Mac benchmarks: Historical data & Moores Law

[Supersonic]

I was in the middle of a discussion with some friends recently about the much mentioned "Moores Law" and the increases of microprocessor speed every 18 months or so, and wondered if there are any Mac benchmark figures floating around to match up early mid-80's systems (Mac 128k, Mac SE, etc...) up to the current day MacPro, etc to prove my point.

I.e, if a Mac 128k system had a value of =1, where would Mac II, '040 systems, PPC and Intel kit lie in comparison???

Its been almost 25 years since the launch of the Mac 128k in 1984, so has performance followed Moores Law in that time?

Cheers in advance!

 

[JNB]

"Moore's law" has nothing to do with speed, but the number of transistors on a chip. And it was doubling those every two years, not 18 months.

So looking at it from that perspective, you had 68,000 transistors in the original Motorola 68000, and 820-900 million in today's Intel C2D's.

So, we have actually exceeded the original 1965 estimate, but not by a huge margin time-wise.

 

[Supersonic]

"Moore's law" has nothing to do with speed, but the number of transistors on a chip. And it was doubling those every two years, not 18 months.

So looking at it from that perspective, you had 68,000 transistors in the original Motorola 68000, and 820-900 million in today's Intel C2D's.

So, we have actually exceeded the original 1965 estimate, but not by a huge margin time-wise.

Good point...whilst Moores Law is as you mentioned (transistor density) the cmmonly held perception is that it equally means processing power. If we have now nearly a billion in the current C2D processors, does the power it offers correlate to 13,250x as much?

Do CPU benchmarks bear this out too??

 

[theBB]

Good point...whilst Moores Law is as you mentioned (transistor density) the cmmonly held perception is that it equally means processing power.

Actually it is not even transistor density, just the number of transistors on a chip. Some of the gains come from larger chips.