robcat2075
-
Posts
27,754 -
Joined
-
Last visited
-
Days Won
341
Content Type
Profiles
Forums
Events
Posts posted by robcat2075
-
-
the third quarter at AnimationMentor is... acting!
Well, such as it is...my first dialog test:
You're right Mr. Thatcher... (QT 500Kb)
-
I'm trying to think of circumstances in which you'd really want to overlap efforts like modeling, animation, lighting... within one shot.
-
Hi all I am Jof!
Hi Jof!
Can anyone advise on how to change bone Keyframe Interpolation eg., to Linear 'in one go' instead of having to go to one bone channel after the other and changing each one's interpolation one at a time?no, but group selecting all the curves will allow you to set the default to linear in one move as you noted:
Normally what I do is to go to the bone channel drag a selection around the bones and call up the menu to change the Interpolation Method to whatever I wantI presume you are bringing up all the curves at once by clicking on the "Bones" folder, rather than selecting one bone at at time. Then right-click>Curve>Interpolation Method>Linear
If you just do right-click>Interpolation Method>Linear you will only change the currently selected keyframes.
-
I can see the advantage of being able to play a series of shots in one chor.
Since it's possible to animate lights and whatever, it would be possible to use one set for a sequence of shots.
Lighting would still need to be adjusted ("animated") on a per-shot basis none-the-less.
However, I'm not convinced that simple text merging of files would result in animation by different animators coexisting well inthe same chor.
-
The character looks great.
I like the run.
It's a bit hard to judge the jumps because of the viewing angle. If this were a "jump" assignment I'd say show it to us from the side.
However she seems to hang unnaturally in the air at about 03:00. That may be because of the camera tracking, but since there's hardly any background it looks like she's floating.
yeah, the handstand bounce looks too light.
On the last landing she hits the compression pose and freezes. Keep that moving. A "moving hold"
-
If you load it in to Quicktime you can single frame it.Could you post in .mov format so we can step through the frames?It's hard to crit it on a per clip basis because each one has things that are going well and then moments where they are not.
Quickly single framing it, it appears many of the jumping/tossing motions may not be well arc'ed. FOr example the two legged sphere on teh big jump travels straight (well, now that I've tracked it with a dry erase, it actually is arcing upwards a bit) from 08:05-08:10 then right angles down to the wall from 08:11-08:15. I'd defintitely round that out and make sure his horizontal velocity is constant between foot contacts.
Also the contacting foot at 08:12 stay rigid, I would expect the heel to give some to indicate the weight of that action.
I guess all of the issues in the silent clips would be grouped under weight/inertia
On the last clip, the frozen spines really stick out to me. (that's my big snit these days... frozen spines
) I'd start the boy out in a less symmetrical pose away from Harriet so he can move even closer to her to deliver his line.
Those huge heads really need anticipations for their moves, and compensating motion on the bodies.
I like the fact you've packed alot of stuff into only 33 seconds. I think that's a good strategy.
-
I'm told there's a new version of the face rig up somewhere. Not yet?
-
wow, nice easy chair. I hope you're going to add the palm fronds on the back.
-
The spine would only stretch if you had "stretchy spine" turned on. With it off the spine would always maintain it's original length, no matter how you curved it.I will say that the one major difference is that when you translate the upper chest and lower hips bones, you get an automatic stretch, as opposed to spine movement/rotation. This took some getting used to and could be limiting for non-stretchy characters.Actually, now that I re-read this I'm not sure if Zach was refering to TSM or Squetchy Thom. Which was it? I was referring to TSM.
-
I believe King Bal needs to be tied down to his throne. I know they could tie ropes to the existing structure, but some dedicated cleats would be a humorous touch, (easier to animate than threading ropes thru the structure).
-
I like the rotoscope solution for labeling the face controls.
some notes...
-I know the face cam has a bone listing inthe model, but turning that off doesn't hide the camera inthe chor. I'd prefer not to see it and it's scope boxe orbiting the character as he's animated.
-Since the null controls are all 2D, can they not be on the same plane as the rotoscope. Or alternatively, make the rotoscope visible only in the face cam?
-If the rotoscope were to stay visible in the main cam view, I'm presuming there is a way to turn it on and off?
-As far as I can tell, there's no separate left and right control for any of the eye features. Is that planned?
-I'm doubtful about the automated eyelids. Can there be an option to disable that?
-The way the eyelids are pulled left and right with the eyeballs seems rather odd. It probably wouldn't be noticed on an untextured character, but if one had eyelashes, for example, they'd be sweeping back and forth with every eye movement.
thanks for the good work, Mark!
-
Yes, although it's not isolation of individual segments, it's independence of the hips and torso. The rest of the spine is interpolated between them.Robert, are you asking for spine segment isolation à la the TSM rigs?
The spine would only stretch if you had "stretchy spine" turned on. With it off the spine would always maintain it's original length, no matter how you curved it.I will say that the one major difference is that when you translate the upper chest and lower hips bones, you get an automatic stretch, as opposed to spine movement/rotation. This took some getting used to and could be limiting for non-stretchy characters.
-
OK, now I remember who Dascurf is.
Well, I'm not sure I followed it.
It seemed to take a long time to get to where it was going. I think you can do just about anything for five minute and the audience will go along with it, but for a feature I think you'd need to draw them in sooner.
I don't think it's "Sick and Twisted" though, it's more "east european".
The compositing does look better in black and white.
-
Some judiciously applied distortion boxes might help in inflating/deflating separate body parts.
But I agree, there isn't going to be a good-looking simple solution.
-
As i look at that wizard of Oz clip, I think that one of the key elements is the dust kicked up where the funnel contacts the ground.
I'm don't know how big our TWO funnel is intended to be. Person sized... bug sized... monster sized? The amount of dust would depend on the scale of it.
-
Those look real cute.
(I can't figure out what "clap" is about.)
-
A couple notes:
-as far as I can tell, the the z-axis on the finger controllers serves no purpose. It doesn't get me to any pose that the x and y axi didn't. Why not incorporate the finger curling functionality into that axis and eliminate the separate finger curling bone. this would reduce PWS clutter.
- I second Zach's comment aout the naming convention. right_hand_pinky would be a useful order I think. That would be a genuine usability plus for this rig.
- Is there a scheme to the difference between a bone labeled "control" and one labeled "controller" and one without either. If there isn't, eliminating that one word would help PWS readibilty also.
-Heel raising still doesn't leave the toe in place. I really think it's too far forward also.
-I would prefer to see much smaller control bones. they really shouldn't stick out of the character in most instances. From a character pose judging standpoint it's distracting... it's altering the apparent silhouette of the character. I realize they dont' render like that... but we dont' animate in rendered mode.
-I have seen shoulder controls that operate like these in that they arc up for a ways and then sort of extend upwards, but i'm not sure I see the anatomical premise for it. Is there a reason for this convention?My shoulders don't do that; I can hunch them all the way up to my neck.
- It's not an issue on a character like Thom who has only one spine bone, but I"m concerned about how this torso control will translate to characters with more. I don't see the scarecrow doing well with one spine bone. In particular I'd like to see an IK spine; they make good posing way easier. I think FK spines are responsible for a lot of very stiff CG characters .
- Is there a document that explains the proper use of the squetch controls? I haven't had much luck with them yet
Ok, that was more than a couple.
Thanks for your hard work David (and everyone else), on this rig!
-
-
Just for fun... this page has a deleted/test scene from "The Wizard of Oz" showing the development of the tornado effect.
Pretty fierce looking for a miniature!
-
A complex rig, I'm not familiar with it yet, but here are four things i notice in my initial look...
-when lifting the heel (IK foot), the toe doesn't appear to stay put. Part of it may be because the origin of that controller is rather unanatomically too far forward.
-moving the shoulder bones seems to stretch the skin between the elbow and shoulder but doesn't seem to take the arm with it in a natural fashion
-rotating the hips on Y causes the IK feet to roll. The IK feet really should stay put.
-When I rotate the hand on Z the mesh twisting is all happening above the elbow, rather than inthe forearm where it normally occurs.
-
That was cute. I wanted to frame thru it to diagnose some things but I dont have a player that can single frame a divx avi.
The best practice for this sort of test is to do it in quicktime and with a regular quicktime codec. That would be Sorenson 3. the slight additional compression of a 3rd party codec is lost if someone can't actually watch it. I do have divx installed, but divx quicktimes still don't play.
For your next test.
-
What I'd like to know is whether the animators would prefer an on-face setup or an off-face interface, and do you want the Osipa-style rig? Any opinions, animators?
Very impressive demo video.
I would much prefer pose sliders (because they are labeled and always in a predictable place) or at least a set of pose sliders that can control these nulls and the option to hide the nulls while doing so. Once these nulls start getting moved around, and the head gets turned, quickly identifying the one I want will get confusing. picking the wrong one and moving it will creat a key I don't want and would have to remember to delete.
Eye targets would be the one exception, on screen for those. We'll need separate eye targets for left and right, parented to a common eye target.
-
I've been looking into using an expression to decrease amount that the wave material displaces the water plane as the water plane recedes in the z-axis. I can't quite figure it out... any ideas? or better alternatives to create that flat horizon?
it won't be enough for the waves to reduce their height inthe distance, they would need to scale in all dimensions.
One trick I saw in an old 3D book was to make the plane hi density in the foreground and low density farther out. Use hooks to reduce the density of your mesh. the last few patches could be stretched WAY out to the horizon.
-
First thing I notice... the horizon isn't flat. Unless we're looking at the top of a huge wave and not the horizon, the horizon will be flat because it's so far away that the height of any waves on it will be negligible.
But the water looks great!
"You're right Mr. Thatcher..."
in Work In Progress / Sweatbox
Posted
Thanks everyone!
Alas, there are things wrong with it. Perfection is elusive, but "it will have to do" seems to arrive every week.For me, the biggest irritants are the interaction between the hips/torso/head. The line between moving hold and float gets crossed several times. I'm still debugging that.
"Yes" and "sort of". I got a hold of an .OBJ someone had exported from the AnimationMentor version and fortunately it was rather well laid out. With a lot of judicious spline deleteing and re-connecting I got to this A:M model. Many fan bones for joints, CP weighting for the mouth, then rigged in TSM2.
You shoulda heard the howls from some of the students when they got this "lipsynch" character that had no mouthshapes, just an up and down control for the jaw. But the teaching strategy seems to be to minimize the whole mouth thing and stress the body language.
That is looking at Mr. Thatcher, but not clear I guess.
well, I did do the mouth on the paperboy shots, but that wasn't really "synch". Jeff Lee put in the mouth movements on everything else I did.BTW, if anyone hasn't seen ZachBG's Duck Sauce, the movie made in a weekend... you really oughta!