[Soc-2018-dev] Weekly report #08 - Implementing a Hair Shader for Cycles

[Brecht Van Lommel]

On Mon, Jul 9, 2018 at 1:17 AM Leonardo E. Segovia <
leonardo.segovia at cs.uns.edu.ar> wrote:

> I've just uploaded charts for the remaining properties. These are all
> done in Inkscape now, so translations should not be an issue
> (obviously, I'll upload the sources once you approve their look). As
> regards the credits, I'll clean them up ASAP.
>

There is a translation system for the Blender manual, and translating text
in SVG files is not part of that, images are the same for all languages. So
text inside images is still to be avoided.


> The parameter explained in the first part of 4.3 is named
> "Glossiness" in V-Ray, which is appropriate since that's the intended
> use. Wouldn't that be a better name? I agree with the remapping, and
> will update that.
> The second part (i.e. how to actually simulate undercoat) is a
> roughness modifier applied to those hairs in certain parts of the
> surface. This may be achieved by manipulating the relevant sockets
> from an UV node, so I don't think we need to do further work there.
>

Glossiness is often defined as 1 - Roughness, so to me that's confusing.
V-Ray uses Glossiness that way in a lot of their shaders, where we use
Roughness. I prefer Coat to indicate it's meant as an additional layer on
top of the main one, and a bit of consistency with Clearcoat in Principled
BSDF which has a similar function.

I'd still like your word on the new charts I pushed (Roughness,
> Roughness Randomization and Undercoat Roughness). Please let me know
> if there are any improvements to make, as I'll standardize the Melanin
> charts with them.
>

* Remove all the text from the image except the number (maybe in a smaller
font), which also gives more space for the renders.
* Show only 5 values, or even 3 where it's enough, with 11 like roughness
randomization it's much too small to see anything.
* For roughness it may be better to go from 0.1 to 0.9 or so, no need to
show the extremes and noisy renders.

Regards,
Brecht.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.blender.org/pipermail/soc-2018-dev/attachments/20180709/0859bdb3/attachment.html>