Rendering, Encoding and Compression
The 32C/64T Ryzen Threadripper 2990WX enjoys an uncontested core count advantage. Of course, as we've seen in the past, accesses from its remote memory controllers can affect performance in weird ways, depending on the metric. When the 2990WX can really utilize its array of execution resources, though, the competition gets stomped.
Intel's Core i9-9980XE isn't as fast across heavily-threaded rendering tasks, but in our benchmarks that isolate single-threaded performance, it lands much higher compared to other high-end desktop processors.
Encoding and Compression
Our compression and decompression tests work directly from system memory, removing storage throughput from the equation.
Ryzen Threadripper 2990WX stretches its legs in the decompression test, but exhibits its bipolar personality during the compression workload. Core i9-9980XE, on the other hand, is impressive in the compression test, largely because AMD's highest-end Threadripper processors are handicapped by memory throughput issues. The Core i9 fares well in decompression workloads as well.
y-cruncher, a single- and multi-threaded program that computes pi, is a great test for evaluating AVX-optimized performance. Intel’s Core i9-9980XE employs two 256-bit FMA units per core that operate in parallel, whereas Ryzen's Zen architecture divides 256-bit AVX operations across two 128-bit FMA units per core. Intel's AVX instruction support shines during the single-threaded benchmark, though the -9980XE only improves by ~4 percent over the -7980XE. We logged the same performance for both processors during the multi-threaded y-cruncher benchmark. However, both processors dropped to the same all-core 2.8 GHz AVX-512 offset during the test.
While our HandBrake x264 benchmark also leverages AVX instructions, the x265 test has a heavier distribution of the densely-packed instructions. Intel's HEDT Core i9 processors lead the pack in both workloads.
MORE: Best CPUs
MORE: All CPUs Content