Google Tensor vs A12 Bionic
We compared the 8-core Google Tensor (Mali-G78 MP20) with the older 6-core Apple A12 Bionic (Apple A12 Bionic GPU) SoC. Here you will find the pros and cons of each chip, technical specs, and comprehensive tests in benchmarks, like AnTuTu and Geekbench.
Review
CPU Performance
Single and multi-core processor tests
Gaming Performance
GPU performance in games and OpenCL/Vulcan
Battery life
Efficiency of battery consumption
NanoReview Score
Overall chip score
Key Differences
Pros of Google Tensor
- Performs 3.9x better in floating-point computations
- Announced 3-years and 2-months later
- Supports 50% higher memory bandwidth (51.2 against 34.1 GB/s)
- Has a smaller size transistor (5 versus 7 nm)
- Shows significantly better (up to 41%) AnTuTu 9 score – 745K vs 530K
- Has 2 more cores
- 12% higher CPU clock speed (2800 vs 2490 MHz)
Pros of Apple A12 Bionic
- Higher GPU frequency (~33%)
- Better instruction set architecture
We may receive a commission from Amazon for purchases made using the links below.
Benchmarks
Performance tests in popular benchmarksAnTuTu 9
CPU | 196221 | 143402 |
GPU | 297786 | 186891 |
Memory | 107096 | 91638 |
UX | 143527 | 107048 |
Total score | 745917 | 530138 |
AnTuTu results from iOS and Android are not directly comparable! They won't affect NanoReview scores.
GeekBench 5
Single-Core Score
Multi-Core Score
Image compression | - | 132.8 Mpixels/s |
Face detection | - | 14.5 images/s |
Speech recognition | - | 67.8 words/s |
Machine learning | - | 54.1 images/s |
Camera shooting | - | 14.6 images/s |
HTML 5 | - | 3.27 Mnodes/s |
SQLite | - | 745 Krows/s |
3DMark
3DMark Wild Life Performance
Stability | 55% | 69% |
Graphics test | 37 FPS | 31 FPS |
Score | 6222 | 5237 |
Specifications
Full list of technical specifications of Google Tensor and A12 BionicCPU
Architecture | 2x 2.8 GHz – Cortex-X1 2x 2.25 GHz – Cortex A76 4x 1.8 GHz – Cortex A55 |
2x 2.49 GHz – Vortex 4x 1.6 GHz – Tempest |
Cores | 8 | 6 |
Frequency | 2800 MHz | 2490 MHz |
Instruction set | ARMv8.2-A | ARMv8.3-A |
L1 cache | - | 128 KB |
L2 cache | 3 MB | 8 MB |
L3 cache | 4 MB | - |
Process | 5 nanometers | 7 nanometers |
Transistor count | - | 6.9 billion |
TDP | 10 W | 6 W |
Graphics
GPU name | Mali-G78 MP20 | Apple A12 Bionic GPU |
Architecture | Valhall | - |
GPU frequency | 848 MHz | 1125 MHz |
Execution units | 20 | 32 |
Shading units | 320 | 256 |
FLOPS | 2171 Gigaflops | 560 Gigaflops |
Vulkan version | 1.1 | 1.1 |
OpenCL version | 2.0 | 2.0 |
DirectX version | 12 | 12 |
Memory
Memory type | LPDDR5 | LPDDR4X |
Memory frequency | 3200 MHz | 2133 MHz |
Bus | 4x 16 Bit | 1x 64 Bit |
Max bandwidth | 51.2 Gbit/s | 34.1 Gbit/s |
Max size | 12 GB | 12 GB |
Multimedia (ISP)
Neural processor (NPU) | Tensor Processing Unit | Yes |
Storage type | UFS 3.1 | NVMe |
Max display resolution | 3840 x 2160 | 2688 x 1242 |
Max camera resolution | - | 2x 12MP |
Video capture | 4K at 60FPS | 4K at 60FPS |
Video playback | 4K at 60FPS | 4K at 60FPS |
Video codecs | H.264, H.265, AV1, VP9 | H.264, H.265, Motion JPEG |
Audio codecs | AAC, AIFF, CAF, MP3, MP4, WAV | AAC, AAC‑LC, Apple Lossless, HE‑AAC, Linear PCM, AAX и AAX+, HE‑AAC v2, MP3 |
Connectivity
4G support | LTE Cat. 24 | LTE Cat. 18 |
5G support | Yes | No |
Download speed | Up to 1200 Mbps | Up to 1000 Mbps |
Upload speed | Up to 200 Mbps | Up to 150 Mbps |
Wi-Fi | 6 | 5 |
Bluetooth | 5.2 | 5.0 |
Navigation | GPS, GLONASS, Beidou, Galileo, QZSS | GPS, GLONASS, Beidou, Galileo |
Info
Announced | October 2021 | September 2018 |
Class | Flagship | Flagship |
Model number | S5E9845 | APL1W81 |
Official page | Google Tensor official site | - |
Cast your vote
Total votes: 139