Mystery NVIDIA Ampere-based GPU Nears The Top Of Graphics Computing, Second Only To A ‘Record Graphics Card’

As AMD readies its Big Navi RDNA 2-based heavyweights, NVIDIA appears all set with it’s Ampere-based RTX 3000 Series. A mystery NVIDIA Ampere-based GPU recently appeared in a benchmark comparison with its own predecessors, and the results are quite promising. The results strongly indicate that AMD might not yet be able to claim the title ‘NVIDIA Killer’ for any of its top-end graphics cards despite offering hardware-level ray tracing on some.

The upcoming NVIDIA Ampere-based RTX 3000 Series are reportedly much better than the RTX 2000 Series which are already leading the market of high-end gaming as well as purpose-driven graphics-intensive computing. Hence the RTX 3000 Series will not only have to be significantly better to maintain and extend the market leadership position but also ensure the AMD’s rapid ascend in the CPU segment doesn’t disturb the GPU segment as well.

Leaked Benchmark Compilation of NVIDIA Top-End GPUs Confirms Improved Performance Over Turing Architecture and Next-Gen Memory:

A recently leaked Time Spy Extreme Graphics scoresheet contains GPUs that are considered the best. The top-end is populated by two custom variants of the RTX 2080 Ti, including the EVGA RTX 2080 Ti XC as overclocked by Kingpin and the MSI RTX 2080 Ti Lightning, as well as the NVIDIA TITAN V. However, a mystery NVIDIA Ampere-based GPU has managed to firmly place itself in the second position, without even being launched yet.

The unidentified but obviously top-end NVIDIA Ampere GPU scored 18,257. What the list doesn’t clearly mention is the KINGPIN’s 2080 Ti with 2.38 GHz Core and 2170MHz memory, scored 8863, while the KINGPIN’s Titan V with 2.28 GHz Core and 1013 MHz memory, scored 8577. The mystery NVIDIA Ampere GPU, despite its substantially lower 1935 MHz Core, but blazingly fast and reportedly GDDR6X memory, scored in the range of 8700 and placed itself firmly in between the two custom-designed top-end NVIDIA Turing-based GPUs.

How Does The Leaked NVIDIA Ampere GPU Benchmark Translate Into Real-World Performance?

Simplifying the results for real-world performance, the mystery NVIDIA Ampere GPU is about 30 percent faster, even in the engineering sample iteration. It is, however, important to note that the reporting of the numbers seem odd and could be faulty due to an important technical reason.

While the low Core Clock speeds could be due to the GPU being in development and prototyping stage, the insanely high 6000 MHz memory speeds could be an error. This is because there have consistent rumors that NVIDIA is testing two variants of the next-gen Ampere-based GPUs. While one variant is reportedly using the GDDR6 memory, the other is working with next-gen GDDR6X memory. The latter is cutting-edge, and there aren’t confirmed benchmarks or specifications available for inclusion in test benchmark repositories. Simply put, programs that detect this particular data may not be prepared or amended for these new cards.

Despite the anomalous numbers, the leak reaffirms the sheer performance boost that NVIDIA’s loyal buyers have come to expect with every generation’s evolution. It may not be clear which NVIDIA RTX 3000 Series, be it NVIDIA RTX 3080, RTX 3080 Ti/Super or the RTX 3090, but AMD still has a long way before it can claim the title of ‘NVIDIA Killer’.

Alap Naik Desai
A B.Tech Plastics (UDCT) and a Windows enthusiast. Optimizing the OS, exploring software, searching and deploying solutions to strange and weird issues is Alap's main interest.