I have it somewhere, but I'd have to look it up. So does anyone know the refraction number of thick see through glass, like a window?

http://hyperphysics.phy-astr.gsu.edu/hbase/Tables/indrf.html there are a few glasses listed here. I'm not sure which is the one you want.

Refraction on windows is a waist of time. It for the effect to look right there is practically no refraction but it will add hours to your render and make for a very noisy final render. I just use a simple Poser root with transparency and reflect. it renders quick and looks like it should. Bleder Guru talks about this is this video but has a more elaborate node setup. https://www.youtube.com/watch?v=KyXRBu7gn2o&t=5s

Actually, I'm having trouble looking through some glass material. Depending on the angle, the glass is reflective instead of see-through.

That can have something to do with the geometry if you are using a refractive material on it. I think all the refract and PBR stuff needs geometry that mimics the real life object or it won't render right. Glass windows have to be modeled like thin boxes instead of one sided squares.

the top window is a thin cube and the bottom is a one sided square with normals facing the wrong way. Either way that one sided square produces strange results with refract.

I made them into really thin boxes. But I did start off with one sided squares. Reflects either way. Like I said, though, it seems to depend on from what angle you're looking through the glass.

EClark1894 posted at 5:56PM Sat, 02 May 2020 - #4387975

I made them into really thin boxes. But I did start off with one sided squares. Reflects either way. Like I said, though, it seems to depend on from what angle you're looking through the glass.

then try this:

https://marmoset.co/posts/basic-theory-of-physically-based-rendering/

Read the bit on Fresnel in particular - reflectivity of a surface will change depending on the angle of the light falling onto it.

ghostship2 posted at 5:21AM Sun, 03 May 2020 - #4387978

EClark1894 posted at 5:56PM Sat, 02 May 2020 - #4387975

I made them into really thin boxes. But I did start off with one sided squares. Reflects either way. Like I said, though, it seems to depend on from what angle you're looking through the glass.

then try this:

Can't. Just modeling this for now in Blender, so I haven't ported it over to Poser yet.

fair enough. You could always try that node setup that Blender Guru talks about in the video.

Actually, I'm having trouble looking through some glass material. Depending on the angle, the glass is reflective instead of see-through.

You mean...like glass?

Keith posted at 5:21AM Tue, 05 May 2020 - #4388175

Actually, I'm having trouble looking through some glass material. Depending on the angle, the glass is reflective instead of see-through.

You mean...like glass?

No. like a mirror.

Earl you're speaking very strangely. Like you are not in agreement that glass (and all smooth materials) becomes 100% reflective when viewed at a grazing angle from the outside, and 100% reflective when viewed at a lot of angles from the inside.

For purposes of computer graphics, the "inside" is the back side of a polygon. If you're looking at a one-sided prop (not a proper box but just a sheet of polygons that isn't a closed volume) then you will see 100% reflection from a lot of angles. https://en.wikipedia.org/wiki/Total_internal_reflection

EClark1894 posted at 7:36AM Tue, 05 May 2020 - #4388187

Keith posted at 5:21AM Tue, 05 May 2020 - #4388175

Actually, I'm having trouble looking through some glass material. Depending on the angle, the glass is reflective instead of see-through.

You mean...like glass?

No. like a mirror.

But Keith's point is a glass is a mirror "depending on the angle" - your observation is simply confirming for all of us that it is acting "like glass" - period.

This guy explains it well https://www.dorian-iten.com/fresnel/

"I was blind to the Fresnel Effect until someone pointed it out to me — now I can see that it is everywhere! If you’re looking for it, you’ll find it."

You appear to be blind to it, while we were all talking about it for 15 years now.

You might also be using too few bounces - to pass through the glass and see what is on the other side. So you get black for the refracted rays. The appearance can be as though it's just a mirror. You really should show us what you're doing and seeing because your words simply tell us "it is glass" - nothing odd about anything you said.

Sorry, BB, but in my experience, I've noticed that not all glass reflects at 100% regardless of the angle. Some reflection, sure.

EClark1894 posted at 10:48PM Wed, 06 May 2020 - #4388203

Sorry, BB, but in my experience, I've noticed that not all glass reflects at 100% regardless of the angle. Some reflection, sure.

Observe the glass door on the right side of the image. Sure not 100% reflection, but nothing reflects at 100% anyway, so there's that.

Gee, I was right.

I was going to let it go but here we are again.

You're not right - total internal reflection is called total because it is literally 100%. It is not a theoretical phenomenon, but a real one, demanded by physics. It isn't just for idealized surfaces, but any surface where the transition is from higher to lower index of refrection (i.e. speed of light).

I gave links - you could read them and respond with the actual words to show that you know what we're talking about.

But more than that - your problems with the glass shader are unknown to us because all the words you've used are just describing how glass works.

Keith posted at 8:28AM Thu, 07 May 2020 - #4388388

EClark1894 posted at 10:48PM Wed, 06 May 2020 - #4388203

Sorry, BB, but in my experience, I've noticed that not all glass reflects at 100% regardless of the angle. Some reflection, sure.

Observe the glass door on the right side of the image. Sure not 100% reflection, but nothing reflects at 100% anyway, so there's that.

It would be 100% at 90 degrees - this image is far short of that angle.

Lots of real-world glass products exploit total internal reflection to make them work. The most prominent are the prisms in binoculars and fiber optic cables which rely on TIR to deliver the signal through curved glass tubes.

https://courses.lumenlearning.com/physics/chapter/25-4-total-internal-reflection/

"A good-quality mirror may reflect more than 90% of the light that falls on it, absorbing the rest. But it would be useful to have a mirror that reflects all of the light that falls on it. Interestingly, we can produce total reflection using an aspect of refraction.

Consider what happens when a ray of light strikes the surface between two materials, such as is shown in Figure 1a. Part of the light crosses the boundary and is refracted; the rest is reflected. If, as shown in the figure, the index of refraction for the second medium is less than for the first, the ray bends away from the perpendicular. (Since n1 > n2, the angle of refraction is greater than the angle of incidence—that is, θ1 > θ2.) Now imagine what happens as the incident angle is increased. This causes θ2 to increase also. The largest the angle of refraction θ2 can be is 90º, as shown in Figure 1b.The critical angle θc for a combination of materials is defined to be the incident angle θ1 that produces an angle of refraction of 90º. That is, θc is the incident angle for which θ2 = 90º. If the incident angle θ1 is greater than the critical angle, as shown in Figure 1c, then all of the light is reflected back into medium 1, a condition called total internal reflection."

Here's a scene I set up to demonstrate. There is a prism in the center, floating over some colored squares. There's a green bar behind it, and surrounding it all is a forest environment. For this render, the prism has a thin-glass shader on it to ensure you are able to understand where it is in the scene.

This is the basic Cycles/SuperFly shader for solid glass.

And here the camera is looking straight at the prism. Everything you see within its interior is 100% reflection. 100% - total internal reflection.

If I decrease the index of refraction, as here I lowered it to that of water instead of glass, then only SOME of the surfaces are doing total internal reflection. The back diagonal surface angle of incidence is now below the critical angle. It is allowing some light to go through to the other side. We can now see somewhat straight through the prism, via refraction.

Optical glass, as is used in binoculars and submarine periscopes, is made with a higher IOR and as a result gives a very wide field of view with total internal reflection.

This is not a fantasy - any argument that this isn't happening is like arguing the world is flat.

For thin slabs of glass the same shader produces a very different effect. The parallel front and back surfaces conspire to avoid total internal reflection.

But even the external reflection rises with angle of incidence. I have marked the rotation angle of each slab in degrees.

Note that I used a volume absorption here because now it matters. When we look in the end face of the slabs, the internal reflection counts go well over 100 and we have to take into account the long path that the light took through all that glass, from one end to the other. As well, I had to increase my bounces to 128 to get the ends to show like real life.

If you're trying to look at 88 degrees and above on a door or something around your house, forget it. Because of the foreshortened viewpoint, you need a very very long slab of glass to be able to see clearly what is happening. You don't have one long enough. In my scene, the top slab at 89 degrees is over 20 feet long.

You also have to stand far away to have a somewhat consistent angle across the whole slab. Otherwise the closest part of the slab will be very much below 89 degrees.