hinkley 5 days ago

And for the record the A17 Pro chip is 3nm. Used in the iPhone 15 pro and the iPad mini.

But they could make iPhone 14’s and the smaller 15’s.

drexlspivey 5 days ago

So which device will these be for then? I thought Apple stuff are always on the cutting edge node.

  • hinkley 5 days ago

    Their new stuff is. The iPad mini just moved from the A15 to the A17, The first MacBook with Intel processors had access to a bin that was not generally available yet. The yield was too low for it to work for an IBM, a Sony, or a Fujitsu. But Apple was low volume and high margin.

    If I was nervous about a new fab, there’s the iPhone SE, the Apple TV, lots of choices for a less aggressive manufacturing node and less aggressive sales figures. If yield is shit you can still offer a product that isn’t killed by its own success.

    • mcintyre1994 4 days ago

      I wonder if Apple Intelligence is forcing them to create new chips for things like the SE and TV instead of using old chips which I think they’d usually do.

      • hinkley 4 days ago

        I'm a long time AAPL shareholder, admittedly I have not been paying super close attention over the last couple of years. But I'm pretty sure Apple has a history of being perfectly fine with their entry level devices being one killer feature behind the rest of their fleet. If the old hardware can support it fine, and if not then better luck next time (literally and figuratively).

        They don't usually do it for very long, and my recollection is that they usually bring it down market before (maybe concurrent with) introducing yet another new killer feature.

        So you'll be correct soon enough, but there's slack time there that they can use to hedge their bets on the new fab before leaning on it for the WWDC keynote.

  • kcb 5 days ago

    Apple still produces older generation devices long after the latest ones are released. That's their whole strategy to address the lower end market.

dietr1ch 5 days ago

As an outsider that means somewhere in 2nm-10nm as everyone measures different things or have awfully off-standard rulers.