OnePlus 13 Cheats in Benchmarks: Performance Under the Microscope

The OnePlus 13, touted as a flagship powerhouse, has been caught manipulating benchmark scores to appear more powerful than it is under real-world conditions. Korean YouTuber BullsLab exposed this practice by disguising benchmark apps like Geekbench and 3DMark as everyday apps like Instagram or games like Genshin Impact. The results? A significant drop in performance metrics.

The Findings: A Tale of Two Scores

When benchmark apps were disguised, the scores dropped significantly compared to their original results. Here’s a breakdown:

Geekbench 5

• Original:
  • Single-core: 2121
  • Multi-core: 6607
• Disguised as Instagram:
  • Single-core: 1501
  • Multi-core: 5607

3DMark

• Original Wildlife Extreme Stress Test: 6695
• Disguised as Genshin Impact: 4165

Geekbench 6

• Original:
  • Single-core: 2998
  • Multi-core: 9227
• Disguised as Genshin Impact:
  • Single-core: 2474
  • Multi-core: 8460

Opinion: The Ethics of Benchmark Manipulation

1. Benchmarking vs. Real-World Performance
   • Benchmarks serve as a universal yardstick to measure device performance, but they should represent real-world scenarios. By optimizing performance solely for benchmarking apps, OnePlus deceives users into believing the device performs better across the board than it does in daily use.
2. Disguising the Apps
   • BullsLab achieved this by decompiling and recompiling benchmark apps, altering their package identifiers to mimic popular apps like Instagram or Genshin Impact. The fact that performance dropped dramatically in these scenarios proves that OnePlus artificially prioritizes resources for known benchmark apps.
3. Impact on Consumer Trust
   • This practice undermines trust in manufacturers and the benchmarking ecosystem. Enthusiasts, who rely on benchmarks to gauge performance, are misled, affecting purchasing decisions.

Why do OEMs cheat in benchmarks?

1. Marketing Hype: High benchmark scores create flashy headlines, giving devices an edge in the competitive smartphone market.
2. Impressing Reviewers: Many tech reviews prominently feature benchmark results, influencing buyers’ perceptions of a device.
3. Short-Term Gains: Benchmark manipulation can boost sales initially, but the long-term impact on brand reputation is often negative.

Is Benchmark Cheating Common?

Unfortunately, OnePlus isn’t alone. In the past, other OEMs like Samsung, Huawei, and even Xiaomi have been caught employing similar tactics. However, with Snapdragon 8 Gen Elite SoC and custom cooling systems on the OnePlus 13, such practices seem unnecessary and raise questions about actual performance efficiency.

What Can Be Done to Address This?

1. Transparent Benchmarking Standards
   • Benchmarking organizations like Geekbench and 3DMark must evolve their tools to detect such optimizations. Geekbench has blacklisted devices in the past for similar reasons, and stricter measures could deter such practices.
2. Consumer Awareness
   • Educating users about real-world performance metrics versus synthetic benchmarks can shift focus to actual usage scenarios.
3. OEM Accountability
   • Consumers and tech reviewers should hold OEMs accountable for ethical practices. Naming and shaming incidents like this is a step toward transparency.

The Bigger Picture: Why This Matters

• Gaming Performance: Disguising apps as games like Genshin Impact revealed a sharp drop in performance, hinting at potential issues with thermal management or resource allocation. Gamers, who push devices to their limits, are likely to face real-world throttling that benchmarks fail to predict.
• Battery Life: Benchmark cheating often involves running the chip at unsustainable speeds, potentially affecting long-term battery health.

Conclusion

The OnePlus 13 scandal underscores a troubling trend in the smartphone industry. While the device remains a strong contender in terms of hardware, such practices erode user trust and call into question the validity of benchmarking as a performance metric.

For consumers, the key takeaway is this: benchmarks aren’t the full story. Real-world usage and long-term reliability are far more important than inflated numbers on a screen.