HiSilicon Kirin 820 5G vs HiSilicon Kirin 9000 5G
The HiSilicon Kirin 820 5G and the HiSilicon Kirin 9000 5G are two processors with different specifications.
Starting with the HiSilicon Kirin 820 5G, it features a CPU architecture consisting of 1x 2.36 GHz Cortex-A76, 3x 2.22 GHz Cortex-A76, and 4x 1.84 GHz Cortex-A55 cores. With a total of 8 cores, this processor operates on the ARMv8.2-A instruction set. It is built using a 7 nm lithography process and has a thermal design power (TDP) of 6 Watts. The HiSilicon Kirin 820 5G is equipped with the Ascend D110 Lite Neural Processing Unit, which utilizes the HUAWEI Da Vinci Architecture.
On the other hand, the HiSilicon Kirin 9000 5G boasts a more powerful CPU architecture. It comprises 1x 3.13 GHz Cortex-A77, 3x 2.54 GHz Cortex-A77, and 4x 2.05 GHz Cortex-A55 cores. Like its counterpart, it also has 8 cores and operates on the ARMv8.2-A instruction set. However, the Kirin 9000 5G is built using a smaller 5 nm lithography process. It incorporates a staggering 15,300 million transistors and has a TDP of 6 Watts. The Kirin 9000 5G features an upgraded Neural Processing Unit, with the combination of Ascend Lite (2x) and Ascend Tiny (1x), utilizing the HUAWEI Da Vinci Architecture 2.0.
In comparison, the Kirin 9000 5G is a more advanced processor than the Kirin 820 5G. It offers higher clock speeds across its cores, resulting in potentially better performance. The smaller 5 nm lithography process allows for increased transistor count, which could lead to improved efficiency and power consumption. Furthermore, the upgraded Neural Processing Unit on the Kirin 9000 5G, incorporating Ascend Lite and Ascend Tiny, suggests enhanced AI capabilities.
While the Kirin 820 5G is a capable processor, particularly for mid-range devices, the Kirin 9000 5G stands out as a flagship-level processor with superior specifications. Its higher clock speeds, smaller lithography process, and advanced Neural Processing Unit make it a superior choice for demanding tasks and applications.
Starting with the HiSilicon Kirin 820 5G, it features a CPU architecture consisting of 1x 2.36 GHz Cortex-A76, 3x 2.22 GHz Cortex-A76, and 4x 1.84 GHz Cortex-A55 cores. With a total of 8 cores, this processor operates on the ARMv8.2-A instruction set. It is built using a 7 nm lithography process and has a thermal design power (TDP) of 6 Watts. The HiSilicon Kirin 820 5G is equipped with the Ascend D110 Lite Neural Processing Unit, which utilizes the HUAWEI Da Vinci Architecture.
On the other hand, the HiSilicon Kirin 9000 5G boasts a more powerful CPU architecture. It comprises 1x 3.13 GHz Cortex-A77, 3x 2.54 GHz Cortex-A77, and 4x 2.05 GHz Cortex-A55 cores. Like its counterpart, it also has 8 cores and operates on the ARMv8.2-A instruction set. However, the Kirin 9000 5G is built using a smaller 5 nm lithography process. It incorporates a staggering 15,300 million transistors and has a TDP of 6 Watts. The Kirin 9000 5G features an upgraded Neural Processing Unit, with the combination of Ascend Lite (2x) and Ascend Tiny (1x), utilizing the HUAWEI Da Vinci Architecture 2.0.
In comparison, the Kirin 9000 5G is a more advanced processor than the Kirin 820 5G. It offers higher clock speeds across its cores, resulting in potentially better performance. The smaller 5 nm lithography process allows for increased transistor count, which could lead to improved efficiency and power consumption. Furthermore, the upgraded Neural Processing Unit on the Kirin 9000 5G, incorporating Ascend Lite and Ascend Tiny, suggests enhanced AI capabilities.
While the Kirin 820 5G is a capable processor, particularly for mid-range devices, the Kirin 9000 5G stands out as a flagship-level processor with superior specifications. Its higher clock speeds, smaller lithography process, and advanced Neural Processing Unit make it a superior choice for demanding tasks and applications.
CPU cores and architecture
Architecture | 1x 2.36 GHz – Cortex-A76 3x 2.22 GHz – Cortex-A76 4x 1.84 GHz – Cortex-A55 |
1x 3.13 GHz – Cortex-A77 3x 2.54 GHz – Cortex-A77 4x 2.05 GHz – Cortex-A55 |
Number of cores | 8 | 8 |
Instruction Set | ARMv8.2-A | ARMv8.2-A |
Lithography | 7 nm | 5 nm |
Number of transistors | 15300 million | |
TDP | 6 Watt | 6 Watt |
Neural Processing | Ascend D110 Lite, HUAWEI Da Vinci Architecture | Ascend Lite (2x) + Ascend Tiny (1x), HUAWEI Da Vinci Architecture 2.0 |
Memory (RAM)
Max amount | up to 12 GB | up to 16 GB |
Memory type | LPDDR4X | LPDDR5 |
Memory frequency | 2133 MHz | 2750 MHz |
Memory-bus | 4x16 bit | 4x16 bit |
Storage
Storage specification | UFS 2.1 | UFS 3.1 |
Graphics
GPU name | Mali-G57 MP6 | Mali-G78 MP24 |
GPU Architecture | Valhall | Valhall |
GPU frequency | 850 MHz | 760 MHz |
Execution units | 6 | 24 |
Shaders | 96 | 384 |
DirectX | 12 | 12 |
OpenCL API | 2.1 | 2.1 |
OpenGL API | ES 3.2 | ES 3.2 |
Vulkan API | 1.2 | 1.2 |
Camera, Video, Display
Max screen resolution | 3840x2160 | |
Max camera resolution | 1x 48MP, 2x 20MP | |
Max Video Capture | 4K@30fps | 4K@60fps |
Video codec support | AV1 H.264 (AVC) H.265 (HEVC) VP8 VP9 |
H.264 (AVC) H.265 (HEVC) VP8 VP9 |
Wireless
4G network | Yes | Yes |
5G network | Yes | Yes |
Peak Download Speed | 1.6 Gbps | 4.6 Gbps |
Peak Upload Speed | 0.2 Gbps | 2.5 Gbps |
Wi-Fi | 6 (802.11ax) | 6 (802.11ax) |
Bluetooth | 5.1 | 5.2 |
Satellite navigation | BeiDou GPS GLONASS |
BeiDou GPS Galileo GLONASS NavIC |
Supplemental Information
Launch Date | 2020 March | 2020 October |
Vertical Segment | Mobiles | Mobiles |
Positioning | Mid-end | Flagship |
AnTuTu 10
Total Score
GeekBench 6 Single-Core
Score
GeekBench 6 Multi-Core
Score
Popular comparisons:
1
MediaTek Helio G80 vs Qualcomm Snapdragon 820
2
Apple A16 Bionic vs Samsung Exynos 7880
3
HiSilicon Kirin 710A vs MediaTek Dimensity 800U
4
Qualcomm Snapdragon 680 vs Qualcomm Snapdragon 460
5
Qualcomm Snapdragon 765 vs MediaTek Helio G90
6
HiSilicon Kirin 935 vs HiSilicon Kirin 990 5G
7
HiSilicon Kirin 820 5G vs Apple A10X Fusion
8
Unisoc Tiger T710 vs Qualcomm Snapdragon 730
9
Samsung Exynos 9609 vs MediaTek Dimensity 1000
10
MediaTek Helio P35 vs Google Tensor G2