CS2 Mouse Settings: Raw Input, Windows Sens, and Polling Rate Explained

A CS2 mouse config is three levers, not one. How far your crosshair moves for a given hand movement is decided by your Windows pointer speed, by whether CS2 reads the mouse through raw input, and by your in-game sensitivity. Get any one of them wrong and the other two can’t save you.

Among the pros in cs2pedia’s dataset who reported a Windows pointer-speed setting, a clear majority sit at the exact same value - the one that leaves the mouse signal untouched. The other two levers follow the same logic: don’t let the operating system or the game distort the raw motion of your hand.

This guide covers the mechanics of each lever; for the gear itself, see the best mouse for CS2.

Windows Pointer Speed (Windows Sensitivity)

Windows pointer speed is the slider in Control Panel that scales every mouse movement before any application - including CS2 - sees it. It runs from 1 to 11, and the notches are not linear: only the middle notch, 6 of 11, applies no scaling at all.

Every value below 6 throws away mouse counts (your cursor moves less than the sensor reported); every value above 6 duplicates or skips counts to move further. Both directions corrupt the 1:1 relationship between sensor movement and on-screen movement, which is exactly what you don’t want for aim.

So 6/11 is the standard for a simple reason: it’s the only setting that passes the mouse signal through unmodified. Across the pros in cs2pedia’s dataset who reported this field, the overwhelming majority run Windows 6 - it’s close to unanimous among the recorded profiles.

Value

Pros

Share

6

927

95.4%

5

18

1.9%

7

9

0.9%

10

7

0.7%

4

6

0.6%

8

3

0.3%

3

2

0.2%

The companion setting is “Enhance pointer precision,” the Windows mouse-acceleration toggle. Acceleration makes the cursor travel further when you move the mouse faster, so an identical physical flick produces a different in-game angle depending on speed. That destroys consistency, which is the whole point of a fixed sensitivity. Turn it off.

In Windows it lives under Settings → Bluetooth & devices → Mouse → Additional mouse settings → Pointer Options, where you uncheck “Enhance pointer precision.” With Windows at 6 and acceleration off, the operating system is out of the way.

Raw Input

Raw input is the setting that tells CS2 to read the mouse directly from the hardware, bypassing the Windows pointer pipeline entirely - the pointer-speed scaling, the acceleration curve, and any other desktop mouse processing. The console variable is m_rawinput, and with raw input enabled the game ignores Windows mouse settings altogether and uses the sensor’s own counts.

m_rawinput 1

Raw input is the universally recommended standard, and it’s the safe default for every competitive player.

Important caveat on the data: raw input is a console variable, and it is not tracked per-profile in cs2pedia’s dataset - so we deliberately don’t show a usage percentage for it, because we don’t have one. We’re not going to invent a number to dress up a setting we can’t measure.

What we can say is the engineering rationale: raw input removes a layer of OS processing between your hand and the game, and there is no competitive reason to leave it off.

There’s a subtle point worth understanding. With m_rawinput 1 enabled, your Windows pointer-speed setting no longer affects in-game aim, because the game stops reading through that pipeline.

So why does the Windows 6 advice above still matter? Two reasons:

• it keeps your desktop and any non-raw-input application consistent with your in-game feel
• it’s the correct fallback the moment raw input is ever disabled or fails to apply after an update

Set both. They cost nothing and they remove a class of “my aim feels off today” problems.

In-Game Sensitivity and eDPI

In-game sensitivity is the sensitivity value in CS2, and on its own it’s meaningless for comparison - a sensitivity of 2.0 means something completely different on a 400-DPI mouse than on an 800-DPI mouse.

The number that actually compares across players is eDPI: DPI multiplied by in-game sensitivity. DPI (dots per inch) is how many counts the sensor reports per inch of movement; eDPI folds it together with the in-game multiplier into a single figure, which is why it’s the only fair way to compare two players’ sensitivity.

The practical upshot is that 400 DPI and 800 DPI are interchangeable as long as you halve or double the in-game sensitivity to match - 400 × 2.0 and 800 × 1.0 produce identical aim. That’s why the pro field splits between two DPI values rather than scattering: 400 and 800 are the two anchors, with 400 the plurality and 800 close behind among the pros who reported a DPI.

Value

Pros

Share

400

497

50.9%

800

410

42.0%

1600

47

4.8%

3200

5

0.5%

500

4

0.4%

1000

3

0.3%

450

2

0.2%

810

2

0.2%

1200

1

0.1%

1800

1

0.1%

2400

1

0.1%

550

1

0.1%

600

1

0.1%

700

1

0.1%

900

1

0.1%

That’s the mechanics. The full pro eDPI curve - where the median lands, how the distribution spreads, and what “low sens” actually buys you - is its own guide. For the complete distribution and how to find your own target, see CS2 eDPI explained. Pick 400 or 800 DPI based on whichever your mouse sensor was natively calibrated for; modern flagship sensors handle both identically, so there’s no tracking penalty either way.

Zoom Sensitivity

Zoom sensitivity controls how fast your crosshair moves while a scoped weapon - the AWP or the SSG 08 - is zoomed in. It’s governed by zoom_sensitivity_ratio_mouse, a multiplier applied to your normal sensitivity while scoped. A value of 1.0 means the zoomed view tracks at the same angular speed as your hip-fire aim, so your muscle memory carries over unchanged between scoped and unscoped.

zoom_sensitivity_ratio_mouse 1

Most pros leave this at the default 1.0, and unless you main the AWP you should too - changing it means relearning two different sensitivities. Dedicated AWP players sometimes lower it slightly to make scoped flicks more deliberate, but that’s a specialist adjustment. If you’re a rifler, leave it alone and keep one sensitivity in your hands.

Polling Rate

Polling rate is how many times per second your mouse reports its position to the PC; 1000 Hz means a report every 1 millisecond, and it’s the correct default for competitive CS2. Higher tiers - 4000 Hz and 8000 Hz - exist on newer flagship mice and shave the reporting interval further, with real but small returns and a CPU cost worth understanding.

Because polling deserves its own treatment, the full adoption breakdown across the pro field and the 1000 vs 4000 vs 8000 Hz tradeoff live in the dedicated CS2 polling rate guide. If you’re on a stock 1000 Hz mouse, you are not leaving performance on the table.

Quick-Reference Config Block

These are the in-game mouse commands worth setting once and persisting. Drop them in your config so a game update can’t quietly reset them:

m_rawinput 1                     // read the mouse directly from hardware, bypass Windows
zoom_sensitivity_ratio_mouse 1   // scoped aim tracks at the same speed as hip-fire
sensitivity 1.0                  // set to YOUR target - see the eDPI guide

And the two settings that live outside the game:

• Windows pointer speed: 6 of 11. The only notch that applies no scaling. Control Panel → Mouse → Pointer Options.
• “Enhance pointer precision”: off. This is Windows mouse acceleration; uncheck it in the same Pointer Options tab.

To make the in-game commands survive CS2 updates, put them in an autoexec rather than relying on the settings menu - see the CS2 autoexec guide for how to create the file and exec it on launch. To verify raw input took effect, open the developer console and type m_rawinput with no value; the console echoes the current value back.

A Note on the Data

The Windows-sensitivity and DPI figures above are computed at build time from cs2pedia’s dataset of 1,000+ active pro profiles, the same database that powers the rest of the site - not a hand-typed list that goes stale.

Two honesty notes. First, both fields have a recorded denominator smaller than the full dataset: the captions state how many profiles actually reported each setting, and every share is computed against the recorded profiles, never silently against the whole field. Second, raw input is not in the dataset at all, which is why we describe it from engineering rationale instead of a fabricated percentage.

Configs also change between events, so treat this as a snapshot rather than a live-per-event feed.

To go further: compare individual configurations on the player profiles page, browse the aggregate settings stats, dig into the full sensitivity picture in the eDPI guide, settle the polling question in the polling rate guide, and when you’re picking hardware, the best mouse for CS2 ranks what the pro field actually runs.