The Apple M2 is a System on a Chip (SoC) from Apple that is found in the late 2022 MacBook Air and, MacBook Pro 13. It offers 8 cores divided in four performance cores and four power-efficiency cores. The big cores offer 192 KB instruction cache, 128 KB data cache, and 16 MB shared L2 cache (up from 12 MB). The four efficiency cores are a lot smaller and offer only 128 KB instruction cache, 64 KB data cache, and 4 MB shared cache. The efficiency cores (E cluster) clock with up to 2,4 GHz, the performance cores (P cluster) with up to 3,5 GHz and therefore higher than the M1 cores. The architecture should be similar to the A15 (iPhone 13) with Avalanche and Blizzard cores.
The chip features a unified memory architecture for the CPU and GPU cores and supports up to 24 GB LPDDR5-6400 for a bandwidth of up to 100GB/s.
According to Apple, the M2 offers a 18% higher CPU performance at the same power consumption level compared to the Apple M1. In our tests, the MacBook Pro 13 with active cooling was able to reach the 18% in Geekbench Multi. In other benchmarks we measured 12 to 15% gains compared to the M1. Therefore, the performance is now near the M1 Pro with 8 cores. The passively cooled MacBook Air may however suffer from throttling in longer load scenarios.
Furthermore, the SoC integrates a fast 16 core neural engine with a peak performance of 16 TOPS (for AI hardware acceleration), a secure enclave (e.g., for encryption), Thunderbolt / USB 4 controller, an ISP, and media de- and encoders.
The Apple M2 includes 20 billion transistors (up from the 16 billion of the M1) and is manufactured in the second generation 5nm process at TSMC (most likely N5P). The power consumption is rated at 20W what we also measured under CPU load.
The AMD Ryzen 7 7840S is a fast high-end laptop processor of the Phoenix series. It offers 8 cores (octa core) based on the Zen 4 architecture that supports hyperthreading (16 threads). The cores clock from 3.8 (base) up to 5.1 GHz (single core boost). The processor includes 8 MB L2 cache and 16 MB L3 cache. The processor is exclusively available in Lenovo laptops but the specs are similar to the regular Ryzen 7 7840HS.
The performance of the R7 7840S is only slightly below the fastest model, the R9 7940HS, as the clock speed difference is only minimal (e.g. -100 MHz / 2% slower single core boost). Therefore, the CPU should also perform slightly higher than the old top model Ryzen 9 6980HX (8 Zen 3 cores with up to 4.9 GHz) at 54W TDP and Ryzen 9 6980HS at 35W TDP. Compared to the higher end Dragon Range series, the 7840HS should be similar to the Ryzen 7 7745HX (also 8 Zen 4 cores, max 5.1 GHz, 55W TDP, 32 MB L3).
The chip integrates a modern and fasts RDNA 3 graphics card (iGPU) called Radeon 780M with 12 CUs and up to 2.7 GHz clock speed. Furthermore, the Phoenix series include a video engine with AV1 de- and encoding, a new Xilinx FPGA based XDNA AI accelerator (Ryzen AI) that should be faster than the AI engine in the Apple M2 SOC and a dual-channel DDR5-5600 / LPDDR5x-7500 memory controller (with ECC support). The connectivity features includes 2 possible USB 4 (40 Gbps) ports and 20 PCIe 4.0 lanes for a GPU and SSDs.
The Phoenix series uses a single monolithic design (unlike the chiplet design of the 7045HX series) and is manufactured in the modern 4nm process at TSMC. The TDP can be configured between 35 and 45 Watt.
The Ryzen 5 7640HS is an upper mid-range processor (APU) of the Phoenix product family. It was brought to life in H1 2023 and it has 6 SMT-enabled Zen 4 cores (12 threads) running at 4.3 GHz to 5.0 GHz. The laptop chip is one of the first to feature Ryzen AI, AMD's answer to Intel's DL Boost and GNA technologies. The Radeon 760M iGPU is responsible for carrying out 3D calculations and similar duties.
Architecture & Features
Phoenix family chips are powered by the Zen 4 architecture, much like Dragon Range family chips are. The latter however lacks hardware AI workload acceleration capabilities that Phoenix has.
More importantly, Zen 4 introduces AVX512 support (Zen 3-powered chips did not have that) and, thanks to a plethora of other improvements including larger caches/registers/buffers across the board, is set to bring a double-digit IPC improvement.
Elsewhere, the 7640HS has 16 MB of L3 cache and a super-fast DDR5-5600 / LPDDR5x-7500 RAM controller. USB 4 (and thus Thunderbolt) is supported natively and so is PCIe 4. In fact, 20 PCIe 4 lanes are on offer; data transfer speeds of up to 7.8 GB/s will be possible provided a sufficiently fast NVMe SSD is used.
Systems powered by this Ryzen 5 series chip are expected to run 64-bit Windows 11, 64-bit Windows 10, or Linux. Please note that the APU is not overclockable and neither will you be able to replace it with a faster one as it gets soldered down permanently (FP7, FP7r2, FP8 socket interfaces).
Performance
The average 7640HS in our database is about as fast as the Ryzen 9 5900HX, Ryzen 7 6800H, Ryzen 7 7736U and also the Core i5-12500H, as far as multi-thread benchmark scores are concerned. This is a good if not record-breaking result as of mid 2023. Such a CPU will handle pretty much any task you can throw at it, gaming included.
Your mileage may vary depending on how high the CPU power limits are and how competent the cooling solution of your system is.
Graphics
The Radeon 760M has 8 CUs (512 shaders) running at up to 2,600 MHz. While this is not the best graphics adapter for gaming as it's only just fast enough for low quality settings and resolutions such as 1600 x 900, it will let you connect up to four SUHD 4320p monitors and it also HW-encodes and HW-decodes the most widely used video codecs such as AV1, HEVC and AVC.
Power consumption
The 7640HS has a long-term power limit (default TDP) of 35 W to 54 W, giving laptop makers a choice between improving battery life and giving the system they are working on more oomph. Either way, an active cooling solution is a must for any system powered by this chip.
The APU is built with TSMC's 4 nm process for high, as of late 2023, energy efficiency.
- Range of benchmark values for this graphics card - Average benchmark values for this graphics card * Smaller numbers mean a higher performance 1 This benchmark is not used for the average calculation
v1.27
log 16. 03:33:47
#0 checking url part for id 14521 +0s ... 0s
#1 checking url part for id 15098 +0s ... 0s
#2 checking url part for id 14950 +0s ... 0s
#3 not redirecting to Ajax server +0s ... 0s
#4 did not recreate cache, as it is less than 5 days old! Created at Thu, 13 Jun 2024 05:33:43 +0200 +0.001s ... 0.001s
#5 composed specs +0.054s ... 0.055s
#6 did output specs +0s ... 0.055s
#7 getting avg benchmarks for device 14521 +0.004s ... 0.059s
#8 got single benchmarks 14521 +0.023s ... 0.082s
#9 getting avg benchmarks for device 15098 +0.021s ... 0.104s
#10 got single benchmarks 15098 +0.009s ... 0.113s
#11 getting avg benchmarks for device 14950 +0.018s ... 0.132s
#12 got single benchmarks 14950 +0.012s ... 0.144s
#13 got avg benchmarks for devices +0s ... 0.144s
#14 min, max, avg, median took s +0.32s ... 0.464s