Re: [Algorithms] Depth-Sorting Axis-Aligned Boxes

[Pablo de H. C. <pa...@vi...>]

I'm also interested in this. I had a look at the Inigo Quilez grid-aligned
approach but as I can see I can'tget it to work for "arbitrary" grids. In
that apporach he uses 48 pre-sorted directions which seems huge
but maybe there is a smart way of compressing that?

Here is the link to the article by Inigo:

http://iquilezles.org/www/articles/volumesort/volumesort.htm

All the best,

Pablo

On Sat, Aug 1, 2009 at 4:05 PM, metanet software <met...@ya...>wrote:

>
> This is a variation of the famous "isometric depth-sorting" problem: I have
> a 3D scene composed of axis-aligned boxes, viewed with an oblique/cabinet
> perspective ( http://en.wikipedia.org/wiki/Oblique_projection ).
>
> Sadly I don't have a z-buffer, so I need to determine a way to sort the
> boxes back-to-front. The boxes will all be non-overlapping, although they
> may be touching, and will move smoothly (i.e not by grid-sized steps as is
> common in some games).
>
> Is there a simple solution to this problem? I.e a formula which considers
> the extents and/or a reference point on each box to produce a back-to-front
> order? Or am I stuck with the general solution: build a BSP tree?
>
> Would it help to consider individual faces rather than boxes? (It seems to
> me that this wouldn't help, since each face can be considered a box with
> size=0 along one axis).
>
> Any suggestions or references would be most appreciated; all of the
> "isometric" stuff I've found via Google seems to be a big mess -- the
> algorithms which do work only work for objects which are all tile-sized and
> move in tile-sized increments.. which is just a special case of BSP (i.e the
> scene is essentially cut up by the grid).
>
> thanks,
> raigan
>
>
>      __________________________________________________________________
> The new Internet Explorer® 8 - Faster, safer, easier.  Optimized for Yahoo!
>  Get it Now for Free! at http://downloads.yahoo.com/ca/internetexplorer/
>
>
> ------------------------------------------------------------------------------
> Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
> trial. Simplify your report design, integration and deployment - and focus
> on
> what you do best, core application coding. Discover what's new with
> Crystal Reports now.  http://p.sf.net/sfu/bobj-july
> _______________________________________________
> GDAlgorithms-list mailing list
> GDA...@li...
> https://lists.sourceforge.net/lists/listinfo/gdalgorithms-list
> Archives:
> http://sourceforge.net/mailarchive/forum.php?forum_name=gdalgorithms-list