cspear,

You do realize you made that a glowing shader, right? Turn off all your lights and render it.

cspear, that rendered image looks like it could have come out of the video game Fallout 3! That looks great, IMO.

Quote -

You do realize you made that a glowing shader, right? Turn off all your lights and render it.

BB,

I do realise that. It has to be adjusted for different light conditions - any suggestions of where else to stick it (!) are welcome.

Quote - For lake water, change the Diffuse_Color to brown. The deeper the lake, the darker the brown should be.

time out. time out BB. this looks way to real to be from poser. he he :) 

Quote - If I change the airplane to use a chrome shader, now I need two bounces. There are places where the water reflects the plane which reflects the sky (two bounces in one chain) and there are places where part of the plane reflects another part of the plane which reflects the sky (two bounces). In these places, since I rendered with one bounce, the contribution from the second bounce is missing. That means the water reflects a black airplane, and the airplane reflects a black airplane.

2 bounces

1 bounce

1 bounce ( if you know that people wont notice the difference)

Quote - Here is the water shader I was using in these latest renders. It is for deep water, such as ocean or lake, where we cannot see anything under the water.

This is not a gamma-correcting shader, and so I would only use it in Poser Pro with PPro render GC enabled, as I've done here.

The scale of the Fractal_Sum is going to depend on the scale of your prop. I used the ground plane here at a scale of 4000%. If you use some other prop, or use the same prop at a different scale, you will need to adjust the Fractal_Sum scale parameters accordingly.

The color I chose for Diffuse_Color, and also the value of Diffuse_Value, was chosen to match a typical ocean color. It matches the photo I'm using here. If you look near the horizon, you see real water in the photograph. My ground plane 3D water shader is matching it almost perfectly.

this is you new way for calculating the fresnel effect right? 
is there any site where we can dial in the settings for different materials that ahve fresnel effect? 

Yes it is a new way to approximate, with good accuracy, the Fresnel equation using only 4 nodes. The full equation takes many more than that.

However, there is no intuitive way to know what values to use in this node setup. I used an interactive graphic tool to find a curve that matches the Fresnel curve for a particular IOR. (This one is not the IOR of water, but I didn't bother with detail - it's good enough.)

I meant to publish this fast Fresnel approximation in a for-pay shader, but I forgot and showed it to you guys for free. Oh well. :)

dont worry. if you will  realese some shaders for money then belive me we will pay for it.

can i ask what settings do we need to change? the edge-blend white and black colors? 

and could you a little explain why you used the blender node? and 0,4. and what does the POW node do? 

thank you very much.

watching
learning

Bagginsbill

can you a little explain why you used the blender node? and 0,4. and what does the POW node do? 

thank you very much.

Also interested... I thought I saw an explanation of how you arrived at this setup somewhere, but I can't find it now.

 This one deserves a bump

I never answered the questions about how I arrived at the approximation because I already answered. I have a program that graphs functions in real time. It has the ability to graph multiple functions, and use parameters that I can click on to adjust the functions.

I wrote the real Fresnel equation, and I wrote the approximation, and I adjusted the parameters until they pretty well matched. I tried many approximations before arriving at the formula that results in those 4 nodes. It is the smallest set of nodes I can find that is within 1% accuracy across the whole range of values.

Every now and then I open this program up and I work on the approximation some more. I have better parameters than before, now.

I used the Blender node because it implements an expression involving 3 arguments that happen to be a good part of the formula I came up with. Blend(a, b, f) implements:

(1-f)a + fb

The way I'm using it, b is 1, so this reduces to:

(1-f)*a + f

The value I used was .04, but that's not actually the correct number for water. It is the number for glass. I was lazy and just copied my glass shader to do the water.

You want the best numbers I currently have for water?

The graph shows three functions. Red is the real Fresnel equation. Blue is the approximation. Purple shows 10 times the difference between the approximation and the real thing.

On the right are the formulas. Formula y1 (in blue) is the approximation. It uses three parameters, b, s, and p, and another formula, y4.

y4 is actually just the cosine of the angle of incidence and is implemented by the Edge_Blend node in the shader.

Functions y4, y5, y6, y7, and finally y8 implement the true Fresnel equation for unpolarized light. The last one, y8, is what is graphed in red.

Function y9, the purple one, is calculating 10 times the difference between the approximation (y1) and the real equation (y8).

The parameters are at the bottom.

a = the desired index of refraction
b = the blend value for the approximation
s = the subtract coefficient value for the approximation
p = the power value for the approximation

The blending value, b, can also be thought of as the "base" value - i.e. it controls the lowest value produced when the angle of incidence is lined up with the surface normal, i.e. when the difference between these angles is 0.

For IOR = 1.54 (glass) I came up with the values shown. For the water, I just decreased b a little bit, but that was simple laziness on my part.

I have better values for glass, now, as well.

b = .0186
p = 60
s = .104

Put .0186 into the Blender blending value.

Put 60 into the Pow node Value_2.

Put .104 into the Subtract node Value_2.

b = .0424
p = 90
s = .0634

VEry very interesting to achieve realism in render, but just a little question, how i determine b p and s if i dont have your software ?

Quote - VEry very interesting to achieve realism in render, but just a little question, how i determine b p and s if i dont have your software ?

Well, I don't mind coming up with parameters for other IOR values, so just ask.

Or, you could have the software. It's free. It's called GRAPES and is here:

http://www.criced.tsukuba.ac.jp/grapes/

Once you have it, I will give you the GRAPES document that has my setup in it for finding fake Fresnel settings and you can play with it yourself.

Oh yesssssssssss !

I come back cause i'm really interested to get lot of realism on my renders and cause i have a little pb, it seem the formula ( i know it's an approximation) dont like the ior under 1 like gold who is 0,47, impossible to match the real fresnel, but for ice or beer it work very well !

That .47 value for Gold came from the BlenderArtists forum and it's wrong.

All metals have IOR above 50!!

In Grapes, enter a=50 and see what the red (real Fresnel) curve does. It becomes highly reflective at all angles. This is why metals are metals - they have a very high IOR. Some are above 100.

[Edit] The subject of IOR for dialectrics is simple, but for metals, it gets complex, literally. Meaning, the IOR is a complex number involving a real part and an imaginary part.

In practice, the reflectivity of metals based on viewing angle is nearly constant, and does not really need a Fresnel effect to be convincing.

Oh my - I just found this - "Schlick's Approximation of Fresnel" on Wikipedia. I swear, I never saw this before I made my approximation. It is the same form! Isn't that amazing? This Schlick fellow and I solved the same problem the same way. This is one of the arguments why math formulas cannot be patented. We do not "invent" them, they exist already - we simply reveal them.

However, mine has different coefficients and is much more accurate.

http://en.wikipedia.org/wiki/Schlick%27s_approximation

Thx, i send my source ...

http://www.frenchcinema4d.fr/forum/index.php?topic=22562.0

Anth,

Yes, the same list is published in CGSociety and other places. Read the comments on many of these. Each time I find this list, underneath are people questioning the metals. They complain that typing in these quoted values for gold and silver into programs such as Maxwell, Maya, 3DSMax, all produce nonsense. And so it is because these numbers are nonsense, and represent only part of the IOR of metals.

It takes two numbers to specify the IOR of a metal, the real and the imaginary part. And these two numbers are used in a more generalized Fresnel equation that solves for reflectivity based on the frequency of the light, not just the angle of incidence.

Render Settings are same as in page 1, BB post "Posted Fri, Apr 3, 2009 10:15 am, Edited Fri, Apr 3, 2009 10:16 am"  (checked and double checked)

I was using a scaled up ground plane, which starts out pretty small. Since the waves scale with the prop, I had to shrink the waves.

If you're putting that same scale wave on a prop that is already large, they will come out very dense.

You need to adjust the scale on the node(s) making the waves.

Yea, I was trying scaling, but I was scaling the bump along with the Fractal Sum node, and it was getting me just rougher seas, when I was looking for a lake.

Scaled the fractal sum up, and left the bump at 1, and now I got it, or at least good enough for me  :)  Good ol' trial and error!