AMD RDNA 4 Strategy Shift Cancels RX 9080 XT Flagship to Focus on Mid Range RX 9070 XT with GDDR6 Memory
There is conjecture floating around the GPU market that AMD is regrading its forthcoming RDNA 4 hardware series by trimming the A1 plans for a flagship gamer GPU. Based on an industry speculation posted on the Broken Silicon podcast Moores Law Is Dead, the semiconductor designer apparently may put a stop to the Radeon RX 9080 XT project. Reportedly, the shutdown is due to the astronomically expensive production costs of the next generation GDDR7 memory, which is soaring in the market owing to massive demand from the artificial intelligence industry.
Should the rumors hold any truth, AMD could be abandoning the entry level market completely by only focusing on mid range and upper mid range cards, the RX 9070 and the RX 9070 XT respectively. This aggressive retreat would ultimately leave the team with a defense against the newly revealed Nvidia RTX 50 Super series refresh. However, by only offering mid range parts, AMD can preserve the fierce performance to price ratio seen on the team last year.
The previous launch of the Radeon RX 9070 XT gave AMD a large head start against the competition. With 16 GB of VRAM in comparison to the Nvidia RTX 5070 Ti, it was priced at 700 to 750 dollars, whereas the competitor was found to cost 900 dollars on Amazon, a figure more than 25 % higher than the rival devices it was competing against. When it is overclocked to the maximum, the RX 9070 XT attained equality with the RTX 5080 cost wise despite costing only a fraction over fifty percent of what the latter did. This added from the value, leading AMD chief executive Lisa Su to declare that the graphics card had turned out to be the company's most successful Radeon, up to that point of time.
Designed to capitalize on this momentum, the proposals RX 9080 XT was based around applying this Navi 48 silicon chip. The flagship product was supposed to provide up to a 30% performance boost over the RX 9070 XT, surpassing the expected RTX 5080 Super. These early engineering specs, necessitated using the best available GDDR7 RAM options of 24 GB or 32 GB. Though, the current economic burden of GDDR7 prices AMD with price the GPU at a point where it is missing its usual hardware cost advantage against Nvidia.
By using mature GDDR6 memory AMD has weaved its existing range of RX 9000 series around the more volatile market for newer GDDR7 types. GDDR6 is still extremely cheap and according to industry sources is predicted to be ever cheaper, enabling AMD to command a massive price premium over Nvidia, whose entire RTX 50 series does use the costlier GDDR7 rather. To take advantage of the price differential AMD analysts are now calling for a supercharged refresh of the RX 9070 XT with additional capacity of equally high specs GDDR6 memory and faster clock speeds to hit the RTX 5080 without elevating production costs.
Nevertheless, this middle ground is present on the professional market with the Radeon AI PRO R9700. Its architecture RDNA 4 is identical to the computer with a local artificial intelligence development in mind and its clock speed is 2.9 GHz for a drawing power of 300W. To keep the pricing at a reasonable level when it gets a high capacity, the graphics card features 32 GB of GDDR 6 memory with a bus of 256 bit. Whether AMD sticks with GDDR 6 for its consumer gaming range, or postpones its high end desktop card until its RDNA 5 next generation architecture, so far unannounced at the time of writing this article, is still a question of the firm today.
