How to compare SQM settings with a bufferbloat test

You changed a router setting called SQM and want to know if it helped, but you do not want to swap computers or rearrange your network. A fair comparison keeps the device, connection, and test conditions the same, then changes only the SQM setting.

SQM, or Smart Queue Management, manages traffic waiting to use your internet connection. The Bufferbloat Test measures how much response delay increases while the connection is busy, and gives a grade from A+ to F.

Compare the settings fairly

1. Open the Bufferbloat Test on the computer you plan to use for every run. Keep the browser and connection type the same throughout the comparison.
2. Set up a quiet baseline. Pause downloads, cloud backups, video streams, and other traffic on your network. Keep the computer in the same place, and leave any VPN setting unchanged for all runs.
3. Run the test with your current SQM setting. When it finishes, note the grade and the reported delay during download and upload load, if shown. Also note the time and the router setting in effect.
4. Change only the SQM setting you want to compare in your router's settings. Do not change the device, Wi-Fi band, cable, browser, or other router options at the same time. Save the original setting so you can restore it.
5. Wait for the router to apply the change, then run the test again under the same conditions. For a more reliable comparison, do two or three runs per setting and compare the middle result rather than relying on one run.

If you are comparing SQM on versus off, label each group clearly. If you are comparing two shaping rates or queue disciplines, write down the exact values or choices. Router menus use different names, so do not assume that a similarly named option works the same way on every model.

What the result means

The grade improves and loaded delay falls

This is a good sign that the new setting is keeping queues shorter when someone is using the connection heavily. Check both download and upload results. A change may help one direction more than the other, especially if the upload rate is the part that fills up first.

The grade changes, but the delay is similar

A small grade difference can reflect normal variation, particularly when the underlying delay is close to a grade boundary. Compare several runs and look at the reported delay as well as the letter grade. A single grade should not decide the result.

Delay is lower, but regular download speed falls

SQM can trade some peak throughput for more consistent response when the connection is busy. That may be worthwhile for calls and games, but a large speed reduction can indicate that the configured rates are too low or that the router cannot keep up. Compare ordinary speed separately, using the same test conditions.

Both settings produce similar results

The connection may not have been busy enough to reveal a difference, or the router may not be applying the setting you changed. It is also possible that your internet connection has little bufferbloat in the tested conditions. A similar result does not by itself prove that SQM is broken.

How to get a clearer result

1. Stop competing traffic first. Pause large transfers and cloud syncing on every device, not just the computer running the test. Another device can fill the same internet connection and distort the result.
2. Keep the connection path fixed. If you start on Wi-Fi, use the same Wi-Fi network and location for every comparison. If you choose Ethernet instead, use that same cable and port throughout. Switching from Wi-Fi to Ethernet changes another variable.
3. Confirm that the router applied the setting. Return to its settings page after saving and check that the selected SQM option and rates remain in place. Some routers require a separate apply or restart step, and the exact process depends on the model.
4. Change one SQM value at a time. If you are adjusting upload and download shaping rates, change one rate, test it, then change the other. If you change multiple settings together, you will not know which one caused the difference.
5. Try shaping rates below your stable connection capacity. When the router lets you set upload and download limits, a starting point around 90 percent of the connection's sustained rate can be useful, but it is not a universal setting. Test small adjustments because the best value depends on the ISP connection and router.
6. Repeat at a similar time if results vary. Internet use elsewhere on the network, or congestion outside your home, can change results. If grades vary by time of day, see why a bufferbloat grade may vary in the evening.

Why the same device matters

A comparison is only useful if the test conditions are close. A different computer may have a different Wi-Fi adapter, browser, VPN, or background traffic. A different location can change wireless signal quality. Those changes can affect the connection even though the router's SQM setting stayed the same.

Keep the computer and its setup fixed, but remember that the test still measures the path from your device through the router to the test server. The Bufferbloat Test exchanges test traffic with the server to measure response delay under download and upload load. It cannot isolate the router from every other part of that path.

If the results are close, avoid making a strong claim from one run. Use repeated comparisons, keep a note of the setting and time, and restore the original router configuration if the change makes your connection less useful.