I’m doing a rigging project for my Masters at the moment and if possible would like some feedback on the ideas I’ve got. Here’s the model I’m rigging, it’s not the greatest in the world and I need to make some changes to it but it’s useable for now. It’s also video games orientated but that doesn’t hold it back to much.
I’ve already been in contact with a few professionals and there main point generally seems to be you need to focus it towards an animator and what they would expect out of a rig.
I’ve got the basic things like main controls, attaching it to a path, opening/closing windows and flags up/down.
The others I have though I’d like to know if people would find them useful if they’re animating it (any automation I will male so it can be toggled on/off):
Attach the ship to a surface as you move it
Being able to break the mast on any loop of edges and have it create the necessary bones
And a speedo that’ll work in whatever units the user has set and can be changed to others like knots.
And please feel free to say anything you would expect to see in a rig like this. It can be awfully hard to create a rig without a purpose but I guess that’s the research part of the project.
I can post up some of my scripts as well for anything I’ve done if anyone is interested. They’re probably quite simple as I’m a total beginner but whatever.
Focusing your rig to an animators needs is quite important. Also, and often more important are technology constraints, especially in gaming. If an animator needs functionality that blows you joint count, everyone has to work it out to meet in the middle.
So, what game platform do you have in mind? If it’s a PC or console game you’ll have much more freedom with moving and breaking parts than an iOS game for example. What’s your target market.
I like the idea of attaching the ship to the surface since you can do that with one joint and nested transfoms, but is that something that will happen anyway in runtime?
Platform is definitely PC, it’s more about the rig so I’m going for the simplest option platform-wise.
My idea for the attach to surface thing was more for to help to set up a base for various animations the ship may have on the waves but again, it would really depend on what process the animator is using to do that. So it’s an avenue, I’ll see what I can do in UDK regarding attaching objects to surfaces or liquids. PhysX may come into it again in that respect so I shall investigate.
You can also add some dynamics to your ship’s ropes, cloth parts, and all the thing which can have dynamically sense in real.
Also in your tool try to have some controls for driving the ship, like how it can rotate dependent of the location that engine is located, also you can add params such acceleration, speed, backward/forward movements and etc…
And this part might be a little harder, but good to give it a shoot, like having a dynamic pivot that is calculated by different planes around the ship (inheriting the ship movements) which they are individually attached to the water surface and waves and etc can affect them, and finally the pivot can bake by that deltas regarding to the changing of planes in their local positions and rotations…
Haha, completely forgot about the ropes, not sure how, seeing as they’re everywhere.
The other thing, the image was probably a little misleading as I haven’t changed it yet. It won’t have the steam part, it’s just going to be pure wind driven
For the last part about the dynamic pivot, would that be for the ropes? And, unless I did it in UDK, the attaching to the waves would be difficult because water/waves would be made in engine so they wouldn’t be available in Max.
[QUOTE=EhsanKiani;7973]No, it’s just for the parent of your ship, so the parent is linked to the dyn pivot in object, so then that’ll be calculated that way,
Well yeah, I only know the Max or Maya part of it, not the game engine![/QUOTE]
Ah that’s fair enough.
Regarding ropes, Unreal has a built in rope system so I may use that but combine it with some user customisation on how many bones they want in the ropes to control the degree of realistic bending. Or offer another option completely and have Unreal ropes OR custom ropes that are calculated some other way, potentially by speed/surface/gravity.
What is the goal of this project? You can make an absolutely awesome rig by having things work dynamically and pseudo-mechanically, how ‘animator friendly’ the rig is may be secondary, and you can have a kickass project. Animator-friendliness is important for production, but probably less-so for research and education. If you don’t have the experience, or an animator to help you and iterate with, you’re not going to make something animator friendly even if you are focusing on it.
Yeah it’s a tricky one. We do definitely have to produce a rig based on games production pipeline but I think the rig is primary though I would have to point out why I’m doing certain things which is where the animator bit comes in. But it doesn’t haven’t to be majorly animator orientated, just having some animator feedback would be good. =)
It is likely that I will end up with two versions of the rig, one that is the awesome do everything rig, and one which is the one that is somewhat more animator friendly and UDK friendly.
And a couple little updates, been a little busy on another deadline:
UDK also offers a vertex deformer in its shaders, which can be used for things like clothes, telephone wires, flags etc… and is all in the Shader Manager for Unreal. Talk to some of the people at Epic if you need help, or ask on UDK.
Dynamically breaking the mast is going to be messy, pick a couple of points and make that work. Getting the doors on the guns to open and slide out, can be accomplished in a variety of ways, most of which I’d hook into a set driven key node, as the whole thing is one action, opening the door and repositioning the cannon.
You may very well end up with two rings one for game export and one for cinematic, as they both require different things. (Speed and Detail)
For a lot of the character stuff in unreal you have to push back on the engineers, since it’ll be the animators who suffer when you add cloth controls to every rope and sail, and then the animators have to generate all the assets to make that work, while some shader work, and some rigging will make it look more natural, and leave the animators to animate the people on the ship.
