The Glossy Roughness Magic Number

Snuggins

0.0376059468835592235203

If you copy that number and paste it into the  "Glossy Roughness" surface settings, and hit enter, you will (still) see light sources reflected in the surface.

But this is as low as you can go. If you paste the magic number in again (since the GUI only shows "0.0376059"), but change the "3" on the end to a "2" this time and hit enter, then the light sources disappear! So will any light-emitting surfaces on the other side of a Section Plane.

For convenience you could use this already-changed-to-a-2 value:

0.0376059468835592235202

This is the make-the-light-sources-disappear number. Maybe that makes it the magic number, not the first one. Up to you.

Again, for those of you that only use light-emitting surfaces for lights, not those old point/spotlight etc light sources, your surfaces will remain visible, but not if they are on the other side of a Section Plane, a nice place to hide them. Ghost lights are still ghosted, no real change there, opacity is not affected. Environment maps are still reflected in the surface even if "Draw Dome" is off, no matter what you do to Glossy Roughness.

You can keep using 0.1 or 0.05 or whatever works for you. It just always bothered me that "0.0" wasn't what I expected-- the least rough setting on a slider between 0.0 to 1.0 where 1.0 was obviously the most rough. I guess when that happened traditional (if photometric) lights were involved. But why do you need to add a little roughness to make it less rough? This is not an explanation really, just a description (albeit overly detailed) of the point above which Glossy Roughness makes sense. Maybe this helps. Or maybe it just sets you off to write a treatise on why this is obvious or why I completely missed something right in front of my face. I would like to know, but I'm sorry for wasting your time writing that when you could be laughing and dancing with the butterflies. I should let you know I tend to disappear for awhile after big long posts like this, which is bad form, and I am sorry for it. I won't realize how wrong I am and how right you are for awhile. Others are standing by to appreciate your corrections though.

Is this test scene I used so simplistic as to make this information worthless? Could be. I didn't even describe it fully, but in its bare form is just a mirror and a light source and a camera using iRay Uber surfaces and seeing what happens in iRay preview mode. I didn't try rendering to disk, maybe that matters.Could it be an artifact of some factor I overlooked? Sure. I am fully expecting responses like "hey dummy, just move the light source a little closer. You found an unbelievably useless number." (Moving it a little closer doesn't work, unless you move a light-emitting surface over a Section Plan boundary. This happened to me actually and I was about to throw it all away. I hadn't tried the traiditional lights yet.)

Wish I had found a cool value like "e" to the "i" or the sine of the derivative of the light equation, or some multiple of Pi, or equally impressive-sounding value. Nope, no Nobel Prize for this one. I just got what looks like a random collection of digits out after the first 3 that will be debunked soon anyway or explained easily, or more likely, ignored since it doesn't change daily life much.

You are of course dying to know that I played with other surface settings, distance between mirror and camera and/or light source, and absolute positions of them as well in case there is some distance from origin thing going on. Don't think so, but I could have easily missed something. Being in the zone doesn't mean you are in a good zone. Blind spots abound when you are zoomed in on a problem, at least for me.

Of course, you can certainly degrade things to make the visible (>=threshold GR) invisible, but I don't think you can make the invisible (<threshold) visible again by doing anything. That might be too strong a statement but who wants to hear a litany of maybes. Prove me wrong! (I realize the strength of the statement is undermined by all the maybes that immediately precede it. Ouch.)

I was using a primitive point source light with its luminance set 1500000 (add 3 zeros to the default.) Same for the light-emitting surface. Amazing fact: that's out of order. The light emitting surface had that value first! Haha! Then and only then did I set the light source luminance to match! It was in a different location as well -- not a very well controlled experiment at first huh? So many hijinks, and that's just one! (Actually I thought none of this mattered, since I usually only use environmental and light-emitting surfaces and had a Section Plane on my camera with a light-emitting ring light just behind it. Then I tried a point light. That turned into nearly the whole story. No you can't have those 20 seconds back, sorry. I am insensitive, and you risked it.)

Did I mention I set Metallicity to 1? Yes, but not when I originally asked that question. Hope things still make sense. I should have made it shorter and sweeter and clearer but I did get that nice really long number there at the top. Concentrate on that! Not too much though.