We have SQRT. But what is the function for other roots - cubed ro.

[ACC]

I am looking for a function that is the reverse of POWER - in that I can
calculate the root of a number, but specify the root - cube, 4th, 5th etc.
We have a SQRT function, but I don't see anything boynd that.

 

[Jason Morin]

You have to construct it as such:

=value^(1/n)

where n = root. For example, to get the cube root of A1,
use:

=A1^(1/3)

HTH
Jason
Atlanta, GA

 

[IlanR]

Try =Your number^(1/n), n being your root
Ilan

 

[Myrna Larson]

In addition to the formulas given by others, you can also use POWER here. The
2nd argument to POWER does not need to be a positive integer. For the 5th root

=POWER(A1,1/5)

 

[Harlan Grove]

Myrna Larson wrote...

In addition to the formulas given by others, you can also use POWER here. The
2nd argument to POWER does not need to be a positive integer. For the 5th root

=POWER(A1,1/5)

....

Are there any cases in which POWER(x,y) gives a different answer than
X^(y) other than perhaps error values when both would return errors?

POWER is in the same class as CONCATENATE: both are pointless since
there's an operator that does the same thing using fewer characters and
no function calls, which can become an issue in nested expressions.

 

[Dana DeLouis]

For me, sometimes the Power function can make it a little easier to read.
ie.
=-(16)^(1/4)
returns #NUM!
which we all know to be a "issue" with Excel. (we don't want to go there
again :>) )

but =-POWER(16,1/4)
returns -2 as one would expect.
Just my opinion.

 

[Harlan Grove]

Dana DeLouis wrote...

For me, sometimes the Power function can make it a little easier to read.
ie.
=-(16)^(1/4)
returns #NUM!
which we all know to be a "issue" with Excel. (we don't want to go there
again :>) )

Well, 'we' may not if it's a rhetorical ploy in 'our' favor.

but =-POWER(16,1/4)
returns -2 as one would expect.
Just my opinion.

However, the equivalent POWER call would be

POWER(-(16),1/4)

which would return the same #NUM! error. The equivalent operator call
for your POWER call would be

-(16^(1/4))

which returns -2. If you mean POWER eliminates some ambiguity and/or
unexpected functionality from the ^ operator due to Excel's unusual
operator precedence, then fine, but IMO it'd be more useful to learn
that

-(16)^(1/4)

is a mistake in waiting [and only FORTRAN, AFAIK, distinguishes X from
(X) when X is a single variable token or a numeric constant, and then
only in function calls] which should be rewritten as

-(16^(1/4))

That is, the whole point to using the extra set of parentheses is to
overcome Excel's unhelpful operator precedence. It should be obvious
that wrapping 16 in parentheses does nothing useful.

 

[Harlan Grove]

ACC wrote...
....

We have a SQRT function, but I don't see anything boynd that.

Tangential!

SQRT is a hold-over from the bad old days when floating point math was
done is software rather than in dedicated hardware. Prior to 1990 or so
most computers didn't have floating point units/numeric data
processors/math coprocessors, so math libraries included detailed
functions to calculate logarithms and antilogarithms in order to
calculate arbitrary powers. SQRT was so frequently used in physical and
statistical algorithms that it was expedient to make it a separate
function using an algorithm specific to square roots, and therefore
much more efficient in software than the general approach.

These days when almost all computers have IEEE hardware floating point
units, there's nothing gained by having the SQRT function because
SQRT(x) and x^0.5 will make the same call to the FPU and receive the
same result from it. In fact, the SQRT call would require setting up a
call stack unless the compiler were optimized to convert SQRT(x) into
inline ((x)^0.5).