The Mega-Tasking Test: AMD Threadripper 2990WX Heavy Multitasking Benchmark

Yesterday we took a long expect at the new Threadripper 2990WX and 2950X processors. In curt, the 2950X impressed with solid functioning across the lath at a premium still competitive price. The 2990WX was less impressive with mixed performance and a high asking price.

Right now Windows 10 is being blamed for the mixed results and maybe rightfully so, we'll have to expect into this farther and keep our fingers crossed for an update. In the meantime, a large number of people voiced concerns over our testing, and many others, equally we were but testing a single awarding at a time.

We had always planned to brand a separate multi-tasking focused benchmark and nosotros said equally much in yesterday'due south review. While editing and wrapping upwards the 2990WX and 2950X review we began looking into multitasking performance and afterwards additional days of testing we've got a few more than results to talk over.

For this examination the 2990WX has been compared directly to the Core i9-7980XE and we take 5 batches of multi-tasking benchmark results to cheque out. All the testing was conducted using stock out of the box settings with a 240mm closed loop cooler install on each CPU. In that location was some jumping upwards and downwardly about the Wraith Ripper causing the 2990WX to throttle in our benchmarks despite united states of america showing footage that contradicts this. Anyway, it wasn't throttling and we did test both air and liquid while making the original content and establish no deviation with stock settings.

Moving on, both systems have been configured with 64GB of DDR4 memory using eight 8GB modules and this limited the 2990WX to DDR4-3000, while the 7980XE could get much college but nosotros settled with low latency DDR4-3200 retentivity to endeavor and proceed things even, and somewhat realistic.

The following graphs might look a little odd at first, especially when compared to what yous're normally used to see, but we're hoping the format volition be like shooting fish in a barrel to follow as it gives you all the info on a single graph per test.

Benchmark 1

Offset up we're going to wait at how the 2990WX and 7980XE compare when running Blender Open up Data and 7-zip. Annotation that results here show how the CPUs perform in each application individually, so that is to say Blender was executed and completed before moving on to testing vii-Zip, so these are just standard criterion results.

Here the 2990WX completed the BMW27 workload 36% faster and the Barbershop workload 25% faster. And so with 7-Nil we come across the 2990WX is 116% faser for decompression work simply 35% slower for pinch work. Now let'south run both the Blender and vii-Zip applications simultaneously, 7-Zip is on a continuous loop while Blender runs information technology's standard batch of workloads.

Here nosotros see a few interesting things. The 2990WX is now 58% faster for the BMW27 workload and 30% faster for the Barbershop workload. Still the 7-Nada performance is less favorable. Whereas the 2990WX was 116% faster previously when looking at the decompression functioning, it's now 3% slower. In the compression test, it was 35% slower previously and now it's only sixteen% slower.

Basically the 2990WX was already mighty impressive in the Blender rendering test and calculation seven-Zip to the mix just makes it look more impressive, but the massive advantage it had in 7-Zip is now gone, making it inferior in both the compression and decompression tests. The 7980XE certainly appears to be prioritizing 7-Null while the 2990WX is prioritizing Blender.

Criterion ii

Side by side up we're testing with HandBrake and RealBench. For HandBrake we have a H.264 4K 60fps video that we're converting to H.265 4K 60 fps and we're reporting the average frame rate as well as the total render time. And then RealBench will be running its heavy multitasking workload on a loop.

The 2990WX doesn't do well in HandBrake and here it was 31% slower than the 7980XE. It's too not great in RealBench either, taking 10% longer to consummate a single laissez passer. Let'southward now combine these two tests by running them simultaneously.

With both applications running simultaneously the 2990WX is now 17% slower than the 7980XE in HandBrake which is a reasonable improvement, but of course, still slower. However while we saw a reduced margin in HandBrake we see the margin increment a little for RealBench, the 2990WX is now 15% slower. Then the 2990WX wasn't really able to brand upwardly whatsoever ground on the 7980XE in this multitasking criterion.

Benchmark 3

One of the worst results for the 2990XW seen in our day one coverage was found when testing with VeraCrypt, 1GB test. Then nosotros're combining that examination with one of the all-time examination results for the 32-core processor, which was seen in Corona. Here the 2990WX was 28% faster than the 7980XE and as nosotros've seen it always does well in rendering workloads. Nonetheless for VeraCrypt it was 47% slower, non a peachy issue that one. So let's now run these applications simultaneously.

As nosotros saw in the outset serial of multitasking benchmarks AMD seems to prioritize rendering workloads while Intel favors the memory retention sensitive workloads. The 7980XE took almost 6x longer than the 2990WX to complete the Corona benchmark while it simply saw a vii% reduction in decryption operation. Meanwhile the 2990WX saw a 34% reduction in decryption performance and a 61% increment in rendering time.

And so the 32-cadre processor does appear more balanced here so information technology should with xiv more cores. Still we practise have mixed results for the 2990WX, it's amazing in Corona and horrible in VeraCrypt and that's true whether y'all run these applications individually or simultaneously.

Benchmark four

This time nosotros have three applications: HandBrake, CPUz and 7-Nada. Here's how they all perform individually on each CPU. The 2990WX is 22% slower in HandBrake, 95% faster in CPUz and again we run across the aforementioned mixed vii-zip performance that we discussed earlier.

Running all three applications simultaneously does provide some interesting results. Whereas the 2990WX was 22% slower in HandBrake previously, it's now 29% faster, so that's impressive. It does take a big hit in CPUz but even and so was nevertheless 21% faster than the 7980XE.

So fifty-fifty in seven-Cipher we find positive results for the 32-core processor. It managed to maintain a practiced chunk of its decompression operation and the 7980XE took a big enough hit in the compression exam that the 2990WX was able to simply edge ahead. So this is the first truly neat event that nosotros've seen for the 2990WX in these multi-tasking benchmarks equally we haven't included a rendering workload to prop the 32-core processor upwardly.

Criterion 5

Last upwardly we have another triple header using POV-Ray, WinRAR and F1 2022. POV-Ray provided the 2990WX with its best result in our day i review and as you tin can see it's 36% faster than the 7980XE in this test. WinRAR though isn't very core heavy but it is extremely retention intensive so information technology volition be keen to see if that messes with the POV-Ray performance.

The 2990WX was 59% slower than the 7980XE when testing with WinRAR, or another way of putting it, the 7980XE was 142% faster. Finally, we have F1 2022 and here the 2990WX was 24% slower when comparison the average frame charge per unit and 30% slower for the one% depression result.

Equally we've seen time and time again, when multitasking the 2990WX's advantage in rendering applications is only extended. Where it was 36% faster than the 7980XE previously, it's now 45% faster. Meanwhile it was 59% slower in WinRAR previously, now it's just 37% slower. Still slower of form simply a profoundly reduced margin. Even so things do become horribly incorrect for the gaming performance and now F1 2022 is completely unplayable.

Although the 7980XE did produce a 1% low result of 34 fps this is a niggling deceiving every bit the game was nevertheless very playable and but once or twice per lap did we notice a big lag spike. For the nearly role frame rates were upwardly around 100 fps and information technology was a surprisingly good experience. The 2990WX on the other manus was a stuttery mess and would have been incommunicable to play, thankfully though the canned benchmark was taking care of the driving. Then once more we see mixed functioning from the 2990WX in this test.

Closing Remarks

Nosotros're now amend armed with more information in regards to how the 2990WX performs than we were a few days ago. That said, this additional information hasn't changed our perception much. Throwing more work at these CPUs only provided more mixed results, whereas yous'd expect a 32-cadre processor to beat out an 18-core processor nether these conditions, it didn't.

The performance hiccups that we've seen with the 2990WX then far might non all be down to the NUMA configuration, increased memory latency for one-half the cores, or limited per cadre bandwidth when fully utilized. But there'south besides the possibility that the culprit can be found on the Windows scheduler. Information technology'due south certainly possible a futurity Windows update could improve things for the 2990WX and nosotros've already seen this with the first generation Threadripper CPUs.

Bottom line remains untouched from yesterday's review. We'd buy neither the 2990WX nor the 7980XE, instead we would buy the 2950X. The 2990WX seems risky at the moment, the mixed Windows performance and extreme price is sketchy enough, only we're almost certain by this fourth dimension next year AMD volition have something significantly better on offering using the 7nm procedure.

From a consumer standpoint the 2990WX is poorly timed. From AMD's perspective we get why they released such a product. As always we'll continue to benchmark and update you if whatsoever noteworthy changes are made.