[ODE] Re: repeatability in collisions?

[Hampus Soderstrom]

What  does your nearCallback code look like?

Have any globaly or statically defined contacts left over
from the earlier simulation?

If it's just a small difference in initial body position and or angular
velocity. It is hard to notice before the first collisions.

Im not using trimesh collision though. So I can't comment anything
on that.

/Hampa
toribash

2005-03-22 kl. 03.01 skrev Victor B. Zordan:

>
>>> With the same initial state the simulations should end
>>> exactly the same. It works for me and many others.
>
> Yes, works for me too - as long as collisions aren't turned on.
> This makes me conclude it is not my *sim* reset function but the
> collide code which is not being reset...
>
> Has anyone had luck making collisions repeatable?  And if so,
> what (if anything) should I reset in the collide code?
>
> Thx!
> vbZ
>
> On Mon, 21 Mar 2005, George Birbilis wrote:
>
>> ----- Original Message -----
>> From: "Hampus Soderstrom" <...>
>> Subject: Re: [ODE] repeatability in collisions?
>>
>>
>>> Hi,
>>>
>>> This has been discussed before. With the same initial state the
>>> simulations should end
>>> exactly the same. It works for me and many others.
>>>
>>> A suggestion is to print out all joint and body parameters before the
>>> first simulation step
>>> and do a diff on them to find out what is not reset.
>>>
>>> What does your reset function look like?
>>>
>>> cheers
>>>
>>> Hampa
>>> toribash
>>>
>>
>>>>> I am trying to determine whether a repeatability bug is
>>>>> known and if there is a workaround (for example, reset
>>>>> some random seed or ???)
>>>>
>>>> maybe you need to search the source base for any functions relating 
>>>> to
>>>> random number generation and to getting system time or reading any
>>>> system timers
>>
>> _______________________________________________
>> ODE mailing list
>> ODE at q12.org
>> http://q12.org/mailman/listinfo/ode
>>
> _______________________________________________
> ODE mailing list
> ODE at q12.org
> http://q12.org/mailman/listinfo/ode
>