They should - remove the memoized versions (obviously it's faster) - show the executed instructions for each language of the main loop (which would be a nice

They should - remove the memoized versions (obviously it's faster) - show the executed instructions for each language of the main loop (which would be a nice exercise with all the vms and dynamic languages)