Debuggable demo of convex raycast against compound [SOLVED]

A place to discuss everything related to Newton Dynamics.

Moderators: Sascha Willems, walaber

Re: Debuggable demo of convex raycast against compound colli

Postby Julio Jerez » Sun Jun 04, 2017 3:18 pm

ok, I see a different behavior. the report values is always the same whe standing on that box, did you test it to see if it reproduced the bug?
    frame: 496, t: 0.337910, p(6.766435 2.725723 2.218622) n(-0.000000 1.000000 0.000000)
    frame: 497, t: 0.337829, p(6.766435 2.725723 2.218622) n(-0.000000 1.000000 0.000000)
    frame: 498, t: 0.337789, p(6.766435 2.725723 2.218622) n(-0.000000 1.000000 0.000000)
    frame: 499, t: 0.337768, p(6.766435 2.725723 2.218622) n(-0.000000 1.000000 0.000000)
    frame: 500, t: 0.337758, p(6.766435 2.725723 2.218622) n(-0.000000 1.000000 0.000000)
    frame: 501, t: 0.337753, p(6.766435 2.725723 2.218622) n(-0.000000 1.000000 0.000000)
    frame: 502, t: 0.337751, p(6.766435 2.725723 2.218622) n(-0.000000 1.000000 0.000000)
    frame: 503, t: 0.335450, p(6.178999 2.725723 2.019202) n(-0.000000 1.000000 0.000000)
    frame: 504, t: 0.337748, p(6.178999 2.725723 2.019202) n(-0.000000 1.000000 0.000000)
    frame: 505, t: 0.335448, p(6.791345 2.725723 2.220759) n(-0.000000 1.000000 0.000000)
    frame: 506, t: 0.338395, p(6.791345 2.725723 2.220759) n(-0.000000 1.000000 0.000000)
    frame: 507, t: 0.338072, p(6.791345 2.725723 2.220759) n(-0.000000 1.000000 0.000000)
    frame: 508, t: 0.337910, p(6.791345 2.725723 2.220759) n(-0.000000 1.000000 0.000000)
    frame: 509, t: 0.337829, p(6.791345 2.725723 2.220759) n(-0.000000 1.000000 0.000000)
    frame: 510, t: 0.337789, p(6.791345 2.725723 2.220759) n(-0.000000 1.000000 0.000000)
    frame: 511, t: 0.337769, p(6.791345 2.725723 2.220759) n(-0.000000 1.000000 0.000000)

the only difference I see is the Camera jittering even when the report value is the same when the cylinder is on top of that box.
Julio Jerez
Moderator
Moderator
 
Posts: 12249
Joined: Sun Sep 14, 2003 2:18 pm
Location: Los Angeles

Re: Debuggable demo of convex raycast against compound colli

Postby pHySiQuE » Sun Jun 04, 2017 3:20 pm

The box works fine, but it is not a compound collision shape.
pHySiQuE
 
Posts: 608
Joined: Fri Sep 02, 2011 9:54 pm

Re: Debuggable demo of convex raycast against compound colli

Postby Julio Jerez » Sun Jun 04, 2017 3:22 pm

aren't we debugging a compound collision?
Julio Jerez
Moderator
Moderator
 
Posts: 12249
Joined: Sun Sep 14, 2003 2:18 pm
Location: Los Angeles

Re: Debuggable demo of convex raycast against compound colli

Postby Julio Jerez » Sun Jun 04, 2017 3:31 pm

ok assuming the tow console are the same high, I see is a problem

the high when standing of the compound console is
frame: 1543, t: 0.950435, p(6.735193 4.157656 6.559999) n(-0.000000 1.000000 0.000000)

and the high when standing on the box is
frame: 497, t: 0.337829, p(6.766435 2.725723 2.218622) n(-0.000000 1.000000 0.000000)

can you do one more test?

try making that box a compound and see if it reproduced the bug, this will rule out if it is caused by a specific shape in the compound or if is by the compound itself.
basically I want to know is all compounds has that bug.
Julio Jerez
Moderator
Moderator
 
Posts: 12249
Joined: Sun Sep 14, 2003 2:18 pm
Location: Los Angeles

Re: Debuggable demo of convex raycast against compound colli

Postby pHySiQuE » Sun Jun 04, 2017 3:32 pm

Okay, I updated it so single boxes will be forced to use a compound collision. Same download link:
https://www.leadwerks.com/files/NewtonTest.zip

The box has the same bug.
pHySiQuE
 
Posts: 608
Joined: Fri Sep 02, 2011 9:54 pm

Re: Debuggable demo of convex raycast against compound colli

Postby Julio Jerez » Sun Jun 04, 2017 3:54 pm

ok, it most be a compound bug, I will debug it.
I have to go out now, I will do it when I come back.
Julio Jerez
Moderator
Moderator
 
Posts: 12249
Joined: Sun Sep 14, 2003 2:18 pm
Location: Los Angeles

Re: Debuggable demo of convex raycast against compound colli

Postby Julio Jerez » Sun Jun 04, 2017 11:22 pm

ok sync, and try again.
Julio Jerez
Moderator
Moderator
 
Posts: 12249
Joined: Sun Sep 14, 2003 2:18 pm
Location: Los Angeles

Re: Debuggable demo of convex raycast against compound colli

Postby pHySiQuE » Mon Jun 05, 2017 12:15 am

Awesome! The problem is definitely fixed. Thank you! :mrgreen:
pHySiQuE
 
Posts: 608
Joined: Fri Sep 02, 2011 9:54 pm

Previous

Return to General Discussion

Who is online

Users browsing this forum: No registered users and 13 guests

cron