Re: Smoking/Flaming Effects

New Message Reply Date view Thread view Subject view Author view

Marcus (Marcus++at++multigenuunet.UU.NET)
Tue, 8 Nov 1994 16:08:08 PST


        Reply to: RE>Smoking/Flaming Effects
>Subject: Smoking/Flaming Effects
>Date: Tue, 08 Nov 94 16:10:31 -0500
>From: Len Granowetter <lengrano++at++mak.mak.com>

<snip>

>One approach that we are considering is using MultiGen models of
>effects, that can be converted (via LoadFlt) into (perhaps)
>pfSequences of pfBillboards, for instance, a sequence of many
>different flaming textures. Since they are on a pfBillboard, they
>will always face the viewer, and since they can be inserted into the
>scene graph, you only need to know the position of the effect in scene
>coordinates.

This approach should work for axis billboards. I just did a quick
check and it behaved as I expected. Note that point billboards
aren't supported in the loader (yet) and that animation sequences
are automatically started by the OpenFlight loader with:

 pfSeqTime( seq, PFSEQ_ALL, 0.1 );
 pfSeqInterval( seq, group->flags & 0x40000000 ? PFSEQ_CYCLE :
                         PFSEQ_SWING, 0 , PFSEQ_ALL );
 pfSeqDuration( seq, 1.0, PFSEQ_ALL );
 pfSeqMode( seq, PFSEQ_START );

These actions are taken before the loader callback is invoked.

Information about sequences can be gotten via the loader
callback. It'll be a group bead (case CB_GROUP) and a pfSequence
node if the GROUPcb.flags are either:

 0x60000000 == pfSequence with interval PFSEQ_SWING
 0x40000000 == pfSequence with interval PFSEQ_CYCLE

Other runtime parameters for time, interval, and duration
can be entered in the group bead comments as you choose.
Reset the pfSequence as desired before your loader
callback function returns. Position the animation and
enable/disable it from you main loop as appropriate.

>What I am wondering is whether this is a common approach. The fact
>that pfusmoke does this in a completely different manner, and Angus
>Henderson's flaming code also does this in a completely different
>manner, leads me to believe that the "use models" approach must be
>flawed. Is it very inefficient? Lead to worse visuals? Is this a
>feasible approach?

I don't know what the quality/performance of the effect
will be compared to the other approaches ...

<snip>

> Len Granowetter
> MaK Technologies
> (617) 876-8085

Regards,
Marcus Barnes, Member Technical Staff
MultiGen Inc., 1884 The Alameda, San Jose CA, 95126
PH: (408) 261 4118 FX: (408) 247 4329
EMAIL: multigen!marcus++at++uunet.UU.NET


New Message Reply Date view Thread view Subject view Author view

This archive was generated by hypermail 2.0b2 on Mon Aug 10 1998 - 17:50:39 PDT

This message has been cleansed for anti-spam protection. Replace '++at++' in any mail addresses with the '@' symbol.