Kopite7kimi A reliable source of information about Nvidia products has released new details about the specifications of the company’s new generation of GPUs.
While only three months since the release of the series graphics cards Ampere By the way, it seems that now new information has been released about the next generation of Intel GPUs. Apparently the company is in close competition with processors AMD RDNA2 With series Ampere In the field of graphics cards, high-end and not very professional professionals, and we should probably expect the arrival of processors based on architectural style earlier than usual. Lovelace Have. The previous generation of Nvidia processors, the Turing series, has been on the market for nearly two years, while the Ampere series looks set to have a shorter lifespan.
Kopite7kimi, From reliable sources that publish information about Nvidia products, has published new rumors about the details of the new generation of Nvidia GPUs. Based on this, it seems that the processor AD102 As the strongest version in the generation based on architectural style Lovelace With 71% growth in the number of shading units (Shader Units) Will be equipped with 18432 shading units. According to 3DCenter.org, there are 18,432 cores CUDA in this GPU It can mean using 144 units SM Or Streaming Multiprocessors And produce a theoretical clock speed of 1.80 GHz, or in other words, a power generation of 66 teraflops.
Possibly series GPUs Lovelace By using the production process of 5-nanometer silicones and the very high compression of its transistors, they show a big step in the field of technical performance. In other Nvidia-related news, we recently received confirmation of the custom version of the GeForce RTX 3080 Ti and GeForce RTX 3060 Ragstreix graphics cards. The GeForce RTX 3080 Ti graphics card contains 20 GB of GDDR6X memory and the GeForce RTX 3060 is equipped with 12 GB of GDDR6 memory.