Apple has been chastised on a number of fronts, most notably with the release of the much-anticipated iPhone 13. Because of the little improvements, the majority of buyers deemed the handset to be lacking; some even stated it was a downgrade from the iPhone 12. Nevertheless, the device does have certain perks since it is powered by the Apple A15 SoC.
Performance of the Apple A15 SoC
Apple announced their new iPhone 13 series handsets a few weeks ago, a collection of phones that are powered by the new Apple A15 SoC. Now, we’re taking a closer look at the next generation chipset, examining what exactly Apple has altered in the new silicon and determining whether it lives up to its name.
Apple’s PR for the A15 release this year was a little strange, mainly because the firm avoided making any generational comparisons between the new design and Apple’s own A14. The fact that Apple opted to define the SoC in terms of the competition was particularly striking; while this isn’t unusual on the Mac side of things, it was something that stuck out this year for the iPhone unveiling more than normal.
Upgrades worth mentioning
Apple is using new designs for their CPUs, a faster Neural engine, a new 4- or 5-core GPU depending on the iPhone variant, and a whole new display pipeline and media hardware block for video encoding and decoding, along with new ISP improvements for camera quality advancements, according to the few facts about the A15.
On the CPU side of things, Apple claimed to be 50 percent quicker than the competition hence can carry out iPhone transfer efficiently, while the GPU performance estimates were similarly imprecise, with the 4-core GPU A15 claiming to be +30 percent faster than the competition and the 5-core variation claiming to be +50 percent faster. We’ve put the SoC through its paces, and today’s post will focus on the new chip’s precise performance and efficiency measurements.
In comparison to the A14, the new A15 boosts the top single-core frequency of the two-performance core cluster by 8%, reaching 3240MHz versus 2998MHz in the previous version. When both performance cores are active, their operational frequency increases by 10%, with both now running at a blistering 3180MHz, up from 2890MHz in the previous iteration.
Additionally, the A15’s E-cores can now run at 2016MHz, which is a 10.5 percent improvement over the A14’s cores. The frequency is independent of the performance cores, in the sense that the number of threads in one cluster has no bearing on the other, and vice versa. This generation, Apple has made some more fascinating adjustments to the small cores, which we’ll get to in a minute.
Moreover, the A15’s E-cores can now run at 2016MHz, which is a 10.5 percent improvement over the A14’s cores. The frequency is independent of the performance cores, in the sense that the number of threads in one cluster has no bearing on the other, and vice versa. This generation, Apple has made some more fascinating adjustments to the small cores.
A storm of several variables could be to blame for Apple’s more moderate micro-architectural modifications this year — Apple famously lost their main architect on the major performance cores, as well as elements of the design teams, to Nuvia back in 2019.
The switch to Armv9 could also indicate more design work, and the pandemic situation could have contributed to some less-than-ideal implementation. We’ll have to wait until next year’s A16 to see if Apple’s design cycle has slowed, or if this was simply a slippage or a pause before a much more significant change in the next microarchitecture.