Set up your CS2 sensitivity test
Testing for:
Your mouse DPI
Set in your mouse software (Logitech G HUB, Razer Synapse…). Most mice default to 800. This is needed to convert your result into a CS2 sensitivity value.
Current CS2 sensitivity (optional)
If you know it, the search starts from there instead of a neutral default, which usually saves one or two rounds.
How many cycles?
Each cycle runs the three finder drills once (flicks, follow the ball, strafe & track). More cycles average out day-to-day noise; the final recommendation is the weighted mean across cycles.
Checklist for valid results
- Windows pointer speed at 6/11, the default (Settings → Mouse).
- “Enhance pointer precision” disabled.
- Browser zoom at 100% (Ctrl+0).
- Any in-mouse-driver acceleration off.
- Use the same mouse and pad you play CS2 with.
- In CS2, raw input is always on; leave m_yaw at its 0.022 default.
- On Firefox/Safari (no raw input), run the mouse calibration in Settings once; it corrects for OS pointer scaling.
Your browser does not support pointer lock, so the test cannot run. Please use a recent Chrome, Edge or Firefox.
How the CS2 sensitivity test works
You play short rounds of wide flicks in a 3D scene that reproduces CS2’s camera exactly — the 0.022 yaw constant, the same field of view, raw mouse input with no browser smoothing. A 180° flick here is the same hand movement as in-game.
Between rounds, an adaptive staircase reads whether your first ballistic movements systematically fly past targets (overshoot — sensitivity too high) or stop short (undershoot — too low), then adjusts your sensitivity in shrinking cm/360 steps until the bias disappears. The session ends with a blind A/B round: the recommended value against a decoy, scores deciding, so placebo never gets a vote.
The whole test takes about 10 minutes and outputs your result as in-game sensitivity for your DPI, eDPI and cm/360 — with every trial exportable from your history. The full method, including its limits, is documented on the methodology page.
Want the background first? How the measurement works · eDPI vs cm/360 explained · Why you overshoot flicks
Before you start — common questions
How long does the test take?
About 10 minutes for one full cycle: adaptive flick rounds plus the blind A/B validation. Extra cycles refine the result and can be selected above.
Do I need to download or install anything?
No. The test runs entirely in your browser using pointer lock and raw input — desktop browser plus a mouse is all it needs. Nothing touches CS2 itself, so it is VAC-safe by construction.
Which DPI should I enter?
The DPI your mouse actually uses in CS2 — check your mouse software. The recommendation converts to any DPI afterwards (it is really a cm/360 value), but entering the right one makes the in-game number directly copy-pasteable.
What do I get at the end?
A recommended in-game sensitivity for your DPI with its eDPI and cm/360 equivalents, the measured overshoot/undershoot data behind it, and the blind A/B verdict. Give the value at least a week of matches before judging it.