I'm a little confused about this...
On Ron Shimek's web site under the DSB section there is a figure
that suggests a DSB particle distribution size that "should work well".
According to Rob Toonen in this thread
the particle distribution of Southdown differs from this.
If you break it down so as to compare the numbers in like particle sizes I think it goes something like this:
Ron's Site Recommendation:
Greater Than 1/2 mm: 15% ...