Rendering
We expect Ryzen 5 2400G's eight threads to compete readily with the more expensive Core i5-8400's six physical cores in well-parallelized tasks. Although AMD never leads at stock settings, it at least holds its own after some casual overclocking.
The Core i5-8400 wins in every benchmark except one. However, the Ryzen 5 2400G is solid in tasks that fully utilize its SMT-enabled cores, such as Cinebench, Corona, and LuxMark.
As expected, the single-threaded benchmarks go Intel's way. Tuning does help Ryzen close the gap, though.
Encoding & Compression
The Ryzen architecture has always excelled in decompression workloads. This continues with Raven Ridge. AMD's Ryzen 5 2400G shows well in the multi-core compression workload as well. Meanwhile, the Core i3-7100 and AMD A10-9700 demonstrate just how under-powered they are in demanding tests.
The Ryzen 5 2400G performs exceedingly well in LAME, which is primarily single-threaded. This result leads us to believe that the benchmark benefits from low cache latency, as Intel processors maintain their per-core performance advantage.
There's normally a larger delta between Intel and AMD processors during our HandBrake x265 test compared to the x264 benchmark due to the former's heavy use of AVX instructions. But we don't really see that this time around. The Ryzen 5 2400G is much more competitive in AVX-heavy workloads than we expected.
Speaking of AVX, we also provide results from y-cruncher, a single- and multi-threaded program that computes Pi using AVX instructions. We tested with version 0.7.3.9474, which includes Ryzen optimizations. The Intel processors take a big lead in our single-threaded run. Again, though, the delta between Intel and AMD contenders is smaller than expected when we divide this task across available physical and logical cores. We theorize that a lower cache latency helps feed the cores in these data-hungry AVX workloads, thus speeding performance.
MORE: Best CPUs
MORE: Intel & AMD Processor Hierarchy
MORE: All CPUs Content