Les Mikesell wrote:
On Wed, 2006-02-01 at 12:48, Mike McCarty wrote:This is a common misconception. Windows is often portrayed as a cycle hog. Since doing benchmarks is one of my hobbies (I dunno why), I have run benchmarks on about a dozen machines I own, with three or on some even five different OS installed. Windows is not a cycle hog.Until you try to do something... Benchmark the time to create a new process on windows vs.about anything else, or the time wasted in context switching among them.By far the slowest machine/OS combination I have is Linux (FC2) on my fastest (2.7GHz) machine. Windows XP on that same machine is noticeably faster (not just measurably faster). As an example, I just "right clicked" on my desktop, and it took five (5) seconds for the menu to pop up. Selecting "open terminal" took ten (10) seconds before first prompt. I have no unusual scripts which run at terminal startup. Windows XP is much faster in starting a console window. I just opened Open Office "Writer Word Processor", and it took thirty nine (39) seconds to initialize.You are observing disk access time and window creation time, next to nothing to do with CPU time. For a similarly 'look and feel' approach to process creation time, run something like '/bin/echo test' on a virtual console and time it on the 2nd run when the program will be in the disk cache.
I am aware of what I am measuring. Thanks for the reply.
My original point stands. Windows is not a cycle hog. Also, my
second point stands. As far as "until you try to do something",
Word starts much faster on my machine than does Open Office.
So do many other apps, like my web browsers. I think this
qualifies as "creating a new process".
Mike
--
p="p=%c%s%c;main(){printf(p,34,p,34);}";main(){printf(p,34,p,34);}
This message made from 100% recycled bits.
You have found the bank of Larn.
I can explain it for you, but I can't understand it for you.
I speak only for myself, and I am unanimous in that!
