On 17/08/2014 09:37, Jaime Vives Piqueres wrote:
> El 16/08/14 a las #4, Stephen escribió:
>> The car's wheels wobble a bit as well.
>
>    Yes, I still have to investigate why this happens, as the wheels
> constraints on the simulation shouldn't allow this movement. It happens
> much less if I reduce friction, but then it looses too much traction.
>

Can you change the simulation rate (steps per frame) in Bullet Physics 
Playground?
Does changing that make a difference or changing the collision difference?

>> I paid attention to them as I was thinking that if I gave you a series
>> of meshes with the driver turning the steering wheel from -90° ~ 0° ~
>> +90°. You could select a mesh depending on the angle of the car's front
>> wheels.
>
>    Yes, that is totally feasible, as I can export such data from the
> simulation.
>

What resolution would you want?
How many degrees of revolution and the spacing between them?

>> I am taking a couple of months off work so I have the time and I've not
>> used Poser in a while. So I need to keep up my skills.
>> No pressure.
>
>    Unfortunately, I've had only a week of vacations this month, but
> should be able to work on it on the weekends.
>

Once I have made the meshes my part will be done. Then you can work on 
it at your leisure.

>> He is glued to the seat in fact  he intersects with the seat.
>
>    Yes, I had to place it manually, when it should have fit perfectly
> given that you placed it over the actual seat used on the animation... I
> will look at it again to understand why this happens.
>

I should have mentioned that you would need to do that.
The way I normally import meshes into Poser. I loose the positional 
information. I should have kept that and translated the figure to the 
seat. It should be better next time.

The model was just a proof of concept. I would take more care.
To add to the resolution question.
Would you want the driver to lean and look into a turn?
Actually you could have a series of meshes to play with @ the same cost 
as PovRay.
(Dr John, note the usage of the at sign. Points?)
Last thing.
The steering wheel would need to be attached to the driver's hands.


>> Not a public road so no problem.
>
>    Ah... yes, no problem then.
>
>> But a model with one one hand on the gear stick would not be a problem.
>
>    Don't worry... the acceleration is constant and fixed, and I'm not
> going to implement a gear box on the simulation. :)
>
Yes. One step at a time. :-)


-- 

Regards
     Stephen

Post a reply to this message