robcat2075

You can use an extremely long focal length camera which will look almost the same, but for my depth mapping purposes i found the gradient material pretty much got me what i wanted with an orthogonal camera.

Last I tried, Fog and orthogonal views don't work together.

I forgot... i also raised the middle of the jump up. You can see the bottom of the frame move up briefly to simulate that.

the BIG thing... it's too slow. I retimed it a bit in Aftereffects; something like this would feel more natural: JobeJumpRetimed.mov

Possibly, but note that depth map details are typically very subtle shades of gray. I do know that OpenEXR depth buffer do work correctly and my gradient material technique also works, having used it in the past to make displacement maps.. I'd have to see your non-working example to know more.

You mean A:M objects, right? One way is to take the depth buffer that A:M can render in OpenEXR format and scale/clip that in a paint program to extract teh range you need for your object inthe scene. You can also do that in A:M http://www.hash.com/forums/index.php?showt...p;hl=extracting If the objects you want to convert to grayscale map are all in one model you could apply a white to black gradient material on them and render that inthe model window.

A few things about walking.... - we almost always have the toes pointed out when we stand or walk. The straight-forward-feet looks very Frankenstein-ish. (I noticed Daniel Day Lewis was dong something like that for "Lincoln" ) - our feet tend to land much closer to a center line than we imagine. On men the heels will just barely be missing each other as they pass and sexy women will do a catwalk where the feet may even be crossing in front of each other. There are some very wide, stiff people who have to lurch from side to side to crane their legs around but it's a very awkward look. - the slower the walk, the longer both feet are touching the ground. The rear foot will still be pushing off with the toe on the ground after the front foot has landed on the heel.

I would honestly have to watch you do it. One of the things you have described doesn't belong and I'm not sure which one it is without seeing what you are looking at. Either the Chor window is open or it is not. If you don't have the "Workbook" on you should turn it on that makes it easier to bring back background windows.

in the chor window>>View>Camera1 In Tools>Customize you can also turn on the tool panel that has view buttons on it.

Press 1 in the number pad to get the camera view back

You mean you deleted a camera from the chor? you can drag a camera in the Objects folder into a Chor to add it back. If you mean you just closed the chor you can double click on the chor to reopen it.

how about a bit more of a V-torso to match the large arms?

The old legs were a bit on the thin side.

I though of an experimental way to coordinate time in the air and height of jump. Suppose you want him in the air for 12 frames. Get out a video camera, a camera phone will do, and shoot yourself dropping a ball from various heights and count frames until you find one that takes half that time to reach the ground. The time to jump up is always the same as the time to fall back down, so whatever height it took a ball to fall in 6 frames will be the height Thom needs to rise to in the middle to be in the air for 12 frames.

Hold on ... I think it's an update problem, for me anyway. Try this. Make the constraint with no offsets and "Translate Only" OFF. After you choose the Aim Target Bone, something goes wrong and some offsets are created. Set all the Translate and rotation offsets in the Constraint Properties to 0 then set X rotation offset to -90 that makes it work as advertised in v17

Good news and bad news.... I got it to mostly work... in v13 My test in v15 and v17 looks like it's at least partly broken. I'll look at this some more. It's definitely not working properly in v17 for me.

b was better. The essential problem i see is that he's barely jumping up at all. There's no way he could stay in the air that long with having achieved that little altitude.

Also make sure the "compensate mode" button is OFF when you select the target.

I haven't investigated the constraint problem, but a possible alternate solution to your pupil placement problem is to use bitmap images of them in Projection Map materials, whose placement can be animated. see this thread: http://www.hash.com/forums/index.php?s=&am...st&p=313694

also... if you are getting stair stepping from the 8-bit maps, OpenEXR maps work too. Harder to find a paint program for them however.

That's quite amusing! I like that.

I like those jumps. The arc ( more visible on the yellow guy) they go around the first squash pose is a bit unusual. it looks like A but B would better get their mass in front of the legs that have to push them up and forward. They both look to be overshooting their landing. The yellow guy works better because he has that foot out in front to catch him. If you reversed that so the left foot was the one out in front it would be clearer from this angle. The orange guy seems to be bringing his right foot forward to catch himself after landing but I think it's too small and too late to be convincing.

Hey Lloyd, If you are wanting something to do right now.... that squetchy bouncing ball that bounces off a wall would be good to try. (hint: you can turn the ball any direction you need)

That is a great tip.... I just figured it was some sort of limitation with using zbuffered shadows that wasnt ever going to be fixed!! thanks. In v17 the ground has this default set properly along with some other things... http://www.hash.com/forums/index.php?showt...ground++default

Cool tests! I'll note that you can make the shadow on the ground work right if you set the Ground to "Cast Shadows" ON