Mouse polling rate
Polling rate has an exact, calculable effect on latency, which makes it one of the few gear specifications you can settle with arithmetic instead of opinion. The arithmetic says get to 1000 Hz and then stop caring.
The arithmetic
Your mouse sends a report at a fixed interval. Your movement happens continuously. On average, a given movement waits half an interval before it is reported.
| Polling rate | Report interval | Mean added delay | Saved vs previous | Verdict |
|---|---|---|---|---|
| 125 Hz | 8.000 ms | 4.000 ms | - | Audibly dated. Fix this first. |
| 250 Hz | 4.000 ms | 2.000 ms | -2.000 ms | Audibly dated. Fix this first. |
| 500 Hz | 2.000 ms | 1.000 ms | -1.000 ms | Fine in practice, one step from optimal. |
| 1000 Hz | 1.000 ms | 0.500 ms | -0.500 ms | The point of diminishing returns for almost everyone. |
| 2000 Hz | 0.500 ms | 0.250 ms | -0.250 ms | Measurable, barely perceivable. |
| 4000 Hz | 0.250 ms | 0.125 ms | -0.125 ms | Sub-0.13 ms saving. CPU cost starts to matter. |
| 8000 Hz | 0.125 ms | 0.063 ms | -0.063 ms | Marginal gain, real CPU overhead on weaker systems. |
Two things jump out of that table. The step from 125 Hz to 1000 Hz removes 3.5 ms, which is more latency than upgrading a 144 Hz display to 240 Hz saves. The step from 1000 Hz to 8000 Hz removes 0.44 ms, which is about an eighth as much, for a great deal more system load.
The cost nobody prints on the box
Every mouse report is a USB interrupt the CPU handles. At 1000 Hz that is a thousand interrupts a second, which is negligible on any modern processor. At 8000 Hz it is eight thousand, and the handling cost is not linear in a friendly direction: interrupt handling can interfere with the timing of the game's own threads.
The symptom is not a lower average framerate so much as worse frame pacing: 1% lows drop, and the game feels less smooth despite the same average. On a system with CPU headroom you will see nothing. On a system already CPU-limited in a busy fight, which describes a lot of competitive setups, you may lose more than you gain.
Test it, do not assume
Run your usual game with a frame time graph visible. Play a busy scenario at 1000 Hz, note your 1% low. Repeat at 8000 Hz. If the 1% low drops, you have your answer, and it is worth more than 0.44 ms.
Wireless, battery and high polling rates
High polling rates on a wireless mouse cost battery in direct proportion: the radio transmits eight times as often at 8000 Hz as at 1000 Hz. A mouse rated for 90 hours at 1000 Hz may manage 20 at 8000 Hz. Manufacturers publish the 1000 Hz figure.
There is also a stability consideration. 8 kHz wireless requires a clean 2.4 GHz environment and a dongle placed close to the mouse. In a crowded radio environment, an 8 kHz link can drop reports where a 1 kHz link would not, and a dropped report is worse than a slightly delayed one.
Verifying your actual polling rate
The rate configured in software is not always the rate delivered, particularly on USB hubs, front-panel ports and older chipsets. Free polling rate test utilities exist that count reports over time and show a live histogram. Two things to check:
- The average should be within a few percent of the configured value.
- The consistency matters more than the average. A rate that averages 1000 Hz but oscillates between 500 and 1400 indicates a port or hub problem. Move to a rear-panel port directly on the motherboard.
The recommendation
- Set 1000 Hz. This is the single largest free latency saving available to most people.
- Go higher only if you have CPU headroom, verified by a frame time test, and a wired connection or a clean wireless environment.
- Never run below 500 Hz on a gaming mouse. There is no upside.
- Verify with a test utility rather than trusting the configuration software.
Related reading
- Input lag budget, where polling sits in the full chain.
- Wireless versus wired, and why the answer changed.
- The click-to-photon chain in detail.
Frequently asked questions
What is mouse polling rate?
How many times per second your mouse reports its position to the computer. At 1000 Hz it reports every millisecond. Between reports, your movement has happened but the computer does not yet know about it, so on average you wait half a report interval before your input is visible.
Is 8000 Hz polling better than 1000 Hz?
By 0.44 ms of mean delay, which is real and very small. It also generates eight times the USB interrupt traffic, which costs measurable CPU time. On a strong system the trade is mildly positive; on a system already limited by CPU, the frame pacing cost can exceed the latency gain. It is not a free upgrade.
Does polling rate affect aim accuracy or just latency?
Mostly latency, with one secondary effect. At low polling rates each report bundles more movement, so a fast flick is described by fewer, larger steps, and games that sample input once per frame can miss detail within the movement. At 1000 Hz and above the sampling is far finer than the game's frame rate, so this effect disappears.
Why does my framerate drop when I enable 8000 Hz?
Each report is a USB interrupt the CPU must service. At 8000 Hz that is 8000 interrupts per second per device, and the handling is not free. On CPU-limited systems this shows up as lower framerate or worse frame pacing. If you see this, drop back to 1000 or 2000 Hz: the frames are worth more.
What polling rate do professionals use?
Overwhelmingly 1000 Hz, and the reason is stability rather than tradition. Tournament machines are standardised and conservative, and 1000 Hz has no failure modes. Some players use higher rates on their own systems.