We have been hearing rumors about AMD’s next big GPU lineup for a while but details on these new cards have been scarce, until now. A new leak has uncovered detailed specifications and benchmarks for the company’s new RX Gamma XT flagship GPUs and it puts into perspective just how powerful the new ‘Big Navi’ cards could be.
The leaked slides come from an unprotected link on the website for one of AMD’s partners and they reveal the design as well as some of the specifications for the RX Gamma XT GPUs. The new cards will be using a dual-fan design but the appearance will likely get a few tweaks before the final release as these are not the final retail versions.
The compute engine capacity on the new card has been doubled and is now referred to as Quad Compute Unit (QCU). Just as before, the clock speeds are divided into three categories and the maximum clock speed it can reach is 2.633 GHz. The only reason the GPU was able to make such a leap in clock speeds is that the card is based on the 7nm EUV based process.
The slide also reveals that there are 80 compute units, 5120 stream processors and that the card will be able to deliver a whopping 26.6 TFLOPs of single-precision compute. There will be 32 GB of GDDR7 memory clocked at 24 Gbps and double the ray-tracing support with up to 21.2 Gigarays per second.
As seen in the image above, AMD is targeting the most high-end market with the 4K 120Hz gaming standard and the comparisons included in the slide should be taken with a grain of salt as retail performance is likely going to be different.
AMD’s new RX Gamma XT flagships are expected to debut in July this year but only time will tell whether the rumors regarding the release date carry any weight.
This was an April Fool’s joke by AMD. The specifications and claims are fictitious and there won’t be a product launching with such specifications.