Board partner rumors indicate NVIDIA Ampere may indeed be a lot faster

NVIDIA is expected to show its Ampere-based gaming GPUs later this year, perhaps with an announcement as early as at the GTC 2020 in March. While the Ampere codename has been circulating around the web for a long time now, we still know little about this architecture. It's believed to be made on a 7nm process and rumors indicate we're in for a massive performance boost.

Igor Wallossek from Igor's Lab now reports that there are indeed signs that Ampere will deliver a significant increase in performance. Talks with board partners and suppliers reveal NVIDIA's board partners are currently test driving new production methods with the RTX 2080 Ti, to prepare for the changed circuit board technology that will be required for the Ampere GPUs.

Ampere will require something called backdrill, a process that allows signals to flow from one board layer to the other by creating vias by removing the stub in multi-layered PCBs. This technique is more expensive because it's more complex than what's currently used for video cards. Igor's Lab also notes that at this moment, not all of NVIDIA's board partners can even implement this in mass production.

This automatically results in a significantly lower bit error rate, avoids jitter and can reduce the signal attenuation. Conversely, this naturally leads to a higher channel bandwidth and higher data rates. Which brings us back to the starting point of the consideration of performance increase. According to Samsung, there is currently no significantly faster GDDR6 available than the 16Gbps, so that Nvidia’s change in manufacturing technology for the new boards must be viewed as a blanket measure for all areas where signal integrity and bandwidth are important.

We'll have to wait to find out for sure, but it could indeed be that Ampere will deliver a a bigger than expected performance increase.