[Bf-cycles] Global SSS Scale Multiplier

[Matthew Heimlich]

It could work well as a per object parameter, or as a per material
parameter as long as it applied to all SSS shaders in that material.

Another issue with SSS scale that I've noticed while working on this shader
is that we can't input very small numbers. For example, one layer of the
SoG skin uses a variance value of .0064mm. Right now the only way to do
numbers this small is to have both a very small RGB radius, and a very
small scale. Any chance we could have the option to have the radii values
store more than 3 places after the decimal?


On Mon, Aug 19, 2013 at 5:50 PM, Brecht Van Lommel <
brechtvanlommel at pandora.be> wrote:

> I'm not sure about this, in my opinion this is a setting that should be in
> the shading nodes. It seems to me somewhat arbitrary that you would have a
> global setting for this. If it's really needed it could be an object or
> material setting, but I would rather not have it.
>
> In the sample .blend that I linked to in my commit, there is a node group
> with a global scale using some math multiply nodes, I think that works ok?
> http://www.pasteall.org/blend/23501
>
>
> On Mon, Aug 19, 2013 at 11:11 PM, Matthew Heimlich <
> matt.heimlich at gmail.com> wrote:
>
>> Since we now have a proper Gaussian falloff for the SSS shader, I've been
>> playing around with implementing the sum of Gaussians technique for
>> rendering skin as explained in GPU Gems 3 and this year's SIGGRAPH talk
>> "OSL The Great and Powerful". It's very fast to set up and is so far giving
>> nice results. However, it uses 6 layers of SSS data, each with its own
>> individual scale, some of which are very small numbers. Changing these
>> scales one by one proves to be tedious when trying to get the scale of the
>> material to match up with the scale of the object to which it is applied.
>> I'd like to propose a global SSS multiplier, perhaps in the scene panel,
>> that would allow us to quickly find the sweet spot for SSS shading,
>> especially when using multi-layer setups. It would be especially useful for
>> the SoG method where certain values are measured real-world levels, but
>> where modeling scale may not be in real world units. Any thoughts on this
>> matter would be appreciated.
>>
>> Cheers,
>>
>> Matt Heimlich
>>
>> _______________________________________________
>> Bf-cycles mailing list
>> Bf-cycles at blender.org
>> http://lists.blender.org/mailman/listinfo/bf-cycles
>>
>>
>
> _______________________________________________
> Bf-cycles mailing list
> Bf-cycles at blender.org
> http://lists.blender.org/mailman/listinfo/bf-cycles
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.blender.org/pipermail/bf-cycles/attachments/20130819/0365e97f/attachment.htm