Tuesday, October 13, 2020 [Tweets] [Favorites]

A14 Bionic

Chris Velazco (via MacRumors):

At a high level, the A14 seems similar to Apple’s other Bionic chipsets. This system-on-a-chip packs a six-core CPU — two cores high-performance cores and four for lower-priority tasks — just as the A12 and A13 did. The number of GPU cores here has also remained unchanged at four. Don’t be fooled by these passing similarities, though: Because the A14 was designed for a 5nm manufacturing process, there’s more going on in this system-on-a-chip than ever before.

[…]

Unsurprisingly, this year’s Neural Engine is a far cry from the first one we saw in 2017. While that original co-processor could perform 600 billion operations per second, last year’s A13 raised the bar to 6 trillion operations in the same amount of time. Meanwhile, the A14 generally obliterates the bar by performing a claimed 11 trillion operations per second.

[…]

Apple hasn’t yet issued claims about the A14 Bionic’s performance improvements over last year’s A13 Bionic -- expect more on that during the company’s upcoming keynote. (A set of leaked benchmarks suggests some healthy gains over last year’s chipset, though some are less than impressed.) When Apple revealed the new iPad Air, though, it did say the A14’s CPU was up to 40 percent faster than the previous model, and that people could expect up to a 30 percent increase in graphics performance.

Upgrade:

Jason and Myke interview Apple's Tim Millet and Tom Boger about the new iPad Air and Apple silicon.

1 Comment

Now if they would only give us some documentation for the Neural Engine. Right now, it's essentially "use CoreML, and pray".

The capability of 11 gazillion operations per second doesn't help if it's just going to sit idle, and I'm not going to spend the time trying to shoehorn my app into CoreML if I have no assurances it'll even be used.

Stay up-to-date by subscribing to the Comments RSS Feed for this post.

Leave a Comment