I have always had very unsatisfactory results from the "extra fine" side of my diamond stone (DMT Duosharp Fine/Extra Fine w/ Hardcoat, W8EF-H-WB.) It is supposedly "1200 mesh / 9 micron" (green in DMT's color code) but it never seemed to provide the results people claim it should online. The edges it produces feel very rough to the fingernail, and I always have to go to sandpaper afterwards to get anything that looks flat instead of ragged under a loupe.
So, I did a test today, flattening the back of a 1/4" chisel on the extra fine side of my stone (stone was cleaned beforehand), and then doing the same on a 1000 grit piece of 3M imperial wet or dry sandpaper spray-adhesived onto plate glass. I believe that the sandpaper is using the CAMI grit system which means 1000 grit is ~9 micron, but even if it was using FEPA-P that just means it would be even coarser at ~20 micron or so.
As far as technique I am holding the chisel flat on the stone/sandpaper and stroking back and forth with light pressure, holding it at a 45* angle to the axis of motion, lubricating with plain water.
The difference is pretty large to me -- the diamond stone produces a very noticeably scratched surface whereas the sandpaper produces a surface that is MUCH smoother. It's not exactly a mirror surface but it is starting to show a cloudy/blurry reflection.
(This difference in surface quality is similar to the difference in edge quality I experience. It is just much easier to photograph a flattening test than the edge itself. When you run a fingernail along the edge produced by the stone you can feel how rough it is!)
I understand diamond stones would cut more aggressively than silicon carbide, but this just seems like a wild difference. I can think of a lot of reasons why I might be getting this result, but first I want to ask, is this really what a 1200 mesh diamond stone should be doing to my tools?
Comparison of 1200/Extra Fine (left) and 600/Fine (right) on either side of a 2 inch/51mm putty knife for comparison: