This browser is no longer supported.
Upgrade to Microsoft Edge to take advantage of the latest features, security updates, and technical support.
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(128, 8%, 22%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EA%3C/text%3E%3C/svg%3E)
As you see above, the other performance is about the same, but in RND 4K Q32 T16 the speed has dropped significantly. I know that because the two software versions are different and the condition of the computer is different, it is normal for the performance to fluctuate, but this performance also dropped too much, and the performance remained stable low after many tests at different times.
Locked Question. This question was migrated from the Microsoft Support Community. You can vote on whether it's helpful, but you can't add comments or replies or follow the question.
watch out for difference in using crystalmark and asssd.
asssd give me nearly correct drive speed figures in win w11. quite a shock, as crystal mark was showing terrible write speeds.
crystalmark was better several years ago as asssd had a bug. so it was my goto benchmark, go figure....
so try a series of different benchmarks and take the average.
AMD Ryzen 5 5600X, MSI MAG B550M Mortar (BIOS 7C94v19 / AMD ComboAm4v2PI 1.2.0.5), 16GB RAM 3600 C16 GSkill, NVMe Samsung 980Pro 1T (on 4xPCIe4), NVIDIA RTX3070 Founders Edition (on 16xPCIe4).
Windows 11 build 22000.556, all drivers and Samsung firmware (5B2QGXA7) up to date on last version.
Note: my Samsung 980Pro is software encrypted with BitLocker
Windows 11:
Same hardware but booted on an USB HDD with Windows 10:
Edit: Also with Samsung Magician, Random Write: ~200 KIOPS (W11) / ~400 KIOPS (W10)
Good news everyone, this is now fixed by KB5012643, specifically: "Reduces the overhead of resource contention in high input/output operations per second (IOPS) scenarios that have many threads contending on a single file."
Good news everyone, this is now fixed by KB5012643, specifically: "Reduces the overhead of resource contention in high input/output operations per second (IOPS) scenarios that have many threads contending on a single file."
Sigh -- I wish people wouldn't assume it is "fixed". Sure my IOPS on random write are a littler higher with the latest update (incrementally so), but it is not where most benchmarks were under Windows 10. And that has really been the main issue: random write (everything else being equal) is not showing as good results on Win 11 vs. Win 10.
What was you Random write prior to this update? Look at your history under Performance Benchmark in Samsung Magician (you can look at 3 months, 6 months and 1 year back). What were Random Writes previously - and specifically -- most recently prior to this latest update? What OS are you running and are you running the OS on the SSD being tested?
My Random Write was previously 163K IOPS, I have been tracking this since I upgraded to then-prerelease Win11, and across several HW upgrades (motherboard, CPU, SSD and DRAM). To be more precise I had a record of almost 800K IOPS on Win10, so 575K is not technically a full fix of the regression. However, this is a synthetic microbenchmark. And this large change after this particular fix shows that this benchmark does precisely what the KB article mentions: high contention of concurrent writes to a single file. This is not a very realistic I/O pattern even for (well-optimized) database servers, one should use few threads but many I/O queues. Unfortunately I don't have a history with better benchmarks.
Anyway at this point, considering the large changes that Windows 11's I/O subsystem had to enable multiple features from virtualization to DirectStorage, I believe this is as good as we'll ever get for this kind of microbenchmark of traditional I/O APIs.