On Thu, 2003-09-11 at 14:41, Darrell Blake wrote: > > my 500Mhz laptop will be faster than the 2.5Ghz > > I have to disagree. It's a physical impossibility for a 500MHz machine to run > faster than a 2.5GHz machine. Windows doesn't use up than many resources. If > you're planning on running it in a CLI then there's no base for comparison > because you would have to run Windows in a CLI too in which case the 2.5GHz > would still be faster. It seems to me that you're trying to compare a 500MHz > CLI machine with a 2.5GHz GUI machine which isn't a really fair trial. > But if you are benchmarking tasks, and Microsoft Windows only allows a particular task to be executed within a clunky bit of closed-source code, with little (or Microsoft only peer review), and that task happens to also include worthless code for drawing windowing by virtue of the fact it's windows... then it's bound to be slower than some well developed code that's been open to lots of (harsh?) peer review, without needless extensions to update widgets on a GUI (with transparency, fading and all those useless effects).. and code that may be optimised heavily towards a specific processor.... then there could be a huge speed increase. Benchmarking is incredibly difficult to do reliably, but just because Windows (of late anyway) is tightly integrated into a GUI, doesn't mean that Linux also has to use a GUI to run it's task in it's way... benchmarking should be about a specific task using the best available on each platform- not crippling Linux because of a limitation within Windows (having to run a GUI). -- Regards, Adam Allen. adam [at] dynamicinteraction.co.uk pgp http://search.keyserver.net:11371/pks/lookup?op=vindex&search=adam%40dynamicinteraction.co.uk
Attachment:
signature.asc
Description: This is a digitally signed message part