> Was their interest piqued by my SSS work?

Yes, and, by the way I was very impressed too.

> My last technique is fraught with problems. Individual light sources have 
> to be taken into account, the model shouldn't be transformed after the SSS 
> is applied, dark areas on the target object cannot be illuminated, it 
> won't work with radiosity, and the parse times can be very high to produce 
> an effect suitable for something like human skin.

I did take a look at your macros but was unable to follow what they were 
doing. I also took a look at Wikipedia about SSS and, although I understood 
the basic explanation of the principles, I wasn't able to piece together how 
one could start implementing that in code. I should say that I'm not very 
strong on high quality textures in general.

>> Would anyone with SSS POV-Ray knowledge/assets be interested in working 
>> (with my support) on developing such a macro for the generated models?
>
> I would, but all of the afore-mentioned problems will make my last 
> technique difficult for the end-user.

ok.
It would of course be ideal to have a generic, easy to use SSS technique, 
but if that's not currently possible then it may be sensible to scale back 
the objectives. Given that we have control over the exported humanoid figure 
and quite detailed information about the model available within MakeHuman, 
is there anything extra that we may be able to generate into the exported 
model that might assist with implementing the technique?

The next option down the scale could be to work on a standard lighting and 
camera setup that would enable sample SSS images to be generated, although 
the average user may have to fall back on just the uv-mapped texture maps 
for incorporating the models into their own scenes. A small team of artists 
is currently starting work on manually crafted ethnicity, gender and age 
morph targets. A standardised SSS scene could enable a nifty set of graphics 
to be incorporated into the MakeHuman user documentation and, if we could 
incorporate something as a rendering option into exported POV-Ray models, 
this could provide users with something that they could play with and that 
may inspire them to further explore and enhance the technique in the future.

> E-mail me if you are interested in hearing my ideas.

Thanks. I will.

> Is the MakeHuman team willing to wait until POV can natively support an 
> opaque surface SSS effect?

The upcoming release of MakeHuman will hopefully be relatively soon, but 
enhancements can certainly be added to future releases of MakeHuman to take 
advantage of any parallel enhancements in POV-Ray.

Regards,
Chris B.

Post a reply to this message