Apple's custom ARM approach for A6 pays off in JavaScript performance

http://arstechnica.com/apple/2012/09/apples-custom-arm-approach-for-a6-pays-off-in-javascript-performance/

by mrwilliamchang • 14 years ago

5 3 14 years ago

arstechnica.com

ck22• 14 years ago

Also, single-core Atom based Razr shows similar sunspider gains over the dual-core ARM version, very strange:

http://www.engadget.com/2012/09/18/motorolas-razr-i-benchmar...

mtgx• 14 years ago

Sunspider is mostly single-threaded, that's why. Also, Intel is being misleading with their chip anyway, as it will run at its highest clock (2 Ghz) in benchmarks, but will only run at 1.3 Ghz in the vast majority of cases, otherwise it would quickly overheat.

It's Intel's first and only processor that they are advertising with its "turbo-boosted" speed, rather than the default real speed. What makes it not okay for all their other chips to be advertised with the TB'ed speed as the "default" speed in their marketing material, but perfectly okay to do it with this Atom chip in the RAZR?

ck22• 14 years ago

Ah that is a great explanation with details, thanks - first good one I've seen to understand what's happening.

Was worried about ARM there for a day or two.

Apple's custom ARM approach for A6 pays off in JavaScript performance

Related Stories