Am 14.12.2016 um 06:07 schrieb omniverse:
> "[GDS|Entropy]" <nomail@nomail> wrote:
>> clipka <ano### [at] anonymousorg> wrote:
>>> Am 13.12.2016 um 22:15 schrieb [GDS|Entropy]:
>>>
>>>> The Fog is being a little weird, because Media without a container is only
>>>> visible when looking at the plane below it, and it vanishes above the horizon.
>>>> ?! Weeeird. Very counterintuitive.
>>>
>>> Doesn't sound like something that should happen.
>>>
>>> Care to share any details?
>>
>> Sure. :)
>>
>> The plane in there seems to have no effect, commented or no, which seems to be
>> improper based on what Alain was saying. Here is a minimal scene which
>> reproduces the effect being discussed.
> 
> 'Official' and 'alpha' versions render differently so something changed, I guess
> you're not using official 3.7? I get the black void in 3.7.1.1-alpha, while
> official 3.7.0 shows oversaturation of media as would be expected.

Actually, no, that's not what is happening.

Official 3.7.0 shows oversaturation simply because media sampling method
3 had always been buggy, showing somewhat wrong results wherever the
actual number of samples is low (such as in uniform media); this has
been fixed in 3.7.1.

With media sampling methods 1 and 2, POV-Ray has always (since 3.6.2 at
any rate) shown the very same symptoms.

To demonstrate that 3.7.0 sampling method 3 behavior /cannot/ be
correct, just notice how the oversaturation disappears close to the
horizon. With proper computations we should expect exactly the opposite
effect.

Post a reply to this message