robcat2075

Hook's Spring Law? More big government! I suppose a side by side test with one hair might reveal the difference in behavior. My first guess is that Hook's Spring Law is more computationally intensive.

a-ha! What is the intended difference between "constraint" and "spring" for hair?

Increasing the "stiffness" may be what you're looking for.

I don't see anything there that can't be done. That first guy has a bunch of pass thrus.

does it work?

how about this? QuieterRope_500.mov One of the problems A:M has is that people look at a test by a novice and presume A:M is the limiting factor. There's lots to know about getting simulations right and I'm a novice. This clip is VERY near what a finished result should be and I'm just a novice trying things. The simulations you see in movies are done by people who do just that as their day job. They spend lots of time coaxing the simulations to get it right. I was looking at the manual for "nCloth", the simulator you get for "Maya Ultimate" and that seems to have pretty much the same controls and same limitations as our "simCloth". You gotta learn to use it and you gotta start small. They start people out with a flag just like A:M does. I can't think of a scene in a movie where a simulated cloth or rope is thrown out and then the camera just stares at it while it's doing nothing. They cut away as soon as the action is done. I can't find a demo like that on youtube either. Can you?

Stoner Jesus! Is there "damping" on hair? Turning that up may get you less flickery hair.

Here's another experiment, no music or acting, but this tosses the rope around in a bit more complex manner and has a bump map added too. RopeToss_600.mov

It's getting used! I think the twitching is from pulling the IK leg targets too far.

that looks great, very impressive!

There's a few elements we're still looking for , but it's coming.

Thanks! Yes, there are strategies to do that. I'm still experimenting on that. Increased "damping" reduces it quite a bit. 193 via the dup wizard maybe when I write my book.

Yeah I guess you're right. And i thought Stian's map didn't have the problem but when i put it on the rope one side of the rope looks like it doesn't match. But his should tile perfectly. So I don't think it's the map that is the problem.

The pecs seem rather pointy.

Another way to state it is that there should only be four corners per at each CP. Two if it's on the edge of a mesh. But three, five or more are usually a sign that something isn't splined right.

that got better.

when I'm modeling I like to think "peaks and valleys". I try to have splines only on the peaks and valleys of the contours and not on the in-between. The in-between will be naturally smooth if there's nothing in the middle of it. For example, when i look at the contour of the nose I see more splines there than just the peaks and valleys.

My inner German tells me... That one seems to be an eat-in only restaurant, they make no mention of delivery. Also they're only open a few hours a day and only take a particular kind of debit card. But the menu looked good. A place that takes PayPal would be ideal.

that does work! I guess this is a genuine AMReport candidate then. Thanks!

Is there any way we could buy him a pizza?

Just gave it a try, didn't help. But worth a shot.

Here's a turn-around. See how one quarter of the model does not shade like the other three quarters even though they all have the same map on them. Rotate.mov

Dat's lookin' foist class!

My model is 4 patches around. Patch image puts one bitmap on each patch. Three of the instances meet up fine with each other. One does not.

Your image is not exactly "seamless" - I took it, applied it to ground plane, repeat 2 x 2, seamless ON. Looks like there's discontinuity 1 pixel wide?. Set mine to bump and turn the model around . Three of the four quadrants seam up together just fine. One does not. If the one pixel discontinuity is a problem, it would be a problem at every join, not just for one.