At the Allaire CF Developers Conference in DC in 2000, they ran a side by side comparison on identical Compaq computers comparing CF6 (called Neo then) against CF 4.5. The script found all the prime numbers between 1 and 100,000 or something like that. What I do remember is that CF 4.5 took something over 12 seconds and Neo took like 4.5 seconds. Now, it's possible that the test was Neo vs. CF 5 but I can't remember for sure. I do know that CF 5 had not been released yet.
Dave ----- Original Message ----- From: "Ben Johnson" <[EMAIL PROTECTED]> To: "CF-Talk" <[EMAIL PROTECTED]> Sent: Wednesday, July 24, 2002 6:49 PM Subject: RE: ColdFusion MX performance brief now available > > I was comparing the dual Linux to the quad Windows setup. > > I'm not a systems expert, but perhaps someone else can confirm this. As far > as I can remember from what I have read, 4 500Mhz processors do not equal 1 > 2000Mhz. The computing cost to push certain tasks or threads to certain > processors can take quite a bit of cycles. Take a look at the data for the > Win2K boxes. The pages per minute with one processor is half of the number > of pages with four processors. Once you start changing the number of > processors, you add all sorts of factors such as how the SMP is handled and > what not. > > Although I would be interested to see a comparison between Linux and Win2K, > I personally don't think it really matters. If somebody knows Linux, stick > with Linux. If you know Windows, stay with that. A couple hundred extra > page hits shouldn't warrent switching over your entire system to something > different. > > > > Ben Johnson > > ______________________________________________________________________ Get the mailserver that powers this list at http://www.coolfusion.com FAQ: http://www.thenetprofits.co.uk/coldfusion/faq Archives: http://www.mail-archive.com/[email protected]/ Unsubscribe: http://www.houseoffusion.com/index.cfm?sidebar=lists
