Competition between AMD and NVIDIA, as usual. In this case, two chips with a similar power face each other, the AMD Genoa vs an NVIDIA Grace. But even having similar characteristics, one of the two stands out over the other. What do you think it is?
We are going to analyze its performance and reveal which of these two chips becomes the king of the range. Once you see them you will not even think about buying the other one.
AMD Genoa vs. NVIDIA Grace
And it is that, apparently, NVIDIA would have presented the data of its new Grace processor, which offers on multiple occasions up to more than double the performance compared to those of Intel and AMD having the same power.
Powered by Arm Neoverse N2 cores, the Grace CPU will be used in its superchips. NVIDIA recently announced its most powerful GPU for AI workloads known as GH200which also comes with memory HBM3e fastest in the world and that will be adopted by the Superchip Grace Hopper.
Among the highlights of Grace we can see the following:
- High-performance CPU for HPC and cloud computing
- Superchip design with up to 144 Arm v9 CPU cores
- World’s first LPDDR5x with ECC memory, 1TB/s total bandwidth
- SPECrate2017_int_base more than 740 (estimated)
- 900 GB/s coherent interface, 7 times faster than PCIe Gen 5
- 2 times the packing density of DIMM-based solutions
- Twice the performance per watt of today’s leading CPU
- Runs all NVIDIA software stacks and platforms, including RTX, HPC, AI, and Omniverse
In the presentation given at Hot Chips 2023, Bill Dally, one of NVIDIA’s scientists, conducted a test between this new NVIDIA Grace and a competing dual-socket x86, among which we saw AMD’s EPYC 9654, which is its fastest flagship with 96 cores and 192 threads, and also Intel’s Xeon Platinum 8480+, which has 56 cores and 112 threads.
In both cases, the NVIDIA chip doubled the performance and speed data compared to other brands. And it is that this Grace offers us a total of 144 cores (72 Arm Neoverse V2 per chip), supports up to 960 GB of LPDDR5X memory with up to 1 TB/s of raw bandwidth and has a combined power consumption of 500 W .
In short, NVIDIA has placed itself at the top of the ranking, achieving a clear advantage over its rivals, with the same architecture and power. Something that can mean a big change of course. In some cases with up to x2.5.
For now it seems that the company led by Jen-Hsun Huang is doing things quite well, with a projected future in AI and this type of news, we may see a clear winner, a step above its rivals and with a technological future very promising. We will be following it closely and telling you all the news that we continue to have, which seems to be many soon.