As speculation swirls about the next AMD Radeon RDNA 4 GPUs, one rumor insists that AMD will not be taking part in the one-upmanship and will use “classic” GDDR6 memory for the whole of its future RX 8000 range. This decision may come as a surprise to those still hoping to see AMD gain the upper hand over Nvidia (which could equip its very high-end range with GDDR7), but in reality, although we can’t really confirm this rumour, it would make sense in terms of AMD’s strategy.
New RX 8000: lower prices (and lower ambitions)?
If you’ve been following the news and rumors about the development of the RDNA4 range, we learned relatively early on that AMD would not be fighting with Nvidia at the very top end of the market. Initially, AMD could have offered a high-end GPU variant, but this option seems to have been cancelled in favor of mid-range and low-end products. According to leaks, Navi 48 and Navi 44 will make up the future range. The Navi 48 GPU would be the top-of-the-range GPU enabling multiple card declinations, while the Navi 44 chip would do the same for entry-level declinations. If this logic is indeed the reality of AMD’s strategy for its next family of GPUs, then GDDR 6, even if limited to 18 Gbps, shouldn’t be a problem.
AMD RX 8000: GDDR6, a sensible choice
But the main argument for using GDDR6 is not just technical. GDDR6 memory has already been mass-produced for years and is widely available. Current tensions suggest that the second half of the year will be a complex one, and the use of more recent memory could entail a risk not only of availability, but also of price inflation. As far as strategy is concerned, even if we’re still dealing with rumors, what comes up regularly is a desire to offer the same performance as current RX7000 cards, but at a lower price segment. Recent leaks indicate that the Navi 48 GPU will offer performance close to that of the Navi 31 chipset, while Navi 44 will be positioned between the Navi 32 and Navi 33″ RDNA 3 GPUs.
