The author of the NTDEV blog has compared the speeds of different Windows 10 updates on various points: from the time the system was installed and its boot / reboot to the time of the Windows Defender scan, the launch of applications and the like. He did it back in February (the last build he tested was 19041), but his post only became popular in the media now. As part of his experiment, the author used the Hyper-V hypervisor, 4 GB of RAM, 4 processor cores and storage with a fixed amount of 32 GB for all tests. Each time the systems were installed cleanly.
%%TRASH_CHAPTERS%%
1. Installation time
Compared to build 10240, update 19041 is slightly slower to install. However, between these builds, the installation time has repeatedly increased (since the Creators Update) and decreased. Most likely, as a result, the installation began to take a little longer than initially, due to the increase in the weight of the system itself – with updates Windows 10 began to take up more disk space.
2. Boot and reboot times
Compared to other indicators, the system boot time has increased significantly – instead of 13 seconds, the system began to boot in 34 seconds.

The reboot time has also increased, and the graph curve for this indicator is very similar to the download speed curve.

The author of the blog believes that the reason for the slowdown in loading lies in new security technologies that have appeared with updates, as well as in new services.
3. Launching Win32 Applications
The following programs were used for this test: Windows Explorer, winver, Notepad, Internet Explorer, Paint, Registry Editor, msconfig, msinfo32, Wordpad, Control Panel, and Task Manager. When comparing the highest and lowest values, the difference is almost 3 times. Most likely, Win32 applications will continue to load more slowly.

4. Launching UWP apps
Microsoft has relied on the Universal Windows Platform (UWP) on several occasions, a platform that allows you to build universal apps for all forks of Windows 10 (Windows 10 IoT and Windows 10 Mobile in the past). However, Microsoft is now ignoring its own idea, both in terms of the number of UWP-applications, and in terms of their “optimization”.

The test used Microsoft Edge, Settings, Calculator, Calendar, Maps, Movies & TV, Groove Music, People, Microsoft Store, and Voice Recorder.
5. Windows Search
The standard Windows search was used to find msinfo32 on the C drive. Unlike most other cases, the results here are constantly changing: either the system searches much faster, or starts searching much slower.

6. CDI Benchmark
For this test, the author resorted to using ADeltaX REGDI32, which creates 9,900 GDI descriptors (just under the maximum threshold of 10,000). This utility tests the time it takes to render all handles.

7. Stress test “Explorer”
As part of this experiment, the author wrote a batch script that opens Explorer windows until GDI runs out. This benchmark checks not only the rendering speed of GDI, but also the opening times of Win32 applications.

8. Windows Defender Quick Scan
Relatively speaking, with each update, Windows Defender becomes more integrated with the system, receiving new improvements. The result shows that the built-in antivirus has become not only smarter, but also significantly faster.

9. Performance of the I / O subsystem
To test this metric, the enthusiast used diskspd, a free and open source utility from Microsoft. He applied the command diskspd -c1G -d300 -r -w40 -o32 -t8 -b64k -Sh -L, and the graph shows the number of read and written commands.

10. Shutting down the system
This indicator accelerated several times, but generally remained at the same level.


Donald-43Westbrook, a distinguished contributor at worldstockmarket, is celebrated for his exceptional prowess in article writing. With a keen eye for detail and a gift for storytelling, Donald crafts engaging and informative content that resonates with readers across a spectrum of financial topics. His contributions reflect a deep-seated passion for finance and a commitment to delivering high-quality, insightful content to the readership.