PDA

View Full Version : Pool balls driving me crazy!



Johnsongraphix
06-16-2006, 07:15 AM
So I'm using the hard dynamics and collision effects to create a cue ball breaking a rack of pool balls. the calculation is moving fast enough for me to make quick changes and test the motion. My problem is that I'm running out of ideas for settings. Isn't there settings you can add to the dynamics that are truly based off of actual physics besides gravity? After messing with setting for hours upon hours I can't figure out what a 2 weight or 3 weight really is and how is moves over a 2.5 resistance and so on.


Is there a formula out there or do I need to keep pounding my head against my computer screen? I'm sooo close to making this look the way I'd like it to, but it still lacks the realism i'm aiming for.

:)

Johnsongraphix
06-19-2006, 11:47 AM
Does anyone have suggestions about this? I've read a few tutorials, and (seemingly) memorized the section in the LW manual about hard FX and collision. I would appreciate it if someone could at least point me in the right direction or give me a clue.

Dodgy
06-19-2006, 04:11 PM
That's one of the things I've been asking for. The values are somewhat arbitary in the settings, whereas if you know a little maths, you'd understand what all the values are or should be. Bounce should be a percentage of how high an object would bounce compared to it's original height, so 50% should mean it bounces half as high, but 100% in Lightwave's settings just stops the object, and so on. It's just fiddle and play I'm afraid...

Johnsongraphix
06-19-2006, 05:41 PM
I was afraid no one would reply with anything. It's always good to hear from others to let me know whether or not i'm crazy. I've been playing around with this scene for about 10 hours now and even though I feel like I've tried everything I am not going to give up. I'm so close sometimes, yet so far it seems. But however the conclusion unfolds, I'll be sure to post my settings and process, scene setup or whatever incase someones interested. I wouldn't want anyone else to go through this ****.

http://www.newtek.com/forums/images/smilies/brians/confused.gif

ramccombe
06-19-2006, 08:20 PM
Hi, I noticed a couple of things in your vid. 1_balls slide and don't roll. 2_the rails don't seem to bounce. You may want to slow the cue ball down a bit. I made a simple scene and I think the physics are close. I found you must have enough geometry on collision surfaces. I attached a mp4 movie (H.264) and the scene files so you can deconstruct. I hope this helps. You will have to calculate I didn't create a motion file.:dance:

Johnsongraphix
06-20-2006, 06:27 AM
Thanks for your comments and postings, they were truely helpful. I ended up creating an invisible bumper around the edges of the pool table. My problem was that the side bumpers and the table were all the same collision object. If I added more bounce they all jumped out off the table. So that helped. As far as the rolling goes, I noticed that too and then eventually ended up with too much roll, so I messed around with much lower torque setting on the balls and it helped a little. Another problem I had when I finally got the rack to break like I wanted, the balls wouldn't stop moving. The roughness on the bumper would actually cause a ball to increase in speed when it hit. Not to mention when the balls went into the pockets they fell through to the floor. Lot's of funny bloopers with this one.

The file attached was when the balls spun too much- before i turned down the torque.

Johnsongraphix
06-20-2006, 06:44 AM
In my animation I want the cue ball to start moving at frame 300. When I calculate the animation starts at frame 0. How can I control this?

HowardM
06-21-2006, 05:04 AM
heh looking cool! alot better than my test I made when 8 first came out (http://www.3dink.com/pooltable3.mov)

what are your torque settings? you can play with the friction of the collision obj (table) to get them to stop moving, and stop spinning. also you could play with StopByStabilizer. I think you need to have lower Min torque to stop them from spinning like they are...

as said, its all arbitrary and you just need to tweak a setting, see what happens, counter tweak an other setting that may have worked before you tweaked the new one, etc.... youll get the feel for it, but its tough to really know what anything does without just simulating it, and tweaking...

if you want to start at 300 just change your timeline and calculate at 300 :D

Johnsongraphix
06-21-2006, 07:01 AM
Thanks for the comments, I was able to get the roll to slow down and stop using the torque and I was also able to stop the balls from flying around by changing the table surface to advanced object collision instead of regular object collision. It does take some getting use to in the sense that when I changed the surface friction it changed the way the resistance reacted on the hard-body objects. Anyhow, through much pain, and mistakes that would otherwise be funny, I think I've got something really close.

I haven't tried it, but I was wondering if anyone has attempted to created the little metal sphere gadget thing that people put on their desks that move like metronomes where the 4 spheres or so don't move in the middle. Yea, those things, and using only dynamic effects. Anyhow that same physical effect would be a great one for pool balls too. Such as the vibration or impact passes through objects with more stability. My biggest challenge was that the balls were tightly racked and even still I was able to create a realistic break- but slightly believable anyhow.

Johnsongraphix
06-21-2006, 08:16 AM
Oh yea, I'm not sure what you meant by changing the time line, but i did find a solution. I saved the motion files and then offset the number of frames in the dynamics tab so the the motion would start at 300. Worked great! I'll be glad to post the final image I was going for, but the render time will take a long time- even with f-prime. Till then! Thanks and Cheers!