Tech

Windows 11 vs Windows 10: AMD Ryzen Edition

Today we’re looking at the performance of Windows 11 and Windows 10 using AMD Ryzen processors. We recently did the same with 10th and 11th Gen Intel Core processors, but refrained from testing any AMD components due to a known L3 cache latency issue.

When we edited this snippet, Microsoft and AMD released a hotfix consisting of a Windows 11 update and a new chipset driver. Many of you have been waiting for an increase in the performance of Ryzen processors, so here we are using the Ryzen 5 3600 and Ryzen 9 5950X.

Both processors were tested with BIOS 3801 installed on an Asus ROG Crosshair VIII Dark Hero motherboard. For memory we used a Crucial Ballistix DDR4-3200 CL16 kit and an MSI GeForce RTX 3090 Gaming X Trio graphics card. In terms of storage, we used 8TB TeamGroup MP34Q NVMe SSDs, which are fast enough and we’ll include some SSD results at the end of the article.

There are 5 test configurations in total: three with the 5950X and two with the 3600. The Ryzen 5 3600 compares the performance of Windows 10 to Windows 11 with an L3 cache lag fix. Then with the 5950X, we compare Windows 10 with two Windows 11 configurations, one with the L3 fix and one without.

We’ll take a look at the performance of apps, games, storage, and load times. All results are based on averaging over 3 runs, and in some cases we turned off the entire system between runs to avoid caching. Now let’s move on to the graphs …

Application performance

Starting with caching results and AIDA64 memory, we see no real difference between the various configurations when comparing DRAM latency, L1 cache latency, and L2 cache latency. However, here you can see how bad the L3 latency was before the fix, when the L3 5950X latency was increased to 36ns, which is about 3 times what it should be.

Fortunately, AMD and Microsoft have solved this problem, and therefore Windows 10 and Windows 11 should be comparable in this regard.

Moving on to Cinebench, we find some pretty boring results, albeit not entirely unexpected. Basically, it doesn’t matter which version of Windows you are using, the performance will be about the same, and this was true even before the L3 cache latency fix.

It’s the same with the Blender Open Data benchmark, especially when looking at the Ryzen 9 5950X, which has a performance deviation of less than 1%. The R5 3600 was similar as well, although we are seeing a 3% improvement with Windows 11, which I would say is rather marginal.

Moving on to the 7-Zip results, we find that there is little difference in decompression between the two operating systems. However, the compression results are interesting as the Ryzen 9 5950X did lose 7% of its performance due to the L3 bug, which of course is now fixed and with the fix, Windows 11 and 10 remain the same.

When tested with Adobe Photoshop 2021, the R5 3600 saw a 5% improvement in Windows 11 performance, which was interesting to see, especially given the lack of performance improvement with the 5950X.

Then, with Adobe After Effects 2021, we’re looking at a 2% improvement with Windows 11 for both CPU configurations, while the L3 bug dropped 4% performance, so it didn’t really make a difference, although it was consistently slower in our testing.

The last app benchmark we ran was Adobe Premiere Pro 2021, and this time the Ryzen 5 3600 performed nearly identical on any operating system, while the 5950X performed 4% better on Windows 10.

Game tests

Let’s move on to some gaming tests. F1 2021 showed the same performance in all tested configurations with 5950X, even the L3 cache error did not cause performance degradation. The R5 3600 was 3% faster on average with Windows 11, but that’s a minor difference.

These results are really interesting …

Tom Clancy’s Rainbow Six Siege was many times faster when using Windows 11 with either of the two processors, although this was more pronounced in Zen 3. Interestingly, the L3 bug showed Windows 11 performance close to Windows 10, but with the bug fix, Windows 11 does not work. now consistently offers 5% more performance.

We saw a smaller 3% improvement with the R5 3600, but again Windows 11 was consistently faster by a small margin.

The Ryzen 9 5950X was also faster on Windows 11, which again boosted performance by 5%. The R5 3600 was also faster with the new operating system, although here the difference was only 2%.

Cyberpunk 2077 scores are similar to Rainbow Six Siege and Watch Dogs Legion scores. Windows 11 offers a 5% performance boost with the 5950X and a 2% slower performance boost with the R5 3600.

When it comes to game load times – or rather level load times – just as we found with Intel processors, Windows 11 does nothing to cut them down.

The Ryzen 5 3600 took 11 seconds to boot Windows regardless of the version used. The 5950X was slightly faster with Windows 10, it took 10 seconds to get to the desktop from the BIOS boot selection menu. Windows 11, meanwhile, took 12 seconds as it seemed like it took longer to get past the boot screen. These results are in line with what we found when testing Intel processors.

Storage performance

Finally, we have the results for CrystalDiskMark storage, and when it comes to sequential throughput, Windows 11 does not offer a performance advantage over Windows 10. However, as we found in testing Intel processors, Windows 11 seems to significantly improve the performance of an arbitrary reading and writing.

With the 5950X, we expect an increase in write performance of nearly 70% and an increase in read throughput of over 20%. Again, this is the same as what we saw earlier when testing Intel Core processors.

A big gain was also seen with the Ryzen 5 3600, as write performance was increased by 44% and read performance was increased by 30%. I was not sure if the results from testing Intel processors were accurate, but after seeing the same thing on a completely different platform with different installations of both operating systems, I am now convinced that the data is accurate.

What have we learned

These results were only slightly more interesting than what we found when testing 10th and 11th Gen Intel Core processors, which for the most part showed no performance difference between Windows 11 and 10 for apps and games.

It’s the same with Ryzen, especially when looking at application performance. For gaming, it can range from zero delta as seen in F1 2021 to a slight increase of ~ 2-5% with Windows 11. The Ryzen 9 5950X was often ~ 5% faster when using the new OS, and although this is a trivial difference for gamers this is less trivial for product reviews.

We originally planned to take the hands-on path for our upcoming Intel Alder Lake processor reviews, and that was reusing the Windows 10 data we already collected for Ryzen and Intel processors. After all, we only have a limited amount of time to prepare this content, and the thought of squeezing 12th Gen core testing into a week alongside all the other processors was not fun.

But since Windows 11 consistently delivers the best gaming performance for Ryzen, even if it is in the 2% to 5% range for the games we tested, the more we faced the dilemma of what to do, and the clearer it became. that we’d better throw away all existing data and start over.

So there is something to expect in a few days … all of our content and Alder Lake processors will be tested using the latest version of Windows 11, with all updates applied and VBS disabled. The BIOS of all motherboards will also be updated to the latest versions, and we switched from an RTX 2080 Ti for application testing and an RTX 3090 for gaming tests on a Radeon RX 6900 XT, which was used exclusively for all tests.

There is still a lot of testing to go before you see our first Alder Lake processor review and an avalanche of content awaits you. Be in touch.


Source link

Read More

Leave a Reply

Your email address will not be published. Required fields are marked *

Back to top button