AMD's Ryzen 7 2700X shows the value of eight cores armed with simultaneous multi-threading technology through our parallelized rendering workloads. The Ryzen CPU comes out ahead in the Cinebench, LuxRender, and Blender benchmarks.
Meanwhile, Core i7-9700K trails AMD, but does offer a slight step up over the previous-gen -8700K in stock form. After overclocking, Core i7-8700K surprisingly beats the -9700K in our Corona and LuxRender tests.
Single-threaded workloads, rendering or otherwise, continue to be dominated by Intel's capable architectures.
Encoding & Compression
LAME is the quintessential example of a lightly-threaded workload. No surprise: an overclocked Core i7-9700K leads convincingly.
Our threaded compression and decompression metrics work directly from system memory, removing storage throughput from the equation. Ryzen 7 2700X is competitive in the compression test, but really shines through our decompression task.
y-cruncher, a single- and multi-threaded program that computes pi, is a great test for measuring the effect of AVX instructions. We dialed back the Core i7-9700K's all-core AVX frequency to 4.8 GHz after overclocking, so its single-core y-cruncher score at stock settings is faster than the tuned configuration. That's a result of the 4.9 GHz single-core Turbo Boost bin.
Core i7-9700K beats the Ryzen 7 2700X during our HandBrake x265 test, which relies heavily on AVX instructions. Meanwhile, the 2700X is more competitive in our H.264 benchmark. Notice that the tuned -9700K outpaces Intel's stock configuration, despite our 4.8 GHz AVX offset. That's because the stock setup drops to an all-core 4.6 GHz under full utilization.