Flint 2: Speeds Far Below the Line You Pay For
Speeds Far Below the Line You Pay For on the Flint 2: what we found on our own £139 unit, the checks that resolve it, and when it is a warranty case rather than a setting. A 900Mbps line delivers 200M...
Daniel Reyes
The symptom
Flint 2, coverage 160m² single unit. Speeds Far Below the Line You Pay For is the fault being chased here, on a unit bought and tested rather than loaned. It is genuinely good at openWrt underneath, so it does whatever you can configure, and the thing that repeatedly cost it marks was support is community-led rather than official.
A 900Mbps line delivers 200Mbps over Wi-Fi, while the provider insists the line is fine. That is the behaviour; what follows is how it tends to present on this particular GL.iNet, and what to rule out first.
Worth knowing: Work through these in order. The first two resolve the majority of cases and neither risks your data.
Check these first
| Check | Why it matters |
|---|---|
| Is it reproducible? | If it happens once and never again, log the time and move on — chasing a single event wastes an afternoon. |
| Has anything changed? | A firmware update, a new app, a new accessory. The cause is almost always the most recent change. |
| Does it happen on a fresh restart? | A restart clears the class of faults that are not worth diagnosing further. |
| Is it under warranty? | The GL.iNet warranty makes some of the steps below unnecessary — check before you start pulling things apart. |
Working through it
- Test wired first, directly into the router, to separate the line from the wireless entirely.
- Test on 5GHz standing beside the router, which is the best case the hardware can produce.
- Check the client, not the router — a Wi-Fi 5 laptop cannot exceed its own ceiling however good the router is.
- Confirm the WAN port is negotiating at full speed; a gigabit port on a 2Gbps line is a hard cap.
- Look for a device saturating the link, using the router traffic view if it has one.
- Switch channel width to 80MHz or 160MHz where the environment is clean enough to support it.
When that is not enough
Wired speeds well below the contracted rate at the master socket are the provider’s problem and worth a fault report.
Preventing a repeat
Always benchmark wired first. Most "slow broadband" complaints are wireless or client limitations.
The usual cause
In our experience this is rarely a single fault. The common pattern is an accumulation: a background process that was fine on its own, a setting changed months ago, and — on the Flint 2 specifically — the stock interface hides most of OpenWrt behind a simplified layer, which narrows the margin before anything else goes wrong. That is why the steps above start with the cheap reversible checks: on a GL.iNet device the expensive diagnosis is almost never the one you needed.
When it is a hardware fault
If the behaviour persists after a factory reset, it is worth raising with GL.iNet under warranty rather than working around it. Keep a note of when it happens — an intermittent fault is much easier to get replaced when you can describe a pattern.
Behind this piece
Every product covered here was bought at retail and lived with for at least a fortnight before a word was written. Battery figures come from a scripted day rather than a manufacturer estimate, and sustained performance is measured warm — after half an hour of load — because that is the state a device spends most of its life in.
Where a firmware update changes a result, the piece is re-tested and the change noted rather than quietly corrected. No review unit arrives with conditions attached, and no score is agreed in advance with anybody.
Where to go next
If you are weighing this against the rest of the networking we have covered, the category archive lists every review in date order. The closest comparisons we have published are Flint 2: The Router Needs Restarting Every Few Days and Flint 2: Video Calls Stutter While Everything Else Is Fine.
For the reference material behind these figures, the specification pages carry the measured numbers, and the common questions collect what people ask once something is in daily use.
