“On frame rate and player performance in first person shooter games,” by Kajal T. Claypool and Mark Claypool, 2007

Banter by housefly and me on this AcmeCTF thread: http://www.acmectf.com/forum/index.php/topic,4462.0.html

“I didn’t really get good until I got 1000 frames per second.”  –Housefly

“After taking the weekend (and a small trip to outer space) to think about these things, I have decided I am going to use the PC for map-making and h0ps skills video, and continue to use the iMac running Fruitz of the Dojo Quake 2 at a lower resolution to play Quake 2 at or above 1000 frames per second.  Here is the science backing up my decision.  Full disclosure: the game these researchers tested their hypothesis on was actually… Quake 3 Arena.

The deadline and precision requirements for shooting and
movement in a FPS game are directly correlated with the
effects of frame rate on player performance for the particular
action. In general, the further away an action is from the
origin in the Precision–Deadline plane, the less the impact
that frame rate has on player performance. Thus, as seen in
Fig. 2, FPS Sniper is more sensitive to the frame rate than
is FPS Machine gun, and movement while Running is more
sensitive than movement while Walking.

Precision. Consider a shooting action where the player targets
an opponent moving across the field of view from left to
right, depicted in Fig. 3. With a high precision weapon, for
example a sniper rifle, the player on the left sees the opponent
as the solid outline, with the target circle representing
the precision of the sniper gun the player is shooting (see
Fig. 3a). At 60 fps, when the player aims and shoots the gun
will hit any opponent within the circle. With lower frame rates
the movement, and hence the location, of the opponent is not
accurately relayed to the player. For example, at 7 fps the
player sees the opponent at the solid outline while the opponent
has moved to the right to the dashed outline, resulting in
a miss. However, when the player is shooting with a lower precision
weapon, such as a machine gun, the target circle is
larger (see Fig. 3b). In this case, while lower frame rates still
delay the feedback of opponent position to the player, the
opponent remains within the target area, enabling the player
to score a hit.

This example illustrates our first insight: For a given game
action, the higher the precision the greater the impact of
frame rate on player performance.

Deadline. Consider once again the shooting action, and
in particular the time it takes to sight the opponent in the
cross-hairs, fire the weapon and have the projectile reach
the opponent. If this time (the deadline) is short, such as
for a sniper rifle, the delay in providing feedback to the
player caused by low frame rate can be significant. However,
when the deadline to complete the action is looser, the
delay in providing user feedback is relatively less significant.
For example, consider a movement action where the player
must move along a suspended beam from point A to point
B. If the beam is straight, as in Fig. 4a, then the deadline is
large relative to the player initiated command to move, so
additional delays induced by a low frame rate do not impact
performance much. However, if the beam is twisty, as in
Fig. 4b, then there are many smaller move commands each
with a tight deadline and even a small delay induced by low
frame rates will significantly impede user movement or cause
a fall.

This example illustrates our second insight: For a given
game action, the tighter the deadline the greater the impact
of frame rate on player performance.” -Reefer

“Reefer you sounded so confused and stoned last night by all that talk of framerates that the last thing I imagined you were going to do after that would be to start reading peer-reviewed articles on the subject.” –Housefly

“reefer, you should ask them to 1v1 you on penetrator.” -h0ps

“I don’t smoke no more.  I don’t smoke no less either.” -Reefer