Wired, 2.4 GHz or Bluetooth: which mouse connection is actually fastest?

Not against a dedicated 2.4 GHz receiver, and yes against Bluetooth. The question hides that distinction and the answer falls apart without it. RTINGS says both halves plainly in its click-latency methodology, read on 2026-08-28: wired mice and models with proprietary USB receivers generally offer the lowest latency while Bluetooth mice have the highest, and many gaming mice with proprietary receivers now match or even surpass wired models.

Its published thresholds put numbers on that without pretending to describe any particular mouse. RTINGS calls a good value under 10 ms over a cable, under 10 ms over a receiver, and under 20 ms over Bluetooth. The first two figures are the same and the third is twice them, which is the shape of the whole argument in three numbers. The same page is candid about how little the small differences mean: it says a mouse scoring 8.0 and one scoring 9.7 feel identical in practice.

The manufacturers' own numbers point the same way and should be read with more suspicion. Razer's HyperSpeed page publishes motion latency of 0.36 ms for one of its wireless mice and 0.57 ms for another, against about 0.92 ms for what it calls equivalent competitors, and click latency figures of 0.204 ms and 0.291 ms against 0.807 ms. Those competitor figures are the ones to look at hardest, because even the slow entry in Razer's own comparison is under a millisecond.

Take the marketing discount those numbers deserve. They come from the company selling the product, neither mouse quoted is a product on this site, and no input latency has been measured here. What survives the discount is the order of magnitude, and it is fractions of a millisecond for every wireless mouse in the comparison, including the ones Razer is arguing against.

One caution against reading that as a flat rule. Igor's Lab measured cable and dongle on the same mice and found the gap ranging from a fraction of a millisecond on one model to a few milliseconds on another, so it is a property of the implementation rather than of the connection type. A good dongle matches a cable. Not every dongle is a good one.

Logitech makes the same case and publishes less of it. Its LIGHTSPEED page claims the technology beats most competitors' wired connections, which is a strong claim, and then gives no millisecond figure anywhere on the page. Two vendors, same argument, and only one of them shows the number. Where a company chooses adjectives over figures, treat the adjectives as the marketing they are.

The belief that wireless lags comes from a real era. Early consumer wireless mice were genuinely worse, and enough people were burned that the reputation outlived the hardware by about a decade. It is worth updating.

A dongle is a private radio and Bluetooth is a public one. The dongle that ships in the box talks to that mouse and nothing else, over a protocol the manufacturer wrote and tuned for one job, and it occupies a USB port to do it. Bluetooth uses the radio already in your computer, shared with your headphones and your phone and whatever else is paired, over a general-purpose standard that has to serve all of them.

That difference shows up as report rate more than as raw speed. A dedicated link can be driven as hard as the manufacturer likes, which is how a gaming mouse reaches 1,000 Hz and beyond wirelessly. A shared stack negotiates its interval with the host and settles far lower, which is why Logitech MX Master 4, a Bluetooth-first office mouse, publishes 125 Hz. That number is covered properly in DPI vs polling rate: which mouse number actually matters?

Offering both radios and letting you pick per device is the arrangement that has quietly become normal on anything above the budget shelf. A mouse that does Bluetooth and nothing else is the unusual one now.

Bluetooth also wins on things a dongle cannot touch. It uses no USB port, which matters on a laptop with two of them, and it pairs to several machines at once so one mouse can serve a work laptop and a desktop. Neither has anything to do with speed, and for a lot of desks they decide it.

Between half again and two thirds more runtime, and you do not have to take that from us. Razer Basilisk V3 Pro 35K is published at up to 140 hours on Razer's HyperSpeed dongle and up to 210 hours on Bluetooth. Razer Naga V2 HyperSpeed is published at up to 250 hours on the dongle and up to 400 on Bluetooth. Same mouse, same manufacturer, same page, and the slower radio lasts substantially longer both times.

This is the cleanest evidence on this page precisely because nothing else changes. Comparing a gaming mouse's battery against an office mouse's tells you about two products; comparing one mouse against itself on two radios tells you about the radios.

Zoom out and the split holds, though not along the line you would expect. A runtime quoted in months or days and a runtime quoted in hours are describing two different kinds of mouse: the hours figures, from 95 up to 500, belong to gaming mice driving the radio hard, and the months figures, from 70 days up to two years, to office mice that report far less often. Rechargeable and AA mice turn up on both sides of that line, so what separates them is how hard the radio is being driven rather than what is powering it.

None of which makes the short-runtime mice worse. It makes them a different product. If the thought of a mouse dying mid-week bothers you, one quoting two years on a single AA cell effectively never dies, and that is a real feature rather than a consolation prize.

Because USB 3.0 is noisy in exactly the band your mouse is using, and this is documented rather than folklore. Intel published a white paper on it, hosted by the USB Implementers Forum, which sets out the mechanism: USB 3.0 signalling produces broadband noise ranging from DC to 5 GHz, with emissions landing in the 2.4 to 2.5 GHz band that wireless mice, keyboards and Bluetooth devices operate in.

The paper is specific about where the noise escapes. It radiates from the USB 3.0 connector on the computer, from the connector on the peripheral, or from the cable itself, and it degrades the signal-to-noise ratio at the wireless receiver. The paper's own test results describe lagging mouse response and reduced wireless range, which is exactly the symptom people report and then blame on the mouse.

The fix follows from the mechanism and Intel names it: put the antenna as far as possible from USB 3.0 ports and devices. In practice that means moving the dongle to a port on the other side of the machine, or into a cheap USB extension cable so it sits on the desk near the mouse instead of behind a tower next to an external drive. An older USB 2.0 port, if the machine still has one, is not affected in the same way.

Interference can look like a faulty mouse. Before returning it, try moving the receiver away from USB 3.0 equipment and closer to the mouse. If that changes nothing, continue with the manufacturer’s troubleshooting steps.

Several things survive every argument above. It never runs out. There is no dongle to lose or leave in a hotel, and no radio for anything to interfere with. None of that is glamorous, and together it is the reason plenty of mice still offer a cable, including several that also offer a radio.

There is also a price floor. At the bottom of the range you find wired mice, because a cable costs a manufacturer almost nothing while a radio, a receiver and a battery each cost something. If the budget is the constraint rather than the desk, wired is not a compromise, it is where the value is.

What a cable costs you is drag. A mouse cable catches on the edge of the desk and pulls, which is a real annoyance rather than a spec, and it is the reason usually given for competitive players moving to wireless once it was good enough. Given the figures above, it has been good enough for a while.

For a desktop you sit at, a 2.4 GHz dongle, and treat it as equivalent to a cable unless you find evidence otherwise. Put the dongle somewhere with a clear line to the mouse rather than behind the machine, and you have dealt with the failure mode this connection runs into most often. Battery, firmware and how well a given vendor built the radio are still yours to live with.

For a laptop, or for a desk with more than one computer on it, Bluetooth, and specifically a mouse that offers both so you are not locked out of the dongle later. Losing a USB port on a laptop is a bigger daily cost than a report every eight milliseconds instead of every one.

For a competitive game, a dongle, and check that the mouse publishes a polling rate before you pay a premium for the word wireless. Which products market themselves that way, and what else a gaming desk needs, is covered in A gaming desk setup, two ways. If you would rather compare actual mice than connection types, the setup builder has all three columns of the board side by side.