Diffusion Solved?

[mawyatt]

Note the artifact around the bottom left hand corner of the top half.

The bluish ring does not look like any artifact that I am familiar with. Can you show us what that looks like in the source images?

The top half is polished smooth, while the bottom half has a brushed finish.

There is a balancing act here. If you put a mirror ball inside an integrating sphere that is uniformly bright except where the lens sticks through, then what you'll see is a uniformly illuminated circle with a dark spot in the middle of it. That's because at every angle the reflection shows either the integrating sphere or the black hole of the lens. But uniformly bright means that all the modeling goes away -- the mirror ball no longer looks like a ball.

So, to get a good rendering of the mirror ball, what you need is illumination that varies enough with angle to provide modeling, while not concentrating so much light in any one area that you get unpleasantly bright highlights from reflecting that.



--Rik

Here's a couple examples of tiny 100 micron diameter spherical mirrors on chips, called solder balls.



Best,

[ChrisR]

To check to what exrent you have "equal light from everywhere", I'd suggest something approaching a glass sphere as a test piece. Practical items could be
dark colored buttons (as linked earlier)
A dome-headed nut
ball bearing
glass headed pin - dark colors work best
something shiny from the kids jewelry section of a dollar store.
They're quite revealing!

[Dalantech]

...It could probably use a bit more diffusion or moving the lamps away a little more.

Actually the opposite is true -you need to get your diffuser closer to the subject. See the Apparent Light Size article at Strobist. That helped me to solve a lot of my flash photography issues.

You also want to make sure that whatever you're using as a diffuser is actually forcing the light to spread out (and not just block it) and that it's not adding an odd color cast that's gonna be a pain to edit in post. Looks like you're doing studio work shooting inanimate objects, so the duration of your flash might not be too important. I like to keep my flash duration as short as possible just to make sure I can freeze as much motion as possible. The misconception is that the flash duration, no matter how long, will always be short enough to freeze motion and it's not true. A lot of the image softness that gets blamed on diffraction is really motion amplified diffraction. Doesn't look like traditional motion blur, but it will rob you of detail none the less.