SteveDB1 wrote...
Well, I'll tell you what Harlan, if you don't want to believe me, I guess
that's your own problem. . . .

Not exactly a problem. And I don't believe you're consciously lying. I
believe you don't know what you're doing and you don't know how to
report your results accurately. In other words, incompetence, not
dishonesty. I hope that makes you feel better.

. . . But to satisfy your obvious ignorance, . . .

Well, one of us is ignorant.

. . . give me your
email-- one you feel safe with-- and I'll put together a sample worksheet
from a blank workbook, and email it to you. This way you'll have that "live
example" you mentioned.

Use Google Groups and you can see my e-mail address. You'd just need to
click on the ... link in my e-mail address then follow the directions.

I've got better things to do than be insulted by someone who obviously has
no understanding of what I'm trying to accomplish.
....

You believe you're the first person posting here who's had problems
with data type mismatches in SUMPRODUCT formulas, do you?

I know what you're trying to accomplish, and I also know why what
you've tried doesn't work, and I have a shrewd guess why you can't
accept the answer unless someone rubs your face in it, and even that's
unlikely to work.

That MS is, or would be unwilling to resolve an issue that affects all of
their customers . . .

It only affects the pigheaded one who refuse to learn how Excel works.

That you'd act as if I'd insulted your integrity by saying that the example
you'd given did not work tells me that you've taken this far too personally,
when it has/had absolutely nothing to do with you is equally disappointing.

What example did I give? =(12345="12345") returning FALSE? It does. If
it returns anything else on your system, then perhaps your copy of
Excel is corrupted. That's the only example I gave. Or did you mean
formula?

I will however not let this go, and expect a more patient person than you to
deal with this issue.

Yeah, so many have done so already, haven't they? And having seen that
you've been given the usual (and correct) answer to this problem but
that you refuse to accept it, they're just dying to try a warner &
fuzzier approach.

There is nothing wrong with the dataset that is being used. The other
engineer that I'd initially mentioned in my first post to Roger and myself
have both opened completely blank workbooks to run our own tests on this and
have found glaring inconsistencies-- it will work with one row, but not the
next. It then may work with a few rows, and then not work with others.
....

OK, so there are two of you who have no clue.

All we're seeking is a consistently acting "formula." That you'd pull a
semantics response is pathetic. Formulas, equations, and algorithms are all
the same thing-- a mathematical statement designed to elicit an answer. At
least have the decency to answer my question instead of playing language
games with me.

Terminology is important, at least to people who understand it.

Formulas, equations and algorithms are NOT the same thing. Formulas, in
the spreadsheet sense, are like assignment statements in programming
languages. As for algorithms, how would one implement binary search as
a series of equations?

So, as I said, since you believe yourself to be infallible, send me an email
address and I'll get you a workbook with a sample dataset, and the
equations/formulas/algorithms.

Figure out how to use Google Groups, and you'll have my e-mail address.
If you're an engineer, you should be able to accomplish that.