Hi,

I developed a program which models the dynamics of a stellar disk. Rendering is done in 3D with rotation and zoom functionalities. Here’s an image at the start of the simulation :

You can see on top the different statistics. My problem is about the initial value of the scale (100 kpc). Indeed, I use for initialization a perspective projection this way :

```
if (isDisplayFirst)
{
isDisplayFirst = false;
// Initialize View
glViewport(0, 0, w_width, w_height);
glMatrixMode(GL_PROJECTION); // Select The Projection Matrix
glLoadIdentity(); // Reset The Projection Matrix
glMatrixMode(GL_MODELVIEW); // Select The Modelview Matrix
glLoadIdentity(); // Reset The Modelview Matrix
/* perspective */
gluPerspective(45.0f, (float)w_width / w_height, g_nearPlane, g_farPlane);
/* distance z = 3 */
gluLookAt (0.0, 0.0, 3.0, 0.0, 0.0, 0.0, 0.0, 1.0, 0.0);
/* zoom out = 100/3 */
glScalef(0.03f, 0.03f, 0.03f);
}
```

My widget has the following dimensions :

```
w_width = w_height = 600
```

So, from a simple computation, I think that I can have :

```
Scale_value_Init = tan(45) * distance * zoom_out = 3*100/3 = 100
```

with

```
distance = 3 and zoom_out = 100/3
```

To display the HUD (Head Up Display), I use a orthogonal projection and I draw a line of 100 pixels.

When I compare the “Scale_value_init” value with the average position values of each particle, I can see this initial value is a little too big.

There must be an arror in this initial value computation for scale bar.

Unities used in this program are “kpc” (kilo parsec) and I have set : 1 pixel = 1 kpc.

I just want to get a good value which gives a distance scale in (xy) plane.

Could you tell me why this calculus is wrong and give me some clues to get a good estimation ?