Apple A18 vs Google Tensor

SoC Comparison Result

Below are the results of a comparison of Apple A18 and Google Tensor mobile processors based on key performance characteristics, as well as power consumption and much more.

Advantages

  • Higher Technology: 3 nm (3 nm vs 5 nm)
  • Higher Frequency: 3780 MHz (3780 MHz vs 2800 MHz)
  • Higher Max Bandwidth: 60 GB/s (60 GB/s vs 51.2 Gbit/s)
  • Newer Launch Date: September 2024 (September 2024 vs October 2021)

Basic

Apple
Label Name
Google
September 2024
Launch Date
October 2021
SmartPhone Flagship
Platform
SmartPhone Flagship
TSMC
Manufacturing
Samsung
A18
Model Name
S5E9845
2x 4.04 GHz – 4x 2.4 GHz –
Architecture
2x 2.8 GHz – Cortex-X1 2x 2.25 GHz – Cortex A76 4x 1.8 GHz – Cortex A55
6
Cores
8
3 nm
Technology
5 nm
3780 MHz
Frequency
2800 MHz

GPU Specifications

Apple A18 GPU
GPU name
Mali-G78 MP20
1398 MHz
GPU frequency
848 MHz
1.7894 TFLOPS
FLOPS
2.1709 TFLOPS
128
Shading units
64
5
Execution units
20
-
OpenCL version
2.0
-
Vulkan version
1.3
-
Max display resolution
3840 x 2160
-
DirectX version
12

Connectivity

LTE Cat. 24
4G support
LTE Cat. 24
Yes
5G support
Yes
5.3
Bluetooth
5.2
7
Wi-Fi
6
GPS, GLONASS, Beidou, Galileo, QZSS, NAVIC
Navigation
GPS, GLONASS, Beidou, Galileo, QZSS

Memory Specifications

LPDDR5X
Memory type
LPDDR5
7500 MHz
Memory frequency
3200 MHz
4x 16 Bit
Memory Bus
4x 16 Bit
60 GB/s
Max Bandwidth
51.2 Gbit/s

Miscellaneous

Neural Engine
Neural processor (NPU)
Tensor Processing Unit
-
L2 Cache
3 MB
AAC, AIFF, CAF, MP3, MP4, WAV, AC-3, E-AC-3, AAX, AAX+
Audio codecs
AAC, AIFF, CAF, MP3, MP4, WAV
1x 48MP
Max camera resolution
1x 200MP, 2x 32MP
NVMe
Storage type
UFS 3.1
8K at 24FPS, 4K at 120FPS
Video capture
4K at 60FPS
H.264, H.265, AV1, VP8, VP9, Motion JPEG
Video codecs
H.264, H.265, AV1, VP9
8K at 24FPS, 4K at 120FPS
Video playback
4K at 60FPS
-
TDP
5.6 W
ARMv9.2-A
Instruction set
ARMv8.2-A

Benchmarks

Geekbench 6 Single Core
A18
3403 +158%
Tensor
1319
Geekbench 6 Multi Core
A18
7828 +142%
Tensor
3234
AnTuTu 10
A18
1490144 +83%
Tensor
813782
FP32 (float)
A18
1825
Tensor
2213 +21%