November 16th, 2012

I’ve been enjoying Halo 4 since its release last week.  The multiplayer feels especially snappy, and I credit 343i for a job well done.

When I say it feels snappy, I mean that the movement in the game seems quick and not sluggish.  The perceived movement speed is tinkered with for each Halo game, and certain games employ faster mechanics than others.


I had assumed the major components in the perception of a Halo game’s speed are:

1) Base Movement Velocity

2) Field of View

3) “Strafe Time”


I define “strafe time” as the amount of video frames that it takes for the player to go from maximum strafing speed in one direction to maximum strafing speed in the opposite direction. In other words, the time it takes for the player to completely reverse direction.

It has been well documented that the Field of View has changed from game to game.  I decided to run a comparison between five Halo games to see if strafing time had been altered as well.

In each test, I stood on a flat surface and directly faced a flat wall with defining marks and/or patterns on it (i.e. bullet holes) and strafed parallel to the wall.



To execute the strafe, I held left on the movement thumbstick until the player was going full speed laterally, and then quickly flipped the thumbstick to the right and held it there until the player reached full speed laterally in the opposite direction.

After importing the captured video into a motion tracking software (Mocha for After Effects), I tracked the wall that I was facing as I strafed, making sure to ignore the stationary HUD elements.



After transferring the tracking data to After Effects, I counted how many frames it took for the motion trackers to go from max. left speed to max. right speed, and therefore, how long it takes for the player to completely reverse direction.



Here are the results for the strafe times (in video frames) for each Halo game:

Halo 1: 12
Halo 2: 12
Halo 3: 12
Reach: 13
Halo 4: 13

(Note: I performed the tests of Halo 1 and 2 on an original Xbox to avoid any potential influence of the Xbox 360′s emulator.)


In comparing five Halo games, the amount of time it takes to completely reverse directions differs by only a single frame. That’s 0.03 seconds of difference, which I think is much too small to be noticed during gameplay.

So why do the games feel so different? What makes the strafes in different Halo games more or less effective if the the strafe itself lasts virtually the same amount of time?


My first thought was that different games might have variable strafe acceleration rates even though the time spent in the strafe is the same.  In other words, one game’s strafe might slow down/speed up quickly and have a lull in the middle (like a “U” shape).  Or the strafe acceleration might be eased in and out (like a sideways “S”).  So I graphed the velocities of the five strafes, expecting to see different shapes which would indicate different accelerations for each game.


To me, the five charts look almost identical.  They all have a “V” shape, indicating a constant strafe acceleration.  In other words, in each strafe, the velocity changes by basically the same amount from frame to frame.

It appears that strafes in these five Halo games have almost an identical linear acceleration.  So what makes different game’s strafes more or less effective?


My best guess is that it’s a combination of differences in a few areas.

First, changing the Field of View between games will affect the perceived speed of all movements (including strafes), as this video demonstrates.


Second, each game likely has different amounts of Aim Assist which will change how magnetized your reticle is to the target.  If there’s a large amount of Aim Assist, it will be much more difficult to strafe in a way that causes players to miss shots at you, adding to the perception that the strafe is slow.

Third, the animation of the player models might make it look like a strafe is more responsive than it truly is.  For example, one game might start a new walking animation immediately when the player changes directions, while a different game might smoothly transition from one animation to another.  The first way would make a strafe appear to be more responsive since the player model would be twitching back and forth instead of gliding, even though the strafe accelerations are almost identical.

Fourth, if the Halo games altered movement speeds drastically, then the value for the acceleration experienced by the player will differ, affecting the strafe.  More research is necessary, but movement speed will be the focus of my next analysis.


There might be other reasons why the strafes seem so different between Halo games, and the ones I just listed are just my best guesses.

But based on the data collected, it is inaccurate to say that the Halo games differ drastically in strafe times.  In fact, it appears that has remained virtually constant since Halo 1.


If you have any comments or questions, please post them here.  Thanks for reading!