A place to discuss everything related to Newton Dynamics.
Moderators: Sascha Willems, walaber
by SRCoder » Thu Aug 14, 2008 6:21 am
I have had immense fun playing around with the NewtonSDK and Irrlicht but have hit a problem.
The car I have made drives well after playing with the materials for the map and the body but the rigid body is just as sticky as the tyres.
Is it possible to have seperate materials for the tyres/body or do I have to poke around with the NewtonVehicleSetMaxSideSleepCoefficient, which by the way is confusing and seems to make very litle difference for me when I do.
I hope this is easily answered!
-
SRCoder
-
- Posts: 1
- Joined: Thu Aug 14, 2008 5:51 am
by vectrex » Thu Aug 14, 2008 11:17 pm
The easiest answer is to PM Julio for Newton v2 beta, if you don't mind altering some of your code and working with a beta. It's vehicle joint is opensource so you can do whatever you like
-
vectrex
-
- Posts: 21
- Joined: Sat Oct 01, 2005 10:15 am
- Location: Australia
-
by microdot » Tue Aug 19, 2008 5:05 am
Rigid bodies are not suited for accurate vehicle simulation, tire wise. Soft bodies are the way to go. Having said that, I am comfortable with Newton and would also like v2.0 beta for my testing. If Julio see's this he can pass it my way.
Regards,
Sean.
-
microdot
-
- Posts: 26
- Joined: Thu Oct 25, 2007 12:20 pm
- Location: Johannesburg, South Africa
-
by Julio Jerez » Tue Aug 19, 2008 10:47 am
microdot wrote:Rigid bodies are not suited for accurate vehicle simulation, tire wise. Soft bodies are the way to go.
Haa where did that brand new new laws of physics come from?
-
Julio Jerez
- Moderator
-
- Posts: 12249
- Joined: Sun Sep 14, 2003 2:18 pm
- Location: Los Angeles
-
by microdot » Tue Aug 19, 2008 10:51 am
I fail to understand
What I meant is that if you want rubber like traction, flex and tear soft bodies is perhaps the right way to go?
Please do correct me if I am incorrect.
-
microdot
-
- Posts: 26
- Joined: Thu Oct 25, 2007 12:20 pm
- Location: Johannesburg, South Africa
-
by Julio Jerez » Tue Aug 19, 2008 11:10 am
I am adding get raycast car for thex next beta, if you have some experience on that maybe you can contribute to make it behave nice.
I had made ray cast car for comercial game before but It is something I do not like to do.
The framework in the SDK is is already in with two models.
If you do did it before I am sure you can add your knowledge and have a noice looking car model, in fact all o tth ehard worlk is in, suspension, stability, friction, an dteh joint is all open source so anybody can work on it.
as for the question, I do nto wnat to antagonize, but the reason whey tire physics do nto work with rigiod body is because we onle have and emetical law to simulation teh tire behavior as a single body.
The law is taken for a stady state model, whis is meshure of eth tyre unde soem contraoll conditions.
The one thopng you do nto wnat to use to model a tire is a soft body, because simulation object with spring and dampers is a cahitic prossesess taht look good but you cna nor make any prosidition with it.
The rigid body model is by far superior, but for that teh tire have to me simulated as a colsiion of small bodies conecetd together with joisnt the obey the laws of the material the tire is made up.
In the limit the simplea arpoce lead to a very simulation but it is very spensive in calculation.
Any way this is somethongg we will not do any time soon.
-
Julio Jerez
- Moderator
-
- Posts: 12249
- Joined: Sun Sep 14, 2003 2:18 pm
- Location: Los Angeles
-
by microdot » Wed Aug 20, 2008 2:43 am
Really appreciate the feedback Julio, it was insightful. I'll PM you for details on how to get involved with open source elements that can grow newton. I won't be bashing out any solid code any time soon but definitely a platform to get started.
I hear what you are now saying about soft bodies and prediction there of, not to mention processing power. I am very excited about the raycast functionality possibly being worked into next beta.
Last night finally saw a workable gearbox, engine and transfer of power via clutch in my own simulation. Not perfect, but closer to what I'm after
Take care.
-
microdot
-
- Posts: 26
- Joined: Thu Oct 25, 2007 12:20 pm
- Location: Johannesburg, South Africa
-
Return to General Discussion
Who is online
Users browsing this forum: No registered users and 9 guests