Benchmarks: 7-aught, Excel, Handbrake, Adobe Premiere, SoTR

vii-zip pinch performance is an event for the 2990WX, at least when testing on Windows. Thus far this issue has not been addressed and therefore we meet the 2920X merely beating the 2990WX while it smashed the 2970WX. Still when it comes to file pinch Intel does appear to have a reasonably pregnant performance reward.

Withal when it comes to decompression the AMD processors come into their own. The 2990WX is a weapon hither and the 2970WX is also mighty impressive, beating the 7980XE by a 29% margin.

The Microsoft Excel criterion might be a chip redundant at present as most of these loftier-end desktop CPUs take less than 2 seconds to consummate the workload, just we have the results so why not include them. Intel does savor a slight performance reward in this exam but with the Monte Carlo Simulation beingness such an farthermost workload I dubiousness anyone is e'er going to notice Intel's performance advantage in this awarding.

Where you might find Intel's performance advantage is when working with H.265 content, their superior AVX implementation tin exist seen here. AMD besides accept an issue with the Windows scheduler in this test which certainly doesn't help and it's the reason why the 12-core Threadripper CPUs are seen matching the 24 and 32-cadre models. AMD's 16-core 2950X is too to put up a decent fight but even so the 7960X is still 14% faster, though in terms of value it's still much worse.

Since reviewing the 2990WX, Adobe Premiere has seen a major update and with information technology performance of the 32-core processor has been much improved, though it's all the same slightly slower than the xvi-cadre model, still it was considerably slower previously. The 2970WX basically matched the 2990WX and was therefore besides slightly slower than the 2950X. Still while the 2950X offers the nigh bang for your buck of the college-stop Threadripper CPUs in Premiere, the 2970WX even so stacks upward well against the Intel competition equally it was just 10% slower than the 7980XE.

The new 2920X also does well, though information technology is simply vii% faster than the 1920X and at launch does cost considerably more. Yet the 2920X was on par with the 9900K running hot without a TDP limit.

This is our mainstream Warp Stabilizer test running but a unmarried iteration. I haven't had fourth dimension to run our dozen simultaneous Warp Stabilizer examination just withal but I will include that data in futurity content. Once more the 2950X is the best Threadripper has to offering in Premiere while the higher-terminate 24-core and 32-core parts do lag behind.

Gaming

Don't freak out... but for now I've just managed to clasp in a single game test for this review, the focus is on productivity but shortly we will create an article more focused on gaming performance. Providing a sample of that work we have the Shadow of the Tomb Raider results, starting with the 1080p data and again I'grand using the RTX 2080 Ti.

With the Dynamic Local Mode enabled for the 2990WX and 2970WX we still run into sub-optimal frame time performance, though the average frame rate isn't bad. Here we see the 2920X and 2950X matching the frame time of the 2700X while beating it by a reasonable margin. I went dorsum and re-tested the 2700X just to brand sure there weren't any updates that improved performance but I got the same 93 fps on average in our test. Then the 12 and 16-core Threadripper CPUs are offer a performance increase here over the 2700X which is most unexpected.

Still once we move to 1440p the 2700X is at present able to roughly lucifer the average frame rate of the Threadripper CPUs while offer slightly better frame time performance, though overall performance was much the same between the 2920X, 2950X and 2700X. They were all comparable with the 9900K as we are mostly GPU bound here. Again even with Dynamic Local Mode enabled nosotros were seeing some serious frame time issues with the 2970WX and 2990WX processors.

This persisted even at the extreme 4K resolution, as far as I could tell the Dynamic Local Mode was working but AMD has reported a few bugs with enabling it, hopefully those are sorted out with the public release which should exist available at present. Anyway as expected with the exception of the 24-core and 32-core models we're very much GPU bound at 4K and here the 2920X was able to get the about out of the RTX 2080 Ti.