robcat2075

It is difficult to top one's self.

Good to see you didn't get washed away!

I believe that says "wherever", not "whoever"

For a tutorial, I'm looking for example images of the following. Images that in a one-second glance can demonstrate the idea... Motion Blur Depth of Field Shadows Ambiance Occlusion Reflections Particles and Hair Subsurface scattering Fog Toon Render Field Render Stereoscopic 3D Plugin-shaders I have candidates for some of these but perhaps you recall better ones. If you can think of a good image, post a link to it here. Thanks!

Another fine Spleensterpiece!

I don't know how hard that would be. IS there any other 3D program that can genuinely treat opposite sides of a single thickness surface differently?

That does fit better. that shadow on the ground is probably a bit too sharp. You might up the softness on it. And maybe using Ambient Occlusion on them might make them less severe in general.

Too many swinging camera moves. I would consider stabilizing that quite a bit more. At about 0:13... one problem is that the background has brighter and more saturated color than the foreground. In general, objects in the distance will appear more faded. Perhaps that is the backdrop problem you are referring to?

the big five-oh! Happy Birthday Matt!

If you can make the shadow bluer i think that would help the match. Looks nice!

Very cool, Mark! You need a 3D cake printer, or maybe you could put a cake into a computer controlled milling machine.

Yes. Understand that it's not an automatic process... you have to decide where to draw the splines. Since you won't need the actual poly mesh, you can import them as "Prop" which is quite fast for large models. I'm sure there must be an upper practical limit. Let us know when you find it! I don't use porcelain anymore. 5-pointers are best when they are fat and flat. Fairly regular side lengths and fairly little surface curvature There are usually several options for the exact placement of a 5-pointer in a mesh so you can avoid using them in problem areas and make 4-pointers do most of the work. There is a bit of strategery to it. Also, the flaws in 5-pointers typically aren't visible once a texture is on the model.

There are two perspective problems. -the sea doesn't extend to a real horizon line -the sky clouds look tilted into the sea Those are both hard to fix without having a very large sea plane and a very large sky disk. Absent those, regular camera fog could help obscure the detail that is making those problems too visible. Something like this:

Welcome back! Yes, A:M's Global Illumination will accept a High Dynamic Range image in several formats for lighting purposes. There is a thread by jason1025 somewhere around here where he goes thru a case on using it and I'm sure there are many other threads on it also.

What do you want to control in the camera fog that you can't?

I would suggest some fog to diminish the horizon line a bit.

Beautiful hair!

If it's one flower on a stem, a dynamic constraint is a good way to go.

can you show shot of it?

that was extra 70's

Today via Skype I got to visit with forum member Roger. We were trying to ascertain why he was having so much trouble accurately placing new key frames in relation key frames he's already made. What we discovered was that A:M was not updating the view window to match what the time counter was indicating. It looked normal if you just hit play, but if you scrubbed the time control forward or back it would stop with the last frame you scrubbed through and not draw the current one. For example if you were trying to get to the frame where a bouncing ball has just touched the ground you would have to scrub to the frame after that before you would see the ball actually make contact. If you hit the Space bar ("refresh display") the display would update to the proper frame but you had to always do that. I presume this is a rare situation, but I mention it in case anyone else is baffled by why things don't appear to match up. Hit the Space bar after every scrub. Roger has an unusual laptop that has both the usual laptop graphics chips and also An NVIDIA Quadro display "card" built in. Somewhere in all that is probably the reason for this but I don't know exactly what's up with it. Ideas anyone?

I don't speak for Hash but I'd expect supporting it very soon is unlikely because... -it's much more expensive than A:M, probably even more expensive than most of the computers that A:M users own -which means very few A:M users would be likely to purchase such a card -which means the development time spent on supporting it would drive very few sales. And it's not OpenGL so it would require a new pipeline to be supported between A:M and the display card in addition to what is already supported. If they could make that work as a graphics card that is transparently run by OpenGL they would have something. And make it cheaper. Also as I watch the demo video, it's obviously faster than CPU ray tracing but it's still only updating a few frames per second. You'd have to turn it off to work on real character animation. You don't really need to see refraction or reflection to judge animation but you do need to see all the frames. None-the-less, A:M has an SDK available. Encourage them to create a plugin for A:M that would use the card, then we'd just have to get someone to buy the card to try it.

That is full-bodied looking hair, Nancy!

That is curious looking. Groomed hair can't keep its shape with dynamics?