Here’s my benchmark results for each SoC and device.
- 1 List of Benchmark Results
- 2 About Benchmark Boost
- 3 CPU Performance
- 4 Graphics Performance
- 5 Performance for everyday use
- 6 Power efficiency
- 7 Storage and memory performance
- 8 Tap and screen latency
- 9 Audio latency
- 10 Conclusion
List of Benchmark Results
AnTuTu benchmark is a popular Android benchmark app, but often falls victim to benchmark boosts, plus it is now excluded from the Play Store.
To minimize the impact of benchmark boosting, I will add my own measurements on multiple applications in turn.
In addition, the individual articles also include Genshin game performance data measured by WeTest PerfDog (some are in Japanese).
Benchmarked devices (model name numbers match the * numbers in the graph below)
|1. OnePlus 8 Pro (WQHD+)||Android 10||Snapdragon 865||UFS 3.0|
|2. OnePlus 7||Android 10||Snapdragon 855||UFS 3.0|
|3. Xperia 1||Android 10||Snapdragon 855||UFS 2.1|
|4. OnePlus Nord||Android 10||Snapdragon 765G||UFS 2.1|
|5. Xperia XZ Premium||Android 9||Snapdragon 835||UFS 2.1|
|6. Rakuten Mini||Android 9||Snapdragon 439||eMMC 5.1|
|7. VANKYO MatrixPad S30||Android 9||Unisoc SC9863a||eMMC 5.1|
|8. Dragon Touch NotePad 102||Android 10||Unisoc SC9863a||eMMC 5.1|
|9. Galaxy Tab S7||Android 10||Snapdragon 865 Plus||UFS 3.0|
|10. Google Pixel 3a||Android 11||Snapdragon 670||eMMC 5.1|
|11. OnePlus 8T||Android 11||Snapdragon 865||UFS 3.1|
|12. Teclast M40||Android 10||Unisoc T618||eMMC 5.1|
|13. Galaxy Tab S7+||Android 10||Snapdragon 865 Plus||UFS 3.0|
|14. realme 7 5G||Android 10||MediaTek Dimensity 800U 5G||UFS 2.1|
|15. Xiaomi Mi 11 (WQHD+)||Android 11||Snapdragon 888||UFS 3.1|
|16. motorola edge s||Android 11||Snapdragon 870
* GPU 905MHz
|17. Xiaomi POCO F3||Android 11||Snapdragon 870||UFS 3.1|
|18. OnePlus 9 Pro
|Android 11||Snapdragon 888||UFS 3.1|
|19. OnePlus 9 Pro
|Android 11||Snapdragon 888||UFS 3.1|
|20. vivo iQOO Neo5||Android 11||Snapdragon 870||UFS 3.1|
|21. POCO M3 Pro 5G||Android 11||MediaTek Dimensity 700||UFS 2.2|
|22. Lenovo Yoga Pad Pro||Android 11||Snapdragon 870||UFS 3.0|
|23. Lenovo Yoga Tab 13||Android 11||Snapdragon 870||UFS 3.0|
|24. OPPO Find X3||Android 11||Snapdragon 870||UFS 3.1|
|25. OnePlus Nord CE 5G||Android 11||Snapdragon 750G||UFS 2.1|
|26. Xiaomi Mi 11 Lite 5G||Android 11||Snapdragon 780G||UFS 2.2|
|27. OnePlus Nord 2
* High Performance Mode ON
|Android 11||MediaTek Dimensity 1200-AI||UFS 3.1|
|28. OnePlus Nord 2
* High Performance Mode OFF
|Android 11||MediaTek Dimensity 1200-AI||UFS 3.1|
|29. Xiaomi MIX 4||Android 11||Snapdragon 888 Plus||UFS 3.1|
|30. Xiaomi Pad 5 Pro||Android 11||Snapdragon 870||UFS 3.1|
|31. Xiaomi Pad 5||Android 11||Snapdragon 860||UFS 3.1|
|32. TECLAST T40 Plus||Android 11||Unisoc T618||eMMC|
|33. Meizu 18||Android 11||Snapdragon 888||UFS 3.1|
|34. POCO M4 Pro 5G||Android 11||MediaTek Dimensity 810||UFS 2.2|
|35. TCL 10 Pro||Android 11||Snapdragon 675||UFS 2.1|
|36. Lenovo XiaoXin Pad Pro 2021||Android 11||Snapdragon 870||UFS 3.1|
|37. OnePlus 9RT 5G||Android 11||Snapdragon 888||UFS 3.1|
|38. Google Pixel 6 Pro||Android 12||Google Tensor||UFS 3.1|
|39. vivo X70 Pro+||Android 11||Snapdragon 888 Plus||UFS 3.1|
|40. motorola edge X30||Android 12||Snapdragon 8 Gen 1||UFS 3.1|
|41. Lenovo XiaoXin Pad Pro 12.6 (Lenovo Tab P12 Pro)||Android 11||Snapdragon 870||UFS 3.1|
|42. Xiaomi 12||Android 12||Snapdragon 8 Gen 1||UFS 3.1|
|43. realme GT 2 Pro||Android 12||Snapdragon 8 Gen 1||UFS 3.1|
|44. Redmi Note 11 (Global)||Android 11||Snapdragon 680||UFS 2.2|
|45. Xiaomi POCO X3 Pro||Android 11||Snapdragon 860||UFS 3.1|
|46. OPPO Find N||Android 11||Snapdragon 888||UFS 3.1|
|47. Redmi K50G||Android 12||Snapdragon 8 Gen 1||UFS 3.1|
|48. Galaxy Tab S8+||Android 12||Snapdragon 8 Gen 1||UFS 3.1|
|49. OPPO Pad||Android 11||Snapdragon 870||UFS 3.1|
|50. Lenovo Legion Tab Y700||Android 11||Snapdragon 870||UFS 3.1|
|51. OPPO Find X5 Pro||Android 12||Snapdragon 8 Gen 1||UFS 3.1|
|52. Redmi Note 11 Pro+ 5G||Android 11||MediaTek Dimensity 920||UFS 2.2|
|53. Redmi K50 Pro||Android 12||MediaTek Dimensity 9000||UFS 3.1|
|54. realme GT Neo3||Android 12||MediaTek Dimensity 8100||UFS 3.1|
|55. vivo X80||Android 12||MediaTek Dimensity 9000||UFS 3.1|
|56. Google Pixel 6||Android 12||Google Tensor||UFS 3.1|
|57. Black Shark 4 Pro (Japan)||Android 11||Snapdragon 888||UFS 3.1|
|58. vivo X Fold||Android 12||Snapdragon 8 Gen 1||UFS 3.1|
|59. Infinix ZERO X PRO||Android 11||MediaTek Helio G95||UFS 2.2|
|60. Blackview Tab 12||Android 11||Unisoc SC9863A||eMMC|
|61. OPPO Reno7 A||Android 11||Snapdragon 695||UFS 2.2|
* Basically, measurements are taken at the highest resolution and refresh rate that can be set on the device.
About Benchmark Boost
In particular, Chinese smartphones sometimes use so-called “benchmark boosting,” in which the CPU clock is fixed to maximum only during benchmarking to make the numbers look better.
Apart from benchmark boosting, there is also a behavior called “throttling” that limits CPU/GPU operation of apps by identifying them by package name, even though the benchmark results are normal.
I do not consider “throttling” to be a major problem if there is a workaround for it, since it is normal for heat generation to increase as performance improves, but I consider “benchmark boosting” to be a malicious behavior that makes benchmarking meaningless and prevents fair comparison with other devices.
Benchmark boosting on realme smartphones uncovered – “Geekbench mode” to fix CPU frequency to maximum – AndroPlus
Of the benchmark results below, Geekbench 5 makes it easy to confirm whether a benchmark boost is being performed, so if I know that a boost is being performed, I will post the non-boosted score.
If it is boosted in Geekbench 5, it is safe to assume that it is done in other benchmark applications (including AnTuTu) as well.
Here is a list of what is known at this time.
Devices with benchmark boosting👠:
- realme / OnePlus (MediaTek) … 54
- Black Shark … 57
- Infinix … 59
Devices with throttling:
- Samsung … 48
- Xiaomi … 53
- vivo … 55, 58
How to check benchmark boost throttling:
- Install Geekbench 5
- Download and install the version disguised as Genshin from this Geekbench developer’s tweet
- First, the scores are measured in the Genshin disguise version, and then also in the regular Geekbench.
Normally, repeating the benchmark for a short period of time will lower the score, but if the score is clearly greater for the regular version than for the Genshin disguised version, it is likely that benchmark boost throttling is taking place.
If you check the CPU clock using an application that can display the CPU clock in real time with an overlay, such as Cpu Float, and when you open the normal version of Geekbench, if it remains stuck to the maximum value or is easily maxed out, it means the device is boosting instead of throttling.
If the overlay display disappears, turn on “Allow screen overlay in settings” in the developer options.
CPU performance is measured by Geekbench 5.
GPU graphics performance is measured by Wild Life, a cross-platform test of 3DMark.
Performance for everyday use
Comfort in everyday operations is measured with PCMark Work 3.0; the Work 3.0 benchmark checks how well it handles common tasks such as web browsing, video editing, document and data manipulation, and photo editing.
It measures overall performance, including GPU and storage performance as well as CPU.
It needs to be at least 7000 or higher for comfortable everyday use.
The average frame rate when playing Genshin at the highest quality and 60FPS setting divided by the average consumption is used as an indicator of how efficient the device is.
A more efficient device means that it can maintain a smooth display with less battery consumption.
It shows not only the performance of the SoC itself, but also how well it is optimized in software.
Even a device that loses to a high-end SoC in terms of SoC performance alone may be able to deliver stable performance in actual game play.
- AI Frame Rate Stabilizer ON…51
- Ultra-steady frames ON…46, 49, 54
- Performance mode ON…53, 54
(Performance mode is required to unthrottle)
Storage and memory performance
Storage and memory performance is measured by CPDT Benchmark.
Tap and screen latency
I measured tap latency (delay in response to touch) and screen drawing latency with the WALT Latency Timer (developed by Google).
I used the result of about 30 taps and measured the screen drawing latency with maximum brightness at the center of the screen.
The total of the latency between touch to kernel, kernel to Java and screen drawing latency at ACTION_DOWN is the “latency taken until the screen display changes after touch,” and the less the better.
Apparently, devices with MediaTek SoC are prone to delays in screen rendering.
I measured latency during audio output with the WALT Latency Timer.
If the device has a 3.5mm earphone jack, I use the earphone jack; if not, I attach the Plugable USB Audio Adapter to a USB Type-C hub.
Even if the same standards are used, newer models tend to be more improved and optimized, resulting in better performance.
The middle-range models in terms of classification can have performance that exceeds that of high-end models from a few years ago, and it is good that we live in an era when high-performance smartphones are available at low cost.