I remembered hearing people talking about this. That was the only possible theory left.

> but I am quite certain (from the last time I read the documentation) that that cache is associated with the package STASH, rather than each object, since it logically applies to the class (package) and is shared between all objects of that class.

Cool, that's what I needed to know! :)

> Arrays in Perl do have considerably smaller overhead than hashes, although not a factor of 10.

I did some tests for different sizes, yes it's not that dramatic - a 50% win is a very optimistic scenario.°

> changing the internal representation to arrays and using constant to name the fields is likely to help.

Well it's not my project anyway and some of my colleagues have weird strategies when it comes to optimization.

Personally I'm thinking rather in O() categories, and 50% best case wouldn't convince me to refactor.

I'd rather give Toby's suggestion for Insight-Out-Objects a shot.

°) Though I'm not too sure if I got the Powershell and Windows right, since all the spared "keys" should cause a bigger impact. Will need to dig deeper. :)