Sorry your browser is not supported!

You are using an outdated browser that does not support modern web technologies, in order to use this site please update to a new browser.

Browsers supported include Chrome, FireFox, Safari, Opera, Internet Explorer 10+ or Microsoft Edge.

Dark Physics & Dark A.I. & Dark Dynamix / [Dark Physics] The Jiggles

Author
Message
imekon
17
Years of Service
User Offline
Joined: 3rd Jul 2006
Location:
Posted: 6th Aug 2006 23:06
Here's a demo program that demonstrates "the jiggles"



After it finishes creating all four hundred boxes, take a look by wandering around the scene at isolated boxes on their own. They're probably "jiggling", i.e. never completely coming to a rest.

I tried linear/angular damping but no real effect, it still jiggles. My SleepingBoxes function seems to show no boxes are sleeping.

Also, if you watch, some thin boxes hurtle through the floor never to be seen again. I'll see if



fixes this.

"Reality Bites"
HowDo
21
Years of Service
User Offline
Joined: 28th Nov 2002
Location: United Kingdom
Posted: 6th Aug 2006 23:37
looks like the ground is vibrating as the blocks fall.

I'm not getting you down am I, Ho Look! another fancy Door?
imekon
17
Years of Service
User Offline
Joined: 3rd Jul 2006
Location:
Posted: 7th Aug 2006 00:07
They keep on jiggling after the last one has fallen - and the effect only shows up after 400 boxes, reduce the number to say 300 and it stops.

"Reality Bites"
HowDo
21
Years of Service
User Offline
Joined: 28th Nov 2002
Location: United Kingdom
Posted: 7th Aug 2006 00:25
for me to get them to stop jiggling I had to drop it down to 100 (system is a P3 1ghz).

I'm not getting you down am I, Ho Look! another fancy Door?
imekon
17
Years of Service
User Offline
Joined: 3rd Jul 2006
Location:
Posted: 9th Aug 2006 13:26 Edited at: 9th Aug 2006 13:27
I'm still seeing the jiggles above 300 boxes... I've been fiddling with some of the settings, and I tried increasing the mass of the boxes, to something ridiculous...

The boxes start sinking into the ground, then pop back up again. What bizarre behaviour!

"Reality Bites"
Mike Johnson
TGC Developer
21
Years of Service
User Offline
Joined: 13th Sep 2002
Location: United Kingdom
Posted: 9th Aug 2006 13:56
I can't replicate the effect here so I'm thinking it has something to do with the timing. One thing you can try doing is changing the default timing to phy set timing 1.0 / 60.0, 3, 0.

Increasing mass to large values can cause unwanted side effects and instabilities. As a guideline try to use the range 1 - 10 for mass.

A few other things to try include:

* change default sleep linear velocity property
* change default sleep angular velocity
* increase linear damping on the rigid bodies
* increase angular damping on the rigid bodies

Have you tried running the debugger alongside? Some information in there might be useful to understanding this.
imekon
17
Years of Service
User Offline
Joined: 3rd Jul 2006
Location:
Posted: 9th Aug 2006 22:24
Quote: "I can't replicate the effect here so I'm thinking it has something to do with the timing. One thing you can try doing is changing the default timing to phy set timing 1.0 / 60.0, 3, 0.

Increasing mass to large values can cause unwanted side effects and instabilities. As a guideline try to use the range 1 - 10 for mass.

A few other things to try include:

* change default sleep linear velocity property
* change default sleep angular velocity
* increase linear damping on the rigid bodies
* increase angular damping on the rigid bodies

Have you tried running the debugger alongside? Some information in there might be useful to understanding this."


None of the above had any effect - except the debugger, but that doesn't count.

The debugger generated 64MBytes of data - I'll have to see if I can cut it down.

"Reality Bites"
Mike Johnson
TGC Developer
21
Years of Service
User Offline
Joined: 13th Sep 2002
Location: United Kingdom
Posted: 9th Aug 2006 22:36
What frame rate are you getting on this program?
imekon
17
Years of Service
User Offline
Joined: 3rd Jul 2006
Location:
Posted: 10th Aug 2006 00:35
Quote: "What frame rate are you getting on this program?"


60 initially, then it drops to around 34 after 300 objects.

"Reality Bites"
Mike Johnson
TGC Developer
21
Years of Service
User Offline
Joined: 13th Sep 2002
Location: United Kingdom
Posted: 10th Aug 2006 13:06
The phy get rigid body sleeping command has been changed for the update. When the update is released can you see how many rigid bodies actually go to sleep. Maybe some of them aren't going to sleep and this is causing the jiggling.
Mike Johnson
TGC Developer
21
Years of Service
User Offline
Joined: 13th Sep 2002
Location: United Kingdom
Posted: 10th Aug 2006 14:00
Just came back to this and had another look. I was able to get the same jittering effect as you by forcing the sync rate to 20. Turns out it was the command phy set auto fixed timing causing this. If you remove it the program will work fine. An alternative approach with the timing is to use phy set timing 1.0 / 60.0, 60 / 20, 0.
imekon
17
Years of Service
User Offline
Joined: 3rd Jul 2006
Location:
Posted: 11th Aug 2006 01:46
Yeah, that works. Even with 500 objects and a frame rate of 26 still works.

"Reality Bites"

Login to post a reply

Server time is: 2024-04-19 13:39:16
Your offset time is: 2024-04-19 13:39:16