robcat2075

A hobo movie! Ripped from today's headlines! In that tutorial Rodney points you to, there's an "emitter" object that gives off the sprites. Basically you'd put something like that in your trash can. I second Rodney's suggestion that you do that tutorial, it will show you the steps to make such a thing.

If that was with multi pass ON, that would mean no anti-aliasing at all. The regular Final render with multipass OFF has the same anti-aliasing as 16 pass, multipass ON. More passes will mean better anti-aliasing but longer render time. It's possible some pixelization is being introduced by the quicktime codec being set to too much compression, but I don't know what you've chosen there.

yeah, I've always felt that the thing that got me the most result in animating was trying the action myself. And testing a pose out in front of a mirror is big too.

a mere afternoon to make a fine model would be a small price. The good news about the smoothing and respining is you only have to do it to one side, then you can CFA the other.

Sorry, I missed this. I wrote some notes on the frames. Use the cursor keys to step thru them. jump_test_2_BothNotesMP4.mov The second jump works a bit better than the first but the number one problem is still the up and down motion. If it isn't working like this, it's not working: All the other overlapping motions need to be built around that motion.

What Fuchur said. A:M is made to not need polygons at all so it doesn't make substantial accommodation for them. Convert to OBJ, make sure it's quads and not triangles, and give it time. The OBJ importer often looks like it has frozen at "20%", but if you give it time (maybe all night) it will eventually spit something out, RAM permitting.

I sense one of those mournful train wreck songs coming.

Rather than use a model directly from the CD. Copy it to your drive, like to the folder you are going to save your project and the images you are going apply as decals in, and open it from there.

we had this one a few weeks ago... it has something to do with canceling out of a an attempt to find a rotoscope image that is no longer available, like one that was on a removeable memory card. solution: edit in text file and delete the "rotocontainer" tags. http://www.hash.com/forums/index.php?s=&am...st&p=297095 remember to save your new file as .mdl not .txt remember to restart A:M and start a new blank project before you try to reload this edit file.

You are going to have five pointers whenever you have a structure like a finger. this might be cleaner, might not, depends on the contour of the object

Welcome back to A:M! The spline at the bridge of the nose that bulges over the center line is very odd. Probably a sign that it is not properly connected or that it has an unnecessary bias on it's CPs. Without seeing the mesh i'm going to guess the same reason for the crease above the mouth.

It's hard to show "dark" in a render unless there is some "light" to compare it with. Here's simple strategy to start out with. This scene has a strong light ("KeyLight") from above and behind. It's like "the moon". It's slightly blue but it's still 100% intensity. Then add just enough fill light in the front so that some detail can be seen. My one fill light is only at 10% intensity. It is also bluish. Since it is from behind the moonlight leaves the objects mostly dark but puts a bright rim on the edges of the objects to separate them from dark background objects. The Keylight is the only shadow casting light. Even though it is "night", our eyes like to see a full range of values from light to dark. The "moon light" helps do that. I changed the ground from 100% white to middle gray, and made Thom a more pale yellow. I set the ambiance on all the objects to 0%. Rodneys' fog suggestion is a good one too. i added just enough to weaken the horizon line a bit. here's the project night03.zip Lighting gurus will spend much time fine tuning a night scene, adding lots of small lights in places to cheat a few details that the basic lights wont catch, but this a start.

That's starting to work. I think the reflection needs to be broken/subdued quite a bit since the water isn't doing such a mirror-like reflection on anything else.

Select group and go to its properites window. You can transform>scale a group of points to 0% on any axis to align them. You can set the "pivot" location (same window) to any location if they need to end up on a certain x y or Z

I think the notion got started with a post on The Spline Doctors Blog where one of them said he was tired of seeing generic characters on demo reels. Because the generic characters are so widely available there's lots of bad animation done with them and so he was biased against animations with them when they showed up on a demoreel. But that's really a jack-ass stupid thing to say. Follow that notion through to its consequences and it's apparent he wasn't really engaging his brain before he opened his mouth. At the studio level animation is all about using a character that someone besides the animator made, and if you aspire to that level you need to get animating sooner rather than wait until you have made your own character. This is especially true in other programs where modeling and rigging a character is 10x the difficulty that it is in A:M. I suspect the % of Maya owners who ever get a character successfully rigged is microscopic. That probably explains why there are those widely used generic characters circulating.

Of course, don't take me as the last word on jumping. There are as many ways to jump as there are things to jump over and places to jump to and people to do the jumping. This jumping in place is a bit of a rarity in animation. Richard Williams has several pages on jumps in his book (pg 212+) but they are all jumping forward. The concepts are still the same but you have to distill out the elements that are about getting the mass moving forward rather than just up. I imagine there are more story reasons to jump forward than to jump in place so that's probably why he covers it that way. I also meant to mention that videoing yourself dong this jump is a good bit of research. You can compare the way it looks with the way it felt doing it (which I always found most useful of all) and you can count frames to find the timing.

Possibly true, only because a detailed character will need more polishing. But what was her explanation? She isn't really using the word "rig" properly. She's really talking about overused characters. I think. But the overused characters are the minimal ones. "some people" Can she actually name an employer that does that? Probably more urban rumor than actual practice. I've heard "some people" do that , but no one who actually says they do it themselves. Students love to find reasons that they were rejected other than their weak portfolio. Squetchy Sam is minimally modeled but well-rigged.

Will he be performing the song? Or is he more of an interpretive dance?

Here are some things I notice. SteveNotesMP4.mov

I like the design of that. I like the bony plates. I presume they are bony.

yes. odd pivoting. the bone will wander from its original position and back as you go from 0 to 100% enforcement. Consider: - at 0% enforcement a constrained bone makes no attempt to get to the incorrect 0,0,0 offsets, and so stays in its original position -at 100% enforcement the constrained bone will be trying to get all the way to it's full offset position, and so stays in its original position -but at 50% enforcement the constrained bone will be trying to get half way to an offset location that is halfway between the target bone and the correct full offset position. try it and you will immediately see . enforcementoffsets.zip

"Radiosity" is probably the most accurate scheme, although it is very time-intensive to render. note the tutorials and examples in that archived forum. Image based lighting with a high-dynamic-range image with the light source in it might work also. you can get close to the effect of a flourescent light by making an array of point lights: http://www.brilliantisland.com/am/crawl.mov (they're in a single row in this example)

good question! change the "curve" for enforcement from "hold" interpolation to "linear" There's a way to copy and paste values form 100% to 0 but I dont' recall it. You can Edit offset curves manually by sliding the CPs at 0% to match the level at 100% (the red green blue icon at the bottom of the PWS switches it to curve view.)

Welcome back to A:M! Stitch: Hold down the Shift key when you are clicking in new spline CPs. This will make the old splines keep their shape.

It sounds like your hitting the (T)urn view key rather than ®otate key.