If you want an accurate test:
1. Draw two parallel lines 1ft away from each other. Make sure they are parallel.
2. Position your mouse in its neutral position where there is no arcing. If your mouse is not in its neutral position, then the test will be invalid with inaccuracies.
3. Encompass a square around it to ensure straight movement. Line up either the left or right side of the square and match it with one of the parallel lines.
Apply a straight edge under or over the square to ensure a straight distance is being moved.
4. Launch a game and make a 90 degree with two square'd sides. Make sure the square'd sides you are using are parallel to the sides of your monitor.
If everything is calibrated and moved in a straight manner, then there will be no arcs.
5. Move a 360 and land on the reference point you made. Look at where it has landed based on your controlled setup.
6. Use http://www.funender.com/quake/mouse/index.html
and configure your sensitivity and dpi until it reaches 12.0 inches/360. The link and its calculations are only applicable for games that use the 0.022 yaw and pitch scales for mouse sensitivity, such as Source Engine or the ID Tech engine.
This is a way to test your mouse dpi on your current mousepad for its real dpi value with 99% accuracy. This is also a way to test if a mouse has acceleration. Oh, this is also a great way to configure your cm or inch per 360 on a new setup or a new game.
Acceleration amount is:
(change in base distance) / (base distance)
|change in base distance| = Moved amount for 360 > or < than - (minus) the base distance.
So if base distance is 12 inches and moved distance at speed x1 (lower) is 12 inches and at speed x2 (higher) is 11 inches, then the positive acceleration amount is 1/12 or 8.3%. If you get more trials with a ratio of more acceleration for speed moved, then you can calculate it exponentially. Ex: 11 - 12 = |-1| = 1; so 1/12
It is negative acceleration if the distance is higher than the base distance for the 360.
P.S. WMinput vs Raw Input has some minor distance differences.
If you do not want to do it so scientifically, then you can always just line up a thick stack of legoblocks as a straight edge and just line your mouse with the edges of a parallel line. This will not provide accurate results because of arcing and displacement variances because of arcing and non-neutrality. This will get you a rough estimate though. A better rough estimate would be your ability to be robotic enough to move from Line A to || Line B without arcing.Edited by Nilizum - 7/26/14 at 2:37am