log(-0.5)^2 no math error?

[pier4r]

Few minutes ago while tinkering with my trusted El 506W (sharp) I got a surprise.

To my knowledge there is no way log(-0.5)^2 is admissible (in R at least), I checked with a couple of other calculators and it seems I remember correctly.

Despite that I got a value back (-0.602...): https://i.postimg.cc/tCpycV6K/photo-5348...3316-y.jpg

Do you know any other calculator that have such subtle quirks despite being generally reliable? (HP35s excluded) Today is the first time I see my EL-506W answer me unreliably (I have it since 2003 ! Maybe it is simply old like me)

edit: apparently playing a bit more with the Sharp, it seems that it is a quirk of the syntax. log(x)^y is not the same of (log(x))^y while on other systems it is the same.

[Thomas Klemm]

It appears that \(\log\left((-0.5)^2\right) = -\log(4)\) is calculated.
I'm a bit surprised that you are using double parentheses: \(\log((-0.5))^2\).
What happens with: \(\log(-0.5)^2\)?
Does this calculator maybe supports this notation: \(\log 3\)?

[pier4r]

yes, playing around it seems a very counterintuitive notation (at least, if one is not aware of it)

https://postimg.cc/gallery/qyn3tvp

I always used log(x) never thinking that the power would have applied only to x, but it is the case.

The double parentesis sometimes is added when the user selects the negative sign.

[Albert Chan]

It seems Sharp does input in 2 phases. Number entering is separate from the rest.

TOP LINE: log (_
2nd LINE: -0.5

[x^2] [=]

TOP LINE: log ((-0.5)^2=
2nd LINE: -0.60206

(07-07-2022 07:25 PM)Thomas Klemm Wrote:  What happens with: \(\log(-0.5)^2\)?

[log] [+/-] .5 [x^2] [=]

TOP LINE: log( -0.5)^2=
2nd LINE: -0.60206

Display is all messed up.