Application Benchmarks 2
We are continuing our benchmarking with application benchmarks.
We are using SiSoftware Sandra version 2020.5.30.41 for benchmarking. We are going to use the Processor Arithmetic test to test Dhrystone and Whetstone performance. This tests the arithmetic and floating-point performance of the CPUs. The results are in GFLOPS.
In this first graph we are looking at Dhrystone Aggregated integer results. According to this test the Intel Core i5-9400 is faster at 212.2 GFLOPS. The AMD Ryzen 3 3300X is close though, only 3% in difference. Keep this mind, the i5-9400 does have 6 physical cores, while Ryzen 3 3300X has 4 physical cores, not taking into account SMT. This could be the difference in this particular test, even though we do have multi-threading enabled for this test. This is just one specific integer test though, and the Ryzen 3 3300X is very close comparatively. The Ryzen 3 3100 is a lot faster than the i3-9100F in this test.
The tables really turn through when it comes to float performance. Now the AMD Ryzen 3 3300 pushes forward 22% ahead of the Intel Core i5-9400 in float performance. This shows how well Ryzen 3 is suited for such tasks. Even the Ryzen 3 3100 is faster than the Intel Core i5-9400 in float. These are strong CPUs for these types of workloads.
We are using wPrime version 2.10 and running the Run 1024M test with the appropriate threads for each CPU.
The results are in time in seconds, and lower is better. The AMD Ryzen 3 3300X is 16% faster than the Intel Core i5-9400 in completing this test. It saves 30 seconds off the time which is pretty big. What’s really slow is the 4c/4t i3-9100F. The 6c/6t i5-9400 is better, but having 8 threads total is more advantageous with the Ryzen 3 3100 and 3300X.
We are using WinRar version 5.90 and running the built-in benchmark.
The AMD Ryzen 3 3300X really takes the cake in this benchmark and just demolishes everything. It is a whopping 79% faster than the Intel Core i5-9400! It’s also 60% faster than the Ryzen 3 3100. It must have something to do with the unified cache on the single CCX versus splitting the cache on the Ryzen 3 3100.