How to establish a memory performance baseline

If you want to know whether a RAM upgrade, system change, or new slowdown affected memory performance, you need a result from before the change. A useful baseline is not one perfect score. It is a small set of results recorded under conditions you can repeat.

The Memory Test (RAM) measures read, write, and copy speed, response time, and a score out of 100. Use the steps below to record your computer's normal result, then compare future tests with that baseline.

Run a baseline memory test

1. Open the Memory Test (RAM) on the computer you want to measure. Run it in the browser you normally use, and keep that browser the same for future comparisons.
2. Prepare the computer. Close games, video editors, large downloads, and other demanding work. Leave the computer plugged in if it is a laptop, and wait until startup activity or updates have settled.
3. Run the test once and note the read speed, write speed, copy speed, response time, and score. Also record the date, browser, operating system, and whether the laptop was plugged in.
4. Run it two more times under the same conditions. Give the computer time to return to its usual temperature and activity level between runs rather than starting each test immediately.
5. Save all three sets of results. Use the middle result for each measurement as your baseline, and keep the highest and lowest values to show the normal spread.

A note in a document or spreadsheet is enough. For example, record the three scores and speeds along with whether the computer was plugged in and which browser you used. You can compare later tests with the same record instead of relying on memory.

Understand the results

The three runs are close together

This is a stable baseline for the conditions you recorded. Small differences are normal. When you test again, compare the new results with the range of your three runs, not just with the single highest score.

One run is much slower or faster

A temporary task, background update, heat, or a change in power mode may have affected that run. Repeat the test after the computer is idle. If the results continue to vary, keep the spread in your notes instead of treating one run as the computer's true speed. System load can distort memory test results, so note anything demanding that was open.

Read, write, and copy speeds do not match

They measure different kinds of memory work, so they do not need to be identical. Compare each measurement with its own earlier result. A higher score does not necessarily mean every kind of memory operation improved.

The score differs from another computer's result

That comparison may not be useful. The browser, processor, operating system, memory configuration, and current system activity can all affect a browser-based test. Your own results, collected with the same setup, are the most useful comparison.

Keep future comparisons fair

1. Match the conditions. Use the same computer, browser, power source, and approximate workload as the baseline. If you are checking a specific change, change one thing at a time.
2. Let background work finish. Pause large downloads and wait for updates, file copies, or heavy apps to finish before testing. A busy system can make a memory benchmark less repeatable.
3. Check the power mode. On Windows 11, review Settings › System › Power & battery › Power mode. On macOS, review System Settings › Battery. Power options vary by computer, so record the mode you used and keep it consistent.
4. Repeat the same number of runs. Use three runs again, with pauses between them. If the computer becomes unusually hot or the results keep falling, stop and let it cool before testing further.
5. Compare the individual measurements. Look at read, write, copy, and response time separately, then check the score. A change in one result does not automatically mean all memory performance changed.

If you have just changed a memory setting or installed an upgrade, save a new set of results after the computer has restarted and settled. Keep the original baseline too. That lets you see the before-and-after difference without losing the reference point.

Know what this baseline can tell you

A browser test measures memory behavior while the test runs in your browser. It is useful for comparing the same computer over time, but it does not certify that RAM is healthy or explain every cause of a slow computer. It also cannot prove the exact amount of installed RAM: browsers receive only a rounded hint about memory size. For another perspective on one reported measure, see why RAM copy speed differs from read speed.

The test runs on your device, so the best baseline is one you can reproduce with the same browser and everyday setup.