robcat2075

I just tried it with a decal and i got visible stuff on the baked bump map. Something else is wrong.

you did have some bump maps applied before you baked, right? Or a bump material?

Your i7 CPU has the "SSE4" extensions that mine does not and that probably is a bigger advantage than having 64-bit.

I just tried it and I did indeed get a shadow image in the alpha buffer. Possibly your After Effects was defaulting to "ignore" the alpha channel. A simple test is to load the image into Photoshop and see if the alpha channel has anything in it.

I think the relevant values are from .1 to 1.0 Try those on a simple case and see what you get. 1.0 wont' necessarily give the same size bitmap with different models.

On to three!

You need a booth babe to sit at your table and sell DVDs.

I have an Intel Q6600 which is a generation or so earlier than your i7 CPU It has four cores at 2.4 GHz. NetRender takes about 5:16 per frame to do the benchmark scene.

TGA will also work. The shadow info will be in the alpha buffer. That will reveal the portion of the otherwise black RGB image to make a shadow.

"Add-lock" is obsolete. The Current "Add" mode does what Add-lock used to... allow you to create more than one CP without having to repeatedly press the Add mode button. The Add-lock button can still be found inthe Customize window but i find it non-functional.

I would suggest making the clouds opaque so the arrangement of the elements is more clear.

The hyper threading is the culprit. The two virtual cores get you some extra overall performance but not as much as two real cores would and they have the drawback of occasionally getting in the way of the real cores, slowing everything down. In most consumer apps that would never be noticed, but A:M saws away continuously for many minutes at a time which is different than most apps. You might try a test with hyperthreading disabled ( and two nodes running) and see if the result is much different. On my four-core, no hyperthreading machine the render nodes are almost as fast as A:M. They do have some extra overhead in that they need to completely reload all assets for every frame (although that is not a significant issue inthe benchmark scene)

It's one of those things where splines are opposite of polygons. I've used your PRJ to make an AMReport on this since it made such a clear example of the problem.

hmmm... the bug seems to still exist in v17. None-the-less, a simpler grid solves the problem... Cookie_CutSimple.prj (use the same decal as above)

You're still on v15, I recall? I think there was a bug that made alpha channels malfunction if there was no actual image in the confines of the patch. If you use a simple 2x2 grid your decal should work.

If you have a 64-bit PC you can have 32 and 64 installed and use either as needed with one license, but you will need separate Mac and PC licenses. 64 bit does render faster unless you have some situation where 64 bit just doesn't run right on your PC as mentioned above. 64 bit doesn't have Quicktime. You render to something else and convert it later if you need QuickTime. There are a few old plugins not available for 64-bit and a few more not available for Mac.

You can select the CP and the >edit CP weights to see exact percentages. In modern version of A:M the pop-up display will also show that info Approximately so. It's only a rough indication. It will be divied (DIV-eed) up, based on distance again. They retain their default weight, usually to teh model bone. I'm not sure I yet have a good description of how the fall-off envelopes operate. It is not infinite. Cubic has a bell curve influence that falls off slowly in the center, more rapidly half-way out, then slowly again.. Linear is straight falloff by distance. This thread has some demonstrations of different ways of attaching CPs to bones, including falloffs... http://www.hash.com/forums/index.php?showt...amp;hl=falloffs

That's a brush?!? I thought it was a funny cigar. Yeah, the circle will go away.

The falloff numbers are something you leave or set manually and not affected by "Computer Weights" You can see the "envelope" the falloff numbers create by selecting that bone. That gives you a visual approximation of how much a bone will grab of a CP when "Compute weights" is done. "Remove all weights" eliminates all sharing of CPs between bones and leaves each CP assigned 100% to one (the nearest?) bone. If you really want eradicate all CP assignments you can select the black model bone and assign all CPs back to that in swoop.

If you set it as a "cookie Cutter" decal the alpha channel in the decal will also create transparency in the model.

I gave up on trying to salvage a screencapture and redrew it king-size. Then I decided the CRT monitor looked a bit too 80's and made it just a "flat panel" instead.

An image sequence has a frame property that can be keyframed and post extrapolated. If that really is absent from a Layer, remember you can always decal an image to a simple flat plane model that should have all the normal properties available. I looked at the clip but I'm not sure what is "the frame"

Well that worked... on the fonts.

Yes it's off.

I'm wanting to capture the A:M desktop icons but Windows blurs them, even the smallest versions. Is there a way to get Windows to not try to anti-alias the icons?