel operator status to wjp
[20:00:43] <ShadwChsr> are you on MSMessenger or ICQ?
[20:00:55] <SB-X> i think this channel is usually dominated by australians
[20:00:57] <wjp> hi
[20:00:59] <SB-X> and wjp
[20:01:00] <ShadwChsr> hi
[20:01:02] <SB-X> hi wjp
[20:01:43] <ShadwChsr> I gotta go but I want to continue this dicussion later on icq/msm/etc
[20:01:51] <SB-X> UIN: 9788597 but i dont usually use it
[20:01:52] <SB-X> ok
[20:02:01] <ShadwChsr> or here
[20:02:11] <SB-X> ill have to ask you for help if i cant get it to compile
[20:02:14] <ShadwChsr> I'll be on in the evening most likely.... :)
[20:02:24] <ShadwChsr> Yeah, I've never tried to compile under kylix before
[20:02:41] <ShadwChsr> your compiling in linux, right?
[20:02:50] <SB-X> yes
[20:03:00] <ShadwChsr> I'm 100% sure it wont compile.. it uses directx ;)
[20:03:10] <ShadwChsr> So you're gonna need to write an SDL wrapper or something first
[20:03:18] <SB-X> hmm :(
[20:03:22] <SB-X> can that be written in C?
[20:03:26] <ShadwChsr> But I would *love* to have it running under linux
[20:03:27] <ShadwChsr> no
[20:03:37] <ShadwChsr> I need the pascal wrapper so I can pass it commands :)
[20:03:59] <ShadwChsr> we'll talk about it later though... see ya
[20:04:01] <-- ShadwChsr has left IRC ()
[21:13:33] <-- SB-X has left IRC ("gotta go")
[22:41:47] <-- Darke|zzZ has left IRC (capek.openprojects.net irc.openprojects.net)
[22:41:50] --> Darke|zzZ has joined #exult
[23:19:54] <-- wjp has left IRC ("Zzzz....")