I'm doing my master thesis on shooting performance with the Oculus Rift and various interaction methods. The three methods that I'm using are:
1) Aiming with the head, where the crosshair is always in the middle and you press the left mouse button to shoot.
2) Aiming with the mouse, where the crosshair is controlled by moving the mouse, and the left mouse button is used to shoot.
3) Aiming with the gun, where a playstation move with a gun peripheral is used to shoot.
I've chosen the that the virtual shooting gallery will be a neutral one, spawning abstract circular targets with a bullseye, one at a time. The reason I don't use humanoid targets is that they might have too much influence on the results.
My question is, is the use of these targets still relevant enough to measure shooting accuracy?
Or do you think that humanoid targets are absolutely necessary for the results to be relevant?
As an additional question, how important is it that the targets are moving in a horizontal direction? Are static targets enough? Remember, the more factors you add, the more difficult it gets to test them with enough people enough times.
So, what does your experience with FPS games tell you? Does shooting targets at a fireing range or shooting humanoid targets make any significant difference? If so, what is that difference?
Thank you for your input!
Hey there, I know there are several members here who have extensive experience with actual firearms so perhaps they'd be able to go into more details, but here are my limited thoughts.
- I don't think target shape should make much of a difference. It'll utilise pretty much exactly the same skill set unless the skill being tested is threat evaluation using a combination of combatant and non-combatant pictures. Whether you're shooting a circular or humanoid target the main skills involved will be aiming for the centre or bullseye, and tracking (if in motion).
- I think static targets would represent too shallow an investigation. What variables are you currently going to use? Is there any time limit to hit the target? Will you penalise misses? Target movement could be horizontal, vertical, or following a path. Are all targets of equal size? How simplified are your ballistics going to be - instant hit, travel time, gravity, wind...?
- I take it your thesis is investigating shooting skill within the context of computer games? In that case you should try to model (some of) the same kind of skills required in a game, i.e. representing some kind of challenge through moving targets, and a risk/reward mechanic, perhaps in the form of a time limit, or penalising misses, or extra score for a bullseye. You'd need to demonstrate that the control method is applicable to real-world gaming, not just an idealised test situation.
This is kind of interesting. I wonder what exactly it is you're working on. If it's the difference between aiming peripheral devices, I'm almost positive the mouse users will win by a majority.
Overall, I think you're best left to using static, non-humanoid targets. There's no point in using one element of reality (human targets are for serious training, i.e. center mass/head shot) when everything else is simply not realistic. If you eliminate trigger pull, recoil (and sometimes anticipation of recoil), the shockwave (handguns/shotguns), ballistics, misfeeds, and the tremendous report of the blast, well then you're just playing a game. Not to mention the fact that I know some guys that can't even rack the slide of certain pistols (for example, a 9mm Baby Desert Eagle if you're wondering).
Additionally, some people may have an adverse reaction to shooting a target that looks like a person. Once you eliminate the psychology and the reality, then you're left with measuring the speed at which humans can point a cursor at something, which if may be so bold, is probably what you're measuring, correct?
*EDIT* Actually, a moving target poses a good challenge. Here's an example for you. Definitely turn off the sound.