The high demand for the 5nm wafers coming out of TSMC has caused AMD to make a trim on their RX 7000. Specifically in its specifications. The reason behind this is that they share space with NVIDIA and it is important to have enough units produced to be able to sell them later. How have the plans been affected compared to those you had a few months ago?
If a few hours ago we told you that the RX 7900 XT could reach 92 TFLOPS, now we have to contradict ourselves with this news. And it is that the latest information about the RX 7000, based on RDNA 3 architecture, points to significant cuts compared to what was initially planned by Lisa Su’s company regarding its next generation of graphics cards. Of course, it must be clarified that until now the information that had been appearing did not have a reliable authority. The reason? Easy, it does not come from AMD itself, but from the different information leakers.
Although the RX 7000 will receive cuts compared to what was initially planned, they will represent a significant leap compared to the current generation. Also, the two highest performance configurations will be the first two chiplet-based graphics cards. Going ahead of NVIDIA and Intel when launching a graphic disaggregated into several chips. Something inherited from their Ryzen desktop processors, but we have yet to see how it affects performance on desktop GPUs. (www.stocktargetadvisor.com)
Changes in the configuration of future AMD RX 7000 graphics
First of all, we must bear in mind that months have passed since the specifications of the AMD RX 7000 based on its future RDNA 3 architecture were leaked. At the moment we do not have evidence of any graphics card that has been finished, so a launch much later than the NVIDIA RTX 40 is expected.
What has been the surprise? Well, the fact that the different GPUs in the range, Navi 31, 32 and 33, have received a considerable cut. In what? Well, in the number of ALUs in FP32 within each of the shader units. Which means that:
- top of the range, Navi 31has passed from 15,360 FP32 ALUs in total at 12,288. This number means 48WGP throughout the processor spread over 12 shader arrays in 6 Shader Engineyes
- As for the mid range, Navi 32the cut has been from the 10,240 cores of the leaked specifications last year to the 8,192. What translates into 32WGP in 8 shader arrays within 4 Shader Engines.
- And to finish with the low range, Navi 33 would have half the units of the previous case. Therefore, 4,096 FP32 ALUs in 16 WGP within 4 shader arrays that are inside 2 Shader Engines.
This cut in the RX 7000 means that the RX 7900 XT has gone from being rated at 92 TFLOPS to 73 TFLOPS. This if the speed of 3GHz and AMD decides not to make any last-minute changes to the graphics processor, which hasn’t been finished yet. So we should expect changes in that aspect as well. Has AMD relinquished the throne to NVIDIA for the umpteenth time or may there still be any surprises?
