TSE and Trajectory Zone?

#1 I might have read it wrong, but right now it appears that the new TSE code is DX only. In my view (if I read the info correctly), in order for TSE to be a viabale option for crossplatform projects it would need to be usable via OpenGL. I can not see GG supporting winblows only with a DX release of TSE, GG has touted cross-platform forever now.

This would also cripple TZ as well as another other TGE projects (Orbz, TT, GR, Lore etc..) who are looking to or currently support cross platform releases.

#2 I'd like to see a timetable for OGL support, that way if its within my development schedule I can go ahead and start implementing TSE when it comes out with reasonable assurances it will be cross platform by the time our development is finished. If thats possible it sure ebats waiting to implement it until the very end when it becomes cross platform.

#3 My advice : never bank on unreleased tech for existing projects. lf it comes out before completion, and the amount of work and time to put in warrants the perceived added value, then you might consider adding it.
Otherwise, as Forhan said in another thread :
concentrate on gameplay first, eye candy a game doesn't make... (and believe me, I know this first hand ;))

#4 There is always Trajectory Zone 2.0

#5 LOL! I can't wait until TZ is released, sounds like it's gonna be really awesome!

#6

Quote:concentrate on gameplay first, eye candy a game doesn't make... (and believe me, I know this first hand ;))

The gameplay will be the same one way or the other. The design document is finished, and 90% of the gameplay is already coded.

#7 Wow - at this point, I can't think of many compelling arguments for moving to TSE. If TSE currently supported OpenGL, I might consider it. However, even if it was just a matter of 'easy' integration to the TSE, then there's the matter of content. Just having an engine that takes advantage of shaders isn't enough - there's now the changes to all the content. Since it wasn't planned for shaders, there even more work to do compaired to if ya' planned for shaders in the first place. It's got some interesting issues that would occur. But, DirectX only? Not a chance - GG has pulled 60% sales from Mac users at times. Even if the average was only 30% - wow, that's a lot of sales to miss out on! :-)

However - TZ 2.0? Oh heck yeah - I'm pretty interested in the idea of using TSE when it's cross platform (I may or may not be an early adopter - if only for the cost savings, and to get some experementing in before it does get OpenGL support. DirectX only games don't appeal to me from a marketing standpoint.) But I'm more interested in doing a smaller game with it first to get use to the art pipeline issues before doing the Magnum Opus that Trajectory Zone 2.0 would be (Wow - how many TZ2 items are on the list, Eric? We generated a lot of 'em :-)

#8 Lots, Davis. :)

Brian answers the questions here. The answers pretty well are that it would be hard to port an xisting TGE 1.2 game and that the TSE is scalable. So, if you don't have 2.0 pixel shaders, it'll fall back to 1.0 shaders, and if none of those, back to the standard TGE.

He also restated OGL is in the works. ;-)

-Eric F

edited because I just can't seem to get that url right. :)

#9 Aye, but, that does mean that we push back showing it off, etc., and extend our development and beta cycle. Not sure I wanna do that :-)

I should also mention that my comments were written as I was thinking - so ignore the chaoticness of 'em ;-)

#10 Me either. As I said, it's largely hypothetical. I guess I was thinking more along the lines of further down the road...

Pretty neat stuff, though! Just imagine the TROFF levels w/shaders. ;-)

#11 DAMNIT ERIC! Now you got me to thinking about it - I may have to go buy TSE RSN - I want to see what they have done with the terrain lighting, and if I can pull off some tricks I couldn't before using the new system. Heck, grass would look awesome using shaders - and, since it falls back in the absence of a shaders, the performance loss would be minimal. AAARRGGG! You must not put these evil thoughts in my head, Eric ;-)

#12 LOL Davis...

You're right. I shoulda kept my mouth shut. ;-)

I really am such an eyecandy junkie, though. All I can say is what Labby did: TZ 2.0. ;-)

#13 So, if I were going to change Trajectory Zone to TSE, what would I need to do? Well, I've had this thread running in the back of my head since I first saw it, and finally came to some conclusions. Nope, not switching TZ over (but, I can almost garantee TZ 2.0 will use TSE), but I'll post this for anyone who's considering the idea, or wondering what's involved or what design choices could lead to using Shaders to improve performance on low end systems (yes, you read that right.)

1) FXGrass would be gone. Instead, I'd use Shaders for grass. Why? Well, two reasons - first, FXGrass is a bit hacky when you look at it from above, and since I've implemented chase-cam type stuff for weapondry (which ROCKS!) you do get that view from time to time. This design choice would also help reduce horsepower requirements on low-end machines - instead of having grass take up horsepower, people with cards that don't support Shaders get the texture that was holding the shader instead. And since the lower end cards suffer performance problems anyway, it's that many more polys that don't have to be pushed - thier performance improves.

2) Improved explosions. Right now, I use the particle system for explosions. However, explosions could alternately be done using shader fx instead. There's a good example of a kick-butt explosion with shockwave and all at CgShaders.org. In this case, I'd disable the normal explosions on a machine that has Shader access, and only use 'em for people who don't have a card that supports it.

3) A couple of 'utility' items - for instance, cloaking. Imagine being able to cloak your player until the next time you fire (but with a nice little game-ballancing cost) Nice little Shader effect for that on CgShaders.org also.

4) Further enhancing the models. Just some minor tweaks here and there would improve the looks of the player models considerably. Would also be REALLY cool to render just the Tonk Tank (looks like a Tonka-Toy version of a tank) as a cartoon style shader instead, just to make the model a little more surreal ;-) The Hover Tank would also emit a soft glow from under it.

5) The UFO's would get some serious rework, emitting glowing beams of light from various places while they spin. I'd darken the level a little more, and let the whole thing have an incredible Close Encounters meets Trajectory Zone sort of feel. (I can visualize this sooooo easily - it's gorgeous in my mind :-)

6) A couple of weapons would use the shaders instead of the current particle system, just so I could do really funkified stuff with 'em ;-)

There would also be some experementing with the terrain here and there, to see if I could improve upon current deformation techniques and discoloration of the landscape.

All but one effect would make no change to gameplay reguardless of if your machine had a shader 1.0 / 2.0 capable card or an older card that TSE would ignore sending shaders too. Or you could turn shaders off in an attempt to improve performance (though I think that might have a negative effect.)

For TZ 2.0, I'll probably plan on using Shaders from the start, yanking out a portion of the existing game assets here and there, then designing some gameplay elements that utilize shaders as a little more than just eyecandy (which is pretty much all I described in my list)

#14 You should be able to do the cloaking with the origional TGE source.

#15 The main reason we're considering using TSE for Mayhem (if it is available early enough in our cycle) is that Mayhem's world is 99% mechanical. Mechanical works real well with shaders :) All that metal :)

#16 Nod - you can make metal look like metal. Or rusty metal. Or scorched metal. Or metal with dust on it. Or... well, you get the idea. Sure, you can do a lot of that without shaders - but, there's just some minor things that shaders do a little bit better for that than just putting rust marks (for instance) in your texture. All of our models are metal. Some of our bases are metal (though most are concrete or wood - which you can get great fx for them too ;-) However, on the landscape it's self, I'm not sure how I'd go about making the landscape look better. There's some things that could be done in Purgatory (a lava filled level.... MMMMMmmmmm - you could do wonders with that!), but few other levels would benefit. Ok, well, TROFF ( a Tron-esque level) would look awsome with that neon glow thing happening.

#17 Harold: You can do a cloaking effect using TGE. You can't do THAT cloaking effect using TGE ;-) (The one I was looking at does the Klingon Bird Of Prey cloaking effect, instead of a simple fade, which is already there in TGE.) Or were you refering to the idea that I could use the typical TGE cloaking effect as a fallback on non-Shader card systems?

#18 But could it be done without significant changes to the art content?

#19 Yes / no. How's that for an answer? :-) Some shader fx are controlled through scripting (from what I understand - since I don't exactly have my hands on TSE, this is based on what I've heard written in the forums so far), and some are attached to existing textures. Would the models have to be reworked? No. The textures for the models? Yes, if you want to add any shader fx to them. Same goes for terrain, dif, and dts objects. So only a portion of content would have to have any rework done to them - basically adding the shader fx. Now, how that would work exactly - I don't know until I have my hands on TSE ;-)

#20 My modeler said that generally to upgrade our art to use shaders would most likely involve retexturing all our models. I'm hoping that we can throw in at least a few decent effects in script alone without having to retexture. If it comes down to retexturing all our models then it will all come down to when TSE is available for us. If it comes too late in the cycle, we won't upgrade just because i'm not goign to extend the cycle another few months. If it comes early enough in the cycle that most of the art is not yet completed, then we will most likely upgrade.