Re:Ask HN: What is likely to improve with 10/7nm CPUs?
by DSingularity on Monday July 30, 2018 @11:30AM
(#17645107)
Not much for you I’m afraid. L1 size likely to not change significantly to maintain access latencies. Frequencies are not going to increase because of power. Core capabilities will not change I’m any significant ways either. The best bet is more LLC.
[ Parent
| on HN ]
Re:Ask HN: What is likely to improve with 10/7nm CPUs?
by zamadatix on Monday July 30, 2018 @11:34AM
(#17645151)
Don't smaller transistors reduce power consumption?
Re:Ask HN: What is likely to improve with 10/7nm CPUs?
by choudanu4 on Monday July 30, 2018 @11:38AM
(#17645183)
Yes, but for the previous few generations of transistor manufacturing, transistors' power consumption has not scaled down as well as their size. This is known as the failure of Dennard Scaling[0].
Re:Ask HN: What is likely to improve with 10/7nm CPUs?
by gsnedders on Monday July 30, 2018 @11:37AM
(#17645178)
Power leakage is an increasingly large problem nowadays. (Because yes, semiconductors are nowadays suffering from observable quantum effects.)
Re:Ask HN: What is likely to improve with 10/7nm CPUs?
by IshKebab on Monday July 30, 2018 @11:39AM
(#17645201)
Yeah but power consumption increases with something like the 4th power (heh) of frequency so there's not much you can do even with really low power transistors.
Re:Ask HN: What is likely to improve with 10/7nm CPUs?
by calvinlh on Monday July 30, 2018 @12:13PM
(#17645539)
According to this article on Wikipedia, it's a linear relationship: https://en.wikipedia.org/wiki/CPU_power_dissipation
Edit: maybe this takes into account that when you increase frequency you might also have to increase the voltage to keep the processor stable, as well? But a 4th power effect seems quite extreme.