[Fluxus] avoiding glitches and sphere normals

[Kassen]

On 19 February 2010 16:45, Evan Raskob [lists] <lists at lowfrequency.org>wrote:

>   Of course, Dave has his city/Traffic project going on, so it seems that
> building virtual worlds for performances is all the rage nowadays :)
>
>
>
And there is space for many more!

Everyone will have his own perspective which is what makes it all worth it.
I had a big fight with perspectives tonight, trying to work from two at once
and separating normal translations, etc, from  operations on the
transform-matrices to keep them separate. It could probably be cleaner still
but this now works.
http://www.flickr.com/photos/41408203@N02/4381614698/

The interaction between the two is really quite interesting. For a while I
tried to get to spot/state on the sphere-surface with positive "z" pointing
away from the sphere's origin, with x&y oriented according to the
orientation of the field. That turned out to be a bad idea... I'll never use
the word "non-euclidean" casually any more :-).

Cheers,
Kas.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.pawfal.org/pipermail/fluxus-pawfal.org/attachments/20100223/8be5e0ed/attachment-0001.html>