By: Chris Brodersen (cb.delete@this.example.com), September 19, 2012 11:54 am
Room: Moderated Discussions
Mark Christiansen (alialiasundercover.delete@this.nospam.net) on September 19, 2012 11:31 am wrote:
> There is a graph at the bottom of page two which would seem to capture the heart
> of the power performance trade off. I think it would be much better if both
> frequency and power were displayed on the same kind of scale. Attempting to
> relate logarithmic frequency with linear power leaves me puzzled.
Agreed. Also, the interesting figure of merit is actually nJ/cycle. 2mW/3MHz is 0.66 nJ/cycle, 17 mW/100MHz is 0.17 nJ/cycle, while 174 mW/500 MHz is 0.348 nJ/cycle. (And 737 mW/913 MHz is 0.8 nJ/cycle.)
> There is a graph at the bottom of page two which would seem to capture the heart
> of the power performance trade off. I think it would be much better if both
> frequency and power were displayed on the same kind of scale. Attempting to
> relate logarithmic frequency with linear power leaves me puzzled.
Agreed. Also, the interesting figure of merit is actually nJ/cycle. 2mW/3MHz is 0.66 nJ/cycle, 17 mW/100MHz is 0.17 nJ/cycle, while 174 mW/500 MHz is 0.348 nJ/cycle. (And 737 mW/913 MHz is 0.8 nJ/cycle.)