[PD] Bug in spot_light?

[james tittle]

On Dec 9, 2005, at 11:19 AM, Frank Barknecht wrote:
> *However*: I now have a working patch, it's attached and was extended
> to also incorporate the attenuation/cutoff/decay messages. It works
> with current Gem-CVS, only the debug-cone is slightly wrong of course.
> It would be nice, if that cone would also display the cutoff-angle
> btw.

...I've fixed the debug-cone problem:  [spot_light] was initializing  
to the wrong direction vector...also added checks to the light params  
so that they don't cause gl to throw errors for out of range:   
glerrors are silent, and at best just cause the calling function to  
be ignored, but at worst they can screw up glstate and user  
expectations...

...according to the spec:

linearAttenuation should be positive
spotCutoff should be 0-90 or 180 ("special value")
spotExponent should be 0-128

> [spot_light] allows for some really nice effects btw, for example if
> you crank up the colors to "out of range" values like [2 3 0(, then it
> gives a surreal feel.

...yeh, neat kinda saturation along the edges...I've attached a  
modified version of your help patch to work with the new, correct,  
initial direction, and will commit it to cvs later when it's  
prettified ;-)

james
-------------- next part --------------
A non-text attachment was scrubbed...
Name: spot_light-help.pd
Type: application/octet-stream
Size: 3497 bytes
Desc: not available
URL: <http://lists.puredata.info/pipermail/pd-list/attachments/20051210/8bd17e88/attachment.obj>