Nvidia has announced its new RTX 2080 Super and RTX 2070 Super mobile GPUs. These are the company’s top-of-the-line (and priciest) RTX cards for laptops, offering a performance bump over the company’s standard RTX 2070 and 2080 offerings.
First up, as rumored, NVIDIA revealed portable versions of the RTX 2080 Super and RTX 2070 Super Max-Q GPUs for flagship-caliber gaming and creator laptops. These will be NVIDIA’s best laptop GPUs, though they have very similar specs to the older models.The RTX 2080 Max-Q will be discontinued, but NVIDIA will continue to sell the RTX 2070 Max-Q.
The top-end RTX 2080 Super model packs 3,072 CUDA cores compared to 2,944 from before, but the boost clock (1080-1560 MHz), memory configuration (8GB of DDR6) and bandwidth (448 GB/s) are roughly the same. Graphics power, meanwhile, now ranges from 80 to 150+ watts.
The RTX 2070 Super gets an equally modest spec bump, with 2,560 CUDA cores compared to 2,304 on the RTX 2070 Max-Q. The base boost clock now goes from 1,155 to 1,380 MHz, compared to 1,125 to 1,455 MHz on the non-Super model. All other specs are the same.
Both these GPUs will appear in high-end laptops like Gigabyte’s Aero 15 and the Razer Blade 15, but the GPU hardware isn’t the end of the story. NVIDIA has also introduced new Max-Q hardware features for the 2020 laptops that will boost performance significantly.
The first is called Dynamic Boost. Before, the static thermal systems in NVIDIA GPUs would always use 80 watts while allocating 35 to the CPU. Dynamic boost can take that same 115 watts and distribute up to 15 percent more to the CPU or GPU on a frame-by-frame basis.
Nvidia will also offer Max-Q configurations (intended for thin gaming laptops), which it says many upcoming notebooks will feature. The company says its Max-Q systems will double the power efficiency of previous designs thanks to a new feature called Dynamic Boost that automatically distributes power between their GPU and CPU, new low-voltage GDDR6 memory, upgraded voltage regulators, and Nvidia’s new Deep Learning Super Sampling (DLSS 2.0), which uses artificial intelligence to help the GPU render pixels.
Another significant upgrade is NVIDIA’s Advanced Optimus technology, essentially a controller that chooses either the GPU or integrated graphics for a given app. Before, the GPU wasn’t directly connected to the display and had to drop frames onto the integrated graphics framebuffer. Now, it can drive the built-in display directly, improving latency and allowing G-Sync to work at up to 4K 120 Hz. The GPU is also now directly connected to the DisplayPort on your laptop so you’ll see improved performance when using VR or an external monitor.
NVIDIA noted that manufacturers like Razer, Acer and Gigabyte are launching 10 new RTX Studio laptops, including 17-inch OLED and HDR models. There are now 45 apps that support RTX acceleration (to widely varying degrees) including Blender, DaVinci Resolve, Adobe Lightroom and others, NVIDIA said. We should be getting our hands on some of the new models soon to see if they measure up to NVIDIA’s promises.