Unisoc T820 vs Google Tensor G2

SOC Specs Compare

CPU

Architecture
1x 2.7 GHz – Cortex-A763x 2.3 GHz – Cortex-A764x 2.1 GHz – Cortex-A55
2x 2.85 GHz – Cortex-X12x 2.35 GHz – Cortex-A784x 1.8 GHz – Cortex-A55
Frequency
2700 MHz
2850 MHz
Cores
8
8
Instruction set
ARMv8.2-A
ARMv8.4-A
Process
6 nm
5 nm
Manufacturing
TSMC
Samsung
L3 cache
-
0
TDP
-
7 W

Graphics

GPU name
Mali-G57 MP4
Mali-G710 MP7
GPU frequency
780 MHz
850 MHz
Execution units
4
7
Shading units
64
96
Vulkan version
1.3
1.3
OpenCL version
2.0
2.0

Memory

Memory Type
LPDDR4X
LPDDR5
Memory frequency
2133 MHz
3200 MHz
Bus
2x 16 Bit
4x 16 Bit
Max bandwidth
17.07 Gbit/s
51.2 Gbit/s

Multimedia (ISP)

Neural processor (NPU)
Yes
Next-gen Tensor Processing Unit
Storage type
eMMC 5.1, UFS 3.1
UFS 3.1
Max display resolution
3440 x 1440
3840 x 2160
Max camera resolution
1x 108MP, 2x 20MP
1x 200MP, 2x 32MP
Video capture
4K at 60FPS
4K at 60FPS
Video playback
4K at 60FPS
4K at 60FPS
Video codecs
H.264, H.265
H.264, H.265, AV1, VP9
Audio codecs
AAC, AIFF, CAF, MP3, MP4, WAV
AAC, AIFF, CAF, MP3, MP4, WAV

Connectivity

4G support
LTE Cat. 15
LTE Cat. 24
5G support
Yes
Yes
Wi-Fi
5
6
Bluetooth
5.0
5.2
Navigation
GPS, GLONASS, Beidou, Galileo
GPS, GLONASS, Beidou, Galileo, QZSS, NAVIC
Download speed
-
Up to 7350 Mbps
Upload speed
-
Up to 3670 Mbps

Info

Announced
Nov 2022
Oct 2022
Class
Mid range
Flagship
Official page
-

AI

NPU
-
Next-gen Tensor Processing Unit

📊 Benchmark

AnTuTu 10
492,038
907,387
Geekbench 6
764
1,439
Geekbench 6
2,301
3,802
FP32 (float)
435
1,142