Hello all, I'd like a little advice. I represent a team which is gearing up for an independent game project, and we have been scouring the net looking for the right open-source engine to use.
I don't want to give to many details, but it is a third person role-playing game. There are no advanced physics necessary as it is a mostly simplistic interface. We're just trying to determine which engine will be most effective.
Thank you in advance for your input,
Rick
General information request
-
- Posts: 3
- Joined: Thu Feb 21, 2008 8:14 pm
- Location: Titan
- Contact:
Wow, so much information, how could we possibly be mislead!
Whether it's a game engine or a graphic engine you're looking for, multi-platform or exclusive to one, the budget, special effects needed, shaders and other, time frame for the project, number of persons, experience they have and what's not, we surely have enough info to help you make the right choice.
Whether it's a game engine or a graphic engine you're looking for, multi-platform or exclusive to one, the budget, special effects needed, shaders and other, time frame for the project, number of persons, experience they have and what's not, we surely have enough info to help you make the right choice.
-
- Posts: 3
- Joined: Thu Feb 21, 2008 8:14 pm
- Location: Titan
- Contact:
Sorry, I practice being vague in the mirror.
To be perfectly honest, we don't have a lot of information at this time. We're still in the early part of the pre-production stage. We have a design document and a combat and magic system established, but the storyline and the finer points are still being decided.
All right, here's one specific issue: the world area is going to be equivalant to around 5 miles. I would like to not have to arbitrarily divide that up into sections that load as the player crosses different lines on the map. Would Irrlicht be capable of accomplishing this feat? (Please take note of our team mantra: "If at first you don't succeed, redefine success.")
Thank you.
To be perfectly honest, we don't have a lot of information at this time. We're still in the early part of the pre-production stage. We have a design document and a combat and magic system established, but the storyline and the finer points are still being decided.
All right, here's one specific issue: the world area is going to be equivalant to around 5 miles. I would like to not have to arbitrarily divide that up into sections that load as the player crosses different lines on the map. Would Irrlicht be capable of accomplishing this feat? (Please take note of our team mantra: "If at first you don't succeed, redefine success.")
Thank you.
-
- Admin
- Posts: 3590
- Joined: Mon Oct 09, 2006 9:36 am
- Location: Scotland - gonnae no slag aff mah Engleesh
- Contact:
There's a great post-mortem on a commercial RPG using a modified Irrlicht that touches on efficiency and suitability for rendering large areas.
That might help to answer your questions.
The general answer is that your architecture and scene layouts are going to be vital. Whatever engine you go with, you'll want to invest some time in culling and LOD strategies. Irrlicht currently lacks LOD nodes (although it's planned), so LOD switching would have to be done by your app.
That might help to answer your questions.
The general answer is that your architecture and scene layouts are going to be vital. Whatever engine you go with, you'll want to invest some time in culling and LOD strategies. Irrlicht currently lacks LOD nodes (although it's planned), so LOD switching would have to be done by your app.
Please upload candidate patches to the tracker.
Need help now? IRC to #irrlicht on irc.freenode.net
How To Ask Questions The Smart Way
Need help now? IRC to #irrlicht on irc.freenode.net
How To Ask Questions The Smart Way
Also, there is no method to clear textures from memory yet. You can clear unused meshes from the mesh cache, but because of efficiency constraints the materials don't grab the textures, so it's impossible to know if a texture is being used.
This probably doesn't matter so much, because (afaik) Direct3D and OpenGL both page stuff into client memory when they run out of video ram. But if you plan to use a hell of a lot of textures across a massive environment or want to support users with not much system memory, you'll need to implement a new texture cache.
Also, if you want a LOD mesh node Spintz has one in 3demon (an Irrlicht based engine) that you could adapt with minimal work (it's pretty simple and I'm gonna steal it for the IrrExt project anyway). You are going to have to split up meshes and add LOD manually, at least unless I finish my impostors implementation before you start.
edit: also hardware mesh buffers are supported in SVN at present, so it's best to use a nightly build if you do a proof of concept
This probably doesn't matter so much, because (afaik) Direct3D and OpenGL both page stuff into client memory when they run out of video ram. But if you plan to use a hell of a lot of textures across a massive environment or want to support users with not much system memory, you'll need to implement a new texture cache.
Also, if you want a LOD mesh node Spintz has one in 3demon (an Irrlicht based engine) that you could adapt with minimal work (it's pretty simple and I'm gonna steal it for the IrrExt project anyway). You are going to have to split up meshes and add LOD manually, at least unless I finish my impostors implementation before you start.
edit: also hardware mesh buffers are supported in SVN at present, so it's best to use a nightly build if you do a proof of concept