Vehicle model for 3.14

A place to discuss everything related to Newton Dynamics.

Moderators: Sascha Willems, walaber

Re: Vehicle model for 3.14

Postby JoeJ » Wed Apr 06, 2016 3:38 am

Ok, thanks for time and explantation :)

I looked at the vehicle controller and see it fills an array of NewtonUserContactPoints.
That would work for the single contact ragdoll - but you convinced me it's no good idea.

My main interest in user contacts is a better solution for the realtime scaling body contacts.
The scaling adds velocity to the surface Newton is not aware of, so it does not work out of the box.
Hacking contacts and / or tweaking contact velocities did not work well, so i replace the contacts with custom joints, but i won't be be able to make this as correct as your default contacts.

You also talked about your sticky contacts idea, i think that requires the same functionality.
Is this already possible?
User avatar
JoeJ
 
Posts: 1453
Joined: Tue Dec 21, 2010 6:18 pm

Re: Vehicle model for 3.14

Postby Julio Jerez » Wed Apr 06, 2016 8:11 am

JoeJ wrote:You also talked about your sticky contacts idea, i think that requires the same functionality. Is this already possible?


no yet, for that I need to add one functionality to the contact joint so that the lower friction force can be shifter by some value, write not that not possible is a joint row is marked as contact normal.
It will be a material option, that will work with that functionality but I am leaving that for newton 3.15

My plan now is this:
-complete the vehicle and make the best I could possible can.
-fix all the bugs
-complete the voronoi and Boolean destruction

that will be the lest of functionality for 3.14 then start 3.15 right away starting with particle bodies.
Julio Jerez
Moderator
Moderator
 
Posts: 12249
Joined: Sun Sep 14, 2003 2:18 pm
Location: Los Angeles

Re: Vehicle model for 3.14

Postby Julio Jerez » Wed Apr 06, 2016 9:53 am

Actually Joe,
as I keep debugging this, every thing seem to check out as it should be.
so I modified the demo so that I can make assumption of what to expect.
and I now can see what happen.

Basically is the coefficient of restitution in combination with the pick force and the body size.

what happens is that the pick force is supplying the body with a velocity of 365 m/s
the default coefficient of restitution is in 1.4
what this mean is that the body will bound with 365 * 0.4 = 156 m/s

since the body is abnormally big, that seem like a bug but it is in fact the right respond.
we see as a bug because we are no use to see big object moving so quickly, in general when a large object collide, is does not bounce back it simple keep going and getting destroyed.

I change the coefficient of friction to 1.1 and now is right. I check in the test demo so that you can see.

This is now more pronounce that in previous version because remember when I made the default joint stiffness a lot much stronger, this affect the contact too, if anything this is better behavior than before.

I am not ruling out that there still could be a bug but I do not see it.
Julio Jerez
Moderator
Moderator
 
Posts: 12249
Joined: Sun Sep 14, 2003 2:18 pm
Location: Los Angeles

Re: Vehicle model for 3.14

Postby collerblade » Sun Jul 10, 2016 2:00 am

Julio
The new vehicle model is awesome! It is much better than i excepted. Grat!!

collerblade
collerblade
 
Posts: 88
Joined: Mon Apr 05, 2010 12:09 pm

Re: Vehicle model for 3.14

Postby Leadwerks » Fri Apr 07, 2017 1:00 am

Do you think this will be 100% ready to use by May 15? I can keep this in our beta build before then, but it needs to go out to release in our new version May 15.
User avatar
Leadwerks
 
Posts: 569
Joined: Fri Oct 27, 2006 2:54 pm

Re: Vehicle model for 3.14

Postby Julio Jerez » Fri Apr 07, 2017 9:38 am

It is ready now.
the only thing missing is end user tweaks, or calibrations I cannot do myself.
Julio Jerez
Moderator
Moderator
 
Posts: 12249
Joined: Sun Sep 14, 2003 2:18 pm
Location: Los Angeles

Re: Vehicle model for 3.14

Postby Leadwerks » Fri Apr 07, 2017 10:25 am

What demo are you guys running? The super car demo just makes the Leadwerks truck float in the air and it rotates the car when I turn the tires.
User avatar
Leadwerks
 
Posts: 569
Joined: Fri Oct 27, 2006 2:54 pm

Re: Vehicle model for 3.14

Postby Julio Jerez » Fri Apr 07, 2017 10:41 am

oh that ls because like I said on the other post I am now working on the ragdoll system.

to test the car, you can set the gavity back to -10.0 in file
../applications\demosSandbox\sdkDemos\toolBox\PhysicsUtils.h

//#define DEMO_GRAVITY -10.0f
#define DEMO_GRAVITY -0.0f

then you can drive the super car and the tank. for the tank demo I have not plugged the engine yet, but the threads and steering do work as expected. I have not changed the simple car demo yet.
Julio Jerez
Moderator
Moderator
 
Posts: 12249
Joined: Sun Sep 14, 2003 2:18 pm
Location: Los Angeles

Re: Vehicle model for 3.14

Postby Leadwerks » Fri Apr 07, 2017 12:37 pm

This is great. I love the way the tire slip on the 4x4 works. Is this using a multi-body approach instead of raycasting? Feels much more solid.
User avatar
Leadwerks
 
Posts: 569
Joined: Fri Oct 27, 2006 2:54 pm

Re: Vehicle model for 3.14

Postby Julio Jerez » Fri Apr 07, 2017 1:51 pm

glad you like it.
Yes this is all articulated body physics. No only the tires and the chassis are articulated body but also the transmission, the engine and the differentials are joints or rigid bodies.
This allows for end users to attach accessories to vehicles by using normal joints and rigid bodies, which was very hard on previous versions.

I am looking forward some one using your game engine put it to good use.
Julio Jerez
Moderator
Moderator
 
Posts: 12249
Joined: Sun Sep 14, 2003 2:18 pm
Location: Los Angeles

Previous

Return to General Discussion

Who is online

Users browsing this forum: No registered users and 5 guests

cron