
Originally Posted by
SMOKEU
... I've just been studying CPU architecture as part of the degree I'm doing and there's a hell of a lot more to performance than just clock speed and cache.
Yes.
I'm not arguing against that at all.
(And for the record, it's quite possible I've been "studying CPU architectures" (to varying levels of depth at different times) longer than you've been alive.)

Originally Posted by
SMOKEU
The best way to "measure" performance is to go to sites like overclockers.com and tomshardware.com who do many hardware benchmarks using different benchmarking tools, so the end user can make an informed decision in relation to which hardware best suits their needs. ...
I totally agree with that, too.
{strikeout}All I'm saying, is that for Joe Average, that kind of thing just isn't going to be done. Mavis the Granny down the road isn't the type to go to overclockers.com. So some basic measure doesn't seem so bad. It's not totally accurate. Sure. But what (within the grasp of Joe Average) is?{/strikeout}
Actually, the main reason I mentioned this issue at all, was I just don't get why people have a such a strong negative reaction to the whole issue. What's the big deal? If someone is going to do the kind of research you suggest, then they'll understand the clock x cores "measure" is (partly) bollocks anyway. It all seems a minor thing to get worked up about.
Measure once, cut twice. Practice makes perfect.
Bookmarks