HiSilicon Kirin 810 vs HiSilicon Kirin 950
The HiSilicon Kirin 810 and HiSilicon Kirin 950 are both powerful processors designed by HiSilicon Technologies, a subsidiary of Huawei. While they are both capable processors, they have some notable differences in their specifications.
Starting with the HiSilicon Kirin 810, it features an architecture consisting of 2x 2.27 GHz Cortex-A76 cores and 6x 1.88 GHz Cortex-A55 cores. With a total of 8 cores, it offers a versatile and efficient performance. It operates on the ARMv8.2-A instruction set and has a lithography of 7 nm, which means it is manufactured using a highly advanced and energy-efficient process. In terms of the number of transistors, it boasts 6900 million, showcasing its ability to handle complex tasks. Additionally, it has a thermal design power (TDP) of 5 Watt, making it energy-efficient. It also includes Ascend D100 Lite for neural processing, utilizing HUAWEI Da Vinci Architecture.
On the other hand, the HiSilicon Kirin 950 operates on an architecture comprising 4x 2.4 GHz Cortex-A72 cores and 4x 1.8 GHz Cortex-A53 cores. Similar to the Kirin 810, it also has a total of 8 cores. However, it employs the ARMv8-A instruction set and has a lithography of 16 nm, indicating it is manufactured on a slightly older and less energy-efficient process. It possesses 2000 million transistors, which is notably lower than the Kirin 810. This suggests that it may not handle more demanding tasks as efficiently as its counterpart. Like the Kirin 810, it also has a TDP of 5 Watts.
In summary, the HiSilicon Kirin 810 and HiSilicon Kirin 950 differ in terms of their CPU architecture, instruction set, lithography, number of transistors, and neural processing capabilities. The Kirin 810 is built on a more advanced 7 nm process and includes Ascend D100 Lite with HUAWEI Da Vinci Architecture, while the Kirin 950 operates on a 16 nm process and lacks dedicated neural processing hardware. These differences may impact their performance, energy efficiency, and ability to handle complex tasks.
Starting with the HiSilicon Kirin 810, it features an architecture consisting of 2x 2.27 GHz Cortex-A76 cores and 6x 1.88 GHz Cortex-A55 cores. With a total of 8 cores, it offers a versatile and efficient performance. It operates on the ARMv8.2-A instruction set and has a lithography of 7 nm, which means it is manufactured using a highly advanced and energy-efficient process. In terms of the number of transistors, it boasts 6900 million, showcasing its ability to handle complex tasks. Additionally, it has a thermal design power (TDP) of 5 Watt, making it energy-efficient. It also includes Ascend D100 Lite for neural processing, utilizing HUAWEI Da Vinci Architecture.
On the other hand, the HiSilicon Kirin 950 operates on an architecture comprising 4x 2.4 GHz Cortex-A72 cores and 4x 1.8 GHz Cortex-A53 cores. Similar to the Kirin 810, it also has a total of 8 cores. However, it employs the ARMv8-A instruction set and has a lithography of 16 nm, indicating it is manufactured on a slightly older and less energy-efficient process. It possesses 2000 million transistors, which is notably lower than the Kirin 810. This suggests that it may not handle more demanding tasks as efficiently as its counterpart. Like the Kirin 810, it also has a TDP of 5 Watts.
In summary, the HiSilicon Kirin 810 and HiSilicon Kirin 950 differ in terms of their CPU architecture, instruction set, lithography, number of transistors, and neural processing capabilities. The Kirin 810 is built on a more advanced 7 nm process and includes Ascend D100 Lite with HUAWEI Da Vinci Architecture, while the Kirin 950 operates on a 16 nm process and lacks dedicated neural processing hardware. These differences may impact their performance, energy efficiency, and ability to handle complex tasks.
CPU cores and architecture
Architecture | 2x 2.27 GHz – Cortex-A76 6x 1.88 GHz – Cortex-A55 |
4x 2.4 GHz – Cortex-A72 4x 1.8 GHz – Cortex-A53 |
Number of cores | 8 | 8 |
Instruction Set | ARMv8.2-A | ARMv8-A |
Lithography | 7 nm | 16 nm |
Number of transistors | 6900 million | 2000 million |
TDP | 5 Watt | 5 Watt |
Neural Processing | Ascend D100 Lite, HUAWEI Da Vinci Architecture |
Memory (RAM)
Max amount | up to 8 GB | up to 4 GB |
Memory type | LPDDR4X | LPDDR4 |
Memory frequency | 2133 MHz | 1333 MHz |
Memory-bus | 4x16 bit | 2x32 bit |
Storage
Storage specification | UFS 2.1 | UFS 2.0 |
Graphics
GPU name | Mali-G52 MP6 | Mali-T880 MP4 |
GPU Architecture | Bifrost | Midgard |
GPU frequency | 820 MHz | 900 MHz |
Execution units | 6 | 4 |
Shaders | 96 | 64 |
DirectX | 12 | 11.2 |
OpenCL API | 2.0 | 1.2 |
OpenGL API | ES 3.2 | |
Vulkan API | 1.0 | 1.0 |
Camera, Video, Display
Max camera resolution | 1x 48MP, 2x 20MP | 1x 31MP, 2x 13MP |
Max Video Capture | FullHD@30fps | FullHD@60fps |
Video codec support | H.264 (AVC) H.265 (HEVC) VP8 VP9 |
H.264 (AVC) H.265 (HEVC) VP8 |
Wireless
4G network | Yes | Yes |
5G network | Yes | Yes |
Peak Download Speed | 0.6 Gbps | 0.3 Gbps |
Peak Upload Speed | 0.15 Gbps | 0.05 Gbps |
Wi-Fi | 6 (802.11ax) | 5 (802.11ac) |
Bluetooth | 5.1 | 4.2 |
Satellite navigation | BeiDou GPS GLONASS |
BeiDou GPS Galileo GLONASS |
Supplemental Information
Launch Date | 2019 Quarter 2 | 2015 November |
Partnumber | Hi6280 | Hi3650 |
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
Qualcomm Snapdragon 821 vs MediaTek Dimensity 6080
2
MediaTek Dimensity 1000 vs MediaTek Helio G35
3
Google Tensor G2 vs MediaTek Helio G96
4
Qualcomm Snapdragon 480 Plus vs Apple A10X Fusion
5
Qualcomm Snapdragon 662 vs Qualcomm Snapdragon 685
6
Samsung Exynos 7904 vs HiSilicon Kirin 710A
7
MediaTek Dimensity 810 vs Qualcomm Snapdragon 835
8
MediaTek Dimensity 1080 vs MediaTek Helio G36
9
MediaTek Helio P70 vs Unisoc Tiger T616
10
Qualcomm Snapdragon 636 vs Qualcomm Snapdragon 8 Plus Gen 1