Zhaoxin: New Chinese APUs and CPUs based on 16 nm introduced

Zhaoxin: New Chinese APUs and CPUs based on 16 nm introduced


from Maximilian Hohm
With the KX-6000G and KH-4000 series, the Chinese company Zhaoxin has introduced new processors with and without an onboard graphics unit. The chips are manufactured using the 16 nm process and feature very low core clock speeds and support for DDR4 memory and PCI-E 3.0. Read more about this below.

If you want to buy a competitive x86 architecture for the end customer or professional area, you can’t avoid AMD and Intel for many years. The two American companies divide the market between themselves and thereby create one-sided dependencies with other countries. These include nations like China, which could theoretically be cut off from such products due to sanctions and are therefore developing their own alternatives. These are usually not nearly as powerful, but still a first step towards your own security.

AMD Ryzen 7000 Mobile: Asus leaks Ryzen 7 7730U in Zenbook 14

Recommended Editorial ContentHere you will find external content from [PLATTFORM]. To protect your personal data, external integrations are only displayed if you confirm this by clicking on “Load all external content”:I consent to external content being displayed to me. Personal data is thus transmitted to third-party platforms. Read more about our privacy policy.

On 01.11. the Chinese company Zhaoxin presented two new CPU series for different application scenarios. The Zhaoxin KX-6000G series was introduced for compact computers, passive systems and tablets. The quad-core APUs offer an integrated graphics unit and are manufactured using the 16 nm Lujiazui process. The C-1080 graphics unit in particular is said to be significantly faster than that of the previous model. The chips’ base clock is between 2 and 3 GHz, depending on the TDP, while the boost clock is between 3 and 3.3 GHz. In addition to DDR4 memory, PCI-E 3.0 and USB 3.2 Gen1 are supported. Zhaoxin promises four times the graphics performance of the previous model and up to 60 percent higher efficiency. At the same time, the company wants to have reduced idle power consumption by 50 percent.

It also introduced the KH-4000 series based on the Yongfeng architecture, which is designed for embedded systems with a lot of computing power. There should be models with twelve, 16 and 32 cores, but they have to do without hyperthreading. The clock rates are between 2 and 2.2 GHz, while a maximum of 2 TB of DDR4 ECC memory can be used per CPU. In the case of the 32 core, there are also up to 128 PCI-E 3.0 lanes, the smaller models have to make do with fewer lanes.

What performance the chips deliver in comparison to AMD’s and Intel’s offerings is still unclear, but production in 16 nanometers is clearly outdated and can be greatly outbid by both manufacturers. At the same time, the higher transfer rate that PCI-E 4.0 and PCI-E 5.0 could have offered is missing in the compute area.

Source: @Loeschzwerg_3DC



Reference-www.pcgameshardware.de