Why do action objects boomerang?

[Pitcher]

Recently, I was trying to animate bubbles coming from my scuba divers' regulators. I constructed about 50 bubbles and put them inside my diver in an action file. I then tried to move different sized bubbles from inside the diver to just outside the regulator in groups of 12 or so bubbles that I would arrange in some sort of random formation. I then would try to move the group (mixing them up and changing the arrangement) every 10 frames further away from the diver. When I got them about two steps away, they started trying to come back to the diver. I would advance 1 frame and about 1/3 of the bubbles started shooting great distances back toward the diver. I had to chase those bubbles down and move them pretty much frame by frame away from the diver because they kept trying to return. What causes this? Is there some limit in play? Is there some way to press a button and let the bubbles go where I want them to go? Also, in the action file, I tried to make the bubbles leave away from the diver to the right and rear and toward the surface of the water. When I put the action in the choreography file, the bubbles went to the front and left toward the surface. I fought the battle of the bubbles for about two or three days and never won. I have some bubbles, but they do not always go where I want them. Does anyone know why this happens and how to solve the problem? This is not a life or death issue with me, but I thought someone else might have problems with action objects like this also.

[Simon Edmondson]

John

I'm far from expert in these matters but, did you constrain the bubbles to a path?

If not it sounds as though you may have the function curves that govern the motion, moving it around. The way to check that would be to click on the graph option in the PWS. They can be adjusted if needed.

 

If I understand the requirement , what you are after is bubbles arising from the diver ?

A possible way to achieve that might be to model the bubbles as a group, constrain them to a path, scale them as they travel along the path and get the actiuon to repeat using interpolation method..

If you constrain the path to the diver with a translate constraint you should be able to move the diver around with the paths and the bubbles arising as the diver moves ?

If not, I'm sure someone more knowledgeable than me will help.

regards

simon

[Pitcher]

Thanks, Simon. Your suggestion sounds like a good one, and you seem to know quite a bit about this program. I will have to try your idea.

[robcat2075]

It sounds like an interesting problem but i don't understand enough from the description.

 

If you could make an example PRJ that has the problem I could look at that.

 

However... if bubbles are what you want... particle sprites using a bubble image would probably do well for you.

[Pitcher]

Probably, I should do something other than what I am doing to make my bubbles emit from the divers and go where I want them to go. I don't know much about features like particle sprites, and probably I need to learn more about them. Maybe there is a tutorial on that. Simon also had an interesting approach, and when I have enough patience to work with bubbles again, I'll have to try that.

 

I mainly have done lessons from the little book that comes with the program. I also have read quite a bit from the book by David Rogers. Probably, if I had taken a course or several and had progressed through a sequence of lessons designed to teach not only the basics, but some of the fine points, I would not even have gone down the many deadends I have explored.

 

I spent three frustrating days trying to shepherd the bubbles across the screen, and that worked to a point, but about twenty or thirty frames after their departure from the diver about a third of them on their own decided to high tail it back to the diver. I then had to pick them up one by one, frame by frame and try to force them to go away from the diver until the end of the action file. (The action file was really created to animate the diver. The bubbles were an afterthought. I had tried the same procedure with tears coming out of eyes in a different scene, and I had no problem with this, but the tears just rolled down the cheeks and fell into another character. They did not have to travel far from the crying character.)

 

What I keep running into, it seems to me (in all my ignorance), is that there are limits or programming that expects we will use the program in certain ways. I keep trying to use my very limited knowledge to do things in ways that I probably shouldn't. I either need to learn how to use these other techniques or learn how to turn off the limits that keep forcing characters to act in certain ways. Here, I'm referring to limits in the skeleton on the elbows, forearms, etc. that keep characters from doing simple things like raising their hand from a position hanging at the side with the palm facing the rear of the body to a position above the head facing away from the body. The arm twists all sorts of ways to keep from doing that. I had to turn off the limits to even approach it. When I wanted an airplane to do a loop-t-loop on a path I moved into the sky and turned sideways, I discovered that the plane would only go to a certain degree of turn, and then would not go the rest of the way. I finally got it to do something like a loop-t-loop, but it was a huge fight. There seems to be an angle limit on how much a character can turn. Someone suggested that I pivot the plane, but that is not a loop-t-loop. Perhaps, I need to place a bone in the middle of the loop and pivot the plane at a distance or something like that.

 

Thanks, Robert, for responding. I will try to learn more about particle sprites using a bubble image and how they can be guided. I just need to learn more about the program. If you want to see "kind of" what I was trying to do, you might look at the clip at: http://www.screencast.com/users/Pitcher123/folders/My%20Library/media/149fc9d6-a1bb-4e7d-9f53-0e3fd423d50c The divers are on a path. There is a diver action file. The bubbles are in the divers in the action file, and every 10 frames or so, I was going to have the bubble come out of the divers at their regulators and then move in groups up, to the rear and right of each diver. I had not thought of having a path for the bubbles and constraining the bubbles to the path and constraining the path to the moving diver, which seems to be close to what Simon is suggesting. I'm not sure what the interpolation method is. I'll have to research that.

[Rodney]

I hesitated to respond because there are two basic ways to respond to your situation.

The first would be to solve your current issue. The second would be to suggest an alternative approach.

Both of those have been discussed here.

 

Briefly, if following your current course of using models as bubbles it might help to think in terms of a starting point, a middle and an end.

The end is why your bubbles are returning to the diver. Something... perhaps a keyframe... it set to make them animate that way.

So my suggestion there would be to first animate the starting position... then set the end position and with those in place work out the performance (this is the classic animation approach... two key poses... a breakdown... inbetweens) I think this would work out nicely but I perceive you'll want to more fully investigate the Timeline and Keyframes.

 

The other methods would be (as suggested) particles (spriticles... blobbies... flocking). That last one would allow you to use a model.

The standard approach will be to use particles sprites (spriticles).

 

Probably, if I had taken a course or several and had progressed through a sequence of lessons designed to teach not only the basics, but some of the fine points, I would not even have gone down the many deadends I have explored.

 

 

This is one of the reason I wish more A:M Users would throw caution to the wind and start posting here in the forum.

(I wasted the first few years of using A:M trying to go it alone... in isolation,,, so I can sympathize)

But we hope that everyone doesn't have to reinvent the wheel.

We've got A:M... we want to create cool stuff with it... and yet we often don't take advantage of the best information available.

We wont find everything we need in a book.. we won't find everything in a tutorial... we won't even find everything in a class or course on modeling/rigging/animation... although supplementing our learning those resources certainly can certainly help us along the way. We will find the information we need by interacting with others of similar interest and... barring the ability to meet others locally in person... this forum is the place to be. We won't always know the 'best' answers (only you can determine that) but working together most problems can soon be relegated to history.

[robcat2075]

You're doing something wrong that is making this 100x more difficult that it needs to be but words are not telling me what it is.

 

If you could send me the PRJ i would look at it. There absolutely is a reason for what ever is happening but I can't figure it out without seeing it.

 

 

On the bubbles... here's what sprite bubbles might look like:

 

bubbles.mov

 

That's got to be easier than animating bubbles by hand.

 

 

Here's the PRJ that makes that.

 

bubblesdemo03.zip

 

 

[Pitcher]

robcat2075, I attempted to send you the embed all project file again, this time with the bubble action project, and once again when I sent the message, I got an error message that said you were not receiving any more messages and that the message could not be sent. Maybe, I'm going to the wrong place to send my message. This time, I looked in the members area and found you. I clicked under the send me a message area. I wrote a message and clicked to use the full editor so I could attach the project file. I browsed and attached. Then I sent and received the error message that the message was not sent because you are not receiving any more messages. Is there something I am doing wrong here?

[Pitcher]

Thanks, Rodney, for the suggestion of starting with the ends and filling in the middle. I do this frequently with familiar types of animation, but when I am trying something new, I often forget. Usually, I don't know exactly where I plan to end up and how many steps it will take to get there. I try to feel my way along and see what it looks like as I go.

 

I'm really going to have to investigate these particles (spriticles) you and robcat have mentioned. I will checkout the zip file robcat sent asap.

 

I agree that the Forum holds lots of promise. It is lonely and frustrating, howling in the wilderness.

[NancyGormezano]

I constructed about 50 bubbles and put them inside my diver in an action file.

It is hard to know exactly what is happening, but I am going to hazard a guess as to what the problem might be.

 

I believe the problem is that you are trying to animate these bubbles in an action (as action objects?). They will follow along with the diver, within the duration of the action (the action has a start and end frame within the chor).

 

BUT, when the action ends in the chor, the positions of the bubbles are probably being reset to wherever they were before the action begins. It requires you do some "expert" keyframing to keep them under control.

 

But as Robcat suggested, sprites are definitely the better choice, rather than modeling each bubble.

 

However, if you animate the sprites (also as an action object) within the action, rather than just in the chor (constrained to the diver), then you will probably run into the same problem as to having to fight the sprites position, when the action ends.

 

Using actions and action objects is trickier than it should be. Actions should mainly be used for repetitive motions (like walk cycles). It is more difficult to smoothly transition sometimes from actions to choreography actions, unless one understands how to keyframe the transitions, as well as the settings for actions (start, end, duration, repeat count, precedence order and overlap of actions & chor actions).

[Pitcher]

Runs With Scissors, I think you might have the answer as to why they are trying to come back to the diver if that is a characteristic of all action objects. It would be nice to have action objects that could depart in a prescribed way and never show up again until they magically reappear on the inside of the diver as the action begins again in the chor. I thought that was possible. As I said above, I am not sure where the limitations are in the program. Thanks for the input!

[robcat2075]

Pitcher, I guess the test PM i sent you never arrived?

[Pitcher]

It arrived and I responded. You did not get the response? My response was, "I hope you get this response." As directed, I did not send the file as an attachment.

[robcat2075]

It arrived and I responded. You did not get the response? My response was, "I hope you get this response." As directed, I did not send the file as an attachment.

 

Nope, nothing arrived here. Another mystery!

[robcat2075]

John, I'm looking at the PRJ. It has three "Bubble" models in it but the action is still calling to load external Bubble models.

 

did you do "Embed All" before you saved this version?

[Pitcher]

robcat2075, I sent you the version that has the bubbles. I'm not sure what happened the first time. I also played the file with the bubbles emitter you made available. As I said in our messages, the file was a real eye opener. You and Rodney are 100% right! I should be using particles (spriticles). When I did Exercise 16 the first time in the book, I got through it finally, but it was very frustrating for me. I missed the whole point and thought, "Now I have a fire and smoke. This could be handy." Thanks for showing me how important particles and sprites can be.

[robcat2075]

I looked a the PRJ. I'm not sure I see a boomerang effect, the bubbles seem to be going where they are keyed to go.

 

Here is one bubble that does return but it is keyed to do that. (see screen capture)

 

The red channel records the horizontal movement. It starts near zero at the beginning (circled as 1) then moves quite a bit to the right (a larger x value, at key 2) then returns to near zero again because key 3 is near zero.

 

Perhaps you unintentionally made that last key that is dragging the bubble back to its origin? If you didn't know that key was there then it would seem that the bubble was trying to return for no reason.

 

 

 

[Pitcher]

Thanks for responding. I had stopped looking here for a response and thought that I would get a return message.

 

I was not aware of the keys you spoke of or how to analyze them. I will have to look into that. At the peak of the boomerang motion, there were about 30% of the bubbles trying to return, but I kept fighting with them, and by the time I sent the file to you, I thought I had gathered every bubble I could find and sent it out of sight. I didn't understand how you planned to analyze what had happened and thought you just needed to know what I was trying to do. I guess part of the difficulty with analyzing the file is that it is like repairing a car. The car has to be having the problem and making the noise when the mechanic looks at it, and the file has to be doing its thing when you are analyzing. Unfortunately, in this case I kept messing with the file and trying to overcome the problem and obscured the extent of the difficulty. That's a major fault of mine. I never ask for help until I have done everything I can think of myself to solve the problem. By then, a post mortem is much more difficult. Thanks again for looking at it.

[robcat2075]

Problem solved now?

 

I think we got you moved over to particle bubbles instead, right?