Nothing here is to suggest that I dislike unrealistic games. My personal favorite game is Wolfenstein: Enemy Territory, which is by any definition an “arcade” game. But since I’ve always wanted to try a truly realistic FPS game, but have been disappointed with everything I’ve seen so far, I thought I’d write down my thoughts on the topic.
Nearly every “realistic” shooter has been forced to compromise for the sake of gameplay. But I believe that if a truly realistic game was made, the players would be able to adapt, just as real armies adapt to the different strengths/weaknesses of their enemies. So, here are a few ideas which I believe are mandatory for a truly realistic shooter.
- No HUD, and the issue of weapon damage.
A realistic game should have no heads up display (HUD), or an extremely limited one at most. There is no need for an ammo counter if you know how much you started with. Just as the soldier in a real war keeps track of his ammo, the gamer can as well. However, in real life (RL) when a soldier grabs a fresh clip, he also gets a tactile update of his overall ammo count (he can “feel” that he has 3 magazines left). Since we can’t feel the game world, a magazine counter may seem acceptable. Then again, a soldier in RL would likely lose track of his ammo in a seriously heated battle, and may simply run out and be forced to use a backup weapon or scavenge for one. So, I believe a realistic game should require the player to keep track of his ammunition status by memory, and those with the coolest heads would be able to keep track better than others, and thus be able to make better decisions based on that information.
As for health status, again, I can’t feel how healthy I am in the game, so I may think a HUD item would make sense. On the other hand, in a realistic shooter we probably don’t want to base a life on the common “100 health points”. Rather, we should say that being shot in a particular place has a certain incapacitating effect, and reflect that in our ability to continue fighting when hit. If I am shot in the leg - by any caliber round - I am likely to be only able to limp along from that point forward. No amount of battlefield medical care is going to eliminate that limp in the short term. It may stop me from bleeding to death, but I will now be a much slower soldier for the rest of the round. If I am shot in the head, I am dead. While it is of course possible to survive a headshot, we must make certain generalizations for the sake of the game. A headshot from any bullet must be fatal, unless we’re wearing an armored helmet. In which case, we would still be totally unable to continue fighting due to severe concussion. We would be on the wounded list for the rest of the round. And since a realism game would not have respawning, that would be the same as being dead. Therefore, unless our game has some kind of persistent campaign mode which we continue in as long as we are alive, a headshot should be considered game-ending for the recipient.
This relates to the HUD because the extent of our wounds should be obvious by our ability to move and fight. But when do you finally die? When you recieve a fatal wound, which would be determined by the game developer. Until then, you will simply become more and more incapacitated until you can no longer function, in which case you will likely be killed soon anyway. All this reflects RL as much as possible, and eliminates the need for a “health meter”. You simply fight until you are killed or the round ends. Period.
Another thing to consider is bleeding. When shot, bleeding would of course be a problem. While some games such as Urban Terror (UrT) have implemented this, I’m not sure how important it really is for a realistic game. Of course there is a risk of bleeding to death in RL, but in the limited timeframe of a game round, the more immediate problem of losing the use of that limb would be more important. Also, the UrT system implies that a gunshot wound is simply a problem of bleeding, and when the bleeding is stopped, the problem is gone. This is of course not realistic. I would argue that bleeding and mending of wounds in a game is not as realistic as it may at first seem. It is much more important to reflect the degree of incapacitation that a wound would give the player, than to simply slow them down until they are able to recieve medical attention, then continue as if they were never shot.
- No crosshair.
This is along the same lines as the HUD, and should be obvious. There are no crosshairs floating in front of us in RL. Just as I can shoot from the hip in RL, I can do so in game. But I can’t expect to hit anything with any kind of accuracy. Shooting from the hip should be worthless except as a means of scaring the enemy into keeping their heads down while I sprint from one piece of cover to the next. It should be suppressive, but not effective.
- Iron sights.
These are mandatory. Without a crosshair, we need a way to aim. Iron sights are realistic and therefore desireable. What kind of iron sights is also important. I believe that model-based iron sights are far superior to the shader-based variety. An example of model-based sights would be Call of Duty (CoD). An example of shader-based would be America’s Army (AA). With model-based sights, the actual weapon model is lifted up in front of the line of sight until the weapons sights are aligned. In shader-based sighting, a two dimensional overlay is applied to the screen to simulate the sight picture, often with some kind of artificial blurring effect to simulate depth. Shader-based sights are obviously inferior and should not be used.
An example of where games like CoD went wrong, however, is in the common “zooming” effect which is applied when sights are raised. Anyone who has fired a weapon knows that the world does not magically zoom in when you take aim. Even if you haven’t fired a gun, simply raise your hands in front of your face and see if distant objects get bigger. Of course they don’t, any more than looking through a paper-towel roll works as a telescope. This common game “feature” must not be used in a realistic game. This also relates to the next item…
- Fixed field of view.
Just as I cannot alter my field of view (FOV) in RL, we shouldn’t allow players to adjust theirs in a realistic game. What that locked FOV should be is another matter. On a flat screen, it’s hard to duplicate reality. I can see perhaps 180 degrees of periphery in RL, but the farther I get from my line of sight, the blurrier things become. If this could be simulated in-game, that would be great. If not, I’d suggest the common 90 degrees would be a good starting point.
- Eliminate “tweaking”.
Every game has certain settings which can be personalized for the sake of performance. This is mandatory, but those settings must be kept in check. Players should not be allowed to remove anything from the screen, such as their weapon model. Performance tweaks should be based on the dumbing down of textures and geometry, not on the elimination of items or weather effects. Nothing says that a realistic game has to “look” real. That’s where things have been going lately, but it’s the wrong approach. Realism should relate to gameplay, not textures. Allow players to eliminate eye-candy without eliminating core aspects of the game or gaining an advantage through tweaks.
- Hit detection.
Many games, such as Quake III: Arena (Q3), use a simple bounding box to detect a hit, with the weapon/projectile determining the amount of damage recieved. In a realistic game, we need some way to detect wounds on different body parts. A hitbox system along the lines of Counter-Strike (CS, based on the Half-Life engine, HL) would be sufficient. We need to know if a player is hit in the head, torso, leg, or arm. That would be the minimum desireable level of differentiation. Anything more complex would allow more precise hit detection, at the price of more complex gamecode and likely lower performance. If we were to use an engine with a less complex hit detection system, such as Enemy Territory (ET), we would have to use some kind of randomization to determine how damaging a “bodyshot” was. ET uses a body box and a head box. Since we can’t know where on the body a hit was scored, we’d have to randomize it (chest, leg, or arm). Such a system could work, but would be less realistic. Engine choice would be a decision made on several criteria, so realism may be more or less depending on that choice.
- Movement while aiming.
In RL, it is virtually impossible to maintain any kind of accuracy while moving at anything more than a very slow walk. So must it be in a realistic game. Movement should penalize aim to such a degree that is is only done in a suppressive situation as mentioned in the crosshair section. In order to have any real chance at hitting a man-sized target in RL, one must stop, take aim, control breathing, and fire. If one misses, one must repeat all those steps, since recoil will have thrown the shooter completely off target. Making use of some kind of support would make the process easier. So, kneeling should decrease the instability problems, since it would imply using the knee and elbow as support. Notice I said decrease instability, not increase accuracy. Accuracy should be determined by the ability of the player to compensate for sway - by using mouse control - as well as by the ballistics of the weapon being fired. Simply crouching or proning should not make a given firearm more accurate. It should only reduce the stability problems (represented in-game by “sway”) which must be overcome by the shooter.
- Realistic weapons.
This doesn’t necessarily mean no laser guns. If the game is set in the distant future, what is real and what isn’t must be determined by the developer. But he/she should determine which futuristic weapons would be common on the field of battle. In the case of modern-day or historical games, we can look at reality. Many (most) games tend to include weapon choices which allow a new player to survive among veterans. An example is the panzerfaust in ET. While it is certainly a valuable weapon requiring skill to be used properly, in the hands of a newbie it represents a chance to get some kills. If ET only had submachineguns, a new player would be dominated and overwhelmed. The system works great as intended for ET and most other games. The problem from a realistic standpoint is that the panzerfaust was not used as a means to kill footsoldiers. So a realistic game must take reality into account. In the case of WWII, history tells us that the M1 Garand was the standard-issue firearm for American troops. Therefore, in a realistic WWII game, that must be the most common weapon for the American team, as an example. If such a game were to include tanks and other vehicles, the panzerfaust and bazooka would be obvious choices as a means of disabling those vehicles. But the developer of a realistic game would have to somehow control the ability of players to use those weapons against other footsoldiers. While there must surely be a story out there of someone doing such a thing in WWII, it would have been an absolutely last-ditch act, and would have never been tolerated by superiors under regular conditions - if only for reasons of wastefulness.
- Speed.
While it may seem like I’m picking on ET, I assure you it’s my favorite game and I would never change it. But it is also a perfect example for this item. In ET, a player runs 15 miles per hour (MPH) when sprinting, and 10.9 MPH during normal sustained movement. 15 MPH means a 4 minute mile - a feat which is considered the mark of an exceptional athleet. Sustained “normal” speed in ET is equivalent to a mile of 5:30, which is certainly moving along at a good clip. While a 19 year old, well trained footsoldier who was motivated by being shot at would certainly be in peak condition, it’s not likely he’d be able to continuously maintain a 5 1/2 minute mile while fully loaded with weapons and gear. This is just an exaple to illustrate what is not realistic. While seemingly slow and boring at first, a realistic game should not have a fast movement speed with things like strafe-jumping (which allows even higher speeds to be reached). Therefore, a brisk walk or “trot” should be the common speed of movement, with the ability to run (sprint) for short distances. While running, it must be impossible to take aim. Only suppressive shooting from the hip should be available. Also, a very slow, nearly silent mode of walking should be available, for stealthy movement.
- ?
I could keep going, adding more ideas and elaborating on these. But I’d rather hear what other people have to say. Please do elaborate on or criticize these ideas, and add your own. Do you think a truly realistic shooter could work? Why or why not? What would it take? How far should realism be sacrificed for the sake of fairness? If a game goes too far in that direction, when does it become “arcade”?
And finally, what do you consider to be the most realistic FPS game to date?
Thanks for reading 
