I think the fourth parameter contains criteria...

In E2168:E2169, E2168 should contain one of the column heading or column
number in the table
and E2169 should contain the condition...

In second formula you have used E2169:E2170 which makes E2169 as the column
heading or number...

I think that is the problem... I prefer to use E1:E2 for one condition F1:F2
for second and so on
avoiding overlapping ranges...



"herman" wrote:


I try to encourage the use of tables and tablenames in Excel2007.
However I get very strange behaviour using tablenames in dbfunctions like
DSUM,DMAX etc.

During a course I noticed that

=DBMAX(tblTop5[#Alles];"pos2006";E2168:E2169) gives correctly 1890
=DBMAX(tblTop5;"pos2006";E2169:E2170) gives #WAARDE! ---- Why?
=DBMAX(tblTop5;1;E2170:E2171) gives correctly 1890
=DBMAX(C2160:D2165;"pos2006";E2170:E2171) gives correctly 1890


Using the range-adresses always works fine.
When I use the formula-step-by step in the erroneous case , I notice the
tablename gets replaced by "C2160;D2165" instead of "C2160:D2165"!

Why does the second formula fail?
I don't understand - I think it should work.

(sorry - only Dutch version of Excel available for me, so I gueess DBMAX is
DMAX in the English version)