[ad_1]
Among the new iPad and Watch devices released today, Apple made headlines by launching the new A14 SoC chip. Apple’s next-generation silicon design is noteworthy because it is the industry’s first commercial chip to be manufactured on a 5nm process node, marking this as the first of a new generation of designs to be expected. that significantly push the envelope into the semiconductor space.
This year’s Apple event disclosures were a bit confusing as the company was comparing the new A14 metrics to the A12, given that’s what the previous generation iPad Air had been using so far – we’ll need to add proper context. behind the figures to extrapolate what this means.
On the CPU side, Apple is using next-gen high-performance cores, as well as new power-efficient small cores, but remains in a 2 + 4 configuration. Apple here claims a 40% performance increase per part of CPUs, although the company doesn’t specify exactly what this metric refers to – is it single-threaded performance? Is it multi-threaded performance? Is it for large or small nuclei?
What we do know, though, is that it refers to the A12 chipset, and the A13 had already claimed a 20% increase over that generation. So simple arithmetic dictates that the A14 would be roughly 16% faster than the A13 if Apple’s performance metrics are consistent across generations.
On the GPU side, we also see a similar calculation, as Apple claims a 30% performance increase compared to the A12 generation thanks to the new 4-core GPU in the A14. Normalizing this against the A13 would mean only an 8.3% performance increase, which is actually pretty meager.
In other areas, Apple has more significant performance leaps, such as the new 16-core neural engine that now has up to 11 TOPs of inference performance, which is more than double the A12’s TOP 5 and 83% more than the A12’s. 6 TOP estimates of neuronal A13. motor.
Apple announces a new image signal processor among the new SoC features, but otherwise the performance metrics (besides the neural engine) seem quite conservative given the fact that the new chip has 11.8 billion transistors, a generational increase of 38% over the 8,500 million figures of A13.
The only explanation and theory I have is that Apple might have finally lowered its peak power consumption in peak performance states of CPUs and GPUs, and therefore peak performance would not have seen such a big leap in this generation, but it favors more sustainable thermal figures.
Apple’s A12 and A13 chips were huge performance improvements on both the CPU and GPU side, however one criticism I made of the company’s designs is that they both increased power consumption beyond what it was generally sustainable in a mobile thermal envelope. This meant that while the designs had staggering peak performance figures, the chips couldn’t sustain them for extended periods of more than 2-3 minutes. With that in mind, the devices were accelerated to performance levels that were still ahead of the competition, leaving Apple in a leading position in terms of efficiency.
What speaks against such a theory is that Apple did not mention at all the concrete power or energy efficiency improvements of this generation, which is quite unusual given that they have traditionally always commented on this aspect of new designs of the A series.
We’ll just have to wait and see if this is indicative that the actual products haven’t improved in this regard, or is just an omission and a side effect of the event’s new, more streamlined presentation style.
Whatever the performance and efficiency figures, Apple can boast the industry’s first 5nm silicon design. The new A14 made by TSMC represents the cutting edge of today’s semiconductor technology, and Apple made sure to mention it during the presentation.
