Thanks, Steve!
That post was more work than it looks like. The hard part was getting the data into a format and file-size that I could post. It's one of those things that's a piece of cake once you know how, but a piece of something else until then.
In retrospect, I don't think the modelling program was at fault after all. I must have not read the data correctly - I was probably comparing absolute SPL in the bass region instead of comparing the levels relative to the midband efficiency. Anyway when I re-ran the program this morning and carefully compared the relative levels I could see a pattern similar to what I observed, though the increases were still less than what I measured.
Duke
