4 ms·
I am the only one who sees the growing number or cores as proof that transistor density growth is sigmoid and not geometric? At least, not anymore. The fact t
by bioh42_2 16y ago
I am the only one who sees the growing number or cores as proof that transistor density growth is sigmoid and not geometric? At least, not anymore.
The fact that we have entered the sigmoid stage is kind of depressing.
- jbri 16y agoNot at all. Transistor density keeps marching on - it's clock rates that have stagnated, and they did so quite a while ago. So we're getting more and more transistors on a chip, but it's prohibitively expensive in terms of power consumption and other tradeoffs to use them to make individual execution units faster. The solution? Use those extra transistors to cram more cores and more cache on the die.
- bioh42_2 16y agoYes, it is marching on, but not nearly as fast as it used to. And it will continue to slow down. Do you disagree?
- ssp 16y agoIt's still as exponential as ever: http://en.wikipedia.org/wiki/File:Transistor_Count_and_Moore%27s_Law_-_2008.svg http://en.wikipedia.org/wiki/File:Transistor_Count_and_Moore... It will come to an end at some point of course.