Lens Diffraction tests... oooo

So I bought the "Perfect Composition" DVD from Photoshop Cafe, and finally found time to go through it. (It's Australia Day today! No work! Kinda!) There was one section about depth of field and close ups, and it got me thinking about diffraction for macro lenses. And diffraction of my lens. Several online DOF calculators recommended for a 1.5x crop dslr like my D200 to use F8, any more and diffraction steps in to soften the image. How is that in reality? So being uber not wanting to leave my room I did a quick test at iso 100. I've got the 100% crop included once you click on this image:

This was with the 16-85 at 16mm, ISO 100. I'd say while at F22 it does get a little softer, there is no reason not to use it. A little wavelet sharpening in gimp should fix that right up, and seriously, I doubt for the print sizes I work at, it'd make much of a difference. It does indicate though, that the optically best to shoot at is probably the region of F8-11. F16 shows a little softness, but it only shows up in crappy side by side tests like these. F22 again is softer, but it really isn't like "OMG someone ran a blur filter on my image" kind of soft.

At the long end of this lens, it shows the same results. But the lens is able to stop down to F32 at 85mm, and there you can see it's really getting soft, but I think in real life. I wouldn't stop down that much.

Bottom line, I found out what is my probably optimum for my handheld panos, and it reassures me that should I need to stop down for DOF, F22 is not a serious issue.

Oh, just to add. The images are converted raws from Bibble Pro; it is running a colormetric tone curve, which gives the image an extremely low contrast. I usually beging from that point for tweaking.

EDIT: Blogger seems to have scaled down my image :( But it's still good enough to see which one is softer etc. The originals were 1k+ in width! GAH!