Waveform bufferbloat test vs NetSteer — which should you use?
Waveform is the reference bufferbloat test; NetSteer is a free in-browser test plus the OpenWrt app that fixes what it finds. Here is how they compare and which one answers your question. Facts as of 2026-11.
| Waveform | NetSteer (this site) | |
|---|---|---|
| Type | Hosted benchmark (community-run) | In-browser test + OpenWrt accelerator |
| Cost | Donation-backed, free tier limited | Free, unlimited runs |
| Measures | Full-load latency vs your gateway | Idle vs loaded latency, loss, jitter |
| Fix guidance | Links to your ISP's docs | Directs you to SQM/CAKE + NetSteer one-toggle |
| OpenWrt angle | — | SQM + RPS + Fullcone NAT in one LuCI app |
When to use each
Use Waveform when you want a community-trusted, independent number for your ISP line — it is the one every forum thread points to, and its results are comparable across users.
Use NetSteer's test when you want a free, instant, in-browser read and a concrete fix path. It is not a substitute for a hosted benchmark; it is a fast triage that tells you whether bufferbloat is present and points you to the right fix.
Use both. Confirm a Waveform result with NetSteer after enabling SQM/CAKE — the gap between "before" and "after" is your proof that the fix worked.
Fixing what the test found
If either test shows high full-load latency, the fix is the same: enable SQM with the cake plugin and pace it to your measured upstream. On OpenWrt, NetSteer does this in one toggle — see the OpenWrt SQM guide. On pfSense, DD-WRT or a consumer router, the bufferbloat guide has the platform-specific steps.
FAQ
Is the Waveform bufferbloat test accurate?
Waveform is a well-known, well-run bufferbloat test (a non-profit project, widely cited in the community). It measures real full-load latency against your gateway. Its main friction: it is not free forever — the free tier is limited and the service is donation-backed. For a quick in-browser read, a WebRTC-based test like NetSteer is instant; for a community-trusted benchmark, Waveform remains the reference.
What is the difference between a bufferbloat test and a speed test?
A speed test measures throughput (Mbps). A bufferbloat test measures latency under load — how many milliseconds your line adds when it is saturated. Two lines with the same speed can feel completely different because one buffers and the other does not. Both matter; bufferbloat is what makes games and calls lag during a download.
How do I interpret a Waveform result?
Look at the "latency" value in the result chart: a few milliseconds means a well-managed queue (CAKE/SQM active or a fast link). Values in the hundreds of milliseconds are classic bufferbloat — your router is holding a huge queue and every packet pays for it. The fix is the same either way: enable SQM with cake and pace it to your real upstream.