Menu
▾
▴
isugamedev-midworld — Midworld Development Discussion
You can subscribe to this list here.
| 2002 |
Jan
|
Feb
|
Mar
|
Apr
|
May
(1) |
Jun
(95) |
Jul
(20) |
Aug
(4) |
Sep
(23) |
Oct
(39) |
Nov
(29) |
Dec
(4) |
|---|
|
From: Brown J. L <Bro...@Jo...> - 2002-06-07 18:26:05
|
nevermind my questions from before I think I was smoking crack. However I still need to get in contact with Chad about terrain generation (not as it applies to AI since all I have to do is use our handy dandy collision detection system) but just to get it started. And also I would like to know who is taking on developing the interface for an enemy unit. If there isn't anyone doing that I'll take it on, I'll be in control of everything evil ;) Will anyone be around Ames on Sunday evening? If so, maybe we can get together for awhile. cheers Josh |
|
From: Brown J. L <Bro...@Jo...> - 2002-06-06 20:56:47
|
I started working on a design for the AI engine and find myself so confussed
I can't even right a question to ask what I'm wondering, but I'm going to
ask some generic questions about how the game works and maybe that will be
enough :\
What type of AI do we want: Do we want an AI engine that iterates over a
vector of enemies and calculates their next action, and then calls a method
in the enemy to do that? If this is the case the game engine would be
required to call the AI engine and the AI engine would update each enemies?
OOORRRR
Do we want an AI engine that gets queried by enemies and gives them the best
action to perform. And in this case the game engine would iterate over each
enemy and the enemy would query the AI for what to do.
OOOORRRRRR
There is one final reasonable simple way to do AI. You can have each enemy
periodically register his current state with the AI engine and the AI engine
just gives him the best order by calling one of his methods - walk, shoot,
etc. This makes it really easy to fine tune a game because in order to make
an enemy appear smarter you just call the register function more frequently.
and the register function just looks like:
register(){
if I can I see my enimy
shoot him
if I can't
find the shortest path to his current position
go there.
}
My suggestion would be the first senario. It allows for a much less coupled
system and that way our enemy object is a lot more simple. So assuming we
want the first option then, where will the AI query for info from the game.
Will the AI engine be able to query the world and get the positions of each
non-static object along with all static objects that are part of the map, or
will the AI engine have to query each non-static object seperately to find
it's current position.
Finally I need to talk to chad about terrain generation, and how that is
going to impact pathfinding. Well I hope this finds you all well and that
you are as pumped for turning this thing out in 2 1/2 months as I am.
cheers
Josh Brown
|
|
From: Ben S. <bs...@vr...> - 2002-06-06 00:06:29
|
Hey! I'm here. I'm even 5 minutes late. Where is everybody??!! cheers, ----- Ben Scott Research Assistant VRAC bs...@ia... On Tue, 4 Jun 2002, Johnathan G. wrote: > Guys, > > How about a meeting tomorrow (Wednesday) night at 7pm at the VRAC? We were > thinking about having it in the second floor conference room (where Dr. > Cruz-Neira's meetings are). Let me know if it works for you. > > Johnathan > > _______________________________________________________________ > > Don't miss the 2002 Sprint PCS Application Developer's Conference > August 25-28 in Las Vegas -- http://devcon.sprintpcs.com/adp/index.cfm > > _______________________________________________ > Isugamedev-midworld mailing list > Isu...@li... > https://lists.sourceforge.net/lists/listinfo/isugamedev-midworld > |
|
From: Johnathan G. <da...@us...> - 2002-06-04 22:30:03
|
Guys, How about a meeting tomorrow (Wednesday) night at 7pm at the VRAC? We were thinking about having it in the second floor conference room (where Dr. Cruz-Neira's meetings are). Let me know if it works for you. Johnathan |
|
From: Ben S. <bs...@vr...> - 2002-06-04 04:37:27
|
I threw together a simple Midworld web site so we can manage our content a little and so we have something to point people to. It all works directly out of CVS through some fun viewcvs magic. I am not attached to the design, so if anybody who actually has graphical design sense wants to tear it apart ... please do! http://isugamedev.sf.net/midworld cheers, ----- Ben Scott Research Assistant VRAC bs...@ia... |
|
From: Chad A. <ae...@ae...> - 2002-06-04 04:35:22
|
Sure. Evenings are good. Ben Scott wrote: > > Are we going to meet again this week? > > cheers, > ----- > Ben Scott > Research Assistant VRAC > bs...@ia... > > _______________________________________________________________ > > Don't miss the 2002 Sprint PCS Application Developer's Conference > August 25-28 in Las Vegas -- http://devcon.sprintpcs.com/adp/index.cfm > > _______________________________________________ > Isugamedev-midworld mailing list > Isu...@li... > https://lists.sourceforge.net/lists/listinfo/isugamedev-midworld |
|
From: Johnathan G. <da...@us...> - 2002-06-04 04:21:15
|
I'm up for it; we need to meet before Friday though; I have to leave like 4am Friday morning Johnathan On Monday 03 June 2002 21:59, Ben Scott wrote: > Are we going to meet again this week? > > cheers, > ----- > Ben Scott > Research Assistant VRAC > bs...@ia... > > > _______________________________________________________________ > > Don't miss the 2002 Sprint PCS Application Developer's Conference > August 25-28 in Las Vegas -- http://devcon.sprintpcs.com/adp/index.cfm > > _______________________________________________ > Isugamedev-midworld mailing list > Isu...@li... > https://lists.sourceforge.net/lists/listinfo/isugamedev-midworld |
|
From: Ben S. <bs...@vr...> - 2002-06-04 02:59:18
|
Are we going to meet again this week? cheers, ----- Ben Scott Research Assistant VRAC bs...@ia... |
|
From: Ben S. <bs...@vr...> - 2002-06-04 02:57:14
|
The idea behind the design was to make it flexible enough to support an easy iterative development schedule. So brute force could be easily written initially and better methods could be added from there. I'll see if I can improve the document to reflect this. ----- Ben Scott Research Assistant VRAC bs...@ia... On Mon, 3 Jun 2002, Kevin Meinert wrote: > > looks like a smart final design. what about your prototype design? the > one that doesn't take a long time to write? This looks like a final > design, and for the prototype you probably don't want to spend all your > time implementing collision. > > I would just do the brute force method first, super easy, and you don't > have to worry about complex logic throwing you off the "get the prototype > done" path. > > *--*---*---*----*-----*------*------*-----*----*---*---*--* > Kevin Meinert /_/ > http://www.vrac.iastate.edu/~kevn \ / > Virtual Reality Applications Center \/ __ \/ > Howe Hall, Iowa State University, Ames Iowa \__ > \_\ > > On Sun, 2 Jun 2002, Ben Scott wrote: > > > Yeah, I'm still working on it. ;) Basically what I was thinking would be > > good about the design is that it would allow for an easy iteration > > process. So at first we can just use the scenegraph for the first pass and > > implement the second pass by always reporting a collision. Thus we would > > have implemented a basic collision detection system that reports > > collisions when bounding boxes intersect. In the next iteration we could > > use a more advanced intersection algorithm to implement a more exact > > second pass. From there we could work towards detecting collisions in 4D > > by doing sweep tests rather than just testing the final position. > > > > Obviously all that needs to be added to the doc. ^_^ I guess I was just > > looking for initial feedback. > > > > Sorry about the URL ... here's the correct one: > > > > http://cvs.sf.net/cgi-bin/viewcvs.cgi/*checkout*/isugamedev/midworld/doc/CollDetDesign.html > > > > cheers, > > ----- > > Ben Scott > > Research Assistant VRAC > > bs...@ia... > > > > On Sun, 2 Jun 2002, Chad Austin wrote: > > > > > Doesn't it seem a little complex? For now, that is. That URL doesn't seem to > > > work, either... > > > > > > I guess it just looks like a little overkill to me. > > > > > > > I've posted some initial plans for a collision detection subsystem into > > > > the repository (doc/CollDetDesign.html). If you have a chance, please send > > > > me your comments. > > > > > > > > http://cvs.sf.net/cgi-bin/viewcvs.cgi/*checkout*/midworld/doc/CollDetDesign.html > > > > > > > > On another note, Levi expressed interest to me this weekend on > > > > participating in work on the collision detection system. Since he's a shy > > > > emailer, I thought I'd give you the heads up on his interest. > > > > > > _______________________________________________________________ > > > > > > Don't miss the 2002 Sprint PCS Application Developer's Conference > > > August 25-28 in Las Vegas -- http://devcon.sprintpcs.com/adp/index.cfm > > > > > > _______________________________________________ > > > Isugamedev-midworld mailing list > > > Isu...@li... > > > https://lists.sourceforge.net/lists/listinfo/isugamedev-midworld > > > > > > > > > _______________________________________________________________ > > > > Don't miss the 2002 Sprint PCS Application Developer's Conference > > August 25-28 in Las Vegas -- http://devcon.sprintpcs.com/adp/index.cfm > > > > _______________________________________________ > > Isugamedev-midworld mailing list > > Isu...@li... > > https://lists.sourceforge.net/lists/listinfo/isugamedev-midworld > > > |
|
From: Kevin M. <ke...@vr...> - 2002-06-03 13:54:52
|
looks like a smart final design. what about your prototype design? the one that doesn't take a long time to write? This looks like a final design, and for the prototype you probably don't want to spend all your time implementing collision. I would just do the brute force method first, super easy, and you don't have to worry about complex logic throwing you off the "get the prototype done" path. *--*---*---*----*-----*------*------*-----*----*---*---*--* Kevin Meinert /_/ http://www.vrac.iastate.edu/~kevn \ / Virtual Reality Applications Center \/ __ \/ Howe Hall, Iowa State University, Ames Iowa \__ \_\ On Sun, 2 Jun 2002, Ben Scott wrote: > Yeah, I'm still working on it. ;) Basically what I was thinking would be > good about the design is that it would allow for an easy iteration > process. So at first we can just use the scenegraph for the first pass and > implement the second pass by always reporting a collision. Thus we would > have implemented a basic collision detection system that reports > collisions when bounding boxes intersect. In the next iteration we could > use a more advanced intersection algorithm to implement a more exact > second pass. From there we could work towards detecting collisions in 4D > by doing sweep tests rather than just testing the final position. > > Obviously all that needs to be added to the doc. ^_^ I guess I was just > looking for initial feedback. > > Sorry about the URL ... here's the correct one: > > http://cvs.sf.net/cgi-bin/viewcvs.cgi/*checkout*/isugamedev/midworld/doc/CollDetDesign.html > > cheers, > ----- > Ben Scott > Research Assistant VRAC > bs...@ia... > > On Sun, 2 Jun 2002, Chad Austin wrote: > > > Doesn't it seem a little complex? For now, that is. That URL doesn't seem to > > work, either... > > > > I guess it just looks like a little overkill to me. > > > > > I've posted some initial plans for a collision detection subsystem into > > > the repository (doc/CollDetDesign.html). If you have a chance, please send > > > me your comments. > > > > > > http://cvs.sf.net/cgi-bin/viewcvs.cgi/*checkout*/midworld/doc/CollDetDesign.html > > > > > > On another note, Levi expressed interest to me this weekend on > > > participating in work on the collision detection system. Since he's a shy > > > emailer, I thought I'd give you the heads up on his interest. > > > > _______________________________________________________________ > > > > Don't miss the 2002 Sprint PCS Application Developer's Conference > > August 25-28 in Las Vegas -- http://devcon.sprintpcs.com/adp/index.cfm > > > > _______________________________________________ > > Isugamedev-midworld mailing list > > Isu...@li... > > https://lists.sourceforge.net/lists/listinfo/isugamedev-midworld > > > > > _______________________________________________________________ > > Don't miss the 2002 Sprint PCS Application Developer's Conference > August 25-28 in Las Vegas -- http://devcon.sprintpcs.com/adp/index.cfm > > _______________________________________________ > Isugamedev-midworld mailing list > Isu...@li... > https://lists.sourceforge.net/lists/listinfo/isugamedev-midworld > |
|
From: Ben S. <bs...@vr...> - 2002-06-02 21:46:53
|
Yeah, I'm still working on it. ;) Basically what I was thinking would be good about the design is that it would allow for an easy iteration process. So at first we can just use the scenegraph for the first pass and implement the second pass by always reporting a collision. Thus we would have implemented a basic collision detection system that reports collisions when bounding boxes intersect. In the next iteration we could use a more advanced intersection algorithm to implement a more exact second pass. From there we could work towards detecting collisions in 4D by doing sweep tests rather than just testing the final position. Obviously all that needs to be added to the doc. ^_^ I guess I was just looking for initial feedback. Sorry about the URL ... here's the correct one: http://cvs.sf.net/cgi-bin/viewcvs.cgi/*checkout*/isugamedev/midworld/doc/CollDetDesign.html cheers, ----- Ben Scott Research Assistant VRAC bs...@ia... On Sun, 2 Jun 2002, Chad Austin wrote: > Doesn't it seem a little complex? For now, that is. That URL doesn't seem to > work, either... > > I guess it just looks like a little overkill to me. > > > I've posted some initial plans for a collision detection subsystem into > > the repository (doc/CollDetDesign.html). If you have a chance, please send > > me your comments. > > > > http://cvs.sf.net/cgi-bin/viewcvs.cgi/*checkout*/midworld/doc/CollDetDesign.html > > > > On another note, Levi expressed interest to me this weekend on > > participating in work on the collision detection system. Since he's a shy > > emailer, I thought I'd give you the heads up on his interest. > > _______________________________________________________________ > > Don't miss the 2002 Sprint PCS Application Developer's Conference > August 25-28 in Las Vegas -- http://devcon.sprintpcs.com/adp/index.cfm > > _______________________________________________ > Isugamedev-midworld mailing list > Isu...@li... > https://lists.sourceforge.net/lists/listinfo/isugamedev-midworld > |
|
From: Chad A. <ae...@ae...> - 2002-06-02 19:50:45
|
Doesn't it seem a little complex? For now, that is. That URL doesn't seem to work, either... I guess it just looks like a little overkill to me. > I've posted some initial plans for a collision detection subsystem into > the repository (doc/CollDetDesign.html). If you have a chance, please send > me your comments. > > http://cvs.sf.net/cgi-bin/viewcvs.cgi/*checkout*/midworld/doc/CollDetDesign.html > > On another note, Levi expressed interest to me this weekend on > participating in work on the collision detection system. Since he's a shy > emailer, I thought I'd give you the heads up on his interest. |
|
From: Ben S. <bs...@vr...> - 2002-06-02 19:27:29
|
I've posted some initial plans for a collision detection subsystem into the repository (doc/CollDetDesign.html). If you have a chance, please send me your comments. http://cvs.sf.net/cgi-bin/viewcvs.cgi/*checkout*/midworld/doc/CollDetDesign.html On another note, Levi expressed interest to me this weekend on participating in work on the collision detection system. Since he's a shy emailer, I thought I'd give you the heads up on his interest. cheers, ----- Ben Scott Research Assistant VRAC bs...@ia... |
|
From: Chad A. <ae...@ae...> - 2002-05-29 23:00:10
|
First message to the list! If you didn't get a copy of this message through isu...@li... but you want to be on the list, let me know. "Johnathan G." wrote: > > Guys, > > Have we decided upon a final name? Last time I checked, it seemed like > Midworld was the most popular, but please reply with a final decision. We > need to get this thing underway with the repos and sf project. > > Also, would you all prefer to meet again before a design doc draft is > written, or after one is written and then make modifications to the draft? > > Johnathan |
1 message has been excluded from this view by a project administrator.
