Radeon RX 8000 can run on old memory. Even the more powerful models don’t have faster GDDR6

#Radeon #run #memory #powerful #models #dont #faster #GDDR6

During Easter alleged specs for Radeon RX 8000 graphics cards have surfacedi.e. new graphics based on the RDNA 4 architecture for the mainstream and cheaper part of the market (while the high-end model was reportedly canceled by AMD). Now another rumor has appeared, according to which AMD will not switch to the new GDDR7 memory technology for these graphics, which is expected to appear at Nvidia. However, it looks like it won’t even take advantage of the faster GDDR6 memories that are on the market.

According to leaker Kepler_L2, AMD is going to use the same 18.0GHz (effectively) GDDR6 memory that they already had Radeony RX 6950 XT, RX 6750 XT and cheap Radeon RX 6500 XT Two years ago. AT Radeonu RX 7900 XTX a XT at the same time, it already has 20GHz GDDR6 memories, which provide 11% more throughput.

Previous reports agreed that the 18.0 GHz effective memory frequency would be used by a card based on the weaker and cheaper of the two chips, the Navi 44 with 2048 shaders and a 128-bit bus. The throughput of this GPU should then be 288 GB/s, the same as u Radeon RX 7600.

The chip will probably again have a 32MB Infinity Cache, and according to the previous leak, the effective throughput after taking into account its effect will not be much higher than that of the Navi 33. It is a question of how much the performance can be increased, if the efficiency of the memory subsystem does not increase much. But we’ll see from the reviews.

The top model of the RX 8000 series with a throughput of only 576 GB/s

But the news is that the same memories should be used for the more powerful Radeon based on the more powerful Navi 48 chip with 4096 shaders, where it raises more doubts. This graphics could reportedly reach up to 50 TFLOPS of processing power and come close to the Radeon RX 7900 XT (if this part of the leaks is not wrong). The previous version of the specifications assumed that the memory bandwidth would be higher with a frequency of around 20.5 GHz, but according to Kepler_L2, even this GPU only has 18.0 GHz of memory. This would give it a throughput of only 576 GB/s like the aforementioned Radeon RX 6950 XT.

Radeon RX 6950 XT

Autor: AMD

The question is, what are the reasons for this? Leaving aside the possibility that the memory controller implementation proved to be insufficient for higher frequencies (which could be because it was designed to take up less space), then this could either be a measure to reduce power or a reduction production costs by using cheaper GDDR6 chips.

But the question is whether this will not reduce the performance so much that it will be a net disadvantage overall. We may have an example in the form of Radeon RX 7900 GRE, which also only has 18.0GHz of memory (and also on a 256-bit bus) and is quite possibly a limitation that prevents it from achieving better performance after overclocking, which it should otherwise have good potential for. At least in theory, GDDR6 memories would they were supposed to be able to reach frequencies up to 24.0 GHz (about a third higher throughput should be known). But as already said, it might not be advantageous in terms of price and/or consumption.

These graphics are expected to be aimed at the cheaper end of the market, somewhere between the Radeon RX 7600 (the Navi 44 should be faster than this model) and the Radeon RX 7900 XTX (which should remain more powerful even after the launch of RDNA 4). But we still have no idea how hot it will really be with that cheapness, so it’s hard to predict whether the Radeon RX 8000 models will be attractive.

Sources: Kepler_L2 (1, 2), VideoCardz