Hacker News Reader

HNHacker News

Top New Best Ask Show Jobs

Comment by VyseofArcadia | Hacker News Reader

Full thread

VyseofArcadia·

> ARM began in low-power embedded systems, initially facing similar performance limitations.

No it didn't. ARM began on Acorn desktop computers, the Acorn Archimedes. ARM originally stood for Acorn RISC Machine.

View on HN

Author here: I was intending to refer to where ARM got first got traction in the market.

Even there, they gained traction in STBs first, which weren't particularly 'low power', although they were embedded.

I was once in a congress, where one of the first developers of ARM was there and told the history of the company. And I perspicuously remember he saying "at the beginning was all about money, every little transistor was important, every transistor less, was a cheaper system. The same savings in money where the key to later be able to do low power processors, because is basically the same optimization"

So sorry, but I have to partly disagree, they did.

Edit: Interesting... sharing what the designer of the first ARM core said leads to downvoting in HN...HN is a very different place as it used to be...

This is RISC versus CISC again. ARM1's 25k transistors beat the CISCy Intel 80286's 134k transistors. Transistors had a cost at micrometer lithography sizes.

This is true, but it's coincidence. He didn't say they started out with the goal of a low-power processor. He's saying that they optimized for low-cost, and luckily that led to later success in low-power.

He clearly said that soon the priority was power more than money.

Reply on news.ycombinator.com