MASON-INTEREST-L Archives

September 2006

MASON-INTEREST-L@LISTSERV.GMU.EDU

Options: Use Monospaced Font
Show Text Part by Default
Condense Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Sender:
MASON Multiagent Simulation Toolkit <[log in to unmask]>
Date:
Sun, 17 Sep 2006 18:07:09 -0400
Reply-To:
MASON Multiagent Simulation Toolkit <[log in to unmask]>
Subject:
From:
Gabriel Catalin Balan <[log in to unmask]>
Content-Type:
TEXT/PLAIN; charset=US-ASCII
In-Reply-To:
MIME-Version:
1.0
Comments:
To: MASON Multiagent Simulation Toolkit <[log in to unmask]>
Parts/Attachments:
TEXT/PLAIN (144 lines)
Sean, Uwe

Transform3d.setScale() seems more suitable for the task then scale();
the former discards the old scaling factor as it puts in the new, so you
don't have to worry about cummulative effects.

Gabriel


On Sun, 17 Sep 2006, MASON-INTEREST-L automatic digest system wrote:

> There are 2 messages totalling 124 lines in this issue.
>
> Topics of the day:
>
>   1. Dynamic TransformedPortayal3D (2)
>
> ----------------------------------------------------------------------
>
> Date:    Sat, 16 Sep 2006 19:38:46 -0400
> From:    Uwe Grueters <[log in to unmask]>
> Subject: Dynamic TransformedPortayal3D
>
> Dear all,
>
> I am completely new to this list and to MASON.
>
> I intend to develop a model for virtual plants in 3d. Plant attributes
> (height, diameter) are calculated in the simulation and ConePortrayals3D
> for the plants created at runtime shall reflect these attributes.
>
> I found example code in the Balls3D model that does similar things to a
> SpherePortrayal3D. The BallPortrayal inherits from the SpherePortrayal3D
> and overrides the getModel method of the superclass. The if condition
> becomes true when the TransformGroup passed to the method is null (as in
> the getModel method of the superclass) or when the passed in Object casted
> to Ball has changed its mass. In the statements that are executed in that
> case the SpherePortayal3D scale-variable is multiplied by the BallObject's
> radius, that also has changed. So, there is a dynamic connection of the
> attributes of a ball and its portrayal.
>
> I tried to follow this scheme by creating a PlantPortrayal that extends
> TransformedPortrayal3D, overriding its getModel method, putting in ||
> ((Plant)obj).oldDiameter != ((Plant)obj).diameter and trying to change the
> internalTransform using components from the method code for scaling
> (double,double,double) and rotatingY in the TransformedPortrayal3D.
>
> However, when I ran the code with a step method in which the diameter is
> increased by 1 each time step I saw the cone size to increase drametically
> at step 1 and at step 2 the cones seemed to be too large to be displayed.
>
> Has anyone written code to dynamically connect a "TransformedPortrayal3D"
> to attributes of the underlying agent? Is there anyone, who could give me
> some advice in these matters?
>
> Since this behaviour is such a basic requirement for the model development
> I would greatly appreciate any helpful suggestions.
>
> Thank you very much in advance, Uwe
>
> ------------------------------
>
> Date:    Sat, 16 Sep 2006 20:52:57 -0400
> From:    Sean Luke <[log in to unmask]>
> Subject: Re: Dynamic TransformedPortayal3D
>
> Hi Uwe.  The TransformedPortrayal3D class uses a Java3D Transform3D
> object underneath, and you can replace the Transform3D as you see
> fit.  However if you're calling the scale() method built into the
> class, you need to understand that these are further scaling the
> *existing* scale -- that is, it's multiplying the scales.  For
> example, if you called scale(2,2,2), and then again called scale
> (2,2,2), your total scaling will be of size 4.
>
> Also, note that performing scales, rotations, and translations, are
> called affine transformations, and are done internally with a 4 by 4
> matrix.  Each operation multiplies that matrix by a new matrix -- and
> matrix multiplications are NOT commutative.  You can't do a rotate
> and then a scale and expect it to be the same as a scale followed by
> a rotate (say).
>
> Sean
>
> On Sep 16, 2006, at 7:38 PM, Uwe Grueters wrote:
>
> > Dear all,
> >
> > I am completely new to this list and to MASON.
> >
> > I intend to develop a model for virtual plants in 3d. Plant attributes
> > (height, diameter) are calculated in the simulation and
> > ConePortrayals3D
> > for the plants created at runtime shall reflect these attributes.
> >
> > I found example code in the Balls3D model that does similar things
> > to a
> > SpherePortrayal3D. The BallPortrayal inherits from the
> > SpherePortrayal3D
> > and overrides the getModel method of the superclass. The if condition
> > becomes true when the TransformGroup passed to the method is null
> > (as in
> > the getModel method of the superclass) or when the passed in Object
> > casted
> > to Ball has changed its mass. In the statements that are executed
> > in that
> > case the SpherePortayal3D scale-variable is multiplied by the
> > BallObject's
> > radius, that also has changed. So, there is a dynamic connection of
> > the
> > attributes of a ball and its portrayal.
> >
> > I tried to follow this scheme by creating a PlantPortrayal that
> > extends
> > TransformedPortrayal3D, overriding its getModel method, putting in ||
> > ((Plant)obj).oldDiameter != ((Plant)obj).diameter and trying to
> > change the
> > internalTransform using components from the method code for scaling
> > (double,double,double) and rotatingY in the TransformedPortrayal3D.
> >
> > However, when I ran the code with a step method in which the
> > diameter is
> > increased by 1 each time step I saw the cone size to increase
> > drametically
> > at step 1 and at step 2 the cones seemed to be too large to be
> > displayed.
> >
> > Has anyone written code to dynamically connect a
> > "TransformedPortrayal3D"
> > to attributes of the underlying agent? Is there anyone, who could
> > give me
> > some advice in these matters?
> >
> > Since this behaviour is such a basic requirement for the model
> > development
> > I would greatly appreciate any helpful suggestions.
> >
> > Thank you very much in advance, Uwe
>
> ------------------------------
>
> End of MASON-INTEREST-L Digest - 2 Sep 2006 to 16 Sep 2006 (#2006-54)
> *********************************************************************
>

ATOM RSS1 RSS2