Yes, that's a problem. You have to have some datasets that you consider
typical. I usually try on random, ascending/descending, and somewhat
repeated data.
In this case, where the performance depends on the instruction mix in
user code, I just guess based on the J code I have written.
Henry Rich
On 7/29/2019 10:33 AM, Raul Miller wrote:
Which reminds me of an interesting question:
How do you adequately benchmark this kind of optimization?
Thanks,
---
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
