Re: DiskBrowser Software - Work In Progress

[David Schmenk <dschmenk@YUCH.gmail.com>]

Michael J. Mahon wrote:
> aiiadict@gmail.com wrote:
> > On Sep 2, 12:15 pm, "Michael J. Mahon" <mjma...@aol.com> wrote:
> > > Agree on all points--now if it would just terminate.  (And, no, it
> > > wasn't doing anything--that *I* asked it to do.  ;-)
> >
> > It will on my machine if you give it a few minutes
>
> Wow!  A few *minutes*?!  That's equivalent to not quitting in
> my book!
>
> What takes so long?  (Task Manager kills it pretty quickly!  ;-)
>
> I just tried it again, and ten minutes after clicking on "close", it's
> still running and responding to mouse clicks...  Since it won't quit,
> I'm quitting.  ;-)
>
> -michael
>
> AppleCrate II: An Apple II "blade server"!
> Home page:  http://members.aol.com/MJMahon/
>
> "The wastebasket is our most important design
> tool--and it's seriously underused."


Well, Java first has to characterize the execution codepath through the 
quit methods.  The optimal native opcodes must then be determined by 
fractal/random sequences until it runs faster than equivalent C code. 
Heuristics must then be applied to keep the working set in the cache 
(but probably gets it wrong and thrashes constantly).  See, a few 
minutes isn't so hard to believe after all ;-)

Dave...