Nobody chooses between these two standards on a purchase order any more. Wi-Fi 5 access points are largely end of sale across the major vendors, so the question arrives as a refresh question instead: there is an installed base of 802.11ac in the ceiling, users are complaining in the rooms that fill up, and somebody has to decide whether an upgrade is worth the capital and what it drags along with it. Framed that way, the useful comparison is not which standard is faster on a spec sheet — it is what 802.11ax fixes, and what it asks of the network underneath.
The single most misread fact about Wi-Fi 6 is that it is a speed upgrade. It is not, really. One client on an empty channel sees a modest gain from 1024-QAM, and only at short range with a clean signal. What 802.11ax actually introduced is a different way of sharing airtime: OFDMA divides a channel into smaller resource units so an access point can serve several clients inside a single transmit opportunity instead of queueing them one behind another, and uplink MU-MIMO and scheduling extend that to traffic flowing the other way. Those mechanisms do very little for one laptop in a quiet office and a great deal for forty devices in a classroom.
The second thing to know is that the access point is rarely the whole bill. Wi-Fi 6 units commonly want PoE+ rather than plain 802.3af, higher-stream models can ask for PoE++, and a busy AP will saturate a single gigabit uplink long before its radios run out. A refresh that replaces ceiling hardware and leaves the switch, the PoE budget and the uplink alone frequently produces a disappointing result and an argument about whether the standard was oversold.
At a glance
Side by side
| Factor | Wi-Fi 5 (802.11ac) | Wi-Fi 6 (802.11ax) |
|---|---|---|
| Bands | 5 GHz only; the 2.4 GHz radio on a dual-band AP is 802.11n | 2.4 GHz and 5 GHz both run the modern PHY |
| Peak theoretical rate | Around 3.5 Gbps on a Wave 2 4x4 radio at 160 MHz | Around 9.6 Gbps aggregated across all streams |
| Channel access | One client per transmit opportunity | OFDMA splits a channel into resource units so several clients share one |
| Modulation | 256-QAM | 1024-QAM — more bits per symbol, at close range with a clean signal |
| MU-MIMO | Downlink only, up to four users (Wave 2) | Uplink and downlink, up to eight users |
| Overlapping cells | Co-channel interference is handled by backing off | BSS coloring lets a radio ignore a neighboring cell's frames instead of deferring |
| Battery-powered clients | No scheduled wake mechanism | Target Wake Time lets sensors, scanners and tags sleep to a schedule |
| Security baseline | WPA2 typical; WPA3 not part of the certification | WPA3 required for Wi-Fi CERTIFIED 6 |
| What it asks of the switch | 802.3af or at, and a gigabit uplink is usually enough | PoE+ commonly, PoE++ on higher-stream units; multi-gig uplinks on dense deployments |
| Where it leads | End of the line — the family stops here | 6 GHz via Wi-Fi 6E, then Wi-Fi 7 on the same wired groundwork |
Stay on Wi-Fi 5 a while longer when
- The coverage areas are genuinely low-density — small offices, corridors, storage, sparse floors — where contention is not what users are complaining about
- The existing access points are still in vendor support and firmware, so the risk of staying is operational rather than security-driven
- The wired side cannot take the upgrade yet: a Wi-Fi 6 access point behind a plain 802.3af port and a saturated gigabit uplink will underperform the unit it replaced
- The budget cycle can fund one refresh, not two, and waiting lands you on 6E or Wi-Fi 7 hardware instead of an interim generation
Move to Wi-Fi 6 when
- Density is the complaint — lecture halls, classrooms, conference rooms, clinics, warehouses, anywhere the trouble ticket says 'it's fine until the room fills up'
- The estate carries a lot of 2.4 GHz devices, because this is the generation that finally brings the modern PHY to that band rather than leaving it on 802.11n
- Battery-powered endpoints matter and Target Wake Time would extend the service interval on scanners, sensors or tags
- A security baseline requires WPA3, which arrives as part of Wi-Fi 6 certification rather than as an optional extra
- You are already refreshing switches or cabling, so the PoE budget and multi-gig uplinks the access points want can be bought in the same project
Bottom line
Treat this as refresh timing rather than a purchase choice, because Wi-Fi 5 hardware is mostly end of sale and the decision in front of you is when to move and what to land on. The gain from 802.11ax is real but it is a capacity gain, not a speed gain: OFDMA, uplink MU-MIMO, BSS coloring and Target Wake Time all make a cell better at serving many clients at once, which is why the improvement is dramatic in a full room and barely visible in an empty one. Two conditions decide whether you actually see it. The client fleet has to include enough 802.11ax devices for the scheduling to have anything to schedule — legacy clients are still served, but in legacy mode. And the wired side has to supply PoE+ or better and enough uplink to carry what the radios can now move, or the bottleneck simply relocates from the air to the switch port. If the refresh is more than a year out, plan the wired groundwork now and buy 6E or Wi-Fi 7 access points when the money lands, since the switching, PoE and cabling work is the same either way and it is the part that takes the longest to fund.
FAQ
Common questions
- Is Wi-Fi 6 actually faster than Wi-Fi 5?
- For a single client on a quiet channel, only modestly — 1024-QAM packs more bits into each symbol, but only at close range with a strong, clean signal, and the difference is nothing like the gap between the headline numbers. Where Wi-Fi 6 is decisively better is under load. OFDMA lets one transmit opportunity carry data for several clients instead of one, uplink MU-MIMO and scheduling apply the same idea to traffic coming back, and BSS coloring reduces how often a radio defers to a neighboring cell. Those are capacity mechanisms, so the improvement scales with how crowded the room is.
- Do my devices need to be Wi-Fi 6 for any of this to help?
- For the scheduling benefits, yes. OFDMA, uplink MU-MIMO and Target Wake Time are negotiated between the access point and the client, so an 802.11ac laptop connects perfectly well to a Wi-Fi 6 access point but is served in legacy mode and contributes nothing to the efficiency gains. That is not an argument against upgrading early — the infrastructure has a longer life than the endpoints, and the benefit grows as the fleet turns over — but it does mean a Wi-Fi 6 rollout into a fleet of older clients will not show the improvement the datasheet implies on day one.
- Will Wi-Fi 6 access points work on my existing switches?
- They will attach, and that is not the same as working well. Many Wi-Fi 6 units want 802.3at PoE+ rather than plain 802.3af, and higher-stream models can ask for 802.3bt to bring up every radio and every port on the unit — an underpowered access point typically boots in a reduced mode instead of failing outright, which is why the symptom is disappointing performance rather than a dead device. The second constraint is the uplink: a busy access point can move more than a gigabit port will carry. Check the PoE class and the uplink requirement on the specific model before the switch budget is closed.
- Should I skip Wi-Fi 6 and go straight to 6E or Wi-Fi 7?
- Often yes, if the timing allows it, because the wired groundwork is identical and it is the expensive, slow part. Wi-Fi 6E adds the 6 GHz band, which is clean and wide but shorter-range and only usable by 6E-capable clients, and Wi-Fi 7 builds further on the same foundations. The argument for buying Wi-Fi 6 now is cost and coverage in places where 6 GHz would go unused; the argument for waiting is avoiding two refreshes in one hardware lifecycle. Either way, size the PoE and the uplinks for the later generation, since retrofitting that after the access points are hung is the painful sequence.
- Does Wi-Fi 6 help 2.4 GHz devices like barcode scanners and sensors?
- This is one of its underrated wins. 802.11ac is a 5 GHz standard, so on a dual-band Wi-Fi 5 access point the 2.4 GHz radio is running 802.11n — which means the estate of scanners, sensors, printers and building devices that live on 2.4 GHz never saw a generational improvement at all. Wi-Fi 6 brings OFDMA, the improved scheduling and Target Wake Time to 2.4 GHz as well, which matters most in exactly the environments that depend on those devices: warehouses, plants, logistics floors and anywhere with a large fleet of battery-powered endpoints.