From the get-go, the Google Pixel series was never about beating the top-performing competitors in performance benchmarks. The case for the Google Pixel 9 lineup isn't much different, as the early benchmark results show that the Tensor G4 is quite behind the Snapdragon 8 Gen 3 and Dimensity 9300.

But Google says that the Tensor G4 of the Pixel 9 series wasn't "designed to beat some specific benchmark" to begin with. Soniya Jobanputra, current product manager at Google, says that the new SoC aims to offer real-world improvements over the predecessor.

More specifically, the Google executive says that the Tensor G4 is designed to "meet our use cases."Soniya Jobanputragave a couple of examples where the new SoC of the Pixel 9 series brings improvements. Among them are optimization for multitasking and efficiency in opening apps. As for the multitasking part, it seems that Google delivered because a video from In Depth Tech Reviews on YouTube (attached at the end) shows that the Pixel 9 Pro can easily handle heavy workloads.

Another core focus for the Tensor G4 was AI. Soniya Jobanputrasays that the team worked with Google's AI division DeepMind to optimize the SoC for Gemini Nano and the "long range" future of models that may eventually run on the Pixel 9 devices.

On that note, during the launch event, the company noted that the new chip enables the multimodal capabilities of Gemini Nano. It was also said to deliver three times better performance with on-device AI tasks than the Tensor G3. There's support for processing up to 45 tokens per second as well.

With all these, it's pretty clear that Google's goal with Tensor G4 isn't any different from the other Tensor SoCs that launched before. Instead of being designed for "speeds and feats," the Pixel 9 series SoC is more about being optimized for what the company has in mind for its latest smartphones.

Pre-order the Google Pixel 9 Pro and get a free $200 Amazon gift card