How Does Advanced Optimus Work?
When it comes to Advanced Optimus, there are a lot of moving parts, Sharma explains, as it needs various panel makers, chip manufacturers, OEM developers, and laptop designers to all work together. “There’s a lot of players that have to dance to the music together to make this work,” he says.
To start with the basics, your gaming laptop comes with two GPUs: one basic unit, built into the processor for low-powered tasks like editing spreadsheets (the Integrated Graphics Processor, iGPU), and a high-powered one for heavy workloads, like pushing high frame rates in Fortnite, the discrete GPU, or dGPU for short. The dGPU is overkill for simple tasks like email and spreadsheets, and running the dGPU for light workloads would drain your battery for no reason since nobody cares about frame rate when you’re just checking emails, after all.
Now, in the bad old days of gaming laptops, you would have to physically switch between these two GPUs. Which is to say you’d have to reboot if you wanted to take a break from work to play a couple of rounds of Valorant, and you may even have had to fiddle with BIOS settings to make said switch.
This all changed with NVIDIA Optimus, the solution now known as MS Hybrid. NVIDIA found a way to detect if the computer has a graphically intense workload or not, and to automatically switch which GPU was active based on that workload. It was a clever optimization that guaranteed the user better battery life or extra performance.
But there was a catch with Optimus: the iGPU was typically hardwired to the laptop panel. This meant when rendering frames on the dGPU an additional step was necessary to get the image to the screen. This additional step comes with some time and computing costs.
While Optimus was still an improvement over manually rebooting to switch GPUs, the additional step increased latency and decreased framerate. In competitive games like Fortnite and Valorant, latency and framerate can be the difference between winning and losing a match. ,