Hello again. I am just about ready to upload a portrait I have been working on for the last few weeks and, of course, it has to be perfect :). I am trying to get a nice surface for an eyeball. This is what I have so far (attachment).  I don't know if there is a better shader for eyeball's other than the one I came up with.  Help me out guys. :)

It would help to uplaod the image lol!

Here it is.

the math node appears to be missing the two other nodes that were sposeta be connected to it.

I'd drag a wire from image to texture and set to a small # .008 (inches). This will break up the gloss. Anistropic might be a better node for that. You can scale and offset anistropic ...
Reflection can help too. Low grade though. Eyewhite and cornea only.

What about putting a color_math node into alternate spec, with a antstropic node in one value and a glossy node in the other. (Should I put an edge_blend node in one of the math function values?)

Most of the math nodes won't actually do anything on their own & have to be combined with at least one other node to have any effect on the image.

The dangling math node is computing the number 2 (Add 1 and 1). You hardly need a math node for that - you could just put 2 in the Diffuse_Value.

I suspect you started with an eye shader that had a bit more to it and you deleted some? You're questions about putting and Edge_Blend in is a hint.

If you were to implement Fresnel reflection (which you should if you want realism) you'd put an Edge_Blend into Reflection_Value and a math node subtracting that Edge_Blend from 1 (1 - EB) into Diffuse_Value. This is to satisfy the conservation of energy law (diffuse + reflect + refract <= 1) as well as the Fresnel law (reflection decreases when surface points towards the observer).

And as others say - a little bump is needed, otherwise you get a shiny billiard ball look.

Glossy is the best node for this. I know many use anisotropic, but not for the right reason at all.

Here's the background on the aniso thing.

Most people put transparency on the cornea, instead of using refraction.

In poser there is confusion about whether "transparent" means "not there" or means "clear". By this I mean, does transparency SKIP the material or not? For diffuse reflection it absolutely does skip it. Plug any color with any strength into diffuse along with 100% transparency and you won't see ANYTHING from that surface. However, for specular, you will see something but it is drastically reduced. This makes no physical sense, but that's how it was implemented. Specular reflection on a 100% transparent surface is drastically reduced.

Now people want bright shiny speculars on a cornea, but because it is 100% transparent, they don't get bright speculars, but dull dark ones. The answer, which seems to occur to almost nobody, is to compensate by setting the Specular_Value to an appropriately high number to compensate. Numbers like 3 to 6 work.

What some people stumbled onto is using the Anisotropic node instead. The reason is that in its default settings, the anisotropic node produces specular highlights that are about 6 times brighter than those produced by the Specular or Glossy node. But that is not because it is for eye reflections. That is because it is for simulating reflections on a grooved surface, such as brushed steel, or nylon stockings. You are supposed to spread all that light out along one of the dimensions. I dont't have Poser in front of me, but there are U and V scale parameters that smear the highlight out over a long thin area. But if you leave those scale parameters tight and small, then the large amount of highlight gets concentrated in a tiny area, thus producing an unusually bright highlight. Combine that with the automatic reduction in brightness caused by the mis-implemented 100% transparency, and you get something close to the desired amount of highlight.

However, if you simply put a 6 in the Glossy node's brightness or value (whatever its called), you can compensate for the transparency darkening, WHILE GETTING AN ACCURATE HIGHLIGHT.

What I mean is that the anisotropic highlight shape or contour is wrong for a wet, glossy surface. The rate of fall-off of that highlight is not accurate. The contour or shape or rate of falloff coming from the Glossy node is precisely the one you want for a "glossy" surface.

I hope this is making sense. 

Now to perfect the realism, you really don't even want to use transparency on the cornea at all. You want to use Refraction and Reflection and a Glossy node, all controlled in such a way as to obey the laws of physics. Just adding those three nodes doesn't get you realism - they are components in the final simulation but there is more to it than that. That's where the Fresnel effect comes into play - the variation in reflection due to the angle of the surface normal with respect to the camera.

Please have a look at this tutorial I did:

The Fresnel Effect - Glass, Paint, Plastic, and Metal** 

**The ideas presented there are directly applicable to eye materials. You want to use the "paint" like shader for the sclera (eye whites) and the "glass" like shader for the cornea. On the iris and pupil, juse use Diffuse_Color = your color map image, and nothing else - no specular.

And don't forget the bump on the sclera. Of course there is no bump on the cornea. For sclera bump you can use the noise-based nodes, or a supplied bump map that comes with your figure. I suggest you do not use the color map for bump. The color map will produce variations, but not the right shapes, and not in the right places. You'd do better to use a noise-based node (such as Turbulence) for sclera bump than to use the color map.

Attached Link: http://www.renderosity.com/mod/forumpro/showthread.php?thread_id=2052365&page=2#message_2058469

Here's something I found from '04/P5 that was interesting. Anistropic *is* inaccurate but on a cornea,,, it really doesn't matter as the UVmapping(cornea) limits its travel. A hot spot on or near the pupil will get you realism easily. Perhaps a gloss on eyewhite.. as this is where anistropic falls short.

A bumpmap is better but,,, a wire from texture will get you there with little draw on time. If it's not a closeup, you won't know the difference. Really though,, if this is about realism then you have to consider sss. Milk in a glass ball for eyewhites and refraction over the irises so they turn as if under a lens. Specmaps, too.

That is a lot of work. Especially in Poser...

Thanks for this background info, bagginsbill.

BB, I have read that tut on the freshnel about a dozen times. :) I didn't think of using those principles on an eye surface. I came up with this last night, someone suggested using a reflection map, although I probably have some values incorrect.

I added the math node simply to brighten the texture. It was really dark without it.