A followup to this article, titled "Best of Tactical Gaming: Infiltration", was posted on August 19th. Once you've finished with this article, make sure to go check it out. If you were linked to this article from a forum, feel free to post about the new article to help spread the word. There will be a link to the new article on the third page of this one.
The main objectives of this article are to increase awareness of unique or exceptional elements that have been successfully integrated into past FPS games (with an emphasis on the tactical/realistic side of things), highlight and praise implementations of "standard" features that were done particularly well in others, and give me a platform from which I can provide commentary on the various features and options relevant to the future of the genre that I find myself so very interested in.
Some of the areas I will be talking about have very clear-cut winners, and they will be cited prominently. In other areas, things are not so clear, and so a discussion will be made between the various games and their implementations. Some things I've written have yet to be implemented, and while much of that is relegated to the second-to-last 'chapter', it does show up occasionaly prior to that.
I've tried to cover as much material as possible in this piece, but despite my best efforts, I was unable to include everything that I had originally set out to have. Still, there is a massive amount of content here, and I'm very much pleased with how it has come out. The bulk of this article deals with the most basic elements of combat - shooting, moving, and interacting with the environment. With that being said, I will likely revisit this topic in the future to address some of the things that were not possible here.
On the media side of things, I've used a combination of real-world and gaming photography, screenshots, and video to illustrate this piece. Some of the real-world stuff is mine; much of it isn't. However, all of the in-game content (be they screenshots or gameplay footage, but excluding the two VBS1 clips) is of my creation.
I would highly recommend that people use the links available to download the hires versions of the in-game footage and view those when possible instead of the embedded Google Video movies, as the quality in them is worlds better. There is a very good possibility that you'll eventually stop getting sound if you play enough of the embedded videos - if this happens, simply refresh the page and it'll be fixed. If you load the hires versions to begin with, you won't have to worry about that.
Before we get started, I'd like to spend a few moments to establish my general opinions on gaming, as far as they relate to this article.
Anyone who knows me knows that realism is where my gaming passion lies. Operation Flashpoint and Raven Shield are the epitomes of gaming for me, and while the Rainbow Six series is for all intents and purposes dead (thanks to Lockdown, and the decision to appeal to a console audience and abandon the things that made the series so great), Flashpoint is thriving and will be a staple of my gaming for as long as Bohemia Interactive (the developers) continue to exist.
Flashpoint's sequels, Armed Assault and the as-yet-unnamed "Game2", are the Holy Grails of tactical combined-arms wargaming in my eyes. Raven Shield is the king of close-quarters battle, on the other hand. They will be referred to frequently in this article, and it's for good reason - if the best elements of Flashpoint and Raven Shield could be combined, you'd have a game of such strength that it would be nigh impossible to compete with it on the field of realistic simulated combat.
With all of that being said, it's time to get started, and I can think of no better way to kick-off an article like this than with some real-world footage of CQB/MOUT training.
This video is from an Australian unit, and was taken in the early stage of their training, so there are quite a few mistakes made - many of which may not be visible to the average gamer - but all in all it serves as an excellent example of what real-world CQB looks like.
Sit back, relax, and enjoy the show. If we're lucky, this what we'll be steadily moving towards for the future of first-person tactical/wargaming.
|I. Movement and Views
||VII. Visual and Aural Effects
|III. Reloading and Failures
||VIII. Interaction with the Environment
|IV. Weapon Handling
|V. Damage Model, Wounding, and Close Calls
||X. Future Things
|XI. In Closing...
The importance of a player having the ability to move in ways that closely replicate the real world, and have as much flexibility and possibility as in reality, cannot be underestimated. Tactical games should strive to make it so that there is rarely if ever a moment where a player is faced with a situation where his real-world self would be able to do something elementary that his in-game avatar cannot. This is doubly so for life-threatening situations. Nothing is more frustrating in a game than to be killed because you couldn't do something simple that could have saved your life, due to a game either not implementing an aspect of movement or poorly modeling it.
Games should also strive to make the most of all possible combinations of key presses. If a game has a walk key in addition to a sprint one, and the default movement is a third option still, most if not all things the player is able to do should be influenced by what movement modifier (sprint, jog, walk) is in effect at the time they're toggled. Hitting your "prone" key while sprinting should send you into a dive, whereas pressing your "stance up" key while slowly creeping forward should cause you to smoothly rise into an upright position. This is simple enough in concept, and while the actual implementation is no doubt much more difficult to successfully pull off, the reward from getting it right is immense and should not be discounted.
Let's look at some of the noteworthy ways with which past games have handled the issue of realistic player movement.
Fluid Stance and Freelook are two things that you will likely see much more of in future tactical games. They both have exciting potential to be used with NaturalPoint's TrackIR system, and I firmly believe that TrackIR-supporting combat simulations of the future will make people look at tactical gaming in a new light thanks to features like these.
Freelook is simply the ability for a player to look around independent of where they're aiming or moving. Operation Flashpoint has the most robust implementation of this, with the player being able to look all around himself during all manner of movements without forcing his entire body or weapon aim to change. Thus, you can be running in one direction while looking in another - just as most people would be easily able to do in reality. The following video will show off Operation Flashpoint's "true first person" (detailed in the next section) as well as the freelook ability.
"Fluid stance", as shown in Raven Shield, allows you to raise/lower your posture and lean left/right to varying degrees via the mouse. Hold the "fluid stance" key and move the mouse up, and you'll rise, move it down, you'll lower, move it left and you'll lean left, right and you'll lean right. Thanks to this, you can expose a very small amount of your body to fire, and generally take great advantage of cover in ways that other games would never allow you to do. In most games, you have prone, crouch, and standing postures - and that's basically it. If there's a piece of cover that's just too tall to shoot over when crouched, but standing would leave you too exposed, you're basically out of luck. In Raven Shield, with the fluid stance, you can adjust your stance to perfectly match whatever cover you happen to be behind.
See this video for an illustration of how the Fluid Stance feature works in RvS.
Now for the downside. Due to the fluid stance command utilizing the mouse for adjustments, you're unable to aim while changing your posture. Up until recently, because of that fact, I discounted this feature, as it was impractical to use effectively in actual heated combat situations. Why has that changed now, you ask?
If you're familiar with the flight simulation scene, you'll no doubt be familiar with the Track IR system, and what it entails for such games. NaturalPoint, the makers of the devices, have practically revolutionized the flight sim genre with their TrackIR's. A player can now use his head movements, scaled to be as big or as small as he wants, to control his in-game cockpit view. A player can tilt their head back to look up through the top of the canopy, or turn it left to look over their left shoulder, or even move slightly closer to the screen to move their view forward in the cockpit. It's something that really must be seen to be believed, so take a moment to download and view these two demo videos from the NaturalPoint website.
If you'd like to know more, you can find their site here: http://www.naturalpoint.com/trackir/
To get back on the topic of fluid stance, and how this all ties together...
Imagine ducking in and out from cover, in a totally natural way, while exchanging fire with someone else who's doing exactly the same thing thanks to their TIR. Imagine always being able to peek up just enough over a piece of cover to fire, without being restricted to the preset "crouch" or "standing" postures. Imagine totally fluid leaning - choose exactly how far out from cover you want to lean. All of this is possible with Raven Shield's fluid stance feature, but it isn't tied into a control method that makes it practical. If you could take a TrackIR and tie it into this system as well as the freelook available in Flashpoint, you'd revolutionize the way such games are experienced. The first FPS game we're likely to see with any serious level of TrackIR support is Armed Assault, which should be out in Q3 2006. After that, who knows?
The sky's the limit.
Update, November 30th 2006: Armed Assault has officially nailed this feature. I've made a video that shows it off quite nicely - you can find the Google Video version here, or the full-res download here.
What "true first person" means is that the in-game 'camera' is placed in the head of your character's in-game model, such that you see exactly what you should see, to the extent that you can look down to see your arms, legs, weapon, et cetera, and your character's animations and the movement of your view corresponds precisely to what your character appears to be doing. There is no separate set of animations being played only for you - what you see yourself doing is what everyone else sees you doing. True first person is without question the preferred way to model any serious tactical or wargame. Prime examples of it can be found in Operation Flashpoint, Raven Shield, and the upcoming Brothers in Arms 3: Hell's Highway.
One interesting side effect of true first-person is an increased feeling of depth when looking around. In most games, the camera spins on an axis and you get a very stable view of the world when shifting your view around. In a true first-person game, the camera is in the eyes of the player model, which are generally forward, and thus when looking around, turning, et cetera, the world subtly shifts in a fashion that replicates real-world head movements. Your eyes move in a 3d space, and thus your surroundings shift ever so slightly due to that movement. It's a very subtle effect, but it has an immense impact on immersion.
When it comes to climbing ladders, Raven Shield is the game to imitate. For mantling, it's Call of Duty 2. Both games implement these features better than any competitors. Raven Shield has your character holster or sling his weapon before ascending a ladder, the climbing is at a believable pace with accurate animations, and there's even a nice look left/look right first-person view animation built in when you get to the top of a ladder. Descending a ladder is done via the speedy method - no reason to take your time going down it when you have gravity on your side, is there?
See this video for a demonstration of RvS ladder climbing/descending, from first and third person perspectives.
Mantling in Call of Duty 2 is done in an interesting way. If an obstacle is short (below knee height), pressing the jump key will not do a full jump but will instead go just high enough to get over the obstacle. If it's at knee-height, the player will step over it while keeping his weapon ready. If it gets much higher, the player will briefly put away his weapon, mantle over the obstacle, and then bring his weapon back out. Most anything in the game world can be mantled over or onto - you can get out of trenches with it, vault over walls, or climb through windows. The animations, sounds, and camera movements attached to these moves are very well done, giving the player a great sense of weight and gravity. All in all it's a solid and immensely useful feature, and the only game that looks like it might surpass it anytime soon is Brothers in Arms 3, which has the bonus of "true first-person" to give the effect an even higher level of immersion.
See this video for Call of Duty 2 mantling.
The only kind of mantling that hasn't really been covered in first-person shooters is the kind you see in the Thief series. In it, a wall could be as tall as you, yet you'd still be able to climb your way up onto it. While not necessarily easy for your average soldier to do due to encumbrance, it would be nice to address this possibility in some fashion. Assisted-mantling, perhaps, or just make the process a time-consuming one which would leave the player somewhat vulnerable while doing it.
The prone position is one of the easiest to get incorrect in a game. All too often there is no real difference between prone and standing, aside from the obvious fact that your view is lower (thus reducing your target profile), and your accuracy greater. The player typically can rotate at will, with no restrictions, making it possible to spin a complete circle with a flick of the mouse. Because of this lack of realism, the player is able to assume the prone position and utilize the benefits it entails (which are almost always modeled) without having to concern himself with the repercussions and drawbacks from such a position.
Raven Shield is one of the games that models more than just that basic, unrealistic setup. The original Ghost Recon also introduced restrictions on prone, but Raven Shield is more recent and does the job better, so I'll use it for the example.
In Raven Shield, a player is restricted to a small arc of movement once he's gone prone. This movement results in his aim shifting, but not his entire body as is the case in most games. For him to move his body, he must use the strafe left/right keys to shift around. You can see this illustrated in the following movie. This system is an excellent one to replicate in future games - the benefits of the position are retained, while drawbacks are implemented in accordance to how things work in reality.
While Raven Shield gets things right once you're already on the ground, the process of getting there is relatively slow. There is no real diving to prone modeled. For that, we turn to Red Orchestra. In Red Orchestra, the process of diving to the prone is automatic if the player is moving forward when the key is pressed. There is a solid feeling of weight, and it's responsive and fluid enough that one can use it in situations where other game implementations would get one killed. For instance, diving into a shell hole to take cover from a machinegun firing at you - there's no time to spare, you need to go from "standing, sprinting" to "on the deck" as rapidly as possible to avoid being ventilated. RO allows for this nicely, and it should be a feature for any serious combat simulation.
The speed at which you assume the prone position should depend upon your movement - if at a sprint or a run, it should be a dive. If moving at a normal rate, or a jog, you should go prone hastily but not outright dive. If walking or creeping, it should be a smoother, slower transition. You could tie this into noise as well - a dive is noisy, a hasty transition less so, and the smooth, slow transition would be the stealthiest of all. The speed and smoothness of the transition would also play an effect on short-term accuracy, such that after a dive the player might not instantly have the steadiest sight picture, with him being able to acquire his sight picture faster the less rough the transition is.
Once in the prone, the player should be able to choose the speed at which he wants to move. Operation Flashpoint has this, as does America's Army. In Flashpoint's version (which I consider to be superior), you can either move at a normal rate or do a hasty crawl which causes you to tire faster. The most logical way to flesh this out would be to add a third movement speed, achieved in a similar fashion to the dive transitions. The third speed would be the slow prone crawl - stealthy, slow, and smooth, and toggled by holding the "walk" key while moving. That would be the kind of crawl that a sniper might use to move into position - less visible movement, less noise, less chance of detection overall.
The final thing to include is rolling. For whatever reason, a player needs to be able to roll while prone. The act itself is speedier than simply shuffling sideways, while being balanced out in a realistic fashion by disorienting the player (which the view effect should do - true first-person would come in handy here) and making it impossible to fire during a roll. Like everything else, there could be three roll types. Hold your "sprint" key and double-tap your strafe left/right key, for instance, and you would do a very rough, very hasty roll, the kind where you would try to clear as much ground as possible as quickly as possible (say, to avoid a grenade). Double-tap your strafe key while holding no movement-modifier key (sprint/walk) and you'd do a normal roll. More control, less disorientation, faster recovery. Hold your walk key before starting the roll and it'll be a slower, quieter, and smoother roll. America's Army makes a good attempt at modeling this, with the roll distance varying based on whether you're in "slow" or "fast" movement mode. "Slow" will roll you a short distance, whereas "fast" will do a dramatic roll that carries you further.
America's Army, overall, has a solid rolling implementation, and there's a mod for Flashpoint that allows for it - while it's not animated all that well, it gets the job done. There's a lot of improvement possible with rolling in the future, and properly integrating "true first person" with it would be a nice start.
The following clip shows a first-person view of the America's Army rolling - it's good, but it would be better if the view was disrupted more during the act of rolling. A player shouldn't be able to see things clearly during the roll , only after he's stopped and realigned his view. Note in this AAO video that the player's view is far too stable during the roll.
Stamina has best been modeled in Operation Flashpoint, for two primary reasons. The game allows a player to "sprint" for an unlimited period of time, with the catch being that they lose speed over time, such that they end up in a run instead of a sprint after a short while. In addition to that, the player begins to breath heavily, and his weapon stability is adversely affected until he stops and regains his breath. These two elements put Flashpoint's stamina modeling above all other games, despite how seemingly elementary they are.
Encumbrance can be thought of the real world's way of "balancing" the amount of ammunition, weaponry, and other gear that can be carried by any given soldier. Many games don't do much to model encumbrance, with the most typical examples being games where players will move slightly slower or faster based upon what weapon they have drawn (i.e. Call of Duty 2, or even Counter-Strike) or what kind of armor they're wearing (i.e. Raven Shield). There are no FPS games to my knowledge that implement a really robust encumbrance system, but if we look outside the genre we can find Brigade E5: New Jagged Union sporting an encumbrance system that would fit perfectly in a game like Operation Flashpoint.
In E5, each item has a set, realistic weight, and the player has an in-depth inventory system (see the section on Inventory/Carrying Capacity later on for more on this) that allows for him to carry a realistic combat load. The catch is that, like in real life, his movement and agility will be influenced by the weight of the gear he's carrying. Consider a typical loadout consisting of an armor vest with ceramic SAPI plates, a rucksack, extra ammo, load-bearing vest stuffed with magazines and grenades, a helmet, rifle, sidearm, extra magazines for the sidearm, a medical kit, and possibly an extra weapon of some sort (like an anti-tank rocket), and you can see how a heavily-outfitted soldier just might not be the fastest sprinter in the world, or the most manueverable. However, when you consider that any items could be left behind to increase agility and speed and potentially reduce the noise signature of the soldier, you start to see how a robust encumbrance system could really add to gameplay. It works exceptionally well in E5, and it would transfer over to a game like Flashpoint quite nicely. Armed Assault is said to be revamping how the inventory system works, so hopefully they'll reproduce something similar to E5's system.
Short of having a "fluid stance" feature, Vietcong's "rising in ironsight" is the next best thing. It's a simple enough concept - when not in ironsight mode, your character crouches or lies lower to best utilize cover and concealment. When it comes time to go into ironsight mode, you rise up a bit so that you can shoot over whatever cover you might have been hiding behind. In this way, you can take cover and only expose yourself when you're ready to fire, without having to raise your stance too high which would result in potentially over-exposing yourself.
Here's a video of this feature in action, from both the prone and crouched postures.
To prevent players from utilizing unrealistic movement techniques like constant rapid side-to-side strafing to avoid bullets, it's important that a game model a player's inertia and momentum. People do not go from "stationary" to "full-sprint" at a moment's notice, neither are they able to to zig-zag strafe while sprinting or even while standing still. The speed of a player should dictate his ability to change direction or utilize different movement types - if at a sprint, a player should be unable to strafe left/right, and should instead only be able to turn via physically turning his body to point in a different direction. Likewise, there should be momentum carried while moving, which should influence how long it takes for a person to come to a stop.
Red Orchestra is one game that has implemented a solid inertia/momentum system in it, and though it is by no means perfect, it is a nice start. More games should invest time in getting the momentum of players modeled properly - the payoff comes in the form of ever more and more realistic movement techniques being used in combat, which in turn results in less "gamey" things like rapid direction changes at a full-sprint and the above-mentioned side-to-side strafing to "dodge" fire.
Leaning is best implemented via the Fluid Stance feature of Raven Shield due to the degree of precision you can attain with your leans. Short of that, there are several games that do leaning well. Call of Duty 2 handles it just fine, as does Red Orchestra (though CoD2's is arguably smoother to work with). Red Orchestra one-up's CoD2 by allowed leaning to happen while prone as well as in the crouched and standing postures. America's Army has a less-than-ideal leaning implementation, due to it being a very small amount of lean, as well as making it impossible to move while leaned without canceling the lean action.
Leaning should also only be possible while stationary or at slow speeds - you should not be able to lean while jogging, running, or sprinting.
Many, many, many games do not model the different speeds of movement possible on slopes of varying grades, or stairs. Because of this, both slopes and stairs become much less of a terrain consideration than they should be. Typical models make it so that a player can race at full-sprint up a set of stairs, with no slowdown whatsoever, and also allow them to sprint up steep slopes without penalty. The only thing that ever seems to happen is that they eventually reach an incline that they simply cannot move up, but until that point they are utterly unimpeded.
When we turn to inclines, there is one game I know of that has truly done it right, and it's Operation Flashpoint. The game has varying speeds possible when moving up varying grades of inclines. The steeper the grade, the slower you go. You can get up small ones at a sprint, moderate ones at a jog, steep ones at a walking pace, and the steepest of all can only be crawled up. It's a great feature, as it gives terrain the ability to impede movement in a fairly realistic manner. The main thing I would change for future iterations is to make it so that moving perpendicular to the direction of slope is slower than normal movement. If you've ever walked on a moderate grassy slope in such a fashion, you'd know that sprinting or any other fast movement is likely to result in a stumble, slip, or otherwise put you on your ass (at best) or rolling down the slope (more likely).
On the subject of stairs, Raven Shield does a good job with them, as is typical of the game's excellent movement model. Stairs slow you down when moving on them, simple as that.
Two nice control touches can be seen in the Call of Duty series and Vietcong. In Call of Duty, a player can map his controls in a variety of ways - he can set his ironsight key to be toggle or hold, and can set similar preferences for the different stances. Crouching can be a toggle or a hold, same with prone, and he can also set stance up/stance down keys. This is a nice configuration option, as it gives players the ability to get the exact control setup they want without being forced to use a system they aren't comfortable with. Vietcong approaches the problem in a slightly different way with how it handles stances. Merely tapping the stance key will toggle it, while holding the key will make it a temporary action until the key is released. Both of these systems are good examples of how an extra bit of thought and effort put into the player's control system can add a great deal to the overall ease-of-use and customizability.
:: Home ::
The TrackIR is a 6DOF (six degrees of freedom) head-tracking device that allows you to control your in-game view via natural head movements, scaled up to requires as much or as little movement as you want. If you'd like to see a video demo of the TrackIR5 in ArmA2, check this out.
I highly recommend looking into getting one of these if you're interested in ArmA2 or flight sims and driving games in general.
You really won't find any other upgrades out that will improve gaming immersion as dramatically as this.