The Radeon RX 7900 XTX, a faster RDNA 3 graphics card that will include the entire Navi 31 XTX GPU core, is also being prepared by AMD. A rumor that AMD would introduce its Radeon RX 7900 XT graphics card with 20 GB GDDR6 memory surfaced last week.

The GPU will employ a scaled-down Navi 31 XT chip and ought to be quite competitive with NVIDIA’s GeForce RTX 4090 graphics processor. We also mentioned that a second graphics card is in development; while we are unable to confirm its exact name at this time, our friends at Benchlife appear to have predicted it.

According to the most recent information, the AMD Radeon RX 7900 XTX will be equipped with the top-tier Navi 31 XTX GPU and 24 GB of GDDR6 RAM. Both cards, which will be at the top of the lineup, will make use of MCM RDNA 3 GPUs. As a result, AMD will still have room for the later release of higher-end SKUs like the RX 7950 XT and RX 7950 XTX. The Navi 31 GPU is said to have a 3D V-Cache design for Infinity Cache in addition to its MCM design. Additionally, a second Navi 3X die has been developed to provide even greater performance, however, it is doubtful that it will ever be used in the Radeon RX 7000 line.

The next-generation enthusiast cards, including the Radeon RX 7900 XTX/XT graphics card, would be powered by the AMD Navi 31 GPU, the top-tier RDNA 3 processor.

It’s reported that AMD would replace its next-generation RDNA 3 GPUs’ compute units with workgroup processors. Dual CU (Compute Units) will be housed in each WGP, but there will be four SIMD32 clusters per CU, as opposed to two on each CU in RDNA 2. According to rumors, AMD has the choice of the 6nm die from TSMC or Samsung.

The AMD Navi 31 GPU with RDNA 3 architecture is anticipated to feature a single GCD with 48 GPS, 12 SAs, and 6 SEs, according to the most recent information. There will be a total of 12,288 stream processors distributed as a result. This represents a 2.4x increase in cores above the 5120 SPs found in the Navi 21 GPU. According to reports, the Navi 31 GCD, often known as the GPU, would be packaged on TSMC’s 5nm production node and measure 308mm2. The reduced version will have 5 MCDs for an 80MB Infinity Cache and 42 WGPs for 10,752 cores on a 320-bit bus interface.

The Navi 31 GPU will also include 6 MCDs, each of which will have a 16 MB Infinity Cache. Additionally, these MCDs are anticipated to include 64-bit (32-bit x 2) memory controllers, giving the device a 384-bit bus interface. There is a 3D-Stacked solution in development that would double the Infinity Cache with 32 MB (16 MB 0-hi + 16 MB 1-hi) capacity for a total of 192 MB of cache, even though this equates to 96 MB of Infinity Cache, which is less than the 128 MB featured on the current Navi 21 GPUs.

As a result, there will be a greater power drain, and AMD appears to have confirmed this. However, their array of next-generation graphics cards will still be more energy-efficient than what NVIDIA has to offer, despite having higher power usage. For a >2x performance increase, the AMD Radeon RX 6950 XT already has a TBP of 335W. The revised triple-fan cooling design, which is a little bit taller than the one currently in use, is expected to remain on the cards, along with its dual 8-pin connector input for power.

Also read:

source