AMD Ryzen 3 3300X vs. Intel Core i3 10100 In 350+ Benchmarks

The Ryzen 3 3300X was also faster for Python workloads including PyBench, Numenta, Cython-Benchmark, Numpy, Mlpack, Scikit-Learn, and PyPerformance.

When testing various CPU-based renderers including Blender, Tungsten, Appleseed, Radiance, AOBench, LuxCore, Smallpt, TTSIOD, V-RAY, and IndigoBench, the Ryzen 3 3300X was nearly 12% faster.

When looking at all of the single-threaded benchmarks, the Ryzen 3 3300X and Core i3 10100 performance was neck-and-neck.

If taking the geometric mean of all the CPU (non-gaming) benchmark results, the Ryzen 3 3300X came out to 6% on average faster than the Core i3 10100.

Considering the Ryzen 3 3300X is currently retailing for a few dollars less than the Core i3 10100, that's another easy win in AMD's favor.

Of all the 318 tests, the Ryzen 3 3300X came in first place 75% of the time.

When looking at the CPU peak frequency polled on a one second interval during the two days worth of testing on each of these CPUs and recording the highest value of any of the cores, the Ryzen 3 3300X had an average clock frequency of 3.95GHz while the Core i3 10100 was at 3.86GHz. Though there may be some subtle issues in Linux's CPUfreq/P-State reporting as both the 3300X and i3-10100 did report peak frequencies at 4.5~4.6GHz rather than 4.3GHz.

At least for what each CPU is exposing via its MSRs for their power reporting interfaces, the Core i3 10100 did have the advantage of consuming less power than the Ryzen 3 3300X. Both CPUs are rated for a 65 Watt TDP while their respective drivers reported peak power usage at 74~76 Watts and the Ryzen 3 3300X had an average power draw of 48 Watts to the i3-10100 at 35 Watts.

The slightly lower power usage of the Core i3 10100 was also reflected in lower temperatures with each of the CPUs stock coolers. The Ryzen 3 3300X ran warmer with a 69 degree average in these hundreds of demanding tests to the i3 10100 at around 58 degrees.

Those wanting to dig through these hundreds of individiual data points can find all of the data via this OpenBenchmarking.org result file.