Post a Comment
We periodically hear announcements like this one from various chip manufacturers, and at the end they result in absolutely nothing. I don't see how this one will be any different given the overwhelming amount of technical explanation on how they will suddenly make things 1000 times better.
Perhaps they licensed some of the technology from this gentleman?
http://www.osnews.com/comment.php?news_id=11784
i agree. and knowing the fact that the cpu power usage is not the bigest battery eater in a laptop, i think even if they decrease power usage of the cpu with 99.9% (1000 times smaller, yes) the whole laptop will only hold out slightly longer. on the other hand, they can increase clockspeeds and diesize this way, and create a more powerfull processor that uses as much or less power than the current Pentium M.
and THAT's interesting. much faster cpu, same power or lower.
i think even if they decrease power usage of the cpu with 99.9% (1000 times smaller, yes)
They're not claiming anything of the sort anyway.
The reduction was referring to leakage only. While that dominates when the processor is in sleep mode, leakage is fairly insignificant compared to the power consumed by switching transistors when the processor is busy.
> It's marketing. They didn't claim 10000 times better processor,
> or 10000 times less power, they claimed 10000 times less leakage,
> but leakage is probably insignificant anyway.
leakage is no way insignificant
With every shrinking step, leakage current increases exponently.
Leakage current is problematic because it increases as transistor threshold voltages (Vth) decrease. The move to the 130-nanometer technology node led to a significant rise in leakage power consumption. At 90 nanometers, leakage power can represent as much as 50 percent of the total power consumed by a chip.
If only they (Intel) woun't play dirty in other areas, I would be even more interested:
http://www.swallowtail.org/naughty-intel.html
But thats just me trolling...
-iMoron
I'm not sure if the inquirer web site is down or not, but I thought a better link may be Intel themselves:
http://www.intel.com/pressroom/archive/releases/20050920comp.htm
A billion is a thousand million. A trillion is a million million.
Depends where you are: http://en.wikipedia.org/wiki/Billion.
A British billion used to be 10^12, but as with many other things they're giving in to the Americans on that.
Intel is American of course, so we're talking 10^9 transistors.
I love these things. How long will it take to Intel until they claim they have supressed quantum tunnelling?
On the battery side, I'd prefer if someone made a screen as bright as a desktop one with less consumtion than today's laptop LCDs.
BTW: Does anybody know how important is tunnelling in current CPU's?
It clearly says 1000 times (1.000), not 10000!
Yes, the articles all linked to say 1,000. However, the leakage current reduction was given as 100nA to 0.01nA, which is 10,000. The Intel PR guy is the one who first said 1,000 which is what all the articles are quoting. I guess Intel's PR guy doesn't bother to double-check his figures with engineering. 



