We see a similar story when encoding with HandBrake, the 2990WX is significantly under utilized in this test, maxing out at about 30% utilization. That said under those conditions you’d expect the 2990WX to roughly match the 2950X and it fails to do so. Here the 2950X was 15% faster and I should also note it offered no real performance gain over the 1950X in this test. Meanwhile the 2990WX was only able to match the 1920X which is obviously a massive disappointment.
When encoding with Adobe Premiere we find a similar situation, the 2990WX took 35% longer to complete the workload when compared to the 2950X. The good here though is that the 2950X was 7% quicker than the 1950X and that made it just 7% slower than the Core i9-7960X. So a pretty great result for the 2950X and an disastrous result for the 2990WX.
Although the export test only utilized the 2950X to the tune of about 90% and the 2990WX to about 60%, the warp stabilizer test where I run a dozen instances simultaneously is able to max out the 2990WX at 100%. Even so the performance was mighty underwhelming and here the 32-core processor was again slower than last years 12-core model and only managed to match Intel 10-core Skylake-X part.
Again while the 2990WX was extremely disappointing the 2950X wasn’t. It shaved 5% off the 1950X completion time to take out top spot so an amazing result for the new 16-core processor.