AMD’s Ryzen 3000 CPUs come with a ton of features. Precision Boost 2 and Precision Boost Overdrive are some of the more well-known ones. Still, there has been a lot of confusion regarding these two boosting algorithms: What they do and how they do it. For a simple theoretical explanation, read this post:
AMD Ryzen Precision Boost 2 vs Precision Boost Overdrive vs AutoOC: What’s the Difference?
In this post, we’ll have a look at the real-world implications of Precision Boost Overdrive and AutoOC (if any). We’ll track the clock scaling, power consumption and temperatures of two high-end Ryzen 3000 parts: The Ryzen 7 3700X and the Ryzen 9 3900X, and of course, see how the resulting performance is affected.
- CPU: AMD Ryzen 7 3700X/3900X
- GPU: NVIDIA GeForce RTX 2080 Super
- Motherboard: ASRock X570 Taichi
- Memory: Trident Z Royal 8GB x 2 @ 3600 MHz
- PSU: Corsair HX1000i
- All the tests were conducted using the stock AMD spire cooler, deliberately, to analyze the impact of PBO and AutoOC under stock conditions
Precision Boost Overdrive and AutoOC: Ryzen 7 3700X
Cinebench R20 Performance
On paper, Precision Boost Overdrive should be slightly faster than stock and AutoOC should be the fastest out of the three. However, that’s not the case. The single-core performance is fastest at stock settings while the PBO and AutoOC scores are identical, both being a tad bit lower than the former.
The multi-threaded scores aren’t that coherent either. At stock, the Ryzen 7 3700X scores 4,900. On enabling PBO, it drops ever so slightly to 4,893. Finally, in AutoOC mode, we see a small performance bump with a score of 4,925 points. Let’s dig deeper and see how the core clocks, power, and temperatures are affected on enabling Precision Boost Overdrive and AutoOC.
Cinebench R20 Single-Core Scaling
In the single-core test, there is no legible difference between the clock scaling at stock and PBO modes. In the former, core 7 and core 4 are the fastest while in the latter core 4 is swapped for core 6. I don’t think this is even because of PBO, just a boost algorithm thing.
On enabling AutoOC, there’s a small but clear increase in the core frequencies. Core 7 spends more time near 4.4GHz, core 3 boosts more often and core 6 also runs at a slightly higher frequency.
The impacts of Precision Boost Overdrive and AutoOC are more pronounced in multi-threaded workloads. Let’s have a look.
Cinebench R20 Multi-Core Scaling
In the Cinebench R20 multi-core test, there are a few subtle differences between stock and PBO. While the average frequency of the eight cores is higher throughout the benchmark in the latter, there are also a few spikes to 4.4GHz that are absent at stock conditions.
AutoOC further pushes the average core frequency up by a notch. Furthermore, the clocks are more consistent now. Towards the end, all the cores get to 4,115MHz and maintain it through the remaining part of the test.
While the power consumption is more or less identical at stock and PBO, there’s a sharp increase when AutoOC is enabled. The TDP surges by 6W from 46W to 53W when it’s switched on.
The temperatures follow the power chart. At stock and PBO, there’s no change but upon enabling AutoOC, the temps cross 80 degrees for the first time and stay there for the remainder of the benchmark.
Benchmarks continue on Page 2