I am sure that I am missing something obvious, but how is
sqrt(144) == 12^2
False
> sqrt(144)
[1] 12
> 12^2
[1] 144
My gut is that it has something to do with precision, but curious as to understand why this takes shape.
Asked
Active
Viewed 80 times
0
-
it would be simialr to [here](https://stackoverflow.com/questions/9508518/why-are-these-numbers-not-equal) – akrun Aug 13 '19 at 14:59
-
3Maybe I'm missing something but it seems like you're asking why `12 == 144` is `False`? – pault Aug 13 '19 at 15:01
-
4Yeah `sqrt(144) == 12`, not `sqrt(144) == 12^2` – MrFlick Aug 13 '19 at 15:02
-
3This should probably have been "closed as typo", not "closed as dupe" (`sqrt(144)==12` is `TRUE`) – Ben Bolker Aug 13 '19 at 15:04
1 Answers
1
Yeah, it's precision...
options("scipen"=100, "digits"=22)
sqrt(144) * 10000000000000
[1] 1199999999999999899362824
> 12 * 100000000000000000000000
[1] 1199999999999999899362824
On R 3.5.2 on Windows 10
cory
- 6,240
- 2
- 17
- 37
-
2on what system? i get `sqrt(144)*10000000000000000000 [1] 120000000000000000000` – Ben Bolker Aug 13 '19 at 15:05
-
-
Interestingly, I get your result when I run the same in R 3.2.3 on Ubuntu 16.04.2 inside Windows 10. – cory Aug 13 '19 at 15:22
-
1OK, chalk it up to "floating point is unreliable ..." - which we knew already – Ben Bolker Aug 13 '19 at 15:23