Tge / Tse Advise

#1

Quote:
If I start with TGE, I'm assuming that any missions/levels I build in TGE will have to be done completely over once I move to TSE - is that correct?

Only the water will change. The terrain will port over fine but you may want to switch to the Atlas terrain (which will require rework). You will also need to recompile your Interiors.

The upgrade path is pretty smooth but it is definitely fun working with TSE =)

#2 Well it all depends. I mean if you want your game to be cross platform, i would highly suggest using TGE. You should be able to get some nice results if you spend your time to make the art good. The lighting pack helps out a lot in that respect. It's possible to get results that look a lot like Half Life 2 just using high quality textures. Of course you won't have the fancy looking shader enhanced floor tiles like in Half Life 2, but the main things that makes Half Life 2 look so good, is that the textures are pretty high resolution. So you could just get a good camera to take pictures for the textures, and it would look pretty sick if the textures are 512x512 pixels or greater. That is if you want a realistic looking stylization for the game.

You can design your game to target TGE and TSE, but it requires a lot of extra work to do it right. If you are not worried about your game being cross platform at the moment, and you aren't worried about the game working on non-shader hardware, I would definitely suggest using TSE. TSE is still missing features because its not done, but if you target the shader based system when doing your art and effects, you can come out with stuff that could look a lot better than TGE. Of course that requires learning how to take advantage of a shader based engine, from both the content creation side and the programming side of things.

For example if you were using TSE, you can have certain textures that are created procedurally, using light and other parameters (like textures) as paramters. You could also do interesting things like use vertex and pixel shaders to do particle effects. On the actual content creation side you could worry about shading the model a little less and use normal maps to do a lot of that work for you. When you use normal mapping you can get away with using pretty low poly count models in some cases. For example if you have a wall that is mostly flat, other than some engravings and the roughness of the wall itself, you could get by just using a normal map on the flat surface, with a texture. If you want to get really fancy, you could parallax map it too, in order to make it look like the actual wall has extra geometry.

So there you go. What you should use is pretty much determined by what your goals are. On thing you should remember though, is that even though TSE can in theory look a lot better than TGE, it will take more time in the content creation process to get it there.

#3 Peter your game looks good. And pretty complete. It's right around the corner. Why do you want to switch engines now?

#4 I just watched the trailer, and i second Ajari... why switch now? (It lookes AMAZING by the way)

#5 Yep the game looks good, my guess would be that you want to use multiplayer support for that game and 3DGS is not the best solution for multiplayer. Most of the models, prefarbs, can be converted to torque, however AI could be a biger problem, scripting is a litle bit different but it shouldnt be too hard. Anyway I would reccomed you to finish this project with 3DGS and release it, then maybe in Orion 2 start with different engine. TSE should be done by then ;)

One more thing. Do you plan to release demo? I would like to see a that game, I was 3DGS user before and I found torque much easier and better for huge/realistic terrains.

#6 Yes, I came pretty far with 3DGS. Thanks for your compliments on the game.

Why change now?

Well, keep in mind that trailers only show off the best stuff. Also, I really only have one level done in the game (with 7 huge levels planned). Also, I'm most concerned about 3DGS's stability across many different PC configurations and video cards. I push the A6 engine to the limit quite a bit. I did some pretty tricky programming to get it so that a large level is broken down into about 10 exterior sub-levels and every building interior is a separate sub-level as well. But with all that level loading and swapping in and out of assets manually, the game is more prone to crashing. I would rather use a AAA engine with proven stability (I'm sure Torque has its own issues, but with a proven large title, I'm more confident I can work out my own future issues with it). And yes, I need to add multiplayer capability for a PC game to really sell well. So, though the trailer looks like the game is far along (and it is in many ways) there's still a long way to go to bring it to market. I can convert much of my code over to Torque I think (NPC dialog system, bartering system, NPC AI code and pathfinding).

3DGS has been great for building a prototype and I've really enjoyed the community, but time to move on to the next level. We have another a little less ambitious title (a Zelda style game) that we'll release using the A6 engine next year which will give us breathing room for porting to Torque and learning the ins and outs.

@Denis - yes I have a demo that I can show you. Let me clean it up a bit (something I'm going to do anyway for the 3DGS folks and let them have my code) and I'll post a link in this thread sometime this week...

#7 Great Peter! I have also a big problems with large terrains in 3DGS, torque is much better at that point. Networking is fantastic in torque (I had a small LAN party with my friends in very early stage of my game) and I am very satisfied with it.

Anyway I found torque more friendly than 3DGS altough I am still a noob (after almost two year) but feel free to ask anything you want to know (you can email me if you get stuck somwhere) and if I can I will help you :)