Engine Question regarding MMORPG
Author
Hi
I''m part of a team and we are planning a MMORPG. I myself am more of the design elements (dungeons, character classes etc...). The team is small currently, tho we do have some really good programmers including a Networking programming specialist, and there are a few more people we are waiting to hear back from.
One of the team members is starting to work on the graphics engine. So far we have a room with boxes in it and some collision detection and camera movement. One of the ideas brough up was using a pre-existing "free" engine. Now i''ve looked around before at some, but from my understanding, these seem to be of FPS type of games, using BSP. From what I know, BSP is not the way to go for landscape ( like the outdoor world itself ), and I read another post which seems to back this up.
I''m just wonidering if there are any other people who can back this up.
P.S. Please don''t bother with the start with pong suggestion, we do have good programmers, the problem is none current ones we have are 3d engine programmers lol... the contact we have who could help with that is currently on duty(army) and we currently dont have much chance to speak.
Thanks
BSP based engines use a pre computed portal system. This will not work for terrain engines because a portal system relys on what can a person not see from a given point. (or rather, they can''t see any of the map except this room and that room. and the rest is culled.)
You would be looking for a quadtree and frustrum culling. Then when you get indoors you load your BSP engine. Or you could drop the BSP engine and use occluders and a quad/oct-tree but your indoor environments will be smaller.
On a side note, I do not wish to discourage you, but if you have never programmed a game before that is a high mark. Even for good programmers. There is a big difference between games and apps. Also, I am assuming that it will be an RPG of some sort here, but RPGs take _alot_ of assests. You will need a large team of modelers and artists to keep up with content.
Pong is too simple but I would suggest, for a first project, you think of something a bit smaller then a MMORPG. I would suggest a first or third person shooter type game. Or maybe a puzzle game. The main reason is you can make your world small and keep your assests down (only a few models, etc.). It also gives a nice manageable environment for testing new procedures and implementations quickly.
I hope that helped some.
Meta_
You would be looking for a quadtree and frustrum culling. Then when you get indoors you load your BSP engine. Or you could drop the BSP engine and use occluders and a quad/oct-tree but your indoor environments will be smaller.
On a side note, I do not wish to discourage you, but if you have never programmed a game before that is a high mark. Even for good programmers. There is a big difference between games and apps. Also, I am assuming that it will be an RPG of some sort here, but RPGs take _alot_ of assests. You will need a large team of modelers and artists to keep up with content.
Pong is too simple but I would suggest, for a first project, you think of something a bit smaller then a MMORPG. I would suggest a first or third person shooter type game. Or maybe a puzzle game. The main reason is you can make your world small and keep your assests down (only a few models, etc.). It also gives a nice manageable environment for testing new procedures and implementations quickly.
I hope that helped some.
Meta_
Author
Excellent information, thank you 
Well for the time being it''s mainly myself and 3 other people. At the current time, with the team we have the first goal is to make the graphics engine (with terrain (outdoors), maybe indoor enviroments, better collision detection, some physics, skybox and basic lighting), being able to have simple models (or avatars) for now, basic client server communication with the ability to move the avatars, and perhaps get a basic chat system inplemented.
See the problem with trying to aim for something less difficuly like a FPS or seomthing, is that the vision for the MMORPG is somethign to aim to achive, doing something that we don''t entirely find interesting, will result in loss of interest then the project will die hehe.
Tho its possible that if we hit too big of snags, we might try aiming for a lesser goal, but the people we do have now are fairly profecient programmers, tho we do have several people with varied levels of talent and area of expertise who have or are going to be approached, so that''s not really what we are worring too much about for now.
Thanks again.
Well for the time being it''s mainly myself and 3 other people. At the current time, with the team we have the first goal is to make the graphics engine (with terrain (outdoors), maybe indoor enviroments, better collision detection, some physics, skybox and basic lighting), being able to have simple models (or avatars) for now, basic client server communication with the ability to move the avatars, and perhaps get a basic chat system inplemented.
See the problem with trying to aim for something less difficuly like a FPS or seomthing, is that the vision for the MMORPG is somethign to aim to achive, doing something that we don''t entirely find interesting, will result in loss of interest then the project will die hehe.
Tho its possible that if we hit too big of snags, we might try aiming for a lesser goal, but the people we do have now are fairly profecient programmers, tho we do have several people with varied levels of talent and area of expertise who have or are going to be approached, so that''s not really what we are worring too much about for now.
Thanks again.
Author
Err double posted, ignore this reply hehe.
[edited by - CorpseX on May 29, 2003 3:43:31 AM]
[edited by - CorpseX on May 29, 2003 3:43:31 AM]
Honestly that was what I started with about a yr ago. For me and my team the snags were more of a team downer then the fact that we have since stopped the project in favor of a less complex one for our first project.
It''s not the we are abandoining that game, we still have the Design Doc and all that and we may go back to it some day when we feel more prepared.
However I am not saying that you will run into the same issues. If your team can work through the tough parts without getting down troden.....
What province you from CorpseX?
Meta_
It''s not the we are abandoining that game, we still have the Design Doc and all that and we may go back to it some day when we feel more prepared.
However I am not saying that you will run into the same issues. If your team can work through the tough parts without getting down troden.....
What province you from CorpseX?
Meta_
I think the main concern is more on the art side. A MMORPG requires some very nice models, prefabs, animations. You need some real talent. The programming, I would think, would not be much more dificult than a FPS (which ain''t easy by all means).
As far as engines go, there is not a definitive free engine for massive outdoor games, although there are a few for indoor sceneries. Have a look at Cube engine (http://wouter.fov120.com/cube/), based on OpenGL. There are loads more renderers, more or less professionnal.
Also have a look at cheo.resnet.wayne.edu/miguel/. He works for a video game company, and works on a ROAM engine for terrains. I helped him implement terrain collisions. He is working on terrain engines, and he is competent, and really helpful. Anyway, for terrain collisions, there are loads of optimisations, simplifications to reduce the physics overhead. I can help you out a little. Well, there are loads of terrain examples all over the web. Should be easy to find something that would suit your requirements.
for Ragdoll physics, you might be interested in http://www.gamasutra.com/resource_guide/20030121/jacobson_01.shtml
Simple system, not a great deal of maths, fast and can do inverse-kinematics (impacts on body parts and stuff). It explains the ragdoll physics for Hitman 2.
For BSP collision, have a look at http://www.melax.com/bsp/, or ftp://ftp.fluidstudios.com/pub/FluidStudios/CollisionDetection/Fluid_Studios_Generic_Collision_Detection_for_Games_Using_Ellipsoids.pdf. This is a good algorithm.
For the physics, I''d recommend you have a look at ODE (http://opende.sourceforge.net/), although it might be an overkill for a MMORPG.
You''d want to look at 3DS, md3 importers for your models and animations too.
loads of links Hope that helps.
As far as engines go, there is not a definitive free engine for massive outdoor games, although there are a few for indoor sceneries. Have a look at Cube engine (http://wouter.fov120.com/cube/), based on OpenGL. There are loads more renderers, more or less professionnal.
Also have a look at cheo.resnet.wayne.edu/miguel/. He works for a video game company, and works on a ROAM engine for terrains. I helped him implement terrain collisions. He is working on terrain engines, and he is competent, and really helpful. Anyway, for terrain collisions, there are loads of optimisations, simplifications to reduce the physics overhead. I can help you out a little. Well, there are loads of terrain examples all over the web. Should be easy to find something that would suit your requirements.
for Ragdoll physics, you might be interested in http://www.gamasutra.com/resource_guide/20030121/jacobson_01.shtml
Simple system, not a great deal of maths, fast and can do inverse-kinematics (impacts on body parts and stuff). It explains the ragdoll physics for Hitman 2.
For BSP collision, have a look at http://www.melax.com/bsp/, or ftp://ftp.fluidstudios.com/pub/FluidStudios/CollisionDetection/Fluid_Studios_Generic_Collision_Detection_for_Games_Using_Ellipsoids.pdf. This is a good algorithm.
For the physics, I''d recommend you have a look at ODE (http://opende.sourceforge.net/), although it might be an overkill for a MMORPG.
You''d want to look at 3DS, md3 importers for your models and animations too.
loads of links
Hope that helps.
Author
Im in Montreal Quebec, as is one of the other people, the one doin the engine part. The other 2 are in the states i think.
Oliii: Thanks for the links ill check those links out...
the one cheo.resnet.wayne.edu/miguel doesn''t seem to be up at the moment.
Oliii: Thanks for the links ill check those links out...
the one cheo.resnet.wayne.edu/miguel doesn''t seem to be up at the moment.
This topic is closed to new replies.
Advertisement
Popular Topics
Advertisement