Linux vs. Windows Benchmark: Threadripper 2990WX vs. Core i9-7980XE Tested - myerstoop1998

Due to touristed necessitate, we're looking into AMD's Threadripper 2990WX performance victimisation Linux. Apparently this is something we should have done since day one, according to many AMD fans, although of course this is not part of our usual battery of tests.

Just in case you missed it, we did an extensive 24-hour interval-one review of the Ryzen Threadripper 2990WX and Threadripper 2950X processors. The latter is the drop-in renewal for the previous flagship Threadripper and an overall great performer for the price. The 2990WX however is a different kind of beast, more expensive and offering more cores, but with mixed results. That's why we followed awake with the mega-tasking examination, a separate multi-tasking focused bench mark in an attempt to research workstation carrying out of these high-end CPUs more closely.

Today we have the Threadripper 2990WX place against the Core i9-7980XE headland to head exploitation both Windows 10 and Linux, many especifically Ubuntu 18.04.1 LTS. Both in operation systems were proved in their out-of-the-box configuration, so no optimizations were made.

Both test systems take up been configured with 64GB of DDR4 memory, the 2990WX is noncomprehensive to 3000 MHz, while the 7980XE happily accepted 3200 MHz and could go high, but for the sake of disagreeable to keep things as apples to apples as possible we went with low latency CL14 DDR4-3200 memory.

Benchmarks

M-Queens is part of the Phoronix-Test-Suite and it was one of the few benchmarks in the suite that we could get to work on Linux and Windows as well equally AMD and Intel hardware. The test cortege was a bit of a nightmare to exploit with to be reliable.

The bench mark measures the time to solve the n-queens problem, m-queen just uses a more larger board qualification it a yearner more complex problem. Here we are measuring the time to solve and as you can consider the 2990WX does this in round half the time of the 7980XE victimisation either Windows or Linux which is exceedingly impressive.

Stockfish is a justify and unsettled-source chess engine, usable on single platforms. It's consistently ranked as uncomparable of the best chess-engines and is the strongest acceptant-seed chess locomotive in the world. The speed of this test is measured in nodes positions per moment, essentially positions per s and this is unregenerate all by the processors carrying into action.

Once again we see when it comes to chess the 2990WX is a beast exploitation Windows or Linux. That said this sentence we do see a 23% performance uplift for the 32-core processor when using Linux. In the meantime the 7980XE only saw a 4% performance increase when testing with Linux.

The last cheat benchmark we're going to look at is Crafty and again we're measuring performance in nodes per secondment. Interestingly, the Nub i9-7980XE wins out Here and saw the biggest performance uplift when moving to Linux, a 5% carrying out increase was seen opposed to just 3% for the 2990WX and this made the Intel CPU 12% faster overall.

John the Ripper is a password cracker that's currently available on bigeminal platforms including Linux and Windows. Exploitation the traditional Data Encryption Casebook we meet that the 2990WX and 7980XE execute roughly the same amount of Crypts per forward when using Windows. However victimisation Linux it's a completely different ball secret plan. Hera the 7980XE is a whopping 80% quicker merely it's the 2990WX that's genuinely staggering delivering 3x more performance making it roughly 80% faster than the Heart and soul i9 processor.

We find a similar story when using the Blowfish cipher. Here the 7980XE is 7% faster on Windows but the 2990WX is almost 70% faster when using Linux. So piece the 32-core processors looks less than impressive using Windows 10, it looks incredible using Linux.

GraphicsMagick is a simple but highly efficiency software package for viewing and manipulating images. It supports a massive array of formats and image processing is to a great extent multi-rib. IT's used past single websites to process large numbers of uploaded images. Hither we are looking at the sharpen carrying into action which is measured in iterations per minute, and so higher is better.

Using Windows 10, the 7980XE is 26% faster than the 2990WX, however using Linux turns that around and right away the 32-core processor is 10% faster. Not a massive win only the AMD processor does come from well rear end on Windows to beat the Sum i9 on Linux.

The enhanced operating performs about 30% fitter with Linux but we ascertain similar gains for some the AMD and Intel processors, the end result it a minor win for AMD.

Next up we have the Apache Benchmark which measures the operation of HTTP web servers and as you can see performance is extremely poor on Windows. Windows Host does legible a bitty amended, but inactive gets smashed about by Linux, which is why almost entanglement servers run Linux (e.g. TechSpot uses nginx instead of Apache happening Linux boxes). Anyway, on Windows 10 the 7980XE was 4% quicker than the 2990WX, but on Linux the 18-core processor was almost 40% faster, so not a great result here for the 2990WX.

We've seen previously that encoding performance with the 2990WX isn't great and again we undergo that the 7980XE is 6% faster using Windows. We see much the same exploitation Linux, the 7980XE is able to butt out the 32-core processor. So for encoding workloads Linux appears to do little to assist the 2990WX dead.

The 2990WX also sucked in our VeraCrypt bench mark and we find approximately odd results with Linux. For the 50MB test the 2990WX was slightly slower than the 7980XE when exploitation Windows 10. But then Linux tanks operation, particularly for the 7980XE.

Simply where the 2990WX really struggled connected Windows was on the memory intensive 1GB test. Here IT offered fractional the operation of the 7980XE. This result is improved on Linux, but even off so, the 7980XE was still 35% faster. So the 1GB results are roughly what we were expecting to find, we're not indisputable what's going happening with the 50MB results using Linux though.

We launch some odd results victimization 7-zip on Linux as well. The 2990WX's exceptional decompression performance was reduced by 13% when using Linux opposing to Windows 10. The 7980XE too saw execution drop off, though IT was same peanut.

Condensation performance for the 2990WX was drastically improved, here we experience a massive 54% performance upheaval. That said the 7980XE besides enjoyed a public presentation further, though it was a much smaller 10% increase. Still this meant overall the 7980XE was still 11% faster than the 2990WX.

Last raised we'Ra going to run off the Blender Surface Data bench mark suite and we'ray expecting confirming results for the 2990WX here since it's already proven to be a rendering power house. Here we have the Barbershop Interior test and we're mensuration completion time in seconds.

The 7980XE was 23% faster victimisation Linux, while the 2990WX was 31% faster. This meant whereas the 2990WX was 21% faster than the 7980XE on Windows, it's 29% faster using Linux.

We see a slightly different scenario with the BMW27 test though the final result still sees the 2990WX well forwards. Still where as the 32-core processors was 52% quicker using Windows 10, that margin was reduced to 45% with Linux.

This time we see similar scaling using either CPU, some were 48% faster using Linux for the Schoolroom workload.

This time Linux slightly best-loved the 2990WX fashioning information technology 59% faster where as IT was 52% faster using Windows.

We see a similar story with the Koro workload, Linux slightly improves the 2990WX's put across, only overall both CPUs see good gains on Linux.

Then we have the Pavilion Barcelona workload and we find out more of the unchanged, so time to wrap this up.

Closing Remarks

So is Windows 10 gimping the Threadripper 2990WX's performance? Unquestionably yes, simply it's gimping the 7980XE to a certain degree as well. If you were to just compare the 2990WX using Linux and Windows 10 the site would appear a whole sle more extreme than it in truth is. For instance, you'd conclude that the 32-core processor is going to wreck the 7980XE past an even more uttermost margin when in point of fact both CPUs see an almost identical performance uplift.

Meanwhile, programs that were problematic connected Windows such as 7-zip and VeraCrypt still provide us with rum results. For representative the 2990WX's compression carrying into action was strengthened, merely the decompressing performance was impaired. Still overall Linux did provide a improved balance.

Though we could stock-still observe the bear on of the multi-die design connected the 2990WX and its "limited" computer storage bandwidth to commit it in some way, in the 7-null compression test, even along Linux and performance jumps all over the place during the examination, but the norm ended up being very similar after each run.

Encoding performance likewise stiff lackluster, though we have only managed very qualified testing happening Linux so far. As for for non-memory light-sensitive workloads like the Blender renders, Here the 2990WX is a wildcat. This also includes the chess and password-cracking benchmarks.

So again, depending on what you plan on using the 2990WX for will square off just how useful it really is. Information technology's our opinion that there's elbow room for Windows 10 optimizations in order to better utilize, or finer manage, these core-heavy CPUs. Improved thread scheduling isn't loss to make the 2990WX a weapon in workloads where it's struggling, merely it's our hope that it can at least match the 2950X under those conditions, rather than get into much slower.

Results like what we sawing machine in the GraphicsMagik sharpen test were promising American Samoa the 2990WX went from being much slower than the 7980XE on Windows to slightly quicker on Linux, and we saw the Saame thing with John The Ripper. Still, this is only when a bittie sample of applications and doesn't really do much for content creators.

We'd love to psychometric test impermissible Premiere connected Linux to see if our custom warp stabilizer prove running play a xii instances simultaneously provides better results. Happening Windows this test maxes out the 2990WX, but the resulting performance is very poor.

Wrapping this one up we have to enjoin, the few years we exhausted messing more or less with Linux were stimulating, but ultimately ready-made us more discriminating of Windows 10's desktop Bone feel.