[Feature] What is the performance difference between Windows 11 and 10? 11 may be dominant in games-PC Watch
Feature
Windows 11
Although Windows 11 started to be provided on October 5th, there is still not enough information about Windows 11, so many people are probably holding off on upgrading to wait a little longer. Apart from the appearance of the UI and operability, I would like to check whether there is a performance difference between Windows 10 and 11 here. Previously, similar verification was performed at the Insider Preview stage, but this time it is a comparison using a proper product version.
Article table of contents(1) Verification with current high-end gaming PC (2) Application performance is even. Storage may increase due to maintenance (3) Game performance is slightly higher on Windows 11 (4) Summary-If you want to improve game performance even a little, upgrade is an option
Verification with a current high-end gaming PC
The gaming PC "G-Tune XN-Z" sold by Mouse Computer was used for this verification. I borrowed the OS when it was equipped with Windows 10 Home. Of course, it is a model that can be upgraded to Windows 11.
Mouse computer gaming PC "G-Tune XN-Z" used for verificationFirst, let me introduce the hardware specs: CPU is Core i7-11700K (8 cores 16 threads), GPU is GeForce RTX 3080 , The chipset is Intel Z590, the memory is DDR4-3200 with 16 GB (dual channel), and the storage is NVMe compatible with 512 GB. It is a high-end gaming PC and the price is 307,780 yen. It's comfortable enough as a PC.
In the verification, Windows Update was first performed on Windows 10 to bring it to the latest state, and after completing a series of measurements, a manual upgrade was performed using the Windows 11 "Installation Assistant". Of course, I also check with the "PC normality check application".
G-Tune XN-Z has TPM 2.0 enabled in the initial state, and passed all items without the need for the user to make BIOS settings.
Next, let's explain the verification method. This time, all tests are performed 10 times, and the average value is indicated by the blue diamond while showing the red circle on the graph. Although it depends on the type, benchmarks have fluctuations, so if you measure multiple times and take the average, the difference will be reduced and the accuracy will be increased, and outliers can be repelled.
And this time we are increasing the number of measurements more than usual reviews. In cases like this one where the hardware configuration is the same and only the OS is different, the score difference is expected to be small in the first place, so the purpose is to increase the accuracy of the score by increasing the number of times.
Also, please note that the graph is not a normal zero start. This is because the difference is so small that it is difficult to grasp with a zero start, so when looking at the graph, please look at both the maximum and minimum values of the graph and the average value of the scores. An arithmetic mean (Average function in Excel) is used to calculate the average value.
The application performance is even. Storage may grow due to maintenance
Let's start with Cinebench R23. The average score for Multi Core is 13,688 for Windows 11 and 13,722 for Windows 10. The score difference is 34 points, and even if you look at the score bias of 10 times, Windows 10 seems to be slightly higher, but the size of the deviation of the score on the Windows 10 side is also worrisome. Whether or not this can be called a decisive difference is questionable.
Cinebench R23's Single Core average score is 1,566 for Windows 11 and 1,565 for Windows 10. There is only a difference of 1 point, which can be said to be an error.
![[Feature] Windows 11 and 10 How much is the performance difference? 11 may be superior for games - PC Watch</p><p> Next, PCMark 10. The overall average score is 7,940 for Windows 11 and 7,905 for Windows 10. There is a difference of 35 points, and the difference in the scores on the Windows 10 side is also worrisome, but it cannot be said that there is a decisive difference.</p><p> Each scenario doesn't show much difference, but if there's one thing that bothers you, it's the Essentials score. Windows 10 has a higher average score, and the difference is even greater if we look at the one low score on the Windows 10 side as an outlier.</p><p> Therefore, I tried to graph the Essentials for each test. The difference is caused by App Start-up, which has an average score difference of 378 between Windows 11 and Windows 10. On the other hand, the others are within plus or minus 30 points, so it can be said to be an error.</p><p> I tried CrystalDiskMark 8.0.4 to see if there is a difference in storage performance from the results of App Start-up. This is not 10 times measurement but 1 time (However, CrystalDiskMark itself is a specification that adopts the maximum value during 5 times measurement).</p><p> Looking at the results of CrystalDiskMark's C drive, most of them are even on the same level of error, but Windows 11 is inferior to Windows 10 in sequential read (Q8T1) and random 4K read (Q1T1), while random 4K write It beat Windows 10 at the same time (Q32T1) and (Q1T1).</p><p> As the name suggests, PCMark 10's App Start-up is a test that affects leads, so it seems that the score has been affected. However, it is difficult to determine whether this is the effect of the driver or the effect of the upgrade (backup is created and puts pressure on the storage). So let's take a look at the measurement value of the D drive.</p><p> The result of the D drive was about 200MB/s better for Windows 10 in sequential read (Q1T1), but otherwise they were almost equal. Judging from this, the suspicion is stronger that the previous App Start-up was caused by storage pressure due to upgrades.</p><p> If the transfer speed is close to the same with a clean install or deleting the backup even after the upgrade, Windows 11 and 10 Essentials are the same or Windows 11 is slightly ahead of the other two scenarios. there is a possibility.</p><h2 id=](https://website-google-hk.oss-cn-hongkong.aliyuncs.com/drawing/article_results_7/2022/3/3/7677d057ed179475c4e95efd80be2c76_1.jpeg)
Game performance is slightly higher on Windows 11
From here on, it's a 3D test. First, 3DMark. I made a comparison between Time Spy and Fire Strike.
First of all, Time Spy's Overall score was about 100 points higher for Windows 11. As you can see from the graph, it seems that Windows 11 clearly has a higher score without overlapping the score band of the 10 tests.
Now, if you look at Time Spy's Graphics test and CPU test, you can see that there is a clear difference mainly between the Graphics test. The difference is 135 points, which is reasonably large.
On the other hand, Fire Strike's overall average score is 217 points higher for Windows 10. However, it cannot be said that it is an absolute difference because the original score is large.
In addition, the Graphics test shows the same trend as the previous Time Spy, with Windows 11 taking the lead. Physics and Combined tests are within margin of error. You can think of it as having little impact.
The FINAL FANTASY XV WINDOWS EDITION benchmark results are almost even as shown below.
Assassin's Creed: Valhalla has a difference of about 3.2 fps in average fps, and Windows 11 is higher. Even at the low 1% fps, Windows 11 tends to be higher.
Watch Dogs Legion, which was subsequently tested, also had an average fps of 2.8 fps, and even at a low 1% fps, Windows 11 was higher by 1.8 fps. These built-in benchmarks truncate decimals and are less accurate. However, it seems that Windows 11 is at least equivalent, or about 1 to 3 fps is better.
Tom Clancy's Rainbow Six Siege, which is lightweight and has a high frame rate, was about 1.4 fps higher on Windows 11. However, there is also a blur width, so I think it's actually almost even.
Summary - If you want to improve game performance even a little, upgrading is an ant
So far, we have verified the performance of Windows 11 and Windows 10 product versions on the same hardware. In my memory, I think it was about Windows Vista that the OS was upgraded and the operation felt heavy. The bottom line is that it's about the same, or, as the benchmark scores show, the gaming performance is slightly better, and the storage may be slightly lower (due to the upgrade approach).
Regarding gaming performance, DirectStorage is also involved, as described in the related article below. Windows 11 may be able to use this function with full performance, so the point of Windows 11 > Windows 10 will become clearer.
Related articleDirectStorage to speed up game loading, works on 10 but Windows 11 is faster
If you upgrade, you have a grace period of 10 days to return to Windows 10, so you can try Windows 11 and check if there are any problems before considering full-scale operation.
![[Excel] How to paste images such as photos and diagrams](https://website-google-hk.oss-cn-hongkong.aliyuncs.com/drawing/article_results_9/2022/3/9/72539ecbf7413c05e4465b39ca06e8e0_0.jpeg "[Excel] How to paste images such as photos and diagrams")
