Search the Community
Showing results for tags 'input'.
Found 1 result
My goal here is to discuss, "What if the tried and true two-part manner of representing recoil is actually wrong, and has spoiled us after all this time? What if the efforts EFT is making in regard to recoil are actually more appropriate, but that after all this years of playing FPS's we're just simply not used to it?" Interested in looking into this further? Then come on in, let's get to it! ----- Traditionally, FPS's have relied upon two mechanics to represent aiming and recoil. The first we'll look at, is what I'm going to call the "bloom" or "dispersion" area. This can be visualized as a shape that widens and contracts over time, representing where the player's bullet might land for any given shot. Each "click" of the mouse generally widens the shape (or, for automatics, the longer the mouse is held down). This "bloom rate" is balanced against the shape's rate of contraction. The more the user waits between each click of his mouse, the more time to circle has to "shrink back down" to it's initial size. The balancing act that goes on here can be imagined as a sort of "user input race". How quickly the shape vs how quickly it returns to normal can help give the "feeling" of more or less recoil control. Notice too how, in the illustrations above, playing with different "shapes" can help give the sense that weapons may buck or kick in different directions, with different patterns, or signatures. The second mechanic often used is what I'll refer to as cursor displacement. This is the game moving the player's mouse position. This effect is also used primarily as a means to simulate the weapon violently "jerking" upward after breaking the shot. Games move the player's mouse upward and possibly slightly to one side or another and then, either leave the mouse at this new position or add a slight "drift" back downward to represent a sort of "cushion" effect. In each of these cases, though, it is important to note that the cursor position is directly influenced. Here's an illustration: Combining these two mechanics together, then, results in something similar to this: Note how the cursor's position is moved away from its original location and that the weapon's intended point of impact tracks the cursor position throughout. Now, Escape From Tarkov caught on to something very interesting here. And that is, the role of the user's cursor position. In traditional FPS games, the user's cursor position is used as a point of feedback as to where the weapon is currently pointed (as we discussed above). But is this the only way of interpreting it? What if we take a step back and instead view the cursor as representing the player's point of aim and allow the weapon to act independently? Something wholly different begins to emerge! For example, in real life, if I draw down on a target with my pistol, every time I break a shot, I'm trying to hit the same area. But, due to a multitude of factors we'll simply refer to here as "recoil", whether or not I'm able to accurately place those rounds on target is a different matter. Generally, the faster I shoot, the closer I start to approach the boundary of my capabilities. Shoot too quickly, and the gun will simply start to "get ahead of me", so to speak. As I get better at shooting, this threshold can be pushed further and further, allowing me to retain control at various rates of fire. It's important to distinguish here that during rapid fire, my intended target does not jerk ever-upward. The firearm is what is moving, not my "focus" or intent. In the same way, EFT seems to interpret the user's cursor position to represent where the user wants his avatar to aim, not necessarily where the weapon's muzzle is currently pointing. Just as in my real life example above, the player's character is always TRYING to hit a certain target, and the weapon then, kind of takes on a life of its own. To accomplish this, EFT introduces a third element of recoil control that I'll refer to here as the "weapon aim point". Now, we have three dynamics in play, rather than the usual two: 1) Bloom 2) Cursor position 3) Weapon aim point. How might that look? Note how the user's cursor position remains stable, and how the bloom spreads in relation to the weapon's aim point, which itself moves off target, then resetles close to the user's cursor position. In this way, EFT's recoil "patterns" (for lack of a better word) are more akin to a spray of water from a wale's blow hole. The water shoots up, spreads out, then falls back downward. The "patterns" are not static and inline with the cursor, but "float" relative the weapon's dynamic point of aim. In my opinion, this is brilliant and is a much more nuanced and accurate way to represent what goes on and what it feels like to shoot an automatic weapon. This leaves the developers a wider range of ways to represent both how different weapons feel and to also the different levels of skill in managing recoil. A tighter bloom shape may represent the user controlling the weapon better (think of trying to stabilize a rattling jackhammer). How far the weapon aim point shifts away from the user's cursor position may relate to how stable the shooter's platform is in relation to the weapon's kick. The rate at which the weapon's aim point returns to the user's cursor position is also important here. For example, a common complaint of EFT's current implementation is that semi auto fire is discouraged. One way BSG might approach this is to return the cursor position on a nonlinear scale that coincides with the weapon's automatic rate of fire. Break one shot, WHAM the muzzle jerks offline then begins to resettle. The start of the weapon's muzzle coming back down could be quite fast, but then it could taper off and slow down as it nears the cursor's position. This way, if a shooter rips of two single shots in rapid succession, unless he "times it" to his skill level, each successive jerk offline of the muzzle will only further add to the distance between the weapon's aim point and his intended target. Just as in real life. The main point here being that, perhaps a weapon should respond more or less consistently to the rate of fire, not necessarily to the fire mode. This way, if a weapon fires at 900 rpm, and the avatar's skill level is currently near, say 750-800 rpm, he should be able to single fire at 700 rpm and OVERALL be more accurate than full auto. I say "OVERALL" because again, just as he will quickly "get used to" a stream of fully automatic fire, he will also quickly "get used to" a stream of rapid semi automatic fire. The main differences being perhaps the magnitude and rate of muzzle aim change, as well as the size of the "bloom" shapes. Thoughts? Improvements? Thanks for reading, interested in hearing your feedback! ----- Post-script: In real life, if I'm taking multiple shots in rapid succession, I don't necessarily "fully re-aim" before each and every shot. There's no time for that. Instead, I trust in my training and focus on firing "within my cadence". The moment the front sight seems to be drifting back on target, BOOM, I take another shot, trusting that the literal act of breaking an intentional shot will "do the rest" and bring it on target. Once these underlying mechanics are mastered, I explore instinctive shooting at close to the same rate of fire, but that's a wholly separate topic. Case and point: Watch a shooter take multiple shots at a moving target in real life that's outside of, say, 30 meters. Very rarely will his shot times even come CLOSE to what we're used to see in a first person shooter. It's not BLAM BLAM BLAM BLAM BLAM!! It's BLAM *slight pause* BLAM *slight pause* BLAM! This is what I mean when I say we've possibly been "spoiled" over all these years by having accurate fire be overly controlled by nothing more than a few small mouse movements. In real life, it's much more difficult to place accurate shots. And in this way, I feel EFT is on the right path (again, not yet fully arrived, but in my book, definitely getting there).