[
https://issues.apache.org/jira/browse/NUMBERS-156?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17349732#comment-17349732
]
Matt Juntunen commented on NUMBERS-156:
---------------------------------------
bq. I still think that that it can be useful that an application is able to
switch implementations
That would be nice but I personally don't have an interest in pursuing this
right now. My thought is just to have a single, accurate (and hopefully
performant) implementation, similar to how the JDK has a single {{Math.hypot}}
implementation.
On another note, if we use extended precision for the Euclidean norm, we should
probably do the same for the Manhattan norm. Perhaps we could pull out some of
the dot2s methods and create a {{Summation}} class, similar to
{{LinearCombination}}?
> SafeNorm 3D overload
> --------------------
>
> Key: NUMBERS-156
> URL: https://issues.apache.org/jira/browse/NUMBERS-156
> Project: Commons Numbers
> Issue Type: Improvement
> Reporter: Matt Juntunen
> Priority: Major
> Attachments: performance-all.png, performance-len-1-5.png,
> performance2-1-5.png, performance2-all.png, performance3-1-5.png,
> performance3-all.png
>
>
> We should create an overload of {{SafeNorm.value}} that accepts 3 arguments
> to potentially improve performance for 3D vectors.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
