How to Use AS SSD Benchmark: Complete Beginner Tutorial

AS SSD Benchmark is a small, freeware Windows tool by Alexej Schepeljanski that measures how fast a solid-state drive reads and writes data. The final version is 2.0.7316.34247, built in January 2020. It does not need to be installed, it does not use the operating system cache for its synthetic tests, and it shows its results in four clear rows plus a total score. This tutorial walks you through the complete workflow: getting the file safely, extracting it correctly, choosing the right drive, running the full test set, reading the four result rows, and saving your result so you can compare it later.

If you have never run the tool before, do not worry about changing any settings on your first run. The defaults are sensible for most drives: a 1 GB test file, all synthetic tests selected, and results shown in MB/s. You can change the test size and the display unit later once you understand what each row means. For a guided first run with extra preparation tips, see our step-by-step guide to running your first benchmark.

Step 1: Download and extract the portable package

Start from the download page, which links to the file served by the official site, alex-is.de. The download is a small ZIP archive. Save it to a folder you can find again, such as Downloads, and check the file size and hash listed on that page if you want to confirm the file is intact before you open it.

Right-click the ZIP and choose Extract All, then pick a plain folder such as Desktop\AS-SSD or C:\Tools\AS-SSD. Inside you should see the AS SSD Benchmark executable together with a language folder named en-US. Keep those items together in the same folder. The program is portable, so there is no installer and no setup wizard. If you move or copy the tool later, copy the whole folder, not just the executable, otherwise the interface may start in German. That portable layout is explained in detail in our guide to using AS SSD Benchmark as a portable tool.

Step 2: Start the program and check the top information

Double-click the executable. No installation happens; the main window opens directly. If Windows asks for confirmation, allow the program to run. At the top of the window you will see a drop-down list of drives, the drive model and firmware, the controller or driver name, the partition alignment status, and the total capacity of the selected drive.

Read that top area before you press Start. Confirm the drive letter and model match the SSD you actually want to test. It is easy to leave an external USB drive or a secondary hard drive selected by mistake. Also note the alignment and controller lines: green OK indicators are what you want to see. If something there looks wrong, finish the run first and then investigate, because the benchmark numbers alone will not tell you whether alignment or controller mode is the cause.

Step 3: Choose the drive and the test size

Use the drive drop-down at the top left to select the target drive. Only select a drive that has enough free space for the test file. The default test size is 1 GB, and you can choose 1, 3, 5, or 10 GB from the size selector. For a first test, and for most comparisons with results you find online, leave it at 1 GB. Larger files take longer and write more data to the drive, and they are mainly useful when you want to see how a drive behaves during longer, sustained writes.

Make sure the checkboxes for the synthetic tests are selected if you want the standard score. The full set includes sequential, 4K, and 4K-64Thrd tests for both reading and writing, plus access time. The separate Copy-Benchmark and Compression-Benchmark are found under the Tools menu and are not part of the main Start button run. Leaving them for later is normal.

Step 4: Run the full test set

Click Start and leave the computer alone while the tests run. The program creates a temporary test file on the selected drive, runs six synthetic tests, and fills in the read column and the write column as each test completes. A typical SSD finishes in a few minutes. A hard drive can take much longer, because the same tests that are quick on flash storage are slow on spinning media, especially the random 4K and access time tests.

Do not use the drive heavily during the run. Opening documents is harmless, but copying large files, installing a game, or running another benchmark at the same time will distort the numbers. If a test seems stuck, give it time first, particularly on a nearly full drive or a hard drive. You can stop a run with the Abort button, but a stopped run does not produce a comparable total score.

Step 5: Read the four result rows

When the run finishes you will see four rows. Seq is sequential read and write speed in MB/s, measured with the 1 GB test file by default. It describes large, continuous transfers such as copying a large video file. 4K is random reading and writing of 4 KB blocks at a queue depth of one. It is closer to the small, scattered access that happens when Windows and programs load many small pieces of data. 4K-64Thrd spreads the 4K work across 64 threads, which shows how the drive handles many simultaneous requests. Acc.time is access time in milliseconds; lower is better here, unlike the speed rows where higher is better.

Below the rows the tool shows a read score, a write score, and a total score. As a simplified guide, the read score is described as Seq Read × 0.1 + 4K Read + 4K-64Thrd Read, with a similar simplified pattern for write. Treat that as a simplified explanation, not a precise formula to reverse-engineer every point. The important lesson is that the total is weighted heavily toward random performance, so a drive with a spectacular sequential number can still receive a modest total if its 4K results are weak. For a fuller walkthrough of every row and score, read how to read AS SSD scores.

You can switch the display between MB/s and IOPS from the View menu. MB/s is easier for most people to compare. IOPS, meaning input/output operations per second, is useful when you are comparing random performance figures quoted by drive makers or other tools. The underlying result is the same; only the unit changes. Note which unit a screenshot uses before you compare it with someone else’s result.

Step 6: Save and export your result

Save the result before you close the program. From the File menu you can take a screenshot of the result window and export the result as an XML file. The screenshot is the easiest format to share and to keep in a folder with your build notes. The XML file stores the numbers in a structured form. When you share a result, include the drive model, capacity, how full the drive was, and the test size, because all of those affect the numbers. Our dedicated guide to exporting AS SSD results covers the menu options and sharing checklist in detail.

What to do with your result

A single run is a snapshot, not a verdict. Run the test two or three times at different moments and compare the pattern rather than obsessing over a small difference in one row. If sequential speed is close to what you expect for your interface but 4K is unusually weak, look at controller mode, alignment, free space, and background activity before you assume the drive is failing. AS SSD Benchmark measures performance only. It does not show SMART health, temperature, or remaining life, so use a health tool alongside it when you are worried about the condition of a drive, not just its speed.

That is the complete basic workflow: extract the whole portable folder, select the correct drive, run at 1 GB, read Seq, 4K, 4K-64Thrd, and access time together, and save a screenshot plus an XML export. Once that feels familiar, explore the Copy-Benchmark and Compression-Benchmark from the Tools menu to see how your drive handles real file patterns and compressible data.