The next-gen NVIDIA graphics cards are reportedly packing the equally futuristic GDDR6X memory. Now a mystery, premium GPU, manufactured by NVIDIA Corporation has appeared in a leaked 3DMark database indirectly confirming the existence of the high-end, next-gen memory onboard the graphics card.
A confidential 3DMark benchmarking database result strongly indicates testing of the next-gen NVIDIA graphics card is in the final stages. Incidentally, the memory bandwidth, clock speeds as well as the Boost Clocks of the unknown NVIDIA Graphics Card fits well with the earlier reports about the NVIDIA GeForce RTX 3080 Ti, 3090 or 3080.
3DMark Database Leak Confirms GDDR6X Memory And Other Details Of Top-End NVIDIA GeForce RTX 30 Series Graphics Cards Family?
The 3DMark benchmarking database result is marked ‘Hidden’. In other words, the results are ‘Private’ and hence not visible for the public, and furthermore, cannot be linked. However, the results are quite mysterious even for the comprehensive 3DMark database.
A released or launched graphics card would have been identified correctly and 3DMark would have tagged the result correctly with the name of the graphics card model. However, this is clearly not the case. The results identify the GPU as ‘Generic VGA’, which is common for as-yet-unreleased SKU. Nonetheless, the results do mention the manufacturer, which is NVIDIA Corporation.
Una GPU Nvidia Ampere aparece por 3DMark siendo un 31% más rápida que la GeForce RTX 2080 Ti FE https://t.co/t5MzqnPr0p
— Laro (@Laro_99) June 21, 2020
The leaked result indicates a graphics card has a 1935 MHz boost clock and 6000 MHz memory clock. 3DMark software lists the memory as 6 GHz, which is rather strange as 3DMark mentions 1750 MHz for NVIDIA GeForce RTX 2080 Ti. Simple multiplication by 8 gives the actual speed as 14 Gbps for the 2080 Ti GPU.
The 6 GHz is most likely an error and indirectly indicates that even the 3DMark benchmarking database is not yet aware of the memory type. This means NVIDIA is truly using a new type of memory for its upcoming GeForce RTX 30 Series Graphics Cards, which is GDDR6X. Still, it is not clear if the memory frequency is 12 Gbps or 24 Gbps.
Mystery NVIDIA Graphics Card In-Line With Performance Improvements Over Generations:
Overall, the mystery NVIDIA Graphics Card certainly offers a higher than 30 percent performance compared to the premium or top-end stock NVIDIA GeForce RTX 2080 Ti. Needless to add, the 30 percent upgrade path for the next series is an average performance boost per generation.
The unknown Ampere variant is:
- 30.98% better than a stock RTX 2080 Ti Founders Edition
- 21.07% better than a stock MSI RTX 2080 Ti Lightning Z
- 22.14% better than a stock Nvidia Titan RTX
- 8.30% better than the best Nvidia Titan V result under LN2
- 2.18% less than KINGPIN’s overclocked EVGA RTX 2080 Ti XC
Утечка результатов NVIDIA GeForce RTX 3080 Ti в 3DMark — 2080 Ti глотает пыль https://t.co/SZYwEiO3mh
— 3DNews.ru (@3D_News) June 21, 2020
Incidentally, these results could very well be for an engineering sample or early production prototype. In other words, there’s a very high chance the mystery GPU is nowhere near final production-ready.
The results are rather strange and interesting because it appears NVIDIA appears to be trying to accommodate efficiency as well as performance parameters instead of being obsessed with the highest possible performance as it is anyways going to infuse a premium $150 cooling solution.