Steve Furber was the head of the Computer Science department at Manchester University when I studied there 10 years ago, I believe he still is.
He took us for a course about ARM processors and how to program them via assembly code. It was tough to say the least but interesting, the text book was the one mentioned in the article. At the time we realised that ARM was a big thing, but looking back we definitely didn't realise how big it would become.
If the power lines to the processor were not connected, then what leakage was powering it? Was it running only off the minimal electricity transmitted when data was moved into/out of the processor?
usually CMOS chips have input protection diodes which prevent the input pins from rising too high above the power lines or too far below ground, which causes SCR latchup - essentially a short from power to ground. This also means that CMOS chips can often be powered by input pins alone.
I've even experienced this with a stepper motor driver which didn't have the main power lines hooked up (but did have logic power), where 3A were going through these diodes to run the motor. So I'm not sure how that particular anecdote says much about the power usage of the ARM chip.
I was lucky enough to have lunch with Steve Furber and my two of my old college lecturers a few months ago.
A very fascinating man with lots of history.
This "accident" reminds me of what I read a week ago in the article "How geniuses think".
And it was this: "GENIUSES PREPARE THEMSELVES FOR CHANCE. Whenever we attempt to do something and fail, we end up doing something else." So when you're up to something interesting drop everything else and study it. It might pay off.
He took us for a course about ARM processors and how to program them via assembly code. It was tough to say the least but interesting, the text book was the one mentioned in the article. At the time we realised that ARM was a big thing, but looking back we definitely didn't realise how big it would become.