If you’ve been wondering when gaming laptops will take a more serious hit on 1440p panels, this week’s CES revelations from Nvidia are aimed directly at you. See: a generational leap in the company’s portable GPUs, this time with Ampere architecture and RTX 3000 series branding.
Three GPU models have been announced at all, and they are named after the GeForce RTX 3080, 3070 and 3060. They are scheduled to launch in “70+” laptops from January 26th. Nvidia has listed “top OEMs” such as Acer, Alienware, ASUS, Gigabyte, HP, Lenovo, MSI and Razer with upcoming RTX 3000 series notebook laptops, along with “local OEMs and system builders.”
Name convention double check
Nvidia’s sales position positions the RTX 3060 laptop variant as “faster than laptops with the RTX 2080 Super”, although this model may land more specifically in 1080p systems. The two advanced models are often referred to as part of 1440p systems, a resolution that has long been left in the gaming laptop litter (and will no doubt benefit greatly from Nvidia’s proprietary DLSS upscaling solution). While Nvidia’s latest marketing materials mention an upgrade for the money, compared to the latest generation of portable GPUs, we’m still waiting for OEMs to roll out specific pricing and specifications for the late January models. (We also wonder if these laptops will sell out too fast for average people to get them.)
Nvidia continues its naming convention to lend desktop model numbers to its smaller portable variants, and in some ways this is checking out. The laptop side jumps to the RTX 3000 series is met with the inclusion of second generation beam cores and third generation tensor cores – something that has been shown on RTX 3000 desktop models to offer performance upgrades outside of pure rasterization, while reducing space on the physical hardware.
But if you put the specifications of desktops and laptops side by side, the downgrades for each vary a bit more wildly than you might expect from the names alone. In particular, the RTX 3080 sees a nearly 30 percent decline in CUDA cores, compared to 13 percent for the RTX 3070. (Oddly enough, the RTX 3060 laptop variant will have 6.6 percent more CUDA cores than its desktop variant, but 21.1 percent less than the desktop 3060 Ti.)
Honestly, the portable version of the RTX 3080 probably deserves a different name, as it is also downgraded to a 256-bit memory bus and GDDR6 VRAM (instead of the glowing GDDR6X variant found in the desktop model, attached to a 320-bit bus). Not that these are not necessary downgrades to operate within the thermal and power constraints of a portable chassis; we just say. Speaking of which, the clock speeds of RTX 3000 laptops will also vary across the board, and Nvidia’s announced series will certainly vary further when in the wild and wrapped in a range of OEM chassis.
And in a surprising turn, the RTX 3080, despite having a VRAM downgrade, gets an optional capacity Upgrade to the tune 16 GB GDDR6. It’s not entirely rumored RTX 3080 Ti (which reportedly can come with as much as 20 GB of GDDR6X VRAM), but if you’re looking for serious VRAM in a studio laptop in 2021, you might want to crack a few more piggy banks. ahead of January 26th.
Define “whisper”, Nvidia
The rest of this week’s announcement mentions a few efficiency-oriented gains that may prove difficult to test, as their boosts are about claims such as “balancing power … on a frame-by-frame basis” and “access to all GPU VRAM simultaneously.” (The latter, called Resizable BAR, will not be ready at launch and will require driver update.) This is also the case with “WhisperMode 2.0”, which apparently will allow users to select a noise level as desired and see system performance adjusted accordingly. Exactly how performance will shift from “hear myself think” to “stand in an airport hangar” remains to be seen.
We look forward to going hand in hand with these GPUs to see exactly how these promises play out, and from the looks of things, Nvidia will once again beat AMD to the market in a given generation, with Team Red only offering one vague “first half of 2020” promise of its series of RDNA2 portable GPUs.
Listing image of Nvidia