, how much performance improvement will Windows 11 have compared to Windows 10? With this question, the technology media Techspot conducted actual measurements on several Intel Core processors. This test is not only to collect some benchmark data, but also to compare with the 12th generation Alder Laker CPU on the Windows 11 platform.
On the community and social platforms, the answers to the above questions are not the same. Some claim that Windows11 is much faster, while others believe the opposite is true. To find out, Techspot tested configurations using the flagship Intel Core i9-11900K and the entry-level Core i3-10105F, an upgraded 10th-gen processor with 4 cores and 8 threads. The purpose is to see if there is any performance change using low-end or high-end hardware.
In addition to a clean install of Windows 10 and Windows 11, Techspot also tested a clean install of Windows 10 before upgrading to Windows 11. In the chart below, [Fresh] represents a clean install from a system image, while [Updated] represents a clean install of Windows 10 followed by an upgrade to Windows 11.
and the fourth item configures [Fresh] VBS, which means enabling Virtualization Based Security (VBS) in Windows 11 systems. It was first introduced in Windows 10 as an optional security layer for enterprise PCs. What VBS does is allow Windows 11 to take advantage of the hardware virtualization features found in modern CPUs to isolate secure regions of memory and host security features such as Hypervisor-Enforced Code Integrity (HVCI).
VBS and HVCI prevent hackers from running malicious code on your system with trusted applications and drivers, as it will fail code integrity checks.All of this sounds great on paper, but early testing has shown that it can impact performance in some cases, most notably gaming, where performance is discounted by up to 28% in AMD partial processor usage scenarios.
Users with 1st Gen Ryzen CPUs or 10th Gen Intel CPUs and above will experience this performance regression. For those using newer hardware, the overall performance hit was closer to 5%. Microsoft recommends that OEMs enable VBS and HVCI by default on new PCs, but they allow both to be disabled by default on gaming PCs.
This test platform includes:
Processor: Intel Core i9-11900K and entry-level Core i3-10105F
Hard Disk: Team Group 8TB MP34Q M.2 NVMeSSD
Motherboard: ASUS ROG Strix Z590-E Gaming WiFi, BIOS Version 1007
Memory: Crucial Ballistix DDR4-3200 CL16 Kit
Graphics Card: MSI RTX 3090 Gaming X Trio
As for the test itself, we're looking at performance for apps, games, storage, and load times. All results are based on an average of 3 runs, in some cases we power down the entire system between runs to avoid caching results. Well, let's get into the test results.
Cache and Memory Latency Performance
Before we jump into application benchmarks, let's take a look at cache and memory latency performance. In short, we see no, or little, change in L1, L2, or L3 cache performance on Intel processors. We see a noticeable performance difference when looking at DRAM latency.
For 10105F, Windows 10 and Windows 11 installations provided the same results, but enabling VBS did incur a 7% performance penalty, which can be noticed in memory-sensitive situations like games.
For the more powerful 11900K, we saw that Windows 10 gave the best results, as upgrading to Windows 11 added 5% more latency, although a fresh install of Windows 11 only added 3% more latency. However, after enabling VBS, we again saw a notable increase, this time with a 9% increase compared to the Windows 10 configuration.
Application Benchmark
● Cinebench R23
On Cinebench R23, we found that the 11900K produced the best results on Windows 10, although the difference was small, an increase of about 1% over a freshly installed Windows 11. Enabling VBS has a very small impact on performance, only 2%, with a difference between the slowest and fastest configurations, and the same is true for a Core i3 system.
● Blender Open Data
Blender Open Data results are very similar to Cinebench data. The results for Windows 10 and Windows 11 are basically the same, but with VBS enabled there is a slight performance penalty, this time around 3% compared to what we saw with Windows 10.
-Zip
Continuing to look at the 7-Zip file manager results, we see that in most cases the Core i3 processor is able to provide the same level of performance when using Windows 10 or 11, the Windows 11 configuration with VBS enabled is The only exception was a 3% drop in performance.
11900K dropped even more, and we found that Windows 10 provided the best results. After upgrading from a Windows 10 installation to Windows 11, there was a 3% performance drop, and a reinstall restored some performance. Then, when comparing a fresh Windows 11 install with VBS enabled, we again saw a 3% performance hit.
● Adobe Photoshop 2021
Testing with Adobe Photoshop 2021 showed similar performance across all four tested configurations when using a Core i3 processor, with VBS enabled resulting in a 3% performance hit.
That said, we saw a more pronounced drop in performance on the 11900K, at 7%, when using VBS. However, when comparing new Windows 10 and 11 installations, performance was largely the same.
● Adobe After Effects
In the Adobe After Effects benchmark, both the Core i3 and Core i9 parts provided the same performance using Windows 10 or 11, although this time we saw a noticeable 10% performance drop with VBS enabled.
● Adobe Premiere Pro
The last application benchmark was based on Adobe Premiere Pro, and we saw almost the same performance using Windows 10 or 11. The only exception is the VBS-enabled configuration, as this dropped performance by 12-13%.
Game Benchmarks
Time for some gaming benchmarks. Note that we are not testing dozens of games, but only a few games. In the article, Techspot shows the test results of 4 games.
● "FI 2021"
Using a Core i9-11900K, we saw that fresh installs of Windows 10 and 11 provided the same performance. Updating a Windows 10 installer to 11 isn't that fast, but we're talking about the minuscule difference here.
These differences increase slightly in the Core i3 section, while Windows 10 does offer the best results at 3%.
● Rainbow Six: Siege
The results of Rainbow Six Siege are interesting, as VBS reduces framerate by 11-12%, although this may be exaggerated due to the extremely high framerate. In addition to these results, we saw that Windows 10 and Windows 11 performed identically with either CPU.
● Watch Dogs: Legion
Enabling VBS reduces the performance of Watch Dogs: Legion by 8%, which can be seen when looking at 1% low data.Other than that, fresh Windows 10 and 11 installs provided the same results, with a very small performance hit when simply upgrading 10 to 11.
● "Cyberpunk 2077"
The last game tested is "Cyberpunk 2077", here we see that with VBS installed, the performance is affected by ~8%. Other than that, when comparing Windows 10 and 11, the performance is the same.
Below is the load time for Cyberpunk 2077 using TeamGroup 8TB NVMe SSD, here we see no more than 1 second difference between the various configurations, so Windows 11 won't speed up load times for today's games.
Interestingly, Windows 11 was also slower to load compared to Windows 10, although the difference wasn't noticeable when running the 11900K. The difference was even more pronounced with the Core i3 processor, as Windows 11 took almost 40 percent longer to load, although it was only a 3-second change. For our configuration, Windows 11 did not improve boot times.
Storage performance
Before ending the test, let's take a look at the storage performance of CrystalDiskMark. The sequential performance for reads and writes is the same regardless of the configuration, so there's nothing to report here. However, the results for a random queue depth of 32 are interesting.
Compared to Windows 10, Windows 11's write performance is a huge improvement. To be completely honest, Techspot is no longer familiar with the SSD test, Techspot hasn't run CrystalDiskMark in years, but I can tell you the results are reproducible after multiple system resets, I did fully test before starting the test Filled with brand new solid state drives.
So I believe the 20-26% performance gain we're seeing here on the 11900K is accurate, I'm just not sure why we're seeing such a massive performance gain.There's even a 10% boost from Windows 10 to 11 with a Core i3 processor, although the results are much lower overall. Another notable issue with
is that VBS impairs storage throughput, reducing read and write performance. On the 11900K, we saw a 21% drop in read throughput and a massive 58% drop in write throughput, or to put it another way, Windows 11's random write performance was almost 2.5x faster with VBS disabled.
Summary
For the most part, Windows 11 and Windows 10 with Intel's 10th and 11th Gen Core series processors showed little difference in gaming, app, and storage performance. If you experience a noticeable performance drop after upgrading, it's worth checking to see if VBS is enabled and doing something as we explained in the introduction.
It's also worth noting that for maximum performance, you'd be better off starting over with a fresh install of Windows 11. Aside from a small performance hit after an upgrade install, we also had occasional random blue screen crashes when loading Windows 11, which never happened with a fresh install.
In short, Windows 11 offers no real performance advantage over Windows 10, maybe the only exception is random read/write storage performance, but we won't know if this is limited to us until we do further testing test configuration. At this point, we're eager to see what Windows 11 does to Intel 's upcoming Alder Lake CPUs, and what to say about Ryzen, which recently received a patch that addresses L3 cache performance issues.
