Wolfire Games Forums

My what big teeth you have.

I mean, what's even creepier than switching to that in a few seconds is doing that in the blink of an eye (it is a facial expression after all >:D)

But, umm, is his fur supposed to change colors? Or did the lighting source move or something?

I'd love to see facial expressions, especially surprise from a split second before inserting your knife into their throat .

This is good news, not only this cool morph system, but the programmers as well. I certainly hope they know what they're doing. Then again, knowing David (which i don't much...) he wouldn't have hired them if they didn't .

Progress is always good, keep up the good work... because... there's cookies waiting for you?

EDIT: Snap, their eyes will be able to look around, right?

Finally, a graphics update that's really useful. But don't listen to me, I'm a troll.

The shot of the morph looks like it's taken from Lightwave or so, not from the engine?

The morphs are shown from inside Max, but the same morphs are working fine in the engine and look great. This was a test Ian did when morph targets where being tweaked.

Nice!

Just as a sidenote, when a wolf shows his fangs, the fur rises. If you ever considered to include realistic fur, that's a way to really benefit from it.

Have you thought about simulating muscle contraction and release with morph targets?

That looks totally badass, but personally, I think the wolf looks too different from its original Lugaru counterpart and almost 'cute and fluffy' until it growls

By the way, what are morph targets? I didn't quite glean enough from the blog and the posts contained herein ...

I guess that if you have 2 models for 1 3D object with vertices at different locations, you could say "move vertices from model-A-vertice-locations to model-B-vertice-locations, adjust the normals and stretch the textures accordingly with seamless tradition if this-and-that happens". That would make model B a morph target for model A, and probably the other way around if you want to have a way to go back.

Seems like it would be easier to have 1 model except have it so that a certain vertex can move to different locations depending on the facial expressions (I think this is kinda what HL2 has?)

In a frozen state, the model with the moved vertex is not the same model as the model with the vertex at the original location. Of course, model B could be a transformation of model A, but the resulting matrices are different and model B doesn't have to be a transformation of model A, which for me makes it 2 models in each case

This is basically combining the best aspects of traditional vertex interpolation (ala pioneered by Unreal and Quake 2) and modern skeletal animation, so we can combine bone-level pose animation with per-vertex soft tissue animation.

The new programmers are really fantastic; they are better at their respective fields than I am, which is why I asked them to work on this project My specialty is game/engine design and rapid feature prototyping, so it is best if I can focus on that as much as possible.

That's good news, could you tell or describe to us what the other 3 specialize in?

For me, it's more interesting who they actually are. Like, you know, their names, history in game development, how they were actually chosen and stuff.
Yeah, I'm curious

PS: I vote I be added to the development team, for purposes of creating cutscenes/intros/etc. Just think, it's a beautiful, ultra-advanced-graphics-engine world with a rising sun that illuminates the vast rolling grassy hills, and suddenly BAM! CUTSCENE! Scenery disappears, everything goes black and white, and looks at though it were drawn by a third grader with a Sharpie!

It's win-win!

It can have voice acting from Microsoft Sam as well

I've stepped into Bizarro world.

But seriously yeah, give us some info'mation on the programmers.