Nvidia promised something Super is coming well over a month ago, and on Tuesday, the wait paid off. The graphics company unveiled not one, not two, but three “Super” variants of its existing GeForce RTX 20-series graphics cards, along with a new FrameView tool intended to give enthusiasts and reviewers deeper insight into the performance and power consumption of GPUs. And get this: The new cards don’t really cost more than their slower, older counterparts.
The timing probably isn’t coincidental, as AMD’s rival Radeon RX 5700 and 5700 XT—the first consumer GPUs built using an advanced 7nm manufacturing technology—launch in mere days, on July 7.
The GeForce RTX 2060 Super starts at $399; the GeForce RTX 2070 Super at $499; and the GeForce RTX 2080 Super at $699. We’ve already reviewed the GeForce RTX 2060 Super and RTX 2070 Super, and well, they live up to their moniker, essentially bumping each card up a performance tier. The GeForce RTX 2060 Super falls within spitting distance of the original $499 RTX 2070, while the new RTX 2070 Super greatly closes the gap with the original $699 RTX 2080. The RTX 2060 Super also shores up weak points in its $349 non-Super cousin, bumping the memory capacity from 6GB to 8GB and the memory interface from 192-bit to 256-bit, justifying its $50 price premium.
We haven’t tested the RTX 2080 Super yet, as it’s launching later in July. Nvidia says that it “offers more performance than the RTX 2080 at the same price. Memory speed has been cranked up to 15.5Gbps. It’s faster than Titan Xp.”
Nvidia isn’t overhauling the entire RTX 20-series lineup. Tony Tamasi, VP of content and technology at Nvidia, told reporters that while the company expects market forces to replace the older RTX 2070 and 2080, it plans on keeping the 6GB RTX 2060 non-Super around at its original price point (presumably because the RTX 2060 Super costs more). The ferocious GeForce RTX 2080 Ti is also sticking around in its original ass-kicking form.
Here’s a look at what Nvidia’s RTX 20-series lineup will look like by the end of July:
- GeForce RTX 2080 Ti—$999 (but really $1,100+)
- GeForce RTX 2080 Super—$699
- GeForce RTX 2070 Super—$499
- GeForce RTX 2060 Super—$399
- GeForce RTX 2060 Super—$349
The Super cards don’t affect Nvidia’s more affordable GTX 16-series graphics card lineup, which lacks the dedicated ray tracing and AI hardware found in RTX GPUs.
To drive home the real-time ray tracing lead blazed by its RTX graphics cards, Nvidia equipped the Super cards with more RT and tensor cores than their predecessors. Nvidia will also bundle two games that feature the futuristic technology for free with all Super purchases: Remedy’s Control and Bethesda’s Wolfenstein: Youngblood. Even better, those games genuinely have the potential to be great and are some of our most anticipated upcoming releases.
The GeForce RTX 2060 Super and 2070 Super are scheduled to launch a week from now, on July 9, with both custom and Nvidia Founders Edition models available. Nvidia’s own boards will cost the same as those baseline prices. The GeForce RTX 2080 Super will launch July 23.
These Super variants should crush the perception that Nvidia’s ray tracing cards sacrifice performance to enable a ray-traced future. The Super cards pack much more firepower for both ray tracing and traditional games now, with little or no price increase compared to their non-Super predecessors.
Nvidia’s Super cards could ruin AMD’s long-anticipated debut of “Navi” 7nm consumer GPUs, built using a new graphics architecture. There’s a reason the RTX 2060 Super and RTX 2070 Super are being announced today, complete with reviews that herald their significantly enhanced performance. While we’ve yet to review the $380 Radeon RX 5700 or $450 Radeon RX 5700 XT, AMD’s public teasers for the duo at E3 showed them performing just slightly ahead of the RTX 2060 and 2070 Founders Edition cards at 1440p resolution, respectively. You can see AMD’s performance graphs above.
Compared against the original GeForce RTX 2060 and 2070, the new Radeons would be very competitive.
Well, Nvidia says that the RTX 2060 Super is up to 22 percent faster (avg 15 percent) than RTX 2060, and the RTX 2070 Super is up to 24 percent faster (avg 16 percent) than the GeForce RTX 2070 for the same price. Our testing confirmed a substantial leap in performance—and only Nvidia’s RTX 20-series cards contain dedicated hardware for real-time ray tracing. Unless AMD’s been sandbagging performance or has a trick up its sleeve, Nvidia just spoiled what should’ve been a triumphant Radeon release. Stay tuned for July 7.
Alongside the Super cards, Nvidia is also announcing a new performance monitoring tool for enthusiasts dubbed FrameView. Nvidia made the tool available to reviewers, but we haven’t been able to validate it yet due to the RTX 2060 and 2070 Super review deadlines. It sounds similar to AMD’s OCAT tool, but with additional power measurement features.
Here’s how Nvidia described FrameView in a guide provided to reviewers:
“FrameView is a software tool designed to capture and measure performance and power utilization of PC-based graphics hardware. It’s especially useful for measuring frame rates and GPU power usage when running stressful “real world” PC gaming scenarios. FrameView captures performance and power data with minimal overhead so as not to impact frame rates or gameplay. FrameView includes an overlay that shows performance and power metrics as a game is being played. It also allows benchmark runs to be captured and charted in detailed reports.
“FrameView captures game performance metrics including average and percentile frame-per-second (FPS) data for both single- and multi-GPU configurations. Percentile FPS data is valuable for illustrating the severity and frequency of stutters that can interrupt gameplay. FrameView has been optimized particularly for detailed frame time, present, and display scheduling metrics for measuring stutter.
FrameView captures real-time power measurements for both total board power (including graphics memory) and GPU chip-only power through application programming interfaces (APIs), which is publicly available software that communicates with the hardware and returns data. This removes the need for additional, and sometimes costly frame capture and power measurement hardware.”
FrameView works only on Windows 10 currently, but it can monitor games built on OpenGL, Vulkan, or DirectX 9 through 12. The overlay feature doesn’t work on DX9/10 or Vulkan games, however. And Nvidia warns that comparing power results from its GeForce cards and AMD’s Radeon cards is not apples-to-apples.
“While FrameView reports both chip and board power for Nvidia graphics cards, it currently only reports what appears to be something in between chip power and TGP for AMD graphics cards because this is all AMD reports in their API,” the user guide states. Because of that, the company says FrameView shouldn’t be used to compare the power draw between the rival cards. (At PCWorld, we test whole-system power draw at the wall for an absolute power view.)
Nvidia plans to make a beta version of FrameView public on the GeForce website on July 9, in a similar fashion to its FCAT tool. I’m looking forward to giving it a proper poke and prod.