Another rumoured Pixel 6 Pro benchmark score has surfaced on Geekbench with improved performance.
Google’s semi-announced Pixel 6 series previously showed up in Geekbench listings in September with a disappointing single-core score of 414 and multi-core score of 2,074. However, this new listing spotted by PhoneArena shows a score of 1,034 and 2,756 in single- and multi-core respectively. The listing also confirms there will be a 6 Pro configuration with 12GB of RAM. It backs up previous information that the Tensor chip will use two ARM Cortex-X1 cores at 2.80GHz, two intermediate cores at 2.25GHz and four power-efficient cores at 1.80GHz.
That’s a significant jump over the previous Pixel 6 Geekbench test, but still not as good as other Snapdragon 888 flagships. The Samsung Galaxy S21, for example, scored 1,112 and 3,378 in our tests. The Z Fold 3 scored 1,109 and 3,185. The OnePlus 9 Pro also sports a Snapdragon 888 and scored 1,071 and 3,282.
It’s also worth comparing the alleged Pixel 6 Pro scores to previous Pixel devices. If accurate, the Pixel 6 Pro outperforms the Pixel 5’s Snapdragon 765G (589 and 1,603) and the Pixel 4’s 855 (695 and 2,339). In other words, it looks like the Pixel 6 series may offer improved performance over older Pixels, so if you’re still holding onto a Pixel 4 or older, you’ll get a speed bump this time around.
Still, this all indicates Google’s customer Tensor chip may actually hold its own against Qualcomm’s venerable Snapdragon line. We also haven’t seen much in the way of machine learning or artificial intelligence performance, which is supposedly where Tensor will shine.
That being said, until the Pixel 6 and 6 Pro actually launch, it’s not worth putting much stock in benchmark numbers. For one, things can clearly change — the previously leaked benchmark scores are much lower than these new ones. The performance may even increase before these phones launch. But we won’t know until reviewers get hands-on with devices and put them through their paces.
And the second issue with benchmark scores is they often don’t reflect real-world use. Day-to-day performance using apps, making calls, sending texts can have a different impact. Plus, I regularly see ‘budget’ devices that score poorly on benchmarks operate without issue in real-world testing (for example, the Pixel 4a). So, don’t read too much into the benchmark scores.