Tensor G3 Benchmarks: Better Than Snapdragon 8 Gen 2?
The Google Pixel 8 series has the new and upgraded Google Tensor G3 chipset. How does it fare in benchmarks such as AnTuTu, Geekbench, and 3DMark? Let’s discuss.
KEY TAKEAWAYS
- Google Tensor G3 scores 1,081,944 in the AnTuTu benchmark.
- Tensor G3 scores 1760 as the single-core score and 4442 as the multi-core score.
- It’s a good leap against the previous year’s Tensor G2, but still no match for other flagship chipsets such as Snapdragon 8 Gen 2 and Apple A17 Pro.
Google Tensor G3 Specs & Features
Let’s first look at the complete specifications of the Google Tensor G3 chipset:
- CPU: 1×3.0 GHz Cortex-X3 & 4×2.45 GHz Cortex-A715 & 4×2.15 GHz Cortex-A510
- GPU: Immortalis-G715s MC10
Google hasn’t revealed the complete specifications of the Tensor G3 chipset at the launch event of the Pixel 8 series. However, we know these specifications from a previous leak on X.
Google doesn’t put the spotlight on raw specifications and performance gains. Instead, the focus has been on AI.
Our work with Tensor has never been about speeds and feeds, or traditional performance metrics. It’s about pushing the mobile computing experience forward. And in our new Tensor G3 chip, every major subsystem has been upgraded, paving the way for on-device generative AI. It includes the latest generation of ARM CPUs, an upgraded GPU, new ISP and Imaging DSP and our next-gen TPU, which was custom-designed to run Google’s AI models.
Google
Google says that compared to Tensor G2, the latest Tensor G3 is capable of running twice as many machine learning models on-device. The Tensor G3 brings more improvements in speech and language, with Google implementing the same text-to-speech model Google uses in its data centers. It also continues to deliver the improvements Google makes with computational photography and tools such as Magic Eraser and Best Take.
But this means Tensor G3 won’t dominate the benchmark charts compared to other flagship chipsets.
Tensor G3 AnTuTu Benchmark Scores
AnTuTu benchmark is the most popular benchmark application for smartphones. AnTuTu benchmark score consists of a total CPU, GPU, Memory, and UX score.
Here are the AnTuTu benchmark scores of the Google Tensor G3 chipset:
CPU Score | 295546 |
GPU Score | 383022 |
Memory Score | 184747 |
UX Score | 225100 |
Total Score | 1081944 |
Tensor G3 Geekbench Scores
The Geekbench benchmark is used to benchmark the CPU performance of a chipset. It is available not only for smartphones but also for PCs and laptops. Geekbench benchmark score consists of two scores – single-core and multi-core scores, both of which benchmark a chipset’s single-core and multi-core performance.
Here are the Geekbench benchmark scores of the Google Tensor G3 chipset:
Single Core Score | 1760 |
Multi-Core Score | 4442 |
Tensor G3 Benchmark Comparison
Let’s compare Tensor G3 with other competing chipsets!
Benchmark | Google Tensor G3 | Google Tensor G2 | Snapdragon 8 Gen 2 | Apple A17 Pro |
AnTuTu Benchmark | 1,081, 944 | 740,256 | 1,226,598 | 1641883 |
Geekbench Single Core | 1760 | 1051 | 2000 | 2846 |
Geekbench Multi Core | 4442 | 3241 | 5556 | 7024 |
As you can see, Google Tensor G3 lags behind Snapdragon 8 Gen 2 and Apple A17 Pro chipset in both AnTuTu and Geekbench benchmarks.
Are Benchmarks Important?
While benchmarks give us a measuring scale for comparing various chipsets, it is not a metric that can be translated into real-world performance. Google Tensor G3 isn’t leading the benchmark charts, but it has many new stuff in AI and machine learning that cannot be measured using benchmarks. Also, benchmarks just give us a glimpse into the performance; real-world usage depends on factors like software optimization, apps you use, usage habits, and more.
Phones With Tensor G3 Chipset
Google Tensor G3 is currently used only on two Google phones: Pixel 8 and Pixel 8 Pro. Google will introduce new devices such as Pixel 8A and Pixel Tablet 2 in the future, both of which will be powered by Tensor G3 chipset.