The successor to Nvidia’s Turing generation of RTX graphics cards is known as Ampere. It will be based on a new process node built by Samsung and is expected to offer performance far in excess of that offered by the current 20-series of Nvidia GPUs.
After somewhat lackluster sales of RTX cards, Nvidia probably wants to change that with its next-generation graphics, along with the current perception of ray tracing. That could mean something quite exciting for the next generation.
Pricing and availability
Nvidia hasn’t made any official announcements about Ampere’s debut just yet, but we have heard from several sources, namely DigiTimes and IgorsLab (via WCCFTech) that Ampere will debut in the first half of 2020. That typically means towards the end of the first half, or at the very soonest, sometime in Q2. We’re expecting a June release, but that may be revised closer to the time.
Pricing remains entirely unknown, but a continuation of the new pricing standards set by Nvidia’s RTX-series seems likely. That puts the most expensive, Ti graphics card (a suggested name could be the 3080 Ti) as somewhere between $ 1,000 and $ 1,500, with more modest cards priced around $ 800, $ 600, and $ 400 respectively.
It is possible, however, that AMD’s competitively priced mid range (and perhaps by mid-2020, high-end) Navi graphics cards could force Nvidia to be more aggressive with its pricing. The new architecture could also allow for price reductions.
Nvidia’s Ampere will be based on a brand new process node, dropping from a 12nm TSMC design for the RTX series, to a 7nm process node from Samsung. It will also be built using Samsung’s extreme ultraviolet lithography (EUV) process, which can help reduce production costs, as well as improve performance and efficiency.
Nvidia nor Samsung have detailed the advancements that can and have been made with the drop to 7nm EUV, but they should deliver higher clock speeds, and greater power efficiency, which Nvidia could then leverage for reduced power draw and heat output, or further performance improvement.
Some reports have suggested that performance could be substantially improved with Ampere over Nvidia’s own RTX cards and AMD’s new and upcoming Navi alternatives. That seems likely, as a drop from 12nm to 7nm, not to mention using the new EUV process, should give Nvidia plenty of room to boost the capabilities of the cards.
That said, it’s worth noting that the very first Samsung 7nm EUV chip, Samsung’s own Exynos 925 SoC, underperformed, delivering improved efficiency over its predecessor, but not much else, according to Toms Hardware. While that chip was only a move from 8nm to 7nm, the 8nm was more like a 10nm process in all but name, so there is some comparison to be made here, even if it’s a somewhat apples for oranges one.
Ultimately, it seems likely that Ampere will deliver at least a modest performance improvement over Turing. Especially with AMD’s rumored big Navi 20 graphics card expected sometime in 2020, with the potential to challenge Nvidia’s most powerful cards of today, like the 2080 Ti.
What about ray tracing?
Although as with performance, we don’t have hard details on what Ampere will do with ray tracing. it seems almost certain that Nvidia will continue to push the new lighting technology’s support in its new-gen cards. It effectively made ray tracing part of the modern gaming conversation with Turing, even if few games support it, more than a year after the release of those cards.
Whether we’ll see more, or faster RT cores alongside general GPU enhancements with Ampere, remains to be seen. But Nvidia will want to remove the stigma of ray tracing killing top card performance outside of 1080p, so we wouldn’t be surprised if Nvidia markets its top cards as being 4K capable with ray tracing, even if frame rates are somewhat limited still.