How would the FPS genre need to be reinvented for VR?

[wavefunction]

I'm thinking primarily about FPS games here. I think racing games and flight simulators would be the easiest to embrace in VR and in often case, wouldn't require much (if anything) to be changed gameplay wise.

1. Most FPS games don't utilize freelook (the only game I play that freelook makes a difference in is Arma II). With VR, freelook should be universal with detached aiming.

2. I think games like the Modern Warfare series wouldn't survive this transition in their current form because these games are all about funnelling a player's experience down a very narrow path where enemies spawn infinitely in front of them like whack-a-mole. With a large FOV, the amount of unused "real estate" in these games becomes a lot more apparent and this is a big reason why the FOV in these games is so low (75-85)... having an FOV of 110+ would destroy the foundation that the games were built upon, in my opinion. Many AAA FPS titles are the modern equivalent of "rail shooters" like House of the Dead (fun but hardly immersion).

3. I think that invisible walls, "on rails" segments, camera effects (blood splatter and floating sprite-based debris), and "cardboard movie sets" (low detail background objects in the distance) that occur in many modern FPS games simply wouldn't translate to the Rift well. Skyboxes might also be a problem (we need real clouds damnit). These things don't ruin your experience in 2D but I can't imagine that they'd translate well into the next dimension (I heard that objects in Doom 3 meant to trick the player into thinking there was depth when there actually wasn't looked like cardboard cut outs in the Rift). Then again, I've never worn a Rift or any other VR device so this is all speculation.

4. The last thing I considered is movement. FPS games are designed around the keyboard with strafing and running at 5+ mph being the norm. Could this continue into VR? I think it might just make a lot of people nauseous if they were to move around on land like a hockey player does on ice all the time. In my opinion, FPS gameplay is going to need to slow down and try to mimick the limits of the human body for the sake of immersion.

What do you think?

[PalmerTech]

I definitely agree with that last one. The stock run speed in Unreal Tournament is something like 30mph!

[brantlew]

4. The last thing I considered is movement. FPS games are designed around the keyboard with strafing and running at 5+ mph being the norm. Could this continue into VR? I think it might just make a lot of people nauseous if they were to move around on land like a hockey player does on ice all the time. In my opinion, FPS gameplay is going to need to slow down and try to mimick the limits of the human body for the sake of immersion.

Definitely agree with this one. The pace of games is going to have to change dramatically - to the point where I'm not sure you can actually port the same game effectively for both monitor and HMD. Distances will need to shrink, enemy counts will have to go down, action will need to diminish, environmental complexity will need to go up.

[Fredz]

I think moving fast detracts from realism but I'm not sure it will really detract from immersion. In many FPS games the environment is not believable at all anyway (shooting zombies and monsters, gravity and BFG guns, etc.), running very fast like a super soldier is just another aspect of that. I also think this speed is there for a reason, if you had to move at a normal pace these games would be plainly boring. It would take an eternity to move from the start to the end of a level.

For 1) it's still left to be seen if freelook support can be added in stereo drivers or not. For now there is no definitive answer about this, but if it's possible current games could be played in VR without the need to reimplement them.

For 3), I don't think skyboxes will be a problem as long as they're rendered at the correct depth. At this distance there is no depth perception, so rendering them in 2D should be enough IMO. Concerning "cardboard movie sets" I'd say it depends on the way they're implemented, for Source games for example they are rendered as correct geometry, so they could still be rendered in 3D if they are placed at the correct depth.

[cybereality]

I like FPS games for what they are, but I do agree that the paradigm needs to shift if they want to bring these game mechanics into VR and make them work. The I biggest thing that bothers me in games, namely first-person titles, is how limited of interaction you have with the environment. For starters, why do doors magically open when you press the 'E' key. Why not model the hand, and make the player actually turn the knob, open the door at a slower speed (ie stealth). How about opening drawers this same way? There have been a few games that explored this, but very little. I remember the Indigo Prophecy (aka Fahrenheit) had some interesting control mechanics. And Penumbra had some cool interactions (especially with the Novint Falcon). But largely this has been totally ignored. And we have the technology to do it, developers are just not trying or they think that consumers are not ready. Hopefully, with the success of the Rift, this will change.

[FingerFlinger]

A quick side-note: there needs to be a standardized interface for character movement because, between devices like the WizDish and outdoor freemotion set-ups, players will be using be an assortment of solutions. A properly agnostic game will need to support keyboards, controllers, gesture-based input, motion vectors, on up to accepting the actual pose of the player's feet (I recently realized that matching the foot-fall sound effects to the player's actual feet will also be important for immersion).

[brantlew]

I also think this speed is there for a reason, if you had to move at a normal pace these games would be plainly boring. It would take an eternity to move from the start to the end of a level.

That's just because the levels are designed for fast movement. If the level was scaled so that instead of clearing a whole kilometer, you were just clearing a few blocks then true walking speed would make more sense.

[JanVR]

I've recently been experimenting a lot with walking in a game (via a Kinect walking recognition app) and observe that a higher-than-natural walking speed is not necessarily an issue. The brain gets used to it pretty quickly (like walking on one of those flat escalators), so it doesn't diminsh the immersion all the much.

[rhinosix]

I'm not a big fan of military style FPS games, except for when they're really immersive, atmospheric, and story driven. That fast style of gameplay which relies on twitch speeds and constant shooting makes me feel exhausted.

If I was making a FPS game, I would make it more about solitary exploration, observation, survival and strategy.

To increase immersion I would probably use sound a lot - the player would have to listen for targets, and avoid making unnecessary noise. Like Bushido Blade, it would be one shot and you're severely disabled, or dead. No respawns.

Levels would have lots of vertical space: cliffs, buildings with windows, narrow bridges, tall grass, office buildings. And multiple paths to the goal. So instead of walking on a flat horizontal space shooting guys, you would constantly need to be looking and moving up and down. Sometimes you'd need to wait silently. When you shoot someone you'd have to move on immediately, or find a place to hide because enemies would come looking for you. You'd always need to be watching up and down for enemies, and to plan your approach through the level. And you'd need to create distractions, or be quiet to get through open areas.

It might be cool to have a player move through a swamp. Using logs and plants for cover. Crouching underwater, seeing fish swim by as they hear the muffled voices of guards talking above them, or someone being shot and splashing into the water.

Basically I would want to slow down gameplay completely and reduce the amount of enemies, but increase the intensity of encounters. The player would always be observing up and down, listening - trying to take in as much information as possible, and advancing forward. They would need to have their exit strategy planned before shooting someone.

I think games like Deus Ex: HR, Fallout or Metal Gear Solid would make better starting points for FPS in VR rather than Quake or Call of Duty.

Using a regular controller should be fine as long as the player doesn't have to be backtracking, or getting in action-heavy encounters.

[2EyeGuy]

A quick side-note: there needs to be a standardized interface for character movement because, between devices like the WizDish and outdoor freemotion set-ups, players will be using be an assortment of solutions. A properly agnostic game will need to support keyboards, controllers, gesture-based input, motion vectors, on up to accepting the actual pose of the player's feet (I recently realized that matching the foot-fall sound effects to the player's actual feet will also be important for immersion).

I agree with this part. I have experimented with this a bit. I haven't really come up with a perfect solution though. It's complicated a bit by the fact that not all of your body moves together, and you need to not be able to walk through things. But I think that a VR engine should tell the game where the character is, and attempted movement triggers callbacks to the game to see if you can walk there.

Another issue is that games are partly characterised by their movement system. Wolfenstein has very different movement from Doom, even though both are using the same keys. It gives them different flavours. I experimented with a VR version of Wolfenstein that I made, where I could change the movement code to match different games. And when you use the Doom movement system in Wolfenstein, it feels quite different.

I also calculated how fast BJ Blaskowitz was running in Wolfenstein, and it is absurdly fast, totally super-human, I can't remember the exact figure though.

[Flassan]

I've had as few vitriolic reactions from hardcore gamers after Ben's article was picked up by other sites. Only a couple of dozen but it was interesting to see how some people react to the thought of changing the way games are played. Others have commented that there are millions of hardcore gamers but there are hundreds of millions of other gamers. I can understand the need to chill and be lazy but that doesn't mean everyone all the time. I believe in a kind of 'Field of dreams' approach: "If we build it, they will come". We should stop thinking of VR as a minor niche within hardcore and discover what everyone else wants. It's normal for new tech. such as the Internet to take 30 years to reach the wider public and VR is just reaching that point now. There they had the mantra 'rough consensus and running code'. So my stance is yes, don't be afraid to reinvent FPS. The growth of computer games came from the audience wanting to try new ideas so we have to keep building things and trying them.

Julian

[BillRoeske]

The biggest thing I've been considering the past few months is how to give the player a sense of place in the virtual world, because I think that's what's going to ultimately define the first-person VR experience (beyond very important things like hardware). Broadly speaking, I've been breaking it down into two categories: player is literally visible in the world (to himself and others), and the world shows a record of the player's interactions.

Some examples of the visible player:
* Player casts a shadow in the world.
* Player can see portions of their body when appropriate.
* Player is visible in reflections.
* NPCs are capable of tracking the player and making direct eye contact (Story-driven Valve games from Half-Life 2 and newer are particularly a good examples).

Some examples of the visible record:
* Objects are generally dynamic and respond reasonably when disturbed. That's a very broad statement, so I'll give a more specific example to go along with it: a basketball will roll away from me if I accidentally kick it, and some plant branches will bend away from me as I walk into them. I'd also expect to see them bend away if I kick the ball into them.
* Player's interaction modifies surfaces: leaves behind footprints where appropriate, leaves cuts and nicks when hit with another item, can draw on solid surfaces.
* Player's relationship to other entities over time (think Skyrim, Fable, or any game with a faction system, really).

Pretty basic ideas, right? Almost all of these are implemented already to varying degrees in simulations/games already, and often in ways that suit the particular experience just fine. With VR, though, the player's view of the world is far more realistic, and I think that will quickly make the quality of the user's interactions with the world the biggest threat to immersion. Not every simulation needs to be an unlimited sandbox, either, but I do think that careful choices need to be made about what to present to the user and how they might try to interact with what they're given. I love playing Rage, and understand the technical reasons for most everything being glued down to the surface it was placed on, but it's also really silly to see a coffee cup survive a direct rocket blast. On the flip side, I can totally see the tech behind Rage enabling me to go in with VR setup and spray paint my own graffiti anywhere I want on any surface of the Wasted Garage (or anywhere else in the world, for that matter). In fact, that may be an interesting workflow for artists at some point.

Anyway, there's a pretty clear path of how advances in VR technology will improve the "visible player" items. Reasonable body tracking alone will be a revelation when used in tandem with the player's shadow and visible body parts. I think each new "generation" of VR hardware will have a few new features (absolute positioning, finger tracking, gaze tracking, facial capture, misc. refinements) that are pretty obvious slam-dunk advancements for this category for a while. Interestingly, the second category is almost divorced from VR hardware features, but is probably where the sense of uncanny interaction (and more rewarding development) will come from. Since it's not tied to the VR interface as much, advancements here will also be valuable to players preferring not to make the jump to VR.

I'm not really trying to present this as a set of authoritative, exhaustive, or even completely thought out conclusions yet. It's just what I've been thinking about and where I feel the best place to put effort is for a software guy like myself. It seemed to line up with the spirit of the original post, so I thought I would share.

[BillRoeske]

On a side note, VR as an input mechanic will have its own set of weaknesses as well as strengths, and obviously each set of hardware within that category will have its own tradeoffs as well. I think it's roughly analogous to motion control as a category, and successive generations of Kinect, PlayStation Eye/Move, and Wii Remotes as specific hardware approaches and generations. Likewise, I see a key part of designing a satisfying simulation as a bit of a balancing act between what the hardware can reasonably do well and what they brain will let you get away with before objecting.

For example, given all of the current prospects for home-use walking input, I think the better choice right now would be a Wii Remote Nunchuck in one hand and your trusty blaster/sword/paint brush/whatever in the other. It's an analog to walking that people are already comfortable with, and from my experience with my own foot controller, feels less weird to the brain than having your motion not line up exactly with how you are walking. If the next generation brings reasonable body tracking and absolute position tracking within a room-sized area, then perhaps it'd be appropriate to do long-distance locomotion with a joystick and local, precise corrections (like dodging) 1:1 with the user. I'm sure there are holes there, but you get the idea.

The general point being that I think we'll most likely see a gradual evolution of how we interact with first-person simulations, rather than a hard-and-fast input revolution.

[wavefunction]

Lots of great points, Bill. I have no hands on experience with VR so a lot of what I say is speculation based on people's accounts. In your case it sounds like you've got a firm grasp of things.

Anyway, regarding your first post, fidelity is going to be so much more important in VR. I've heard a couple of accounts from people who played games intended for a PC in a VR environment and could see aspects of the game that simply didn't survive the transition which is what influenced my original post. When you're surrounded by the game world, you can see the seams that hold the game together all the more acutely and that means developers really have to start focusing on minute details (like being able to interact with objects as you mentioned) and forgo the kinds of "wizardry" used up until now that took advantage of the limited way players could interact with the game. With good VR and a matching input solution, we'll actually be in these worlds for the most part.

VR adds a whole new and complex layer to game development, one that requires more artwork, attention to detail, and more of a focus on the virtual world as a real space that people can go to. Developers will no longer be able to get away with not giving the player a body, not giving players control over their appendages, not using proper physics, and not animating things meant to appear off camera (to give a few examples). Actually, I think this is going to mean that production values have to go up a lot for titles designed around VR (and consumers will have to pay more for one title). The big thing is that visual and spatial fidelity will be so much more important than the amount of weapons or game modes, especially in the beginning. Advertisements will begin lauding how much interaction and immersion a game has over how many weapons, game modes, and vehicles it has.

I'm probably going to wait out on input methods until a standard begins to become apparent. Between the Leap, the WizDish, the Razer Hydra and whatever else, something is bound to become the standard that is universally supported by all games. I doubt we'll see people putting ODTs in their houses. Like any other input device, people will try them out in various configurations and gradually one will prevail to be dominant for a time.

[Flassan]

A quick side-note: there needs to be a standardized interface for character movement because, between devices like the WizDish and outdoor freemotion set-ups, players will be using be an assortment of solutions. A properly agnostic game will need to support keyboards, controllers, gesture-based input, motion vectors, on up to accepting the actual pose of the player's feet (I recently realized that matching the foot-fall sound effects to the player's actual feet will also be important for immersion).

Have you seen this? http://imi.aau.dk/~sts/?page_id=207
I know from TV how important sound is to a production and this demonstrates the immersive value extremely well.

Something to consider: in full body VR the equivalent of the 5.1 centre channel might be the sound that you make yourself, such as your footsteps? I've casually wondered if an advanced version of the WizDish could have in-built speakers and sub-woofer. I'm not sure if the dish itself could form part of a speaker diaphragm.

[Flassan]

I agree with JanVR and think you could even have a walk speed 'volume control' to suit your taste or a game could modify your speed according to the environment. This would need to be tried to find out how it affects you.
Another factor, assuming you are using an HMD for surround vision, is do you sit or stand and what impact that has? If you stand a wireless connection is better, so I'd like to ask if anyone has any more thoughts regarding WHDI ? (http://www.whdi.org/) as it was recently mentioned in this Wired article http://www.wired.com/gamelife/2012/10/c ... prize/all/

[Flassan]

I realise WHDI has been discussed but would love to hear any thoughts.

[Syntax]

I think it is widely underestimated how far of an perceptional impact it makes when your entire vision is cut off the "real" world. It is no more "playing" games watching at a screen which has no close dettachment to your body.
With an HMD you CAN NOT escape the screen, it takes 100% of your attention while wearing it. Therefore it is crucial that the virtual environment is something you can relate to, it should have a longer introduction to it.
I noticed some people here reflecting their views from the standpoint of todays standart of "gaming", well its a striking word: gaming, sadly it does nail it for the most part of interactive entertainment that exist today.
Videogames are percieved in large parts as "toys", something you have fun for a while but cannot gain experiences like you would get from a novel or a movie. This has to do with the mindlessnes and repetitious unninovative gameplay publishers dictate to ensure big money success This has to change and I think VR can help building better experiences.
In my mind VR experiences should aim the early gaming scene of the 90s. In that time Adventure Games were a big success such as day of the tentacle and Myst. The average gameplay was much slower but back then it was considered as: FUN to actually THINK about that particulary game. A well known gamedesigner was a guy named Ron Gilbert which created monkey island and he gave a keynote at pax one time where he reflects the past & present standard of "gaming": http://www.youtube.com/watch?v=wvixZvSs ... ure=relmfu (watch the hole keynote its worth it !!! ) Todays games only aim for the adrenaline kick, mundane dialog and childish storytelling nothing more! Also a viable point is characktere facial animation, technology has made big advances in this field for games such as L.A Noire... it adds much emotional bounding to the virtual charackters. see some exsamples here: http://www.youtube.com/watch?v=fBR4cT-0sKY
Ok I think I made my point.

[JanVR]

Syntax: Monkey Island - where is the time! They don't make 'em anymore. Agree with you that VR could bring some of those experiences back.

[Conan]

Everybody keeps emphasizing realism but my view is the exact opposite to that


gaming is an escape from the everyday world, the more immersive the gaming world and the technology the further away from the real world we can go so why restrict ourselves to the world we know

I say let the artists go nuts in creating worlds we will never see in reality

Not to mention the technology is just not there yet, Lighting is hacked together and GPU's just dont have the power to model fine details in full 3D

[Syntax]

Monkey Island - where is the time! They don't make 'em anymore.

actually this is not true, some month ago the company "DoubleFine" where Ron Gilbert and Tim Schafer (Day of the Tentacle, Full Throttle) work, raised 3,3 Million on kickstarter for a new adventure game, with over 87.000 backers!
---> http://www.kickstarter.com/projects/dou ... -adventure

@Conan
True, there arent many surreal games out there like for instance Tron 2.0 with stunning visuals and indepth storytelling.

[JanVR]

Nice! The number of backers seems to indicate that there is still demand for this kind of gaming.

[wavefunction]

Everybody keeps emphasizing realism but my view is the exact opposite to that


gaming is an escape from the everyday world, the more immersive the gaming world and the technology the further away from the real world we can go so why restrict ourselves to the world we know

I say let the artists go nuts in creating worlds we will never see in reality

Not to mention the technology is just not there yet, Lighting is hacked together and GPU's just dont have the power to model fine details in full 3D

I think plausibility is a better word to describe what I'm after. I think that as long as you can genuinely be fooled by whatever reality is presented to you, then that's enough for immersion.

[Leahy]

I think plausibility is a better word to describe what I'm after. I think that as long as you can genuinely be fooled by whatever reality is presented to you, then that's enough for immersion.

I like that. I think the term virtual reality has acquired too many connotations over the years, plausible reality just has a nicer ring to it.