New GA144 Info

The GA144 data sheet has been updated at GreenArrays. They now include instruction timing and some interesting power consumption info. One graph shows execution time, power consumption and energy used for a single instruction as a function of temperature. Interestingly the power consumption and energy used goes up quite dramatically with temperature, about 30% at 70C and over 140% increase at 125C!

I'm not familiar with a mechanism that would cause that. I am aware that leakage increases dramatically with temperature, but I didn't realize the dynamic current was also a strong function of temperature. I can't imagine that the authors would not have corrected for leakage since they are only measuring the dynamic power of a single node and the static power would be for the whole chip. If the static power is not subtracted out it would greatly skew the data for applications using all the processors. But even so I don't think the curve can be explained by the leakage current.

Is this temperature dependance typical? I don't recall seeing this before.

Rick

Reply to
rickman
Loading thread data ...

It's interesting that they characterize at 125C. They probably intend automotive use, which is great for a low power chip. How many processors are in a car these days? There could be a dozen just under the hood.

The increased power consumption at high temperature is typical of CMOS chips. It's not specific to Greenarrays. I can think of some obvious factors, such as:

- Increased on resistance of the transistors

- More overlap in a gate's linear region, where both P and N transistors conduct because the input voltage is halfway between hi and lo. Slower edge rates (due to higher on resistances) would exacerbate the problem.

Reply to
Brad

Certainly, modern very small feature size CMOS circuits' power dissipation increases with temperature, because a major part of that power dissipation is now due to leakage, which is governed by Arrhenius' Equation.

--------------------------------------- Posted through

formatting link

Reply to
RCIngham

n
n

I don't think you read my post carefully and you don't seem to have looked at any of the data on the GreenArrays chips. They are extremely low power devices and do not use "modern" (meaning smallest possible) feature sizes. They are using a process somewhere around

120 nm if I remember correctly, or maybe even larger! I am pretty sure they have corrected for the static current since they provide that info separately and it does not account for the huge increase in dynamic power.

Rick

Reply to
rickman

Have you told the folks at Green Arrays these "not familiar withs" and "can't imagines" and asked THEM your questions? They don't monitor clf and won't answer here because they want to build a FAQ on their website where the answers will be easily searchable indefinitely.

Contact them at for real answers.

-John [disclosure: I'm a contractor who's worked on/with the chips.]

Reply to
John Rible

increase

dissipatio=

dissipatio=

As someone who started off with 5 micron feature size CMOS, 120 nm IS "very small feature size".

YMMV.

--------------------------------------- Posted through

formatting link

Reply to
RCIngham

Actually I cross posted this to c.a.e and expected someone to know enough about semi-conductor physics to know the answer. I don't think the GA144 is so unique that the basics are different. I see that not everyone there cross posted back to c.l.f

Rick

Reply to
rickman

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.