Hellcat

[Corsair(pilot)]

HCl's multiplayer Prophecy news has got me all excited. So much that I've started doing something I don't usually do; make a game model. The ultimate goal here is to get this eventually into the vision engine. So far, here's what I got... not much yet. No texturing, just the bare mesh.

 

[Lt.Death100]

I thought the Hellcat Vs cockpit was a bit bigger.

 

[Corsair(pilot)]

Well, from the reference I have compiled I don't think so. I'm looking at both the in-game and fmv models from WC3 and WC4, the WC Academy Cartoon, and the CCG cards.

There are discrepencies between the sources, though the cockpit size doesn't really stick out in my mind... the biggest is the engine array. I tried to amalgamate some of the different styles to come up with something that looks nice. Also, the underside of the fighter seems to vary drastically from source to source.

 

[AD]

I thought the Hellcat Vs cockpit was a bit bigger.

Actualy no... I think the specs say the thing is 27 or 29 Meters long. If anything that cockpit is to wide and high.

 

[Death]

The problem with those specs is that the FMV sequences of WC3 and 4 don't show them to be that big, comparing the humans approaching the fighters during the launch video clips. We know Blair is somewhere around 5'6" or so (or whatever Hammil's actual height is), so we do have a baseline for comparison.

 

[Corsair(pilot)]

Alright, I've got some questions perhaps some of the more seasoned vision engine editors could help me out with.

The model right now sits at just over 2000 tris, over the assumed 1200 limit, so I was trying to think of ways to get around this before I start texturing... which leads me to my first question;

#1: Is it possible to have fighters consisting of several different components, like a capship would? And would this be a way of sidestepping the single object poly limit? So I could, for example, have seperate main fuselage, left wing, right wing and cockpit models, all well under 1200 tris.

#2: What kind of texture resolution would I have to deal with? Are multiple texture image maps for a single object possible, and if so, what kind of impact does it have on the engine during gameplay?

#3: I use Softimage XSI as my primary 3D software. Has anybody ever tried a conversion from XSI to the vision engine? Would it be possible to convert from XSI, or would I need to export to a different package? The only other tools I have access to at work are Lightwave and Deep Exploration... argh.

And I think that's it for now, I appreciate any help anybody gives.

 

[SabreAce]

I'm pretty sure you can do fighters with multiple components - IIRC, the Vampire and Panther had their engines as seperate components, due to the need to have them rotate in flight.

As for the other two, have you looked Killerwave's site?

http://killerwave.solsector.net/Model_Tutorial.htm

 

[Quarto]

#1: Is it possible to have fighters consisting of several different components, like a capship would? And would this be a way of sidestepping the single object poly limit? So I could, for example, have seperate main fuselage, left wing, right wing and cockpit models, all well under 1200 tris.

Yes, it is possible. If you'd like to see an example of how this works, check out the Movie Rapier that Eder made for the CIC birthday two or three years ago. I don't think we've ever really checked how much of an impact on performance this might have, but given that multiplayer wouldn't involve as many ships as a normal mission would, I'd say you're safe, especially because fighters use collision spheres instead of mesh-based collision-trees.

#2: What kind of texture resolution would I have to deal with? Are multiple texture image maps for a single object possible, and if so, what kind of impact does it have on the engine during gameplay?

The best texture resolution to use is 256x256, because this keeps your ship compatible with 3Dfx. And yes, you can use multiple textures on one ship - for example, just about every ship model converted directly from WC3/4 to WCP used multiple textures (so, the UE Banshee, for example). As near as I can figure out, this doesn't noticeably affect performance.

#3: I use Softimage XSI as my primary 3D software. Has anybody ever tried a conversion from XSI to the vision engine? Would it be possible to convert from XSI, or would I need to export to a different package? The only other tools I have access to at work are Lightwave and Deep Exploration... argh.

No idea, you'd have to check KillerWave's model conversion tutorial, maybe it says something about that.

 

[Ptarmigan]

Looking sharp.