Are switch-case statements optimised for up to 10 cases?

[pawel.b...@gmail.com]

Hi,

I noticed a performance cliff when using switch case statements longer than 10 cases. I did a bit of poking around and noticed that different instructions are used to handle cases above the 10th. Until the 10th its "JumpIfTrue" and above "JumpIfTrueConstant".

Two questions:

 - are people aware of performance optimisation where switch-case statements might deteriorate after 10 cases?

 - might those different bytecode instructions be causing this?

Thanks,

Pawel