-
Posts
28,206 -
Joined
-
Last visited
-
Days Won
394
Content Type
Profiles
Forums
Events
Everything posted by robcat2075
-
Quick (incomplete) stand-up animation test
robcat2075 replied to mouseman's topic in Work In Progress / Sweatbox
that's not a bad start. i can see you are thinking about the problem of showing weight and mass in motion. I agree that he's not getting his weight over his feet soon enough to stand up. When he's pushing off with his right hand he can be leaning farther forward and over his knees. When he's standing up his whole spine can bend to lag behind his hips. A core challenge of all 3D animation is trying to make it not look like a set of parts flying in rigid formation. We look for opportunities to overlap motion so things are not leaving and arriving all at the same time and to not move them in straight lines. For example when he moves his right hand from his knees to the floor, the hand could lag behind the wrist initially. The wrist is moving down in two straight line paths (with a lurch in the middle), an arc that lifts off the knee then around down to the floor would look nicer. When the hand lands it shouldn't all stop at once. You could have the fingers contact first, then spread as the palm pushes down, or have the heel of the palm land first and the fingers flop down a moment later. There's something about that guy's upper arm that bothers me. It has an natural forward bend to it. I'd go into the model and fix that. When he glances to the side at the start, dont' just pivot the head, the shoulders should contribute to that gesture. In Hanna-Barbera animation they can get away with moving a head and leaving everything else locked in place but it looks odd with more realistic characters. If the face isn't rigged yet, at least rig the eyeballs so you can pose them other than straight forward, which is almost always too neutral. Something I'd experiment with is the way his feet react when he's standing. They look a bit too static for all the weight that's being thrown around on top of them. I'd think about having them nudge forward and down when he's shoving his weight against them from behind, like they are getting pushed into the soft carpet. -
variation view AuroraB000.mov
-
Here's a go at the curtain effect... http://www.hash.com/forums/index.php?s=&am...st&p=366288
-
This isn't quite what Rusty is looking for in his other thread but I'll show it none-the-less... a fly-over of the Aurora curtain effect. Basically two gradient combiners are used to fade out the top and bottom and in the middle is a turbulence combiner, stretched vertically and animated in the chor to move through the mesh of the curtain to give it the undulating look. AuroraE000.mov PRJ done in v17 but it will work in earlier versions Aurora06_animated_flyby.prj
-
That's a great idea!
-
-
The view from space looks like a cloud-like material applied to a transparent shell that is slightly larger than the planet surface.
-
Wow! I didn't think that would work at all! Idea.... maybe you've thought of this already, but if you have credits at the end of your DVD, include everyone who contributed to the Kickstarter. Small type. "Sita Sings the Blues" has such a list at the very end.
-
IBL HDR environment map projection map Tutorial
robcat2075 replied to jason1025's topic in A:M Tutorials & Demos
Is there something missing from the very beginning? It seems to starting mid-sentence and I'm not sure what we're trying to do there. -
IBL HDR environment map projection map Tutorial
robcat2075 replied to jason1025's topic in A:M Tutorials & Demos
I'd recommend splitting part 2 into two parts, then you could upload them. The download from Youtube is turning out to be 346MB -
IBL HDR environment map projection map Tutorial
robcat2075 replied to jason1025's topic in A:M Tutorials & Demos
How big was the Part II QuickTime file? -
Nice Orba-tron! (That's what I would have guessed it was called.)
-
However, a better first gambit would be to move the three pointers away from the edge of the snout by inserting a spline like this: three- and five-pointers are happiest when they can be a flat as possible and not trying to span sharp curves.
-
I like to stretch the magnitude out a bit then drag my cursor on the gamma property to rock the bias handle back and forth and see if the crease gets better or worse in either direction.
-
The three pointers on the snout will be troublesome. It's possible that if you stretch the bias on them they may look smoother.
-
Don't do the dead-end spline right below the corner of the mouth. That will crease.. Continue it across the next patch and hook it or join it with it's mate from the other side.
-
If you can recall it, do let me know. So far, in my spot testing, making wider looks OK.
-
Doing things inside rooms is problematic in terms of lighting Day scene with interior lights OFF with interior lights ON
-
It seems to be related to the resolution of the z buffer map. Here's a test with a flat surface with an overhanging edge in the scene: in the original scene the res is 4096x4086 If I reduce that to 512x512 the banding becomes more obvious... If I go back to 4096x4096 and increase the width of the sun to 5000cm (it was zero originally) that seems to help a bit...
-
I'm testing out camera angles... the banding on smooth surfaces bothers me, I think it has something to do with "bias" on the z buffered lights but I haven't ascertained what light is doing this. I was sure it was main "sun" light but changing the bias on that had no effect. any ideas?
-
Maybe... a very determined person could use a distortion box as an action object in an action, that woudl let you key the mesh of the distortion box and then somehow use this action for the functionality of a Pose slider...hmmm... tricky...
-
Once you leave distortion box mode you can't get that old distortion box back. Returning to D mode creates a new distortion box unaware of all previous adventures.
-
That is true. That's a limit of distortion boxes.
-
I'm able to get an expected result. Can you show your example more?
-
Yup, I have no idea what that's about But it looks cool.