Apple Unveils New A15 Bionic SoC
As part of its iPhone 13 launch event this year, Apple has unveiled the A15 Bionic, the next-generation SoC that powers the iPhone. Like last year’s CPU, the A15 will be built on TSMC’s 5nm process. Apple didn’t clarify if it’s using the company’s N5 or N5P node, but we’d assume its N5P to take advantage of in-generation improvements.
The A15 keeps the same core configuration as the A14, with two high-performance cores (Avalanche) and four high-efficiency cores (Blizzard) replacing the FireStorm/IceStorm cores used in A14. Typically Apple compares performance with its own previous generation, but this year the company made claims relative to “the competition.” Specifically, the A15 is supposed to be 1.5x faster than its unnamed competitors, while the GPU is supposedly 1.3x faster.
While this framing would be normal from AMD or Intel, it’s not normal from Apple. Apple doesn’t compare against other companies, typically, because Apple products are fast enough that they generally only compete with themselves.
In tests like Geekbench 5, the existing iPhone 12 scores ~1550 – 1600 points in the single-core test and 3850 – 4000 points in multi-core compared with the Samsung Galaxy S21 at 800-1000 points in single-core and 2800 – 3300 in multi-core. GeekBench 5 is just one application test, but the point stands in other tests as well.
The reason Apple can get away with fielding a 2+4 core array against the (1+3)+4 or 4+4 chips in modern Android devices is that Apple doesn’t need more high-end cores to beat the competition. It’s the same reason why the M1 represents a long-term threat to AMD and Intel while no one has been talking about a Cortex processor in quite the same way.
The fact that Apple has pivoted to talking about the competition, however, may be a way of acknowledging that this chip doesn’t actually pack much of a punch over and above the A14. Anandtech notes that the A14 is ahead of its existing competitors by 1.41x and 1.18x in CPU and GPU workloads, which means a 1.5x and 1.30x projection for the A15 works out to a 1.06x gain for CPU workloads and a 1.1x gain in GPU performance. The faster variant of the A15, reserved for the iPhone 13 Pro Max, would deliver a more significant 1.28x performance increase.
It’s possible that the A15 is focused more on improving efficiency than on boosting absolute performance. Apple did note that it doubled the system cache to 32MB, which would deliver its own performance boost.
Even if the A15 isn’t a tremendous leap above the A14, however, Apple’s position at the top of the product stack is secure. There are no other mobile ARM devices that match Apple’s single-core performance or overall power efficiency and that won’t change even if the A15 is a modest step forward compared with Apple’s typical generational improvement.
Continue reading
The Xbox Series S Is Handicapped by Its Storage Capacity
The Xbox Series S has been favorably received, for the most part, but the console's low base storage makes the Xbox Series X a better value for a lot of people.
In Massive Shift, Apple Announces New Macs With ARM-Based M1 Chip
Apple saw huge success the last time it switched architectures to Intel, but this time? The jury's still out, but one thing is certain: Apple is about to make a lot more money.
Apple’s New M1 SoC Looks Great, Is Not Faster Than 98 Percent of PC Laptops
Apple's new M1 silicon really looks amazing, but it isn't faster than 98 percent of the PCs sold last year, despite what the company claims.
What Does It Mean for the PC Market If Apple Makes the Fastest CPU?
Apple's M1 SoC could have a profound impact on the PC market. After 25 years, x86 may no longer be the highest-performing CPU architecture you can practically buy.