Just figured out how to integrate bump-mapping into SSLT:

Post a reply to this message

Attachments:
Download 'sslt_bumps.png' (420 KB)

Preview of image 'sslt_bumps.png'

Am 08.01.2013 05:44, schrieb clipka:
> Just figured out how to integrate bump-mapping into SSLT:

I guess I should mention that this, once again, is a patch to POV-Ray. 
(Not scheduled for 3.7.0, unfortunately.)

Post a reply to this message

clipka <ano### [at] anonymousorg> wrote:
> Am 08.01.2013 05:44, schrieb clipka:
> > Just figured out how to integrate bump-mapping into SSLT:
>
> I guess I should mention that this, once again, is a patch to POV-Ray.
> (Not scheduled for 3.7.0, unfortunately.)

What do the various spheres represent? various scattering amounts?

I don't really see any bumpy surface in the part receiving more light,Maybe the
picture is too small. I do see scattering occur, but this is already the case in
current version, maybe the fact that there is normal perturbation at grazing
angle is what you mean? would it be possible to see it with less fresnell
falloff, so that the actuel lit part gets some normal bumps?

I would love to try it for character skins (but i'm patient). Maybe this means
that slope effect could be used for some peach fuzz as well?

Post a reply to this message

Am 15.01.2013 16:58, schrieb Mr:
> clipka <ano### [at] anonymousorg> wrote:
>> Am 08.01.2013 05:44, schrieb clipka:
>>> Just figured out how to integrate bump-mapping into SSLT:
>>
>> I guess I should mention that this, once again, is a patch to POV-Ray.
>> (Not scheduled for 3.7.0, unfortunately.)
>
> What do the various spheres represent? various scattering amounts?

Yes. Leftmost one is without subsurface scattering.

> I don't really see any bumpy surface in the part receiving more light,Maybe the
> picture is too small. I do see scattering occur, but this is already the case in
> current version, maybe the fact that there is normal perturbation at grazing
> angle is what you mean? would it be possible to see it with less fresnell
> falloff, so that the actuel lit part gets some normal bumps?

I'm afraid I don't know what you're /not/ seeing there. The spheres 
obviously don't look perfectly smooth, do they?

Post a reply to this message

clipka <ano### [at] anonymousorg> wrote:

> I'm afraid I don't know what you're /not/ seeing there. The spheres
> obviously don't look perfectly smooth, do they?

I guess it's the particular shader or lighting context, then: it just looked to
me as if the bright part with SSS was like media rather than a surface finish,
as if specularity was interrupted in that area. Then again it's probably
accurate, sorry. I have a few more question that I hope can be related:

-will this solve the problem of bright edges with SSLT giving characters a
wireframed edge? (maybe by allowing SSLT to really consider smooth a meshe's
edge when its normals are specified as such)

-Will the modifications allow to do some UVmapped tangeant normal perturbations?
(maybe using the slope map with a syntax to say that channels RGB corresponds to
XYZ/UVW)

Post a reply to this message

Am 18.01.2013 15:03, schrieb Mr:

> -will this solve the problem of bright edges with SSLT giving characters a
> wireframed edge? (maybe by allowing SSLT to really consider smooth a meshe's
> edge when its normals are specified as such)

Not directly; but while working on the bump-mapping support, I also 
tackled the related problem of those bright wireframe edges. (See my 
latest posts in the "Young Teens 5" thread for a sample.)

> -Will the modifications allow to do some UVmapped tangeant normal perturbations?
> (maybe using the slope map with a syntax to say that channels RGB corresponds to
> XYZ/UVW)

No. Such a feature is on my to-do list, but I haven't tackled it yet.

Post a reply to this message

clipka <ano### [at] anonymousorg> wrote:
> Am 18.01.2013 15:03, schrieb Mr:
>
> > -will this solve the problem of bright edges with SSLT giving characters a
> > wireframed edge? (maybe by allowing SSLT to really consider smooth a meshe's
> > edge when its normals are specified as such)
>
> Not directly; but while working on the bump-mapping support, I also
> tackled the related problem of those bright wireframe edges. (See my
> latest posts in the "Young Teens 5" thread for a sample.)
>
> > -Will the modifications allow to do some UVmapped tangeant normal perturbations?
> > (maybe using the slope map with a syntax to say that channels RGB corresponds to
> > XYZ/UVW)
>
> No. Such a feature is on my to-do list, but I haven't tackled it yet.

You're simply the best ! :)

Post a reply to this message

clipka <ano### [at] anonymousorg> wrote:
> Am 08.01.2013 05:44, schrieb clipka:
> > Just figured out how to integrate bump-mapping into SSLT:
>
> I guess I should mention that this, once again, is a patch to POV-Ray.
> (Not scheduled for 3.7.0, unfortunately.)

no prob, you're still awesome

Post a reply to this message

Am 19.01.2013 12:30, schrieb Mr:
 > You're simply the best ! :)

I don't know about being the best, but...

Am 20.01.2013 02:35, schrieb nemesis:
> no prob, you're still awesome

... yeah, sometimes I do think I am :-)

Thanks guys. Let's not forget the rest of the dev team though, who are 
doing an equally awesome job in the background.

Post a reply to this message

On 20-1-2013 6:08, clipka wrote:
> Am 19.01.2013 12:30, schrieb Mr:
>  > You're simply the best ! :)
>
> I don't know about being the best, but...
>
> Am 20.01.2013 02:35, schrieb nemesis:
>> no prob, you're still awesome
>
> ... yeah, sometimes I do think I am :-)
>
> Thanks guys. Let's not forget the rest of the dev team though, who are
> doing an equally awesome job in the background.
>

Absolutely! And all in their free time too! Thanks guys!

Any girls involved? if so: thanks girls! ;-)

And not forgetting wives/companions and children who have to sacrifice 
some free time too...

Thomas

Post a reply to this message