Setting up a thread for the March Cycle which you can see on a path here in the attachment.



I don't know what you mean by this. It's an action so I haven't moved the model bone.

Hi Ken,

I guess I can explain it better....

marchcycle01commentsmp4.mov

you'll let me know if i didn't.

Oh OK....got it now.

Robert.....I've updated the march on svn. See how that bites you.

Just bumping this up in case the last post was missed.

It takes me while to examine these.

But here are the long-awaited comments...

winkiemarch01Commentsmp4.mov


thanks, Ken!

Hi Robert. Good points. However, maybe you can help me with this bit....it's driving me nuts. As you can see in the timeline, I have the x/y values of the hips constant with the z axis varing yet, his hips aren't level in the y axis if you compare them on the model. Why is this? Is there something set that I don't know of?

PS A clue might be when I move the hips down it effects the z axis too and vice versa.

I just took a look at the model. I think the fact that the hip null inthe model itself is not level creates that problem. I guess all translations are relative to the bone's original orientation, when you try move it "forward" you actually have to "raise" it on Y just to keep it from sinking.

??

I dont' know if we can fix that by just fixing it, or if this involves rigger work.

I notice the model seems to be slightly leaning forward; they were probably trying to accomodate that. If we leveled the Hip null then Z rotation onthe null would no longer be aligned with the torso.

Second complication... a version of this model is being textured out of the SVN loop which means any change we make to this one won't get made to that one...

I'm leaning toward leaving the null the way it is and working around it. What do you think?

If the character had a dynamic simulation or hair on him the sudden shift of that null when the cycle repeats might be a problem. Fortunately he doen't have any of that.

Physically, if the bone's origin is being moved only in one axis (using the arrow keys), it still shows the values of two axis being changed. It doesn't seem like that's something to do with the rig. Yet it also happens in other TWO models so that indicates that it is. I'm lost.

You've already tested the hip null so yeah, I'll move other body parts around it.

Just as you mention dynamics....it got me thinking about motion blur. I wonder if that's going to be effected using this method?

The translate directions are based on the orientation of a bone or null...so, the easiest thing for inputting values on the hips is have to that null level.

To level the hip null, set the orientation you want it at in Bones mode (0, 0, 0), then reset the compensate on the "spine_1_stomach" bone in the "Rig_Components/torso/neck/head_constraints_folder/torso/neck/head_constraints" Pose and the "right_thigh_base" and "left_thigh_base" in the "Rig_Components/leg_constraints_folder/leg_constraints" Pose. That should make everything work like it was originally set up.

Hope that helps.

I recall something like this on models that were boned with the feet not pointing directly ahead. But it's been awhile.



The first iteration always seems to get a glitch, but after that it works smoothly. We can work around that.

Click to view attachment

Cool. Thanks guys.

Updated the march cycle again. I think I actually worked out all the stuff in the last video. I'm getting some pausing and snapping in the hips/chest bones when I look at it face on. But I think it's to do with the curves (which are cut off at the end). Not sure exactly how to fix it.

March

Looks good...Onward and forward...

winkiemarch3commentsMP4.mov

thanks!

Yay! Movin', movin', movin'.

I've done a trumpet holding action along with a walk. On the walk, I can only get him moving along the global z axis....don't know how to having him move along his own model bone axis at any global rotation. Anyway, I've uploaded the project and cleaned everything into a separate winki folder in the main actions folder. You should update from the action folder and you'll find a new folder in there located here:

data/Shared Data/Actions/Winki_Cycles/Winki_Cycles.prj

In the project, you'll see I added a bone to the trumpet and constrained that to the head. No problems there. But when I constrained the hand nulls to the trumpet bone, they wouldn't follow it (I think they should). So, I had to add another translate constraint to them connecting them to the head bone. Delete that constraint and you'll see him arms don't follow in the chor.

On a side note, I seem to remember that it's possible to manipulate the model bone in an action (non chor action). I can't remember how exactly, but I wonder if that would allow you to just snap back the model bone once in the chor and not multiple times in the chor. But I've a feeling it might cause other problems....foot slippage and the like.

To translate the model bone in an action, go to the model's properties and modify the transform translate values.

Ah nice one Mark. You have to have the shortcut to the model selected to see the Transform options. I don't know if that would help in this action Robert....

BTW here's the trumpet movie link:

Trumpet

Just to let you know.....I uploaded some new head swinging changes in spite of the constraints working like they are.

Is his right eye wandering around or is just too small to render right?



If we could constrain the hands to the instrument it would be easy to add a small bit of overlap in the side-to-side motion. Maybe what we'll need to do is add a null at the "embouchure" in the Winkie model itself, and constrain both the instrument andthe hands to that.

But when we do the trombone we'll definitely want to be able to attach a hand to the instrument. hmmmm....

His eyes are transparent in real time (on my system) so it looks like it's moving around.

Edit: Problem solved. I re-boned the trumpet and didn't rename it. Now it seems to work ok. I don't know if it's because I didn't rename it or it may be because I hadn't selected the model bone in bones mode before I made it. Too tired to explore it now.
Update the trumpet model and the trumpet action.

Hey Ken, where do I find the new chor you made that has the series of cycles in it?

It's embedded in the proj file located here:

data\Shared Data\Actions\Winki_Cycles

Hey, Ken, the trumpet march cycle looks spiffy!

As an experiment I made the hips go down for one more frame before going back up, to try to make the walk weightier. Not sure yet if it succeeds or not. it's Generic Winkie March 2. I also lagged the toe of the foot more as it comes off the ground to loosen up the appearance of the foot. The difference is quite slight.

Go ahead and make actions for the rest of the instruments. I want to be able to animate the trombone slide, see if you can come up with a pose slider for that. A slider that went from 1 to 7 would be great as there are 7 "positions" on the trombone slide. Look that up on the web for proper placement.

Okeydokily. I changed over to your new action.....I also updated it so that the hips don't have an extra keyframe past the 1 sec mark.

Is this how the drum should look? It doesn't seem correct.

The shoulder harness would need to be scaled down. But instead of that lets make it a huge drum that is on someone's back like in this drawing:

Click to view attachment

Can you make a cycle for the guy carrying that drum? It won't be a high stepping march like everyone else.

All cycles started and uploaded. I was having difficulty keeping the trihorn on the ground. Maybe you have a suggestion about the constraints. Check out the video of them all here:

Band

PS I put a pose on the trombone....couldn't find anything about the placements so I just devided up the length into seven even sections. Every 3% of the pose slider moves to the next section.....all the way to 21%.

That looks great! we definitely need a few more goofy instruments to fill this out. There are still people wanting to model, we might be able to snag a few.

-The bass drum is a bit too far back. Bend the guy carrying it over some more to get it more balanced on his back.
-it also looks like it's waving side to side more than the back of the winkie it's resting on. The bass drum waver should probably be very small since it's such a large object.
- can you add a guy perched on top of the bass drum playing it, like in the drawing? You might want to size him down a little.

-lets move the clarinet down into a more classical playing position



How about if there was a bone on the trihorn the originated where the wheel contacts the ground and stretched all the way up to the mouthpiece. the origin could be animated to stay a certain distance in front of the winkie and the tip could be kinematic constrained to his mouth. I'm not sure if that would be a circular constraint or not...

The drum is constrained to his chest bone so it's not rotating any more than that. It's probably an illusion since it's so large. I'll see if I can counter animate it.
Which on is the "bass drum waver"? The guy on top?

Hmmm.....adding a guy on top is difficult. I can't put him in the same action as AM doesn't allow duplicate models in an action where the model is the main model. If I save a separate winki with a new name it would work though....
Also, how many times should he hit the drum?

I'll try that on the trihorn.

no. waver= amount if waves back and forth



How about sitting him a bit off center, to the rear. He could be hanging on with one hand and holding the beater with the other.




beat, rest, beat, rest, beat, beat, beat, rest

It's possible the reverse might work too. There may be some use of the Translate Limits constraint that would be of use in keeping the wheel on the ground.

Huh? In frames man! I'm not a (good) musician. So you want multiple beats in that second cycle. I'm thinking he hits it every 6 frames.....or not.

I adressed most of the issues. I tried the trihorn as you suggested, but couldn't get it to orient like with the translation limit constraint in place. So, I reverted (and improved abit) the way it was.
I tried the guy on the the drum lying down as he made the drum looked unbalanced sitting perched on the end. Also, it was difficult to get him to hit between his legs. See what you think.
Let me know about the drum beats....

Band

The trihorn player's head seems to have an odd shake to it that the others don't and the trihorn doesn't seem to be staying on his mouth. I thought the previous version was working better



The drum itself still looks like it's too far back. The winkie carrying it needs to bend over more.


here are some notes...

2_01_80Bassdrumpose.ASF

OK so the drum beats go on for more than the regular cycle. I'm not sure how I'd do that....

At the moment I have the drum and beater as action objects in the drum walker action. To do the beats that long, I'd need to have the beater in a separate action which would have the walker as the base object. But then I won't have the drum to put him onto or constrain him to. Ahhhh....head exploding!

Any ideas?

hmmm.....

How about if...

-you deleted the drum and drum player from the winkie-carrying-the drum action. That would leave a winkie crouched over as if he were carrying the drum (24 frames)

-make an action for the drum that puts the drum player on top and animated him (96 frames)

-then in the chor constrain the drum back onto the drum carrier's back and then add the drum player action to the drum

That'll work. I may be able to make the drum constraint in the chor into an action that you can drop onto the drum in the real chor too.

The new scheme should get us back to just one version of the Winkie, right? If we weren't there already.

Yes.

Check out the update:

Band

Somehow the second three beats don't seem fast enough, but they're evenly spread over 2 seconds.

ok... there's 8 counts in 96 frames and they go like this...

beat, rest, beat, rest, beat, beat, beat, rest (repeat)

Are you saying the guy should start with his arm hitting the drum (beat) rather than with his arm in the air (rest?)? I thought I had it similar to how you did it in the video.....but maybe the last hits not as fast. I'm not sure.

He should be playinging in the same tempo that they are marching.

It looks like the first foot contact hits about frame 6 so that would be the first musical beat and he hits the drum there.

The next foot lands about on frame 18 so that would be the next musical beat. but thats also a "rest" so he doesn't hit the drum

the next foot lands about frame 30. that's the next beat and he does hit the drum there.

the next foot lands on frame 42. thats a rest for the drummer.

the next beat,frame 54 is a hit
the next beat,frame 66 is a hit
the next beat,frame 78 is a hit
the next beat, frame 90 is a rest

since this is a looping cycle its possible that some frames before frame 96 may be part of the action leading into the first drum hit on frame 6



Do you recall how drill sargeants call out "left... left... left, right, left..."? That's the rhythm that the bass drum is playing.

I think I got the timing right now. Check it out:

Band

That looks terrific. Nice work Ken.

That's the right rhythm!

now, can you loosen up the action? Lag the drum stick on the way down and lag it on the way up. Also he won't be able to do as big a wind up when he has to strike the drum just one beat after having already struck it. When he's got a beat rest inbetween he can do a big anticipation, but when hit again right away he can't do that. Try the action yourself to feel how that works.

The clarinet players head will face forward like all the others even though the clarinet is angled down.

example:

http://www.flickr.com/photos/beckzaidan/1325895333/

Thanks Dhar!

Robert: Any better?

Band

yes that works better!

-What bone is the bass drum actually constrained to ? It still looks like it's wandering.

-Can you get the clarinet players finger tips on the barrel?

-You can get more reach on the trombone by holding the slide with the 2nd and 3rd fingers (no thumb) instead of a fist. throwing the shoulder forward will help too.

We definitely need a few more percussion instruments. Cymbals woudln't be hard to make but I wonder if there's a snare drum around somewhere already made?

Well, the next thing to do is to lay out a grid of Winkies marching from left to right as we look at the front view.

How about 12 rows of 6 winkies in each row. It look's like you've already figured out how to quickly duplicate the path, but I can explain that if not.

Don't put any instruments on them yet, lets just get them marching.

The chest control bone. Not sure why it's wandering.....I saw it too. I'll have another look.



OK



Are these instruments in the song? Is there a place you would hear the cymbals?



I've just exported the path step motion from the chor of the first one and dropped it on the rest. It's only as long as the animation length above. If there's an easier way let me know. As I said, I don't know how to move them in any other direction other than positive global z axis (with that action).

Do you have a chor you want to do the grid in? Or just start in a frest unlit one. It will be a pain to have to re-do it again for all those characters.

So, is this the idea for the grid? Path is from top to bottom. 72 characters altogether (hope my computer is up to it):

......
......
......
......
......
......
......
......
......
......
......
......

You may have to find the actual geometry bone for the torso. The chest controller doesn't directly control and CPs

Pretty much all through the segment.






I've uploaded a prj for 2_01_80 that has an action that will move the bone inthe X direction. Update your Winkie Cycles folder too.

The chor has one winkie set up with it working for 36 beats or so. His chor action is set to "add" which allows him to be positioned to any starting point.

Do the other Winkies like him and they can all be re-positioned without messing up their cycle.

as you add winkies rename them "Bandrrx" where rr is the row number and x is A B C D or E counting across.

Just set up 3 rows to start and we'll see what spacing looks right.

thanks!