PDA

View Full Version : BSDF, nodes and IOR

Pavlov
07-02-2013, 04:01 AM
Hi,
i think i've asked this before but i didnt get there yet.
Question is: how do you guys manage to control Delta/Sigma nodes ?
A common BSDF node should control Fresnel (change of transparency and reflection at 0 and 90°) with index of refraction. I.e. Maxwell's materials are like this, and they are incredibly powerful.
We still seem to be forced to use gradients, which are fake.
Does someone knows a smart setup to manage the entire fresnel thing (both transparency and reflection change) with a simple IOR value ?

Paolo

RebelHill
07-02-2013, 04:07 AM
use the fresnel node.

Pavlov
07-02-2013, 04:24 AM
I did, but what if you have other nodes plugged into Delta's specular ? Multiplying the outputs of previous node flux with Fresnel keeps the physical accuracy of IOR, or is there a better way to "insert" Fresnel node into the flux ?

Thanks,
Paolo

Pavlov
07-02-2013, 04:33 AM
And again, question is: normally i can chose a specularity of 30%. This means material is 30% reflective and 70% diffuse. If i plug Fresnel node in specular, behavior changes with incidence, but i'm not able to set a starting reflection level anymore.
Does IOR imply automatically that reflection amount is managed by incidence ? Again, Fresnel node seems to output something different from the value i put in, like if it precalculates something. How does it work exactly ?

thanks
Paolo

RebelHill
07-02-2013, 05:35 AM
The IOR controls the % of reflection. You dont set any "base" amount (strength of reflection)... fresnel IOR represents the strength of reflection at different angles of view.

Pavlov
07-02-2013, 05:39 AM
So its the common fresnel curve, which ranges from 0 to 100% and change the in-between values.. thanks, i should have thought of it :)
Paolo

RebelHill
07-02-2013, 05:56 AM
Well no... it ranges from x to 100%. What x is (and the rate of change across the angles) is determined by the materials IOR.

Pavlov
07-02-2013, 07:36 AM
Ah. Interesting, i assumed minimum value was 0, and jumped to high values fastly with high IOR values.
Maybe IOR curve display in FRYrender, when it still exhisted, fooled me :)

thanks
Paolo