Re: [Plib-devel] sending objects to hell

[Steve B. <sjb...@ai...>]

> Dave McClurg wrote:
> 
> In tuxkart/src/herring.cxx i noticed that when a herring is eaten
> it is subsequently sent to hell.  (hell means z=-1000000).

Yes - that's how it works.

> Is this
> preferred over removing the herring from the scene graph?

Not really :-)

I lifted the herring code more or less unchanged out of
Tux_AQFH.  Bear in mind that quite a bit of TuxAQFH was
written BEFORE SSG - and subsequently ported onto it!

> Is the culling in ssg efficient enough to quickly throw this
> out without much processing?

Yes. I'm really rather proud of the efficiency of field of view
culling in SSG.

There are rarely more than a dozen herring in a single level - so
even if it was inefficient, it wouldn't make a dent in performance.

There is also an argument of the form:

  * When they ARE visible (before a certain gluttonous
    Spheniscidae Linuxia gets to them), not having to
    test whether they are eaten or not saves time.

  * There are generally more uneaten herring than
    eaten....certainly in TuxKart where they reappear
    within a few seconds of being eaten. So it's
    better to optimise the case when they are NOT eaten.

But that's not why I do it...it's just ancient history.

-- 
Steve Baker   HomeEmail: <sjb...@ai...>
              WorkEmail: <sj...@li...>
              HomePage : http://web2.airmail.net/sjbaker1
              Projects : http://plib.sourceforge.net
                         http://tuxaqfh.sourceforge.net
                         http://tuxkart.sourceforge.net
                         http://prettypoly.sourceforge.net