(double – edits below)
M.E.L (& others),
When you have time, there’s a few things that I’d like you to take a look at (and I’ve got a couple questions of a TD nature).
- I ran through some quick blocking with the existing rig for one of the scenes, and I find myself wanting to arch the top edges of Orange like eyebrows occasionally when he’s looking at stuff (curious) or thinking (furrowed brows). I can get away with using the arrow controls for quick accents, but I can’t push those very far before they look awkward – basically I can’t get the punch I want for, say, a head tilt and quizzical look. Blendshapes might be the best solution for this one, because we’re talking a ton of clusters otherwise, I suspect.
Is it cool for me to just sculpt & add one-off blendshapes as I need them to my scene files? (there’s a lot of splat gags in my section in particular) Is this going to mess up the pipeline?
The ‘hand/eyebrow’ arrow controls feel a little weird because they don’t follow the top ring.
The scaling of the final room file and the rig are off. The room is approximately a factor of 10 times bigger than it should be in relation to the R&D rig.
I’ve been assigning temp lambert shaders to the room file and orange as it helps me get into the scene and composition a little better (they’re all flat gray). Is this going to throw anyone off down the pipe?
I’ve been temping in camera framing, movement and lens stuff (to exaggerate distance, for example) while I’ve been running tests. I’m using the 3-node camera and just naming them scene22cam etc for the benefit of the camera and rendering peeps down the pipe. Should I be doing this differently?
Hmm…i can take a look at adding more influence to the top clusters if you’d like.
Should be just fine, remember to just dupe the actual character geo and apply the blendshape (dupe and delete history to be safe).
Ahh…easy fix there, I will reparent those tonight
No worries, that rig is totally scaleable…i have a globalScale node in there just for that.
Shouldn’t be an issue there, the animation stage really doesn’t need any crazy fancy shader stuff, once that goes to rendering and texturing is where they handle all that. No need to clutter the scenes at this point.
Hard to say right now, the cameras will need to stay fairly consistent and a set naming convention will have to be in order for when we render (say if we launch renders via script we can just batch stuff in and someone like myself can easily just go through a list of shots with camera names to throw into the batch).
I completly agree
yea by the way If you are too busy with your animations I can make some blendshapes for you. Now those which will be in the rig are just basic, simple, almost all of them made with nonlinear deformers blens shapes! so if you need something just post here and I´ll check what I can do for you!
and feel free to make blendShapes by yourself if ya want I am not offended
Wow, super quick responses!
Thanks to both of you guys for being so on top of this and helpful.
We should probably start hammering out naming conventions, camera specs and other potential pipeline issues pretty soon, so that animators can start adhering to them from the start instead of retrofitting stuff.
I’ve got my scenes set to 1.85:1 (640x355 blasts, if I recall), I’m using a 3-node (camera, aim, up) camera, and I’m varying between 22mm (x-wide) and 56mm (POV) with my lenses for initial composition & blocking purposes. The camera is called scene_<boardnumber>_camera, although there are some instances where a camera flows through two or more boards.
(Edit: We’re continuing this discussion over in Damnation Alley, the TD forum.)
what comes to naming conventions as far as those are clear and not too complex, and everyone is following those strictly! it really dont matter atleast for me how its done!
and so the heavens parted and there was light!
Any word on the ETA of the updated rig? I was hoping to get in some serious work this weekend, and there’s only so far I can go with the current one
(Trying not to be a pest)
As it stands… my post above was the farthers I got.
The ear controller issue hasn’t been dealt with yet mainly because the fact that I am using clusters on it and the parenting will cause double transforms (can’t parent constrain, orient or connect…setting cluster to non relative will cause it to convert back to worldspace and rocket off). Basically, it may just have to be worked with.
The scale issue, this is a HUGE problem because if the room is 10x the size of our character, we have a serious dilemma. I was not made aware that our room would be 10x the size! Again, the big deal here is that now by scaling our clusters, we now lose the transformation matrice that they were created with (scale value of 1 as compared to 10 which now distorts and lessens the alotted transforms).
As for the shapes, totally make whatever you need to…just fire us a copy of the stuff so we can keep tabs on it all
Last note, I am super extremely insanely unbelievably busy right now…it is beyond with work and other things. I seriously hope that the rig will be able to pass through the animators blocking and stuff, as it stands right now…I have a milestone delivery to hit at work in less than 2 weeks now and there’s way more than I thought needed to be done! Feel free to email me, PM me or ask questions here…I’ll try my absolute best to answer them but no guarantees.
the room was imported from max and current scale is because of that … so I would just scale down that room.
and MEL if you´r too busy and If you think that I can help you with something just say so… and Ill check If I can
M.E.L… understand completely about getting slammed at work. No worries… the rig as-is is workable for blocking.
I just scaled the room down to taste with the existing files (.1 in xyz), and I’ve translated Orange’s rings a little in Y to make him slightly taller (like in the storyboards and pre-vis).
I’ll just keep with the blocking of my scenes, and avoid polishy stuff like working with the blends and clusters until the rig is updated then.
how do I duplicate the cube? I need to have like 10 grey cubes in my shot but i cant duplicate the cube no matter what i have tried.
Also the cube is very small in comparison to the room so I had to eyeball scale it. none of the animators file will match.
To duplicate the cube, just create more references to it in your master scene file… you could duplicate the whole hierarchy with upstream connections as well, but that would work out rather messy further down the pipe.
Secondly, the room was a factor of ten times larger than it should have been based on the storyboards. I referenced it in my scene and then just scaled it to .1 in X,Y and Z and it’s working just fine.
Let me know if you’ve got any questions about using the rig… I’ve been pushing it pretty hard for the last week or so.
If you need to dupe the cube, just unlock it in the Layer Editor (it’s set to Reference) and dupe it that way
Thanx for the tip MEL!
Big problem… I
ve got the positions down that I needed but now Im putting in blendshapes in my blocking but Orange’s geometry is blending from his rig to 0,0,0… any ideas?
The blendshapes don’t currently work in the rig, and it’s on the list for fixes in the next rig revision.
In the meantime, I’d suggest animating the rings to start with your squash & stretch, and concentrate more on the overall body language of the characters than their blends.
It may take a couple of days to get used to it, but you can get a really good performance out of the rig once you’ve gotten comfortable with it. It was designed to be pushed really, really hard.
having been mainly doing forward motion I now need to rotate Orange in Z… and its all rotating wrong… when I rotate in Z in rotates in X aswell… unless I`m supposed to be setting keys for rotating on a different node other than the translate one?