HiSilicon Kirin 810 vs HiSilicon Kirin 9000E 5G

The HiSilicon Kirin 810 and the HiSilicon Kirin 9000E 5G are two processors designed by HiSilicon, a subsidiary of Huawei. While both processors offer impressive specifications, there are some notable differences between them.

In terms of CPU cores and architecture, the HiSilicon Kirin 810 features two Cortex-A76 cores clocked at 2.27 GHz and six Cortex-A55 cores clocked at 1.88 GHz. On the other hand, the Kirin 9000E 5G boasts a more powerful configuration with one Cortex-A77 core clocked at 3.13 GHz, three Cortex-A77 cores clocked at 2.54 GHz, and four Cortex-A55 cores clocked at 2.05 GHz. This suggests that the Kirin 9000E 5G offers better performance and multitasking capabilities compared to the Kirin 810.

Both processors have eight cores and support the ARMv8.2-A instruction set. However, the Kirin 9000E 5G is manufactured using a more advanced 5 nm lithography process, while the Kirin 810 utilizes a 7 nm lithography process. A smaller lithography process generally allows for better power efficiency and performance.

In terms of transistor count, the Kirin 9000E 5G has 15,300 million transistors, which is more than twice the number found in the Kirin 810, which has 6,900 million transistors. This indicates that the Kirin 9000E 5G is a more advanced and sophisticated processor.

In the field of neural processing, both processors utilize HiSilicon's AI technology. The Kirin 810 incorporates the Ascend D100 Lite neural processing unit (NPU), while the Kirin 9000E 5G employs the Ascend Lite and Ascend Tiny NPUs. The Kirin 9000E 5G's neural processing capabilities are likely to be superior, allowing for more efficient and powerful AI functionalities.

Lastly, in terms of thermal design power (TDP), the Kirin 810 has a TDP of 5 watts, while the Kirin 9000E 5G has a slightly higher TDP of 6 watts. This suggests that the Kirin 9000E 5G may generate more heat, but also provide better performance.

In conclusion, the HiSilicon Kirin 9000E 5G surpasses the Kirin 810 in terms of processing power, lithography process, transistor count, and neural processing capabilities. However, it's important to note that these processors are designed for different purposes and target different market segments.