Path: utzoo!censor!geac!maccs!cs4g6ag From: cs4g6ag@maccs.dcss.mcmaster.ca (Stephen M. Dunn) Newsgroups: comp.sys.ibm.pc Subject: Re: Souping up my 286/10Mhz Message-ID: <25FFD0E8.7382@maccs.dcss.mcmaster.ca> Date: 15 Mar 90 17:29:12 GMT References: <25fb9dc7@ralf> Reply-To: cs4g6ag@maccs.dcss.mcmaster.ca (Stephen M. Dunn) Organization: McMaster University, Hamilton, Ontario Lines: 18 In article <25fb9dc7@ralf> Ralf.Brown@B.GP.CS.CMU.EDU writes: $Because ATPERF computes the actual speed of the clock signal fed to the $processor, while Landmark and all other benchmarks I know of compute the $"effective throughput" which is good for advertising but not much else. I can't agree with you on that. Personally, I couldn't care less what the actual clock speed of my machine is; it's the performance that counts, and that's what ATPERF does _not_ measure! Given the choice between a 20 MHz 80286 with a disgusting memory subsystem that gives 10 MHz throughput and a 12 MHz 80286 with 0 wait states that gives 13 or 14 MHz throughput, I want the 12 MHz system because it works faster, not the 20 MHz one because its crystal runs at a higher frequency. (Okay, so I'm exaggerating, but you get the idea). -- Stephen M. Dunn cs4g6ag@maccs.dcss.mcmaster.ca = "\nI'm only an undergraduate!!!\n"; **************************************************************************** "So sorry, I never meant to break your heart ... but you broke mine."