Chinese GPU has the same memory as NVIDIA’s next-gen RTX 4090

Chinese GPU manufacturer InnoSilicon has achieved a remarkable feat. In November last year, it launched the Fenghua No.1 graphics card for desktops and data centers that promised leadership in computing performance and technologies. Now we have more details about this model, confirming the specifications. The first and most important revelation is regarding memory, as the Fenghua graphics card uses the same 21 Gbps GDDR6X memory found in the NVIDIA RTX 3090 Ti and its successor, the RTX 4090.

Micron has confirmed on its website that it is working with Innosilicon to improve the PAM4 signaling technology that powers its memory. GDDR6X. The two have jointly developed memory for AI and neural networks:

When combined with a 256-bit bus, this results in a bandwidth of more than 5 Tbits/s, 5 times more than LPDDR4 memory and roughly the same as HBMalbeit at a much lower cost. In addition to state-of-the-art graphics memory, this Chinese GPU also comes with a chiplet design. In addition, Fenghua has the latest I/O technologies, including PCIe Gen 4, Display Port 1.4, HDMI 2.1, etc.

Fuente: Mydrivers