Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : Intel Corporation (INTC) -- Ignore unavailable to you. Want to Upgrade?

To: fingolfen who wrote (127192)	2/12/2001 1:20:43 PM
From: Dan3	Respond to of 186894

Re: It seems like most benchmarks these days are cooked up by gormet chefs with an axe to grind... I agree. That's why the only benchmarks I trust are ones that predate the processors being compared. Such benchmarks also make more sense since they generally better reflect the software that the overwhelming majority of CPUs actually run (most benchmarks are released running the very newest software, to reflect the code that will be actually run by most CPUs during their productive lifetimes, code released 1 to 2 years earlier is more appropriate) So winstone98, Sysmark98 (or whatever Bapco called it then), winstone99, sysmark99, winstone2000, and other benchmarks that were released a year ago or earlier, are much more likely to provide an unbiased presentation of actual performance (they were likely skewed to accomodate PIII, and so will be "honest" when comparing Athlon to P4). And, on virtually all such benchmarks, Athlon substantially outperforms P4. Unless you carefully select which portions of the program's code are executed most often by the benchmark, to use primarily that small subset of X86 instructions instructions it doesn't choke on, P4 almost always loses. P4 may be a nice processor if you primarily run code you develop yourself in assembler, but it's a lousy CPU for running industry standard X86 software. There are a handful of instructions that Motorola's Power PC chips does very well at, as Apple likes to point out. As long as you are allowed to design a benchmark that "takes advantage" of its "advanced features", a 500MHZ Mac can Beat a 1GHZ PIII. Which is almost as nonsensical as the thought that P4 is competitive with an Athlon. Dan