Problem after updating from Newton 2.00 to latest from SVN

Report any bugs here and we'll post fixes

Moderators: Sascha Willems, Thomas

Problem after updating from Newton 2.00 to latest from SVN

Postby TheDushan » Mon Dec 05, 2011 9:05 am

Hello guys.
I am expiring one problem after updating code from Newton 2.00 to latest from svn. [rev.r1155].
While my application is working on Newton 2.00 after updating it to latest revision from svn.
I am experiencing some problem, and I don't know is that Newton problem or problem with my code.

Am I using [coreLibrary_200].

I have uploaded two photos of my problems.

Thanks
Attachments
newton dynamics problem 2.jpg
newton dynamics problem 2.jpg (45.13 KiB) Viewed 3404 times
newton dynamics problem 1.jpg
newton dynamics problem 1.jpg (47.91 KiB) Viewed 3404 times
TheDushan
 
Posts: 3
Joined: Sun Dec 04, 2011 6:43 pm

Re: Problem after updating from Newton 2.00 to latest from S

Postby Julio Jerez » Mon Dec 05, 2011 9:47 am

you are not releasing the collisions afte you asigne then to a body
Julio Jerez
Moderator
Moderator
 
Posts: 12249
Joined: Sun Sep 14, 2003 2:18 pm
Location: Los Angeles

Re: Problem after updating from Newton 2.00 to latest from S

Postby JoeJ » Mon Dec 05, 2011 10:27 am

I had the same as in screenshot #1 after the update.
The app worked, but when closing it i got the same error msg.
When starting to play around with newton i was too lazy to care about proper deallocation.
Adding this 2 lines to my world object destructor solved it for me:

NewtonDestroyAllBodies (world);
NewtonDestroy (world);

The more dense error checking in actual newton is a good thing, as it helps to find your own mistakes.
You'll be warned if you generate NaNs, for example.
User avatar
JoeJ
 
Posts: 1453
Joined: Tue Dec 21, 2010 6:18 pm

Re: Problem after updating from Newton 2.00 to latest from S

Postby JoeJ » Tue Dec 06, 2011 7:29 am

Hi Dushan, i've got your PM.

It seems in CMod_PhysicsAddEntity(sharedEntity_t * gEnt) you are not realeasing the collision, like you do correctly at generation of the static bsp model. Or is the release elsewhere?

Two other things i've found:
You call NewtonCreateBody (something, something, NULL)
If i use NULL in my app, i get a crash. Use the transformation matrix at creation instead.

Most important: It seems you use the timestep for newton based on fps. This is bad, as it needs a fixed timestep.
In practice, you should make steps at 60 Hz, and interpolate between actual and previous physical states for the rendering.
User avatar
JoeJ
 
Posts: 1453
Joined: Tue Dec 21, 2010 6:18 pm


Return to Bugs and Fixes

Who is online

Users browsing this forum: No registered users and 9 guests

cron