Unannounced Performance Jump Leaves Pixel Users Stunned
It appears as if Google might have been paying attention. A recent update shows they've been holding back some serious punch. Pixel users are seeing a startling jump in graphics performance, seemingly out of the blue.
Pixel Users Report Unexpected Graphics Performance Surge
Google hasn't issued any official word on performance boosts in their recent Pixel Feature Drop. But, on Reddit and tech forums, users of Pixels are posting jaw-dplotlib Geekbench 6 scores. These scores tell the picture clearly. The graphics performance of Pixel phones has virtually doubled in a few instances. Imagine – Vulkan test scores increasing by nearly 100%, and OpenCL performance by more than 75%.
And the cherry on top of all this? Older Pixel phones are getting a gigantic performance boost. The Pixel 6a using the original Tensor G1 and the Pixel 7a on Tensor G2 are now not just holding their own. In some graphing tests, they are actually surpassing the new Pixel 9 Pro with its Tensor G4 chip. This performance bump isn't unique to one device. Everyone is noticing improvement on nearly the entire range of Tensor-powered Pixel smartphones.
Driver Magic: Unlocking the Performance Mystery
It seems this performance boost isn't device-specific. It's happening on Android 15 and Android 16 both. This means the operating system itself isn't the driving force behind this. Android gurus at Android Authority believe they've cracked it. The magic lies with updated drivers for the ARM Mali GPUs used within Tensor chips.
It seems Google snuck in very fresher drivers in the recent Pixel Drop update under the radar. For Tensor G1, G2, and G3, the fresh drivers are dated February 2024. Curiously, the Tensor G4 in the Pixel 9 Pro still has older drivers from December 2023. This driver update may be the magic bullet that unlocks hidden potential within the Tensor architecture.
Unexpected Boost Raises Questions
No matter why, Pixel owners are getting a welcome and unexpected performance boost. It raises questions regarding how Google works and what else they have hidden up their sleeve when it comes to the future of Tensor.
