[Coco] 6809/6309 Power Analysis

[Boisy G. Pitre]

On Jan 21, 2010, at 5:31 PM, Christian Lesage wrote:

> Lothan wrote:
>> I'm not so sure this presents that big of a problem. If the intent to is to measure pure processor current, you're probably right; however, I think think current sourced to read/write memory and used by the circuits in support of the processor is meaningful if the intent to support greener development methodology as a new branch of science. 
> I was just pointing out that you'd have to take those considerations into account if you want to make comparisons that are meaningful. The address range where your program and data reside is probably a factor in the total power consumption. If you are not careful enough, you could measure differences that are not due to the instructions being executed, but to other factors like this.

These are good points Christian.  For load location, NitrOS-9 loads applications of similar size in the same area of memory (at $E000 if the program is <= 7680 bytes in size).  The processor will be fetching each instruction which obviously causes memory references; the size of the instruction (and any operands it has) is another factor that affects power.

In that case, I think it is important to break instructions down into classes: instructions that operate on register operands, instructions that operate on memory operands, and within those classes you can have subclasses based on instruction byte count, register size, etc.

Thanks for the input.

>> It could certainly lend credence to register-based optimizations to reduce current drain accessing memory if that's the scope.
> Indeed.
> 
> 
> --
> Coco mailing list
> Coco at maltedmedia.com
> http://five.pairlist.net/mailman/listinfo/coco