Apple has made very full-bodied promises about the performance and efficiency of the new system-on-chip M1, but has not published any common benchmarks. The first results in Geekbench 5 have now appeared, which put the M1 well ahead of the current fastest x in multi-core performance 86 – Mobile processors with 15 to 30 Show watts TDP, i.e. Intel Core i7 – 1185 G7 (4 cores) and AMD Ryzen 7 Pro 4750 U (8 cores). But also knowledge about the Apple A 14 from the iPhone 12, which is closely related to the M1, reveal more about the M1 technology.

Eight cores , GPU and KI According to Apple, the M1 contains eight computing cores, a GPU with eight cores or shader clusters as well as a neural engine with 16 Cores for AI and machine learning algorithms. The GPU can also be used as an accelerator for some computing tasks.

The system-on-chip (SoC) M1 also contains controllers for I / O interfaces (Thunderbolt / USB4, PCIe 4.0) as well as functions that were previously handled by the separate T2 security chip with built-in SSD controller, including a Secure Enclave (SE). The latter is important for biometric authentication (Face ID, Touch ID) and payment functions.

In addition, the M1 has a RAM controller that can handle up to 16 GByte memory connects, probably with two 64 – bit channels for LPDDR4 or LPDDR4X.

The system- on-Chip Apple M1

(Image: Apple)

Fire and Ice An internal “fabric” links the individual function blocks as well as an additional cache that intercepts memory accesses. The CPU cores are divided into two groups, each with its own L2 cache. Each individual CPU core in turn has L1 cache, as with x 86 Usually separate for commands (Instructions), abbreviated L1I, and data (L1D).

The M1 has four particularly powerful and particularly efficient CPU cores. Apple has not officially confirmed it, but experts say the code names for the strong cores are “Firestorm” and the more efficient ones are “Icestorm”.

According to Apple, each Firestorm core has 192 KByte L1I cache and 128 KByte L1D cache. These data buffers are much larger than for example Intel’s current “Tiger Lake” CPUs (48 KByte L1I / 32 KByte L1D) and AMD Zen 2 (each 32 KByte). However, the x 86 – processors also so-called Micro-Op- (µOP-) caches, which work particularly efficiently.

Apple’s information on the strong and efficient ARM cores in the M1

(Image: Apple)

Apple’s different ARM core groups can each use a common Access L2 cache: The four Firestorms on together 12 MByte, the four Icestorms on 4 MByte. In addition, there is the fabric cache, the capacity of which Apple has not revealed and which, from the point of view of the CPU cores, serves as an L3 cache. Bionic is very similar to the M1 and has as many efficient Icestorm cores as well as a neural engine with also 16 cores. In A 14 but only two strong Firestorm cores with a smaller L2 cache (8 MByte) and a GPU with 4 instead of 8 cores. In addition, the arithmetic units clock in A 14 is generally lower because it can dissipate less heat in iPhones and iPad Air 4 and draw less battery power than in the Macs.

Clock frequencies Nominal clock frequencies of A 14 and M1 are not revealed by Apple, nor is the Thermal Design Power (TDP). The latter is found in smartphone SoCs like the A 14 usually around 5 watts, but even then a smartphone would become very hot in the long run and a 11 – watt-hour battery (3.7 volts / 3000 mAh) would be empty in little more than 2 hours. On average, the power consumption of the chip is much lower, especially since the display usually needs more energy and the modem also swallows electricity.

Apple reveals the Thermal design power of the M1 not exactly, but rearranges it Watts on.

(Image: Apple)

When introducing the M1, Apple made comparisons with other chips at 10 Watt employed – so it is probably between Watts in the MacBook Air without fan and 15 to 20 watts for fan cooling. Processors of the “U” classes from AMD and Intel have 15 to 28 Watt TDP, but Here, too, the quality of the notebook manufacturer’s cooling system is crucial. These processors can be set in a wide range via configurable TDP (cTDP) and then deliver more or less computing power because they have to slow down quickly with sustained load. This also applies to the M1, as Apple explained at the launch: In the MacBook Pro 12 inches with a fan, its sustained performance is higher .

Geekbench data Notes on the clock frequencies from A 14 and M1 deliver results from the online database of the benchmark Geekbench 5. There is an A 14 the specification 2, 99 GHz and in the M1 result from a fanless MacBok Air a value of 3.2 GHz. In a MacBook Pro 13 Inch or Mac mini with fan, the M1 may clock a little higher.

However, these clock frequencies are significantly lower than the Turbo frequencies of Core i7 – 1185 G7 (4.8 GHz) and AMD Ryzen 7 Pro 4750 U (4.1 GHz). For higher clock frequencies both ARM and x 79 – Cores have higher core voltages, which is associated with significantly higher power loss. The “wider” data paths are laid out, i.e. the more arithmetic units and transistors run with higher voltage and clock rate, the more the losses grow when clocking up.

The more impressive are the results that A 14 and M1 in Geekbench 5. In the single-core run, which certainly works on a strong Firestorm core, there are 1596 (A 14) or 1732 Points (M1). The M1’s lead of 8.5 percent is largely due to the higher clock rate, the larger L2 cache has no major influence.

Apple M1 and A 14 in Geekbench 5 CPU Information according to the Geekbench database Benchmark results Device Clock Operating system Single-Core Multi-Core Apple M1 MacBook Air 3, 19 GHz macOS 11. 0 1732 7545 Apple A 14 iPhone 12 Per 2, 99 GHz iOS 14.1 1596 4008 Intel Core i7 – 1185 G7 MSI MS – 13 C4 4, 79 GHz Windows 10 1610 6113 AMD Ryzen 7 Pro 4750 U Lenovo ThinkPad T 14 4, 19 GHz Windows 10 1162 6509 Intel Core i7 – 1165 G7 Dell XPS 13 4, 70 GHz Linux 5.8 1726 5313 Compared to a Core i7 – 1185 G7 in an MSI notebook, the M1 in the multi-core evaluation is around 23 percent ahead, with single-core it is 7.5 percent faster. But there are also Geekbench 5 values ​​for a Core i7 – 1165 G7 under Linux, which show it to be almost on par in single-core performance – but here it is much weaker in multi-core.

Intel’s Tiger Lakes like the mentioned Core i7 – 1185 G7 and 1165 G7s only have four cores and Simultaneous Multi-Threading (SMT alias Hyper-Threading), while all eight cores of the M1 work simultaneously – fast and efficient together. Therefore a comparison with the previously strongest multithreading 14 – exciting watt notebook processor, the AMD Ryzen 7 Pro 4750 U with eight cores and SMT, so a total of 16 threads. But even that is in Geekbench 5 multi-core around 15 Percent behind Apple’s M1 – because its Zen-2 cores deliver 33 Percent less single-core performance than an Apple Firestorm at 3.2 GHz. With the Ryzen 5000 U (Cezanne) with Zen 3 this could be 2021 move.