RE: [Algorithms] Scrolling starfield

[Ken C. <Ken@HELIKON.com>]

Instead of moving the Y and Z randomly when you add the cube length
to X, you could do it by a reversible function.  That way if the player
went "back", you could get the original positions back.



Ken Carpenter

-----Original Message-----
From: jeffrey Rainy [mailto:jr...@ma...]
Sent: July 30, 2002 11:12 AM
To: Tom Forsyth; gda...@li...
Subject: RE: [Algorithms] Scrolling starfield


This would work fine, except for the fact that any weird-shaped cluster
of star that you pass by will be encountered again at every cube length
when you move in an axis-aligned direction.  And that could be noticed.

An easy fix for that is to randomize the star position on the plane
perpendicular to the axis value you are changing.  i.e. if you add a
cube length to X, just randomize Y and Z on the cube face.  That way you
get nice random distirbuted stars.  

However, if your players goes back a cube length, the stars will have
moved.  Depends on your game style, I guess. ;-)

J.


-----Original Message-----
From: gda...@li...
[mailto:gda...@li...]On Behalf Of Tom
Forsyth
Sent: Tuesday, July 30, 2002 13:36
To: gda...@li...
Subject: RE: [Algorithms] Scrolling starfield


Probably the easiest way to do this is just to fill a cube surrounding
the
player with N stars, fading out with distance so that they are black at
the
greatest distance. When the player moves, see which stars are now
outside
the cube, and "wrap" them (i.e. add or subtract the length of a cube
edge to
the appropriate coordinates) so that they go back inside the cube. That
way
you get a nice even distribution of stars wherever you go, and all the
rest
(parallax, turning, etc) Just Work.

That's assuming you don't mind drawing about six times as many starts as
will actually be visible of course (for a 90-degree FOV). Usually it's
not a
problem on modern hardware though - a nice big batch, very easy for the
card
to reject.


Tom Forsyth - purely hypothetical Muckyfoot bloke.

This email is the product of your deranged imagination,
and does not in any way imply existence of the author.

> -----Original Message-----
> From: Brian Hook [mailto:bri...@py...]
> Sent: 30 July 2002 18:35
> To: gda...@li...
> Subject: [Algorithms] Scrolling starfield
> 
> 
> Hopefully this question is not as stupid as I fear it is.
> 
> I'm working on a top down display right now, Asteroids style, 
> but in 3D
> (i.e. it's a camera positioned over the scene instead of a pure 2D
> projection).  Doing a regular star field is obviously trivial, but I'd
> like to have a starfield with a certain amount of arbitrary parallax
> (not just parallax layers).
> 
> (Yes, I know that when dealing with distances in light years, that
> parallax is not visible, but it gives a sense of movement to the scene
> =P)
> 
> The obvious way to do this, to me, is to generate N stars inside the
> view frustum.  In a nutshell, the scene is composed as if 
> you're looking
> through a viewfinder straight down.  This is the frustum.  At some
> distance away from the near plane is the "ship plane" where 
> your ship is
> located (always centered in the screen).  
> 
> Filling the volume with evenly distribute points bounded by the "ship
> plane", far plane, and frustum sides seems like a trivial 
> operation, but
> so far I haven't come up with anything that's been blindingly 
> intuitive.
> Everything I've come up with has been grossly complex, in fact.
> 
> That aside, I'm not sure this solves my problem entirely, because
> there's the scrolling/wrapping element as well.  If the 
> starfield was on
> a single plane, I could simply translate all stars by some 
> modded value
> of the ship's translation to get the proper effect.  However, with
> perspective projection thrown in and stars at various values, a single
> translation is impossible for all stars.
> 
> So maybe I'm just going about this all wrong.  Even with all of the
> above working, it would end up causing problems if I suddenly 
> decided to
> go to a virtual trackball interface instead of a top down interface.
> 
> 
> Brian
> 
> 
> 
> -------------------------------------------------------
> This sf.net email is sponsored by: Dice - The leading online job board
> for high-tech professionals. Search and apply for tech jobs today!
> http://seeker.dice.com/seeker.epl?rel_code=31
> _______________________________________________
> GDAlgorithms-list mailing list
> GDA...@li...
> https://lists.sourceforge.net/lists/listinfo/gdalgorithms-list
> Archives:
> http://sourceforge.net/mailarchive/forum.php?forum_id=6188
> 


-------------------------------------------------------
This sf.net email is sponsored by: Dice - The leading online job board
for high-tech professionals. Search and apply for tech jobs today!
http://seeker.dice.com/seeker.epl?rel_code=31
_______________________________________________
GDAlgorithms-list mailing list
GDA...@li...
https://lists.sourceforge.net/lists/listinfo/gdalgorithms-list
Archives:
http://sourceforge.net/mailarchive/forum.php?forum_id=6188


-------------------------------------------------------
This sf.net email is sponsored by: Dice - The leading online job board
for high-tech professionals. Search and apply for tech jobs today!
http://seeker.dice.com/seeker.epl?rel_code1
_______________________________________________
GDAlgorithms-list mailing list
GDA...@li...
https://lists.sourceforge.net/lists/listinfo/gdalgorithms-list
Archives:
http://sourceforge.net/mailarchive/forum.php?forum_ida88