No, but ^ is. That's what I was saying. Depending on the version, %.
might be affected by the math library.
Henry Rich
On 1/13/2021 9:26 PM, bill lam wrote:
Is %. affected by migration to SLEEF?
On Thu, Jan 14, 2021, 9:32 AM Henry Rich <[email protected]> wrote:
The %. implementation may change from time to time, but I don't think it
did in this case.
The implementation of ^. did, and might account for the difference. You
could check.
Henry Rich
On 1/13/2021 7:56 PM, Vanessa McHale wrote:
Hi all,
I perform the calculation
2 4 6 %. 1 2 3 ^/ i.2
for the test suite of my j package
https://github.com/vmchale/j-hs/blob/main/test/Spec.hs#L76
With J 9.01 this gave (5.995204332975845e-15, 1.9999999999999971) as a
result. With J902, it gives (7.105427357601002e-15, 1.9999999999999967).
Since the exact answer is (0, 2), this is ever-so-slightly more
incorrect! Any insights into why?
(I'm on x86_64 and Linux FWIW)
Cheers,
Vanessa McHale
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
--
This email has been checked for viruses by AVG.
https://www.avg.com
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
--
This email has been checked for viruses by AVG.
https://www.avg.com
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
