Brant asked me to take a break from my usual gear porn and serious musings on self-defense to share some thoughts on how weapons and tactics are represented in games. After all, this is primarily a gaming-oriented blog :).
Unlike most of my colleagues here at GrogNews, I am, first and foremost, a console gamer, not a PC gamer or a tabletop gamer. The relative merits of electronic vs. tabletop gaming and of the various electronic gaming platforms can be debated for hours (or megabytes, as the case may be), but suffice to say that there are some great strategic and tactical games on consoles. At the strategic level, Sid Meier's Civilization: Revolution comes to mind, while Tom Clancy's Ghost Recon and Rainbow Six series and Operation Flashpoint: Dragon Rising are among the best tactical games on console platforms. In terms of popularity, Call of Duty: Modern Warfare 2 (MW2) is at the top of the heap right now, with Battlefield: Bad Company 2 and Medal of Honor following in its footsteps with hopes of topping its gameplay and technical accomplishments and at least approaching its commercial success.
In today's posting, I'll touch on some features that I would like to see in a tactical-action game on the PC or next-gen console platforms. Some games do some, but not all of these things, right, while some of the other ideas have not yet (to the best of my knowledge) hit the market.
Distinct Models for Weapons and Ammunition
Terminal effects (e.g., damage inflicted) are fundamentally a property of the cartridge, not the weapon from which it was fired. An M855/SS109 NATO 5.56x45mm, 62-grain penetrator cartridge fired from an M4 is fundamentally the same as one fired from an M16, an ACR, an Tavor TAR-21, an F2000, a FAMAS, or any other weapon chambered for 5.56 NATO. There can be differences in terminal effects based on different muzzle velocities from different barrel lengths, but these effects only come into play at medium ranges (beyond 100 meters).
Yes, MW2, I'm looking at you. Why is someone shot by an M16A4 or a TAR-21 dramatically more wounded than someone shot by an M4?
Magazine Compatibility
While we're on the topic of ammunition, a small improvement that a lot of games (yes, MW2, I'm looking at you again) could make is to recognize compatibility of magazines between various weapons. If my squad-mate is armed with an M16A4 and I'm armed with an M4A1, we can use each other's magazines. In fact, the M4, M16, ACR, TAR-21, F2000, FAMAS, AUG, G36, L85, M249 SAW, L86 LSW, and many other weapons can use the STANAG magazine most commonly associated with the M4/M16/AR-15.
Penetration
This has really improved over the past few years, but some games still don't get it right. The other night, I was playing M.A.G. while my wife worked on her laptop on the sofa next to me. I was playing a sniper role, popping 1-2 enemies at a time and then displacing left or right along a long fence line. I was really racking up the kills when my wife, who is not anywhere near as into weapons and tactics as I am, looks up at the screen, hears the enemy's return fire hitting the fence, and innocently asks, "What's the fence made out of?" "Wood," I responded and then realized what she was pointing out: the game had lulled me into treating a wooden fence as hard cover.
Movement and Positions
For a game to be in any way realistic, it has to model the real biomechanics of a how trained shooter moves, the positions that they take, and the effects of movement and position on accuracy. Ideally, games should also model fatigue and the effect of pulse and respiration rate on the stability of your body as a shooting platform.
I can't tell you how many times I've been killed in on-line games by some 12-year-old bunny-hopping on the approach and circle-strafing for the kill. I've completed some fairly high-speed training (especially for a civilian), but I must have missed the session where they covered bunny-hopping and circle-stafing as individual movement techniques.
Ricochets
Richochets are a real issue with small-arms fire, especially in urban environments that can be used to one's advantage or can present a risk (either from friendly or enemy fire). For example, a skilled machine gunner can "skip" rounds along a wall to hit otherwise protected targets. FM 3-06 discusses this and other factors that designers of realistic tactical action games would be well-advised to study.
Reaction Times
I haven't done any controlled tests on this topic, but it seems that games vary in how they treat reaction times of friendly NPCs and AI opponents. I know that SWAT 3: Close Quarters Battle and SWAT 4, from the now-defunct Sierra Online, did it fairly well, because they actually had a difficulty setting to control the reaction times of AI opponents.
Wikipedia offers a decent overview of the topic and a column in a recent issue of SWAT Magazine also discussed the topic of reaction times with an emphasis on its tactical implications. At a minimum, I think a game should include some delay for the reaction time of the character observing the stimulus, making a decision (do I shoot or not? do I need to re-load? do I need to turn left or right to face a threat? etc.), and then actually executing the action (squeezing the trigger, executing a tactical or speed re-load, turning, moving, etc.).
Go/no-go reactions are also faster than choice reactions. If the AI is already covering a door that it expects the player to come through, it just needs to decide to fire when he sees the player come through the door (like readying an action in D&D). If, on the other hand, the AI is just standing around smoking and joking one moment and in the next moment he's covered in his buddy's blood and brain matter from a suppressed 7.62 round to the head, he has to go through a lot of choices: drop prone? dash for cover? unsling my weapon and return fire (where did the shot come from)? freeze in panic? etc.
There are other areas to consider (like the teamwork between a sniper and an observer that it takes to make a one-shot kill against a fleeting target at 800 yards or a better representation of terminal effects, including the effects of body armor), but perhaps those would push the game too far toward a simulation for realism fanatics like me.
In conclusion, I'd love to hear what you, our readers, think. What makes a good tactical-action game? What games do it really well and, conversely, what are some of the worst representations of combat in games that claim to be realistic?
[Edited to fix a typo.]
By: Guardian
I can't speak to the realism of guns in these games as I haven't shot most of them.
ReplyDeleteHowever, regarding your comment on reaction times, the Rainbow Six series, specifically Raven Shield, addressed this by making each side's uniforms almost identical. Friendly fire was not well received by most of the community and you needed to have a *really* good reason for blue on blue.
The result was often a clash in dimly lit areas where proper target ID took a moment: was that a tinge of red on the uniform, or are my eyes playing tricks on me?
Itchy trigger fingers usually got kicked from servers rather quickly, but being too slow meant you ended up dead.
The balance made for some really fun dynamics once you learned the game.
Guardian,
ReplyDeleteIMHO guns, bullets, ballistics, explosions, and Other Bad Stuff in almost every FPS shooter will be subject to a number of simplifications and abstractions. Some of these are made in the sake of game play; others are made because they are simply too complex to model. Programming ricochets requires a fairly complex geometry and arithmetic processing model - running and firing 72 rounds from a PPsh against the side of a Panzer IV means lots of real-time math;
Modelling penetration requires that all in game objects will need to have hardness values associated with them, and the bullets will need have a penetration value that is subject to range and angle. In the real world, it is conceivable that a bullet could pass through three or four (or more) sheets of drywall in a house - unless the bullet hit one of the framing studs. So - how does one model that? Anything you do will must needs be an abstraction. Having said that, SWAT 3 modelling penetration reasonably well, although it was at times frustrating to see team mates get hit from who-knows-where. Interestingly, I think SWAT4 dumped penetration from the game.
Other real-life situations that never see light-of-day in FPS games are things like weapon realiability (namely - jammning and fouling). Although it would be totally reasonable to model "the dead man's click" in an FPS game, (it happened in real life to one of the SAS guys in the Embassy raid in London), only a few serious, verisimilitude oriented FPS grogs would want to see this feature implemented (although interestingly it has been a part of flight sims since Red Baron).
I could go on, but other recommendations I would make for good FPS experiences are:
Sniper Elite (ballistics, breathing, optics)
Red Orchestra/Darkest Hour (weapon modelling, movement)
Rogue Spear (weapon modelling, ballistics, movement)
Operation Flashpoint (weapon modelling).
Regards,
Jack Nastyface
Great article!
ReplyDeleteSome additions to the Jacks post above:
Operation Flashpoint (DR and ARMA1/2 at least) all have breathing, ballistics and movement covered.
IIRC there also are weapon jams modelled in the game.
What still is left to get into a FPS is the ricochets and shrapnel. "Bad Company" at least have collapsible buildings and environments featured. I am however not sure on how the damager around this is modelled.
Thank you all for the feedback!
ReplyDeleteRegarding target ID, I had forgotten about Raven Shield but that is a very nice touch. Target ID is certainly part of my OODA cycle when I play these games and similar uniforms are a nice, subtle but effective way to encourage it.
As a software developer, I think there is sufficient processing power and storage in today's PCs and game consoles to model each bullet and apply various properties (such as hardness) to materials. There will always be abstractions and simplications in the underlying models and there may be some trade-offs with eye-candy, but I'd like to see developers push more in these directions, preferably with scalable realism (which Forza Motorsport 3, in another genre, does very well) so they don't go broke pleasing the 10% or so of us who would appreciate the realism while turning off the ever-popular "casual gamers."
Personally, I would love to see the "dead man's click" in an FPS. I suppose its because one of the pearls of wisdom I've picked up from real-world training is that "the most terrifying sound in the world is a click when you expected a bang."
I don't think I gave Operation Flashpoint the credit that it deserved. Dragon Rising (at least on consoles) came out right before COD: MW2, which has cast a huge shadow in the tactical FPS genre, but DR did so much right. The movement feels quite realistic and, best of all, I could take shots in DR the same way I would take them IRL and actually get hits, instead of missing because the ballistics model wasn't up to snuff. The only compliant I had with the game is that some of the scripting was touchy: you could get stuck if you didn't approach the mission exactly the way the designers expected you to.
As for the other games mentioned, Sniper Elite is on its way from Amazon Marketplace. It'll be a nice change of pace. Thanks for the tip!
Again, thanks to everyone for the feedback!
"
ReplyDeleteThe meaning of "realistic" in FPS games
Grog News wrote long and interesting article about reality of guns, gears and tactics in FPS games. ...
I really would like to have features like ricochets effects, but there is a field to improve simple mechanisms first.
"