Peter Popov wrote in message <7m8igs024ej0nm5t62hpv4r1ob538qv9bt@4ax.com>...
>On Thu, 27 Apr 2000 01:15:06 -0400, "Mark Wagner"
><mar### [at] gtenet> wrote:
>
>>Current progress is that it is almost complete.  However, I'm running into
>>the problem of oversaturation of colors near the horizon.  Consequently,
the
>>image looks just fine -- until you get to the horizon.  At that point, the
>>sky turns white.
>
>Well, maybe you could fake it by decreasing brightness by an
>exponential factor. Not quite realistic, I guess, but it will look as
>expected and that what usually counts.


The problem is figuring out how to do this while computing the media.  It
would be much easier to do this after the image is rendered.

>>I'm waiting for the "film effect" post-processing to be implemented to see
>>if this will fix the problem.  If people want me to, by next Thursday I
>>could release a version that doesn't have this problem fixed.
>
>This would basically take the same approach, right? I mean, applying a
>nonlinear curve to input/output brightness to desaturate the upper
>levels.


Yes, but as I said above, it is easier to implement.

Mark

Post a reply to this message