eDPI, Sensitivity and cm/360 Explained

Everybody compares sensitivities, and almost nobody compares them correctly. "I play on 2" means nothing without a DPI. "I play on 800 DPI" means nothing without an in-game value. And even eDPI, the number most communities settled on, is only comparable inside a single game. The measurement that actually transfers is centimetres per 360 degrees. This article explains what each number means, gives you the arithmetic to convert between them, and shows why any recoil compensation has to be tied to that conversion to work at all.

Table of contents

• DPI, CPI and what your mouse reports
• eDPI: the useful shorthand
• cm/360: the measurement that transfers
• The conversion table
• Converting between games
• Why a macro must match your sensitivity
• Choosing a sensitivity and keeping it

DPI, CPI and what your mouse reports

DPI — more correctly CPI, counts per inch — is how many movement units your mouse reports per inch of physical travel. At 800 DPI, moving the mouse one inch sends 800 counts to the computer. The sensor produces these counts; everything after that is software.

Two things people get wrong here. First, a higher DPI is not inherently more precise. Modern sensors are accurate across their range, and above a certain point extra counts land inside the noise rather than adding information. Second, DPI alone says nothing about how fast you turn, because the game multiplies those counts by its own sensitivity value before rotating your camera.

There is one setting that genuinely matters and is worth checking right now: pointer acceleration must be off, in Windows and in the game. Acceleration makes the relationship between physical distance and camera rotation depend on how fast you moved, which destroys both muscle memory and any fixed compensation curve. Enhance Pointer Precision in Windows is an acceleration setting despite the name.

eDPI: the useful shorthand

eDPI is simply DPI multiplied by the in-game sensitivity:

• 800 DPI at sensitivity 2.0 gives an eDPI of 1600.
• 1600 DPI at sensitivity 1.0 gives the same eDPI of 1600.
• 400 DPI at sensitivity 4.0 gives the same again.

All three turn your camera at the same rate for the same physical movement. That is the whole point of eDPI: it collapses two settings into the one number that determines your turn speed.

Its limitation is that it is game-specific. An eDPI of 1600 in Counter-Strike and an eDPI of 1600 in Valorant are completely different speeds, because the two games apply different rotation values per sensitivity unit. Comparing eDPI across games is one of the most common mistakes in sensitivity discussions.

cm/360: the measurement that transfers

Centimetres per 360 is the physical distance you must move the mouse to rotate a full circle. It is measured in the real world, so it is directly comparable between any two games, any two mice and any two setups.

The arithmetic uses one game-specific constant, usually called yaw: the degrees your camera rotates per mouse count at sensitivity 1. Counter-Strike and other Source-derived engines use 0.022. Valorant uses 0.07. With that, the formula is:

• Counts per 360 = 360 / (yaw x sensitivity)
• Inches per 360 = counts per 360 / DPI
• cm per 360 = inches per 360 x 2.54

Worked example, Counter-Strike 2 at 800 DPI and sensitivity 2.0: 360 / (0.022 x 2.0) = 8182 counts. Divided by 800 DPI gives 10.23 inches. Multiplied by 2.54 gives roughly 26 cm per 360.

Because eDPI is DPI times sensitivity, the whole thing collapses to a single division. For Counter-Strike 2 and other 0.022-yaw games, cm/360 is approximately 41564 divided by your eDPI. For Valorant, it is approximately 13063 divided by your eDPI.

The conversion table

Values rounded to one decimal place. The middle column applies to Counter-Strike 2 and other engines using a yaw of 0.022; the right column applies to Valorant.

eDPI	cm/360 at yaw 0.022	cm/360 at yaw 0.07
200	207.8	65.3
280	148.4	46.7
400	103.9	32.7
600	69.3	21.8
800	52.0	16.3
1000	41.6	13.1
1200	34.6	10.9
1600	26.0	8.2
2000	20.8	6.5
2400	17.3	5.4
3200	13.0	4.1

Reading the table sideways is instructive: a Valorant player on 280 eDPI and a Counter-Strike player on roughly 890 eDPI are moving the mouse the same physical distance for the same rotation, despite eDPI numbers that differ by a factor of three.

Converting between games

To move a sensitivity from one game to another, convert through cm/360 rather than through eDPI. Work out your cm/360 in the game you know, then solve for the sensitivity that produces the same distance in the target game.

Between Counter-Strike and Valorant the two yaw values give a clean ratio: 0.07 divided by 0.022 is about 3.18. So a Valorant sensitivity multiplied by 3.18 gives the matching Counter-Strike sensitivity at the same DPI, and dividing goes the other way. A Valorant player on 0.35 is playing the equivalent of about 1.11 in Counter-Strike.

Some caveats that trip people up:

• Field of view changes the feel even at identical cm/360, because the same rotation covers a different fraction of your screen. A matched sensitivity will not feel identical across games with different FOV — it will only be geometrically equivalent.
• Aim-down-sight and scoped sensitivity are separate multipliers in most games, and matching your hip sensitivity does not match your scoped one.
• Not every game exposes its yaw value, and a few apply their own smoothing or acceleration internally. When in doubt, measure empirically: mark a spot, do a full turn, measure the distance.

Why a macro must match your sensitivity

This is where the arithmetic stops being trivia. A spray pattern is stored by the game in degrees. A macro can only send mouse counts. The conversion between the two is exactly the yaw-times-sensitivity factor above, so the counts a compensation profile needs to send are entirely determined by your DPI and in-game sensitivity.

What that means in practice:

• The same profile is wrong on a different sensitivity. Halve your sensitivity and every correction is half as large as it needs to be — the compensation under-pulls by a constant factor.
• The error is proportional, not random. If your rounds consistently land above the target, the scaling factor is too small; if they consistently land below, it is too large. A consistent bias is a configuration problem, not a pattern problem.
• Changing DPI has the same effect as changing sensitivity. Both change counts per degree, so bumping your mouse's DPI button mid-session silently invalidates the profile.
• Acceleration breaks it completely. With acceleration on, the counts-to-degrees relationship is no longer constant, so no fixed profile can be correct.

This is why the NoRecoil app asks for your DPI and in-game sensitivity during setup rather than offering a single generic strength slider: without those two numbers there is no way to convert a pattern in degrees into the movement your specific setup requires. The setup guide covers where to find both values per game, and the FAQ page answers the common follow-up about what to do after you change your sensitivity.

Choosing a sensitivity and keeping it

There is no correct sensitivity, but there are sensible ranges and one rule that matters more than the number itself.

Lower sensitivities — longer cm/360 — spread every correction over more physical travel, so a given hand tremor translates into a smaller angular error. They cost turn speed and demand desk space and an arm-aiming posture. Higher sensitivities turn faster and suit smaller surfaces, at the cost of magnifying every imprecision. Most competitive players across tactical shooters sit somewhere between roughly 20 and 45 cm/360, which is a wide band and should tell you how little the exact figure matters.

The rule that does matter: pick one and stop changing it. Every hour of aim practice is building a mapping between physical distance and rotation. Changing sensitivity resets that mapping. A mediocre sensitivity you have used for six months will beat an optimal one you adopted last week.

If you do change, change once, write down the old and new values, and give it at least a week before judging. Measurable drills — the kind NR Aim is built around — make that judgement far less subjective than going by feel after a bad match.

Frequently asked questions

What DPI should I use?

Anything in the common 400 to 1600 range is fine on a modern sensor. Pick a value, set your in-game sensitivity to reach the cm/360 you want, and leave the DPI alone. Avoid mouse buttons that cycle DPI mid-game.

Is a lower sensitivity objectively better?

No. It trades turn speed for angular precision. The right answer depends on your desk, your grip and your game. Consistency over time matters far more than the specific value.

Does polling rate affect my sensitivity?

No. Polling rate changes how often the mouse reports, not how many counts an inch produces. It affects latency and smoothness, not your cm/360.