Apple M2 vs Google Tensor. Specifications, performance, tests

Comprehensive background on the processors being compared, detailing their series, generation, and targeted market segment.

• Segment
  Mobile Mobile
• Family
  Apple M series Google Tensor
• Generation
  2 1
• CPU group
  Apple M2 Google Tensor

Essential parameters including the number of cores, threads, base and turbo frequencies, and cache size. These metrics provide insight into the processor’s speed—higher values generally indicate better performance.

• Frequency
  3.50 GHz 2.80 GHz
• Turbo (1 Core)
  No turbo No turbo
• Turbo (8 Cores)
  No turbo No turbo
• Hyperthreading
  No No
• Overclocking
  No No
• Core architecture
  hybrid (big.LITTLE) hybrid (Prime / big.LITTLE)

The integrated graphics (iGPU) do not influence the CPU performance significantly; they serve as a substitute for a dedicated graphics card in the absence of one or are utilized in mobile devices.

• GPU name
  Apple M2 (8 Core) ARM Mali-G78 MP20
• GPU frequency
  3.50 GHz 0.76 GHz
• GPU (Turbo)
  No turbo No turbo
• Execution units
  128 20
• Shader
  0 320
• Max. GPU Memory
  16 GB --
• Max. displays
  3 1
• Generation
  2 Vallhall 2
• Technology
  5 nm 5 nm
• Release date
  Q1/2022 Q4/2021

This section details the built-in codecs used for encoding and decoding media content, which significantly enhance processing speed and efficiency.

• h265 / HEVC (8 bit)
  Decode / Encode Decode / Encode
• h265 / HEVC (10 bit)
  Decode / Encode Decode / Encode
• h264
  Decode / Encode Decode / Encode
• VP9
  Decode / Encode Decode / Encode
• VP8
  Decode Decode / Encode
• AV1
  Decode Decode
• AVC
  Decode Decode / Encode
• VC-1
  Decode Decode / Encode
• JPEG
  Decode / Encode Decode / Encode

Overview of the types and quantities of RAM supported by Google Tensor and Apple M2. The supported memory frequencies may vary depending on the motherboard configuration.

• Memory type
  LPDDR5-5500 LPDDR5-5500
• Max. Memory
  16 GB 12 GB
• ECC
  No No
• Memory channels
  2 2

Analyze the TDP (Thermal Design Power) requirements of Apple M2 and Google Tensor to make an informed decision on the appropriate cooling system. Remember that TDP refers to thermal watts, not electrical watts.

• TDP (PL1)
  15 W 10 W
• TDP up
  20 W --
• TDP down
  10 W --

Information on architecture, interfaces, and additional instructions supported by Apple M2 and Google Tensor, including virtual machine technologies and fabrication processes.

• Instruction set (ISA)
  ARMv8-A64 (64 bit) ARMv8-A64 (64 bit)
• Virtualization
  None None
• ISA extensions
  Rosetta 2 x86-Emulation
• L2-Cache
  16.00 MB 8.00 MB
• Architecture
  M2
• Technology
  5 nm 5 nm
• Socket
  N/A N/A
• Release date
  Q1/2022 Q4/2021