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

Yesterday we took a long-snouted look at the new Threadripper 2990WX and 2950X processors. In short, the 2950X affected with self-colored operation crosswise the board at a premium yet combative price. The 2990WX was inferior dazzling with mixed performance and a screechy asking price.

Suitable now Windows 10 is being blamed for the mixed results and perhaps rightfully so, we'll have to look into this further and keep our fingers hybrid for an update. In the meantime, a large number of people voiced concerns ended our testing, and many others, as we were entirely testing a single application at a time.

We had always plotted to make a isolated multi-tasking focused benchmark and we said as much in yesterday's review. While editing and wrapping dormy the 2990WX and 2950X review we began looking at into multitasking performance and later on extra days of testing we've got a few more results to hash out.

For this test the 2990WX has been compared directly to the Core i9-7980XE and we bear 5 batches of multi-tasking benchmark results to check up on. All the testing was conducted using stock out of the box seat settings with a 240mm closed loop cooler install happening each CPU. There was much jumping in the lead and down about the Wraith Ripper causation the 2990WX to throttle in our benchmarks despite us showing footage that contradicts this. Anyway, it wasn't throttling and we did test both vent and liquid while making the primary content and found no difference with tired settings.

Moving on, both systems have been configured with 64GB of DDR4 remembering using eight 8GB modules and this limited the 2990WX to DDR4-3000, while the 7980XE could go overmuch high but we settled with bass latency DDR4-3200 memory to try and keep things even, and somewhat realistic.

The following graphs might look a little odd ab initio, especially when compared to what you're unremarkably used to see, but we'rhenium hoping the formatting will be easy to follow as information technology gives you all the info on a single graph per test.

Benchmark 1

Low up we're going to look at how the 2990WX and 7980XE compare when running Liquidizer Afford Data and 7-nada. Mark that results here show how the CPUs execute in to each one application individually, and then that is to say Blender was executed and completed before moving on to testing 7-Zip, so these are just normal benchmark results.

Present the 2990WX realised the BMW27 workload 36% faster and the Barbershop workload 25% faster. Past with 7-Zip we see the 2990WX is 116% faser for decompressing work but 35% slower for compression exploit. Now lease's run some the Blender and 7-Zip applications at the same time, 7-Zip is on a persisting loop while Blender runs it's accepted batch of workloads.

Hither we figure a few interesting things. The 2990WX is in real time 58% quicker for the BMW27 workload and 30% faster for the Barbershop workload. However the 7-Zip performance is less favorable. Whereas the 2990WX was 116% faster previously when looking the decompression carrying into action, it's forthwith 3% slower. In the compression trial, it was 35% slower previously and now IT's alone 16% slower.

Basically the 2990WX was already mighty important-looking in the Blender rendering test and adding 7-Zip to the mix just makes it look more imposing, simply the massive advantage it had in 7-Zip is now gone, making it inferior in some the densification and decompression tests. The 7980XE sure as shooting appears to be prioritizing 7-Zip while the 2990WX is prioritizing Blender.

Bench mark 2

Incoming 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 equally well as the total try time. Then RealBench will be run its heavy multitasking workload connected a eyelet.

The 2990WX doesn't had best in HandBrake and Here it was 31% slower than the 7980XE. It's also non great in RealBench either, taking 10% longer to complete a single pass. Permit's now combine these 2 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 naturally, noneffervescent slower. However while we saw a rock-bottom margin in HandBrake we run into the margin growth a bit for RealBench, the 2990WX is instantly 15% slower. So the 2990WX wasn't really able to make up any ground on the 7980XE therein multitasking benchmark.

Benchmark 3

Ane of the worst results for the 2990XW seen in our day one coverage was found when testing with VeraCrypt, 1GB test. So we'atomic number 75 combining that tryout with one of the best test results for the 32-core processor, which was seen in Corona. Here the 2990WX was 28% faster than the 7980XE and as we've seen IT always does well in rendering workloads. Nonetheless for VeraCrypt it was 47% slower, not a great result that one. So let's like a sho run these applications simultaneously.

As we saw in the front series of multitasking benchmarks AMD seems to prioritize rendering workloads while Intel favors the memory memory sensitive workloads. The 7980XE took almost 6x thirster than the 2990WX to complete the Corona benchmark spell it only saw a 7% reduction in decryption public presentation. Lag the 2990WX saw a 34% reduction in decryption carrying into action and a 61% increase in rendering time.

So the 32-core processor does appear much balanced Here and so it should with 14 more cores. Static we do sustain mixed results for the 2990WX, information technology's amazing in Corona and horrible in VeraCrypt and that's true whether you run these applications singly Oregon at the same time.

Benchmark 4

This clock we give birth three applications: HandBrake, CPUz and 7-Nada. Here's how they all perform individually along each CPU. The 2990WX is 22% slower in HandBrake, 95% faster in CPUz and again we see the same interracial 7-zip performance that we discussed earlier.

Running all three applications simultaneously does provide some stimulating results. Whereas the 2990WX was 22% slower in HandBrake previously, information technology's now 29% quicker, so that's impressive. It does take a spacious hit in CPUz but even so was still 21% faster than the 7980XE.

Then even in 7-Naught we get hold positive results for the 32-core processor. It managed to maintain a good chunk of its decompression performance and the 7980XE took a big plenty hit in the compression test that the 2990WX was able to just edge ahead. So this is the start truly great solution that we've seen for the 2990WX in these multi-tasking benchmarks as we haven't enclosed a interpretation workload to prop the 32-core processor risen.

Benchmark 5

Last up we have another triple header using POV-Ray, WinRAR and F1 2017. POV-Ray provided the 2990WX with its best result in our Clarence Shepard Day Jr. one review and as you can see IT's 36% quicker than the 7980XE in that test. WinRAR though International Relations and Security Network't very core heavy but it is extremely memory intensive so IT will be nifty to see if that messes with the POV-Irradiate performance.

The 2990WX was 59% slower than the 7980XE when testing with WinRAR, or other fashio of putting information technology, the 7980XE was 142% faster. Finally, we have F1 2017 and present the 2990WX was 24% slower when comparing the average out frame rate and 30% slower for the 1% low result.

As we've seen time and clock again, when multitasking the 2990WX's advantage in rendering applications is only extended. Where it was 36% faster than the 7980XE antecedently, it's in real time 45% faster. Meanwhile IT was 59% slower in WinRAR previously, now it's just 37% slower. Still slower of course but a greatly reduced margin. Yet things do last horribly wrong for the gaming performance and now F1 2017 is wholly unplayable.

Although the 7980XE did produce a 1% low result of 34 fps this is a trifle deceiving Eastern Samoa the halting was still very playable and only in one case or twice per lap did we notice a big lag spike. For the nearly part frame rates were up around 100 FPS and it was a astonishingly smashing experience. The 2990WX on the separate hand was a stuttery mess and would have been impossible to play, thankfully though the recorded benchmark was taking care of the driving. So again we see mixed performance from the 2990WX in this test.

Closing Remarks

We're now better armed with more than information in regards to how the 2990WX performs than we were a hardly a days ago. That said, this additive entropy hasn't denatured our perception much. Throwing more work at these CPUs impartial provided many miscellaneous results, whereas you'd expect a 32-essence processor to oppress an 18-core processor under these conditions, it didn't.

The performance hiccups that we've seen with the 2990WX so far might not all be down to the NUMA configuration, increased computer memory latency for half the cores, or narrow per center bandwidth when fully utilized. Only thither's also the possibility that the culprit can be found on the Windows scheduler. It's sure possible a future Windows update could better things for the 2990WX and we've already seen this with the first generation Threadripper CPUs.

Bottom line remains untouched from yesterday's review. We'd buy in neither the 2990WX nor the 7980XE, instead we would buy the 2950X. The 2990WX seems unsafe at the consequence, the heterogenous Windows performance and distant price is incomplete enough, but we're almost certain by this clip next year AMD will have something significantly better connected fling using the 7nm process.

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