Qualcomm Snapdragon 8 Gen 2 vs Apple M1 Pro

Qualcomm Snapdragon 8 Gen 2

The Qualcomm Snapdragon 8 Gen 2 Mobile Platform is a high-end SoC for smartphones that was introduced in late 2022 and manufactured in 4 nm at TSMC (N4P). It integrates four different CPU clusters. A fast and big prime core based on the ARM Cortex-X3 architecture and clocked at up to 3.2 GHz for 64 bit applications. Two more ARM Cortex-A715 based performance cores clocked at up to 2.8 GHz (also only for 64 bit apps). Two more Cortex-A710 performance cores at up to 2.8 GHz that can also be used for older apps and three efficiency cores (ARM Cortex-A510 with up to 2 GHz). All cores can use the shared 8 MB level 3 cache.

Thanks to the new architecture and high clock rates, the SD 8 Gen 2 is the fastest mobile CPU for Android based devices and is only topped by current Apple SoCs like the Apple A16 (see benchmarks below).

The chip also integrates a Hexagon accelerator for AI workloads (tensor, scalar and vector). Qualcomm states that its one of the first chips with INT4 support and up to 4.35x performance compared to the previous generation. 

The 18-Bit-Spectra ISP is able to process photos with up to 200 MP and work with videos up to 8k30 (including 10-bit, HDR10+, HLG and Dolby Vision).

A big improvement of the Snapdragon 8 G2 is the integrated Adreno 740 GPU, that now supports Hardware Raytracing and tops even the iGPU in the Apple A16.

The integrated AI Engine can offer up to 4x faster AI performance thanks to the bigger Tensor engine. The 18-Bit-Spectra-ISP also saw some big improvements and is now called Snapdragon Sight.

The integrated Adreno 730 offers a 30% graphics boost compared to the previous Adreno 660.

The integrated Snapdragon X70 5G modem is also new as is the FastConnect 7800 modem with Wi-Fi 7 and Bluetooth 5.3 support.

Apple M1 Pro

The Apple M1 Pro is a System on a Chip (SoC) from Apple that is found in the late 2021 MacBook Pro 14 and 16-inch models. It offers  all 10 cores available in the chip divided in eight performance cores (P-cores with 600 - 3220 MHz) and two power-efficiency cores (E-cores with 600 - 2064 MHz). There is no Turbo Boost for single cores or short burst periods. The cores are similar to the cores in the Apple M1. The entry level model offers only 8 cores.

The big cores (codename Firestorm) offer 192 KB instruction cache, 128 KB data cache, and 24 MB shared L2 cache (up from 12 MB in the M1). The four efficiency cores (codename Icestorm) are a lot smaller and offer only 128 KB instruction cache, 64 KB data cache, and 4 MB shared cache. CPU and GPU can both use the 24 MB SLC (System Level Cache). The efficiency cores (E cluster) clock with 600 - 2064 MHz, the performance cores (P cluster) with 600 - 3228 MHz.

The unified memory (16 or 32 GB LPDDR5-6400) next to the chip is connected by a 256 bit memory controller (200 GB/s bandwidth) and can be used by the GPU and CPU.

The integrated graphics card in the M1 Pro offers all 16 cores.

Furthermore, the SoC integrates a fast 16 core neural engine, a secure enclave (e.g., for encryption), a unified memory architecture, Thunderbolt 4 controller, an ISP, and media de- and encoders (including ProRes).

The M1 Pro is manufactured in 5 nm at TSMC and integrates 33.7 billion transistors. The peak power consumption of the chip was advertised around 30W for CPU intensive tasks. In the Prime95 benchmark the chip uses in our tests (with a MBP16) 33.6W package power and 31W for the CPU part. In idle the SoC only reports 1W package power.