Jump to content
Hash, Inc. Forums
Rodney

Exercise 17: Cosmic Bowling

Recommended Posts

Spot on.

 

I never doubted you for a minute. :)

 

Don't get too distracted but when appropriate you definitely will want to explore the Newton Dynamics and the Simulations forums. It has been reported that very dynamic stuff happens there.

Share this post


Link to post
Share on other sites

Name: Tim Roberts

 

Exercise Completed: Exercise 17: Cosmic Bowling

 

Date Completed: Nov 25, 2008

 

Instructor: Manual, forum

 

Comments: Pretty straightforward. Really amazing results. More and more impressed w/ the programming and the results. (and the forum support!).

 

(added) OK, resolved upload problem.

 

Here it is:

 

ex_17_test2.mov

Share this post


Link to post
Share on other sites

Hey! Where's the Exercise? Cosmic bowling is a team sport!

 

Edit: I see it now. Good show! :)

Share this post


Link to post
Share on other sites

Exercise 17: Cosmic Bowling

 

Here is the rigid body sim.

Had some problems with the ball jumping off the lane and over the pins if I tried to put spin on the ball.

The ball also wanted to fall into the lane until I dropped it down in the first five frames to where it was almost released at teh floor level.

 

Bowling_Project2.mov

Share this post


Link to post
Share on other sites

Very nice Tim.

 

Had some problems with the ball jumping off the lane and over the pins if I tried to put spin on the ball.

 

See! You try to get all fancy and show off for the crowd and where does it get you?

Bet you had the audience entranced though. ;)

 

The ball also wanted to fall into the lane until I dropped it down in the first five frames to where it was almost released at teh floor level.

 

As my weary brain recalls this can be an effect of the mesh density of the floor.

That you discovered the solution as you did demonstrates your understanding of how the feature works.

 

Rack them up and try again with Caroline's Newton Dynamics version of Cosmic Bowling.

Its all the rage these days! :)

Share this post


Link to post
Share on other sites
Very nice Tim.

 

Had some problems with the ball jumping off the lane and over the pins if I tried to put spin on the ball.

 

See! You try to get all fancy and show off for the crowd and where does it get you?

Bet you had the audience entranced though. ;)

 

The ball also wanted to fall into the lane until I dropped it down in the first five frames to where it was almost released at teh floor level.

 

As my weary brain recalls this can be an effect of the mesh density of the floor.

That you discovered the solution as you did demonstrates your understanding of how the feature works.

 

Rack them up and try again with Caroline's Newton Dynamics version of Cosmic Bowling.

Its all the rage these days! :)

 

That was way cool!

Easier,

More realistic,

Faster calculating,

the ball spins,

the ball even bites and goes a bit Brooklyn.

.....So why would you use rigid bodies?????

 

I did notice that one pin starts to fall before the ball reaches the pins. Any idea what I may have done to cause that?

 

Bowling_Project_newton.mov

Share this post


Link to post
Share on other sites
.....So why would you use rigid bodies?????

 

Before Newton Dynamics?

Just about anything that had rigid body similuation.

Now we've got another very powerful option to animate with.

 

I did notice that one pin starts to fall before the ball reaches the pins. Any idea what I may have done to cause that?

 

Not enough information to factor in.

I would guess:

- That pin has different settings than the rest.

- Its partially off of the lane already.

Share this post


Link to post
Share on other sites

Although this lesson doesn't explain quite what's going on with rigid body, logic would argue that the computer is using the direction and animation from the first five frames to; in relation with the other bodies in the choreography, extrapolate the action. And it does a pretty good job too.

 

I felt back in my element with this lesson. It felt more like animating. A minor adjustement to the ball and a recalculation and I got my strike. The ball is dipping through the floor based on the rigid body keyframes (which don't seem to be adjustable themselves but you can keyframe over them so just as well)? Adjust the path on the fly and add some keyframes to the axis that's allowing it to go through the floor (I think the Y Axis but like my nephew who has trouble telling his left from his right I've never been great at keeping the X,Y, and Z straight; and I would bet that there are several ways to skin this cat but adding keyframes to the trouble areas seemed the easiest/most obvious/most comfortable path for me).

 

I can see this would be very useful. But here's a few questions.

 

1) Can you animate more than one model in those first few frames and have them interact with each other? For example if I placed another bowling ball at the other end of the alley and animated it towards the first ball in the first five frames would the "Simulate Rigid Bodies" command send them careening at each other only to bounce off when they encounter each other or is the "Simulate Rigid Bodies" restricted to one model in motion and all others to be encountered by the moving model?

 

2) Can a fully boned model work with "Simulate Rigid Bodies" or does it have to be a simple model like the bowling ball. For example; there is a scene in the first short I plan on attempting (I already have the script pretty much planned out) where a cat is thrown at a person. The cat is a super powered side kick to my hero and one of his "attacks" is to roll into a ball (not a real ball just a tightly balled up cat) and slam into an enemy. I plan on cutting from the throw to the look on the guy who's about to be hit's face and then to a side view as the cat whips down the alley and slams into him. Since this model of my cat will be fully rigged, once I curl him up into his position and animated him spinning towards the guy he's going to hit would "Simulate Rigid Bodies" send him down the alley and into the guy for me? Since his skeleton is so much more complicated I'm wondering if that will be an issue?

 

3) There are more than a few pass throughs with the pins and ball in my vid. They aren't INCREDIBLY obvious but I see them. I'm sure you guys will too. Would reducing the Bounce attribute to less than 5% on the pins help with that? I read a few posts up Rodney said something about mesh density or something like that but I couldn't find an attribute called that for the pins that I could adjust so I'm thinking he was referring to the actual number of splines and cp's... as in more = denser or something. Can someone straighten that out for me and give me an idea on adjusting tha animation to correct the pass throughs? Obviously I could do it manually but I really think that would screw up the great flow the "Simulate Rigid Bodies" created in the animation.

 

4) This isn't really a fair question because I haven't read the post yet but I see Nancy has contributed a new version of this lesson using the Newton Physics plug in that one of my fellow newbies is raving about. I plan on investigating that tomorrow as it seems to be the wave of the future. But can anyone explain to me what the difference is, why one might be better than the other and under what circumstances one might be chosen instead of the other? If the answer to those questions are in her post then please feel free to ignore the question. It's 4:30 am here so I need to sleep but I'll definitely check it out tomorrow. (and thanks Nancy for doing that... in advance! ;) )

 

That's it. Vid is attached, I cribbed a little audio from Wii bowling to make it more interesting but skipped the credits as its so late (and so unnecessary).

 

Thanks. ;)

lesson17cosmicbowlingfinal.mov

Share this post


Link to post
Share on other sites

That looks successful!

 

This one is way outside my field so these are general answers...

 

1) Can you animate more than one model in those first few frames and have them interact with each other?
I'd suppose so but easy for you to try

.

2) Can a fully boned model work with "Simulate Rigid Bodies" or does it have to be a simple model like the bowling ball. For example; there is a scene in the first short I plan on attempting (I already have the script pretty much planned out) where a cat is thrown at a person...

 

Vernon "HeyVern" Zehr has done some breakthrough work on "ragdoll" physics with "newton". Check that stuff out. I don't' think he's written a tut on it yet but hopefully will.

 

However, your scene sounds more like straightforward animating rather than a simulation.

 

3) There are more than a few pass throughs with the pins and ball in my vid...Would reducing the Bounce attribute to less than 5% on the pins help with that?
Something is stopping you from trying it? i don't actually know.

 

Simulations work with approximations of the shapes so some pass thrus will happen. In your vid they are so quick that they probably wouldn't be noticed in the context of a longer movie.

 

 

I read a few posts up Rodney said something about mesh density or something like that but I couldn't find an attribute called that for the pins that I could adjust so I'm thinking he was referring to the actual number of splines and cp's... as in more = denser or something.

 

yes he is referring to the spline mesh.

 

 

 

4) This isn't really a fair question because I haven't read the post yet but I see Nancy has contributed a new version of this lesson using the Newton Physics plug in that one of my fellow newbies is raving about...
Yes, the Newton simulations should be quite superior in most ways to A:M's original simulation engine. Steffen Gross wrote the port to A:M and has numerous examples on his site. Plan on using Newton for all your physical simulation ventures.

Share this post


Link to post
Share on other sites
Something is stopping you from trying it?

 

Yes, 4:30 am and a half hour to render. But today is another day. I'm going to take a look at it now.

 

Thanks for the answers. :D

Share this post


Link to post
Share on other sites
Something is stopping you from trying it?

 

Yes, 4:30 am and a half hour to render.

 

render in shaded mode and you can quickly see how things are moving .

 

Actually just scrubbing in the timeline after you simulate will show you if there are pass thrus

Share this post


Link to post
Share on other sites

Just redid the lesson with Nancy's Newton Tut. Big thanks for Nancy for this. I didn't notice it on the scrub through but it looks like the pins aren't perfectly on the lane. I moved the model bone and made sure it was on 0,0,0. I must have screwed something else up. Still it's a nice intro to the plug in. Especially with the provided links for additional info in the PDF. Thanks again for the extra mile Nancy.

 

Although it took a good bit longer to render this definitely looks more "real" and I had no problems with pass throughs with either the pins and ball nor the ball and the lane. Everything worked perfectly. It was a lot harder to line up for a strike. I only had to make 2 minor changes to the rigid body and I got my strike. I had to reload the lesson and try about ten times to get it right with Newton. So..... since I suck at bowling that's more real too. ;)

 

 

 

Vid attached. Onto 18. :)

lesson17cosmicbowlingnewton.mov

Share this post


Link to post
Share on other sites
Just redid the lesson with Nancy's Newton Tut. Big thanks for Nancy for this.

 

Rob - t'wasn't me - It was CAROLINE that did the terrific Tut - Big Thanks to Caroline!

Share this post


Link to post
Share on other sites
Just redid the lesson with Nancy's Newton Tut. Big thanks for Nancy for this.

 

Rob - t'wasn't me - It was CAROLINE that did the terrific Tut - Big Thanks to Caroline!

 

Ok. I am E-Tarded. Sorry about that Nancy... and big sorry to Caroline. Thanks Caroline. I'm going to leave my incorrect posts up there so I can wallow in my shame....

 

 

IT BURNS! The STUPID.... IT BURNS!

Share this post


Link to post
Share on other sites

I suppose an easy fix for the floating pins would be to move the alley up after the simulation has run.

Share this post


Link to post
Share on other sites
I suppose an easy fix for the floating pins would be to move the alley up after the simulation has run.

 

I've got the program up now as I just finished 18. I'll see if that does it. ;)

 

EDIT: Nope. Not an easy fix. FUBAR-ed the ball as it made the ball sink into the lane. Made some passthroughs between the lane and the pins as well.

 

I was kind of surprised when I looked closely at the pins. They are all a little "off" on thier settings. Some a bit higher off the lane than others and almost all of them had to be adjusted.

 

So not an easy fix. But a fix would be to grab each pin before running the sim and make sure it is actually on the lane. Which is what I should have done myself. Or checking it again after the sim and adjusting them down if the sim puts them up a bit after calculating and before rendering. None of which I did.

 

But at least I know how to fix it. I just have no great desire to. But I will if it means I can't get my Rodney certificate. :D

 

EDIT 2: I went back and tried to lower the pins in frame 0 and frame 2:22 (just before the ball hits) but because every frame is a keyfram with the Newton plug in they just pop right back up. So my fix isn't a fix. I would have to adjust every frame or fix the issue before activating the Newton plug in. And since I'm not sure why they weren't situated properly to begin with, I'm kind of at a loss.

 

EDIT 3: Yeah the pins are not actually on the lane when the project loads. They come in at slightly different heights and you have to adjust them manually on fram zero before you run the Newton sim. But once you've done that it's all straight. Added a little reflection to the floor. Didn't bother with the strike this time. But at least the pins are actually on the lane. So done and done. I'll upload it as son as it's done rendering.

 

EDIT 4: Ok that didn't work either. I forgot there were keyframes up to the fifth frame. I foolishly assumed they were related to the ball only but sure enough the pins were back floating in the air by frame 5. I must have to seat them for the first 5 frames to fix it. And that's not something I'm going to go through tonight. Maybe later. -_-

Share this post


Link to post
Share on other sites

Do an Embed All and save the PRJ after the simulation is completed but beforeyou tweaked anythingand post it here. I'd be curious to look at it.

Share this post


Link to post
Share on other sites

Coming right up! :D

 

Although I have to say that I'm pretty sure that if I dropped the pins down on frames 1-5 they will stay put.

 

 

Maybe I'll just do it and see. :rolleyes:

 

EDIT: Ok, well I just forced keyframes on the pins for the first 5 seconds and I forced them on 2:22 just before the ball hits and it didn't matter. None of it made any difference. Once the Newton plug in runs the pins go back to hanging just above the lane. So I must have missed something somewhere. The first thing that comes to mind is that maybe I didn't put the model bone in the right spot or maybe the model wasn't set to 0,0,0 as I thought it was.

 

I've looked I can't figure it out. Let me know what you find. Thanks for the help. :unsure:

Lesson_17_Cosmic_Bowling_Newton_Robcat.prj

Share this post


Link to post
Share on other sites

BowlH264.mov

 

Basically, on frame 0 i moved the alley up to meet the bottom of the pins.

 

That made the ball look like it was sunk into the floor so I went to a frame where the ball had landed and was rolling, turned off the "A" button (so that any change I made would affect ALL keyframes. You need to be really careful you never switch this off accidentally), turned on the WorldSpace button (so that the translate manip would be level), and raised the ball until it was just touching the floor.

 

 

then I rendered. the pins now look like they are contacting the floor properly. This sort of cheat won't work in all situations but in this one it does, so there it is.

Share this post


Link to post
Share on other sites
BowlH264.mov

 

Basically, on frame 0 i moved the alley up to meet the bottom of the pins.

 

That made the ball look like it was sunk into the floor so I went to a frame where the ball had landed and was rolling, turned off the "A" button (so that any change I made would affect ALL keyframes. You need to be really careful you never switch this off accidentally), turned on the WorldSpace button (so that the translate manip would be level), and raised the ball until it was just touching the floor.

 

 

then I rendered. the pins now look like they are contacting the floor properly. This sort of cheat won't work in all situations but in this one it does, so there it is.

 

I didn't even know there was an "A" button. I'll have to check that out. Thanks. Have any idea what was wrong? I mean why they were raising up like that to begin with?

 

Also, I see the fix did away with the little bounce the ball takes in the first few frames (which is fine really because I don't recall bowling balls bouncing like that when thrown properly). Could you animate that back in with the "A" button off and have it work or is this trick basically a very delicate path correction and animating the bounce would send the wheels flying off the bus?

Share this post


Link to post
Share on other sites
I didn't even know there was an "A" button. I'll have to check that out. Thanks. Have any idea what was wrong? I mean why they were raising up like that to begin with?
You'd have to ask a Newton expert like Vern.

 

But beware of the A button. This is one of the few times you'd want to turn it off. Forgetting to turn it back on after making an adjustment like this has resulted in much messed up animation and gnashing of teeth.

 

A button off means nothing you do will create a keyframe. That's usually bad.

 

Also, I see the fix did away with the little bounce the ball takes in the first few frames (which is fine really because I don't recall bowling balls bouncing like that when thrown properly).
The bounce is still there but I left the ball still intersecting the floor a bit too much so it gets covered up.

 

Could you animate that back in with the "A" button off and have it work or is this trick basically a very delicate path correction and animating the bounce would send the wheels flying off the bus?

 

Yes, you could re-readjust the ball to a slightly higher level until it was perfect.

Share this post


Link to post
Share on other sites
Project 17

Complete 15 Feb 10

 

Upgraded to the newest version for this one.

 

Looks like you got that working!

 

BTW, you can make much smaller files by rendering to a different codec. See the notes about Quicktime and codecs in the middle of this page.

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


×
×
  • Create New...