robcat2075

I'm a bit doubtful about what look like specular highlights facing us on the big rocks. How would the sun light those from that angle?

Sorry, didn't know there was a cloth question in this thread from the topic title. My limited experience tells me that cloth needs to be fairly dense... meaning quite a few patches, as Nancy showed. For real world scaled objects about 2cmx2cm patch size is a place to start. Deflectors, on the other hand don't need to be terribly dense. Less dense cloth has the advantage of simulating faster, so you don't want it any denser than necessary. A:M cloth is much more powerful than I thought but you have to be willing to do lots of tweaking and re trying with settings.

Looking more closely at your image, are those minnows under the water?

I like the shot. A real challenging arrangement to pull off. I agree with the others that water/sand transition is the most troublesome. The reflectivity of the water surface seems to be uniform throughout, but in nature water seen at a more oblique angle (down near the bottom of the image) is less reflective than water seen at a glancing angle (out near the horizon). This is true of almost all reflective surfaces except mirrors. The gradiant combiner (with start and end set to 0 and "edge threshold" set to anything other than 0 or 100) can make surface vary from one attribute to another as the angle it presents to the camera changes. This would be one way of making your water reflection behave more naturally. Of course there would need to be some details under the water to see. Also possible would be to bump map some ripples emanating form the waterline to sell the idea that there is some interaction.

Ogre came out corrupted for me too.

whoops.

ok, maybe I've detected a few. There are still a lot of rigid stops though. When the guy on the right gestures with his arm on "you stole form my kin" that would be a great place for overlap on the whole arm motion. I'd also not make both gestures not the same size. First large, second small or first small, second large... it could go either way. There's room for overlaps on just about every body motion, really. If you have "The Animator's Survival Kit" he start into that on pg. 222

I'm not sure I see the moving holds yet...

They're working here in FF

This is a little try at making an AO style render with just z-buffered kleig lights. I chose the "Biobot" character from the extras disk because he has lots of crevices and bumps to challenge the occlusion effect. AOBTurnTestH.mov The result isn't quite as satisfying as a real AO image but it takes only 5 seconds per frame to render. The quicktime compression is making the shadows on the ground look patchier than they really are. In the original render they're quite smooth.

Looks good! What's the gyroscope about?

That's pretty much it. "ease" is a property of the path constraint. It's a percentage of how far the object has traveled the path. If you set the ease to 50% at 3 secs and 50% at 4 secs the object on the path will stay at 50% for that time, then move on (if we presume you have keyed 100% at the end of your animation).

Short answer... yes, a shader could do that, i think, but someone would have to write it. I don't know that it would be any faster. A long time ago Marcel Bricman put for the this technique, but its also a muti-pass technique and requires his gradient plugin which may not be available now. My experiment is based on Jeremy Birn's suggestion of changing the IOR of the glass while the renderer did the R, G and B channels. A:M doesn't actually do the R G and B at separate times so you can't key IOR for them. Using multipass and keying the light to change color is about the same thing in practice.

Dispersion is how refraction bends the different colors of white light differently causing the color fringing you see in prisms, cheap camera lenses and other irregularly shaped transparent objects. This is an attempt to simulate it by changing the Index of refraction of the glass material as the color of the light is swept across the spectrum during a multipass render. I also moved the light to fringe the shadow a bit. This is not a fast process, but if you need dispersion, it can be done.

Are her feet really doing the same arc that the guy in the clip was doing? A side view is better for right now.

You don't have deleted file recovery software on your PC? Now you know an action doesn't contain any model. It's like a small chor that load models it needs. You can put one model in another by importing it into the current model window. But why delete a model? You might want to use it in something else later. Storage is cheap today. No reason not to save everything.

A:M's AO is about occlusion from the sky dome which isn't quite the same as occlusion from indoor surroundings. So here's a couple of attempts to do indoor ambient occlusion. The first is my omni directional light set to be almost as wide as the room and letting A:M jitter that with multipass That averages out to the light originating from the center of the room which isn't right since the ambient light is really from the walls This one has a box of splines crawling along the walls of the room. This is better but I need to find a way to make the light not shine on the wall it is closest to at any point. That is what's making the bright spots.

I hear the beats in the speech as "you STOLE from my KIN" but the motion of the chicken is doing "you STOLE from MY kin". I really sense he should be landing on KIN. Overlap on the head as it goes up and down will make it snappier.

The rim lighting you have that outlines the pharaoh character really helps separate him from the background. The frog would benefit from something like that too. So would the guy with the Don King hair.

I just tried a 4096x1742 (double the Panavision res) render of my lighting test scene and it survived fine. So it's not the res that's the problem but probably a combination of a complicated scene and the hi res. If you submit an A:M Report on it, include a PRJ with all objects embedded and all the necessary texture maps in a zip, so your specific case can be tested.

Jeremy Birn's Digital Lighting and Rendering I'm testing out ideas I get as I read it.

thanks! The appearance of light spill is simulated with lights behind the walls, animated to change intensity as the light in the room moves about. I'm sure a real radiosity render would be more nuanced. Ding, ding, ding ding! Yup, it's a shadow-mapped "bulb light" made out of an array of klieg lights. A square gel on each one keeps them from overlapping. It's only slightly faster than a 1-ray bulb light but much faster than a many-rayed bulb light that would be needed to do soft shadows. I think the unnaturally large hot spot from the light being so close to the wall does that. There's probably a way to not have the hot spot at all.

An object that is off camera is probably not a great render hit. But I think setting the Active property OFF is even easier than making a pose to make something transparent. If your >Constant on the property it wont create any key frames as you turn it on and off as you work.

As the light bounces off the walls it takes on the wall's color. OmniTestEH.mov A typical Cornell box image takes about an hour to render with radiosity, but this takes about 15 seconds per frame to render. This scene also has something in it that technically doesn't exist in A:M. Can anyone guess what it is? There are clues in the render.

I think that's definitely true on the chicken beak. If you enough head nods no one can see mouth shapes anyway. I think you need to add a few frames at the start so chicken's head can wind up into "you" as the anticipation to going down on "stole" You want B, not A