0
mccordia

Wingsuit Proximity Flying Game

Recommended Posts

Another fun project!

A while ago I got contacted by an eager Dutch guy, interested in wingsuits, wanting to develop a wingsuit game, based on proximity flying.

The Russian BASE game also has wingsuits in it, but doesnt feel very realistic, not does it follow actual performance and steering input.

Martijn was interested in real world physics. Drift, flare, diving. All working as they do in real life.
He made a some cool initial tests, and at that point I joined in, helping with feedback and 3D modeling (the current test features no final models, besides a hint at the start on stuff we are currently modeling)

Check out the Video and tell me what you think!

If anyone can help getting me some 3D data on places such as Bispen (the more detailed the better) so I can start modeling a first realistic test-level. Shoot me a PMB|

Anyways..curious to see what everyone thinks..
JC
FlyLikeBrick
I'm an Athlete?

Share this post


Link to post
Share on other sites
Going to be interesting to follow the progress on this project! Im gonna try to get hold of some real world topography numbers for bispen..

I know that Hans and the guys from VKB has done a few test flights with the GGGIII suit. where they had 3D models of the flight path, based on the GPS data.. Maybe they can chip in with some expertize!

Share this post


Link to post
Share on other sites
Looks very cool. The flights look like you took actual video from some of the videoed BASE flights out there and animated it for the game. You should be able to get some good topo info from online or actual maps that can help you develop the terrain relief realistically. Check google earth out if you haven't already as they have some pretty nice 3D updates to the mountains of Europe thats fairly new.
"It's just skydiving..additional drama is not required"
Some people dream about flying, I live my dream
SKYMONKEY PUBLISHING

Share this post


Link to post
Share on other sites

Sadly most known geological informationbanks dont have the requested areas in detail higher than a few pixels per km. But plenty of options left to explore..

And one addition, The footage is not animated, but actual controller based flying.
So thats the best possible compliment possible in terms of look and feel I guess..;)

JC
FlyLikeBrick
I'm an Athlete?

Share this post


Link to post
Share on other sites
Martijn was in the process of creating a username, and can maybe comment on that subject.
On an un related software subject, your input is now included and working in that one as well and will be available for public testing/playing at F&D (will mail you a copy after the event).

But back to the game stuff!
JC
FlyLikeBrick
I'm an Athlete?

Share this post


Link to post
Share on other sites

Hey all, Martijn here!

First off: Man! I'm really glad you guys like it so far. That really validates all the time I've spent obsessing over wingsuits and aerodynamics the past months.;)

About the physics; there's two important factors here:
1. I was a newbie when I started this. I barely had any knowledge of aerodynamics when I started.
2. Similarly, I also did not have much experience with the tools I'm using (Unity game engine, which contains nvidia's PhysX for physical simulation).

All this meant that I had to experiment. And I'm serious about that part. My first prototypes would have made Leonardo da Vinci explode in his grave for sure. Eventually though, I had some code that allowed me to design airfoils very flexibly. From that point on (to brag slightly) my code was already better than D3's BASE, since they don't even simulate lift (and thus also drift and flare) correctly. Even modeling the player as a single, rigid wing behaved better than that game!

But no point in stopping there. I'm now working on simulating all control surfaces (the wings) individually, linking them up to a simulated human skeleton with a simple muscle system. The first prototype of this system is featured in the video. As you move the arms and legs around, the wings move along, and the resulting forces change.

What it does not yet have; and what I'm working on now, is that the lift force should really impact the body so that it will really have to struggle to maintain a good posture. This will make the task of finding balance a lot more complex and dynamic, resulting in a much more deep and fun system. :)
I'm not a BASE or wingsuit practitioner myself (well, not yet at least), so to make sure the game's performance matches the real thing I have a couple of options:
- Base the system om proven aerodynamic models, and speculate how they work together in wingsuit flight.
- Watch and carefully analyse absolutely every skydiving/wingsuit video ever created. Obsessively.
- Converse with Jarno to see if I'm at all on the right track.

It's surprising how much you can learn in a couple of months if you really pour yourself into it. :)
As far as control goes: I'm working on a comprehensive fly-by-wire system. In theory you could control every joint in the body individually using some kind of crazy input device, but for those that want a slightly more playable game I'm building some AI that helps you translate your keyboard/mouse/gamepad input to context-dependent movement of the skeleton system.

If it all works out you'll not only be able to fly normally, but also perform backflying, barrel rolls, flips, etc. Drift, flare and stall should all work as you'd expect.

As you can see in the video, Jarno's working on a beautiful wingsuit model. Right now it's up to me to figure out a good way to link the model up to both an animation driven, and a physics driven system. As soon as I've figured that out, the game will start looking a whole lot better.

There's plenty of other features that I'm planning. Stuff to do with accurate wind, ground effect, etc. But I can't talk too much about those because it is all just speculative.


Anyway, if you guys have any suggestion about what you'd really like to see in a wingsuit flying game, give us a shout, we'd love to hear !B|

Share this post


Link to post
Share on other sites
Yo!
Finally got around to looking at your video...
At the risk of sounding repetitive, thats a fuckin' MASTERPIECE!
I mean ok, its a work in progress, flyer model looks rigid and is obviously still under construction, but DAMN... the camera shake, windblast and sheer speed of it are VERY convincing. Reminds me of some racing games I've played.
Guys, that is really, really awesome!
Class job, guys. I hope you'll release it for the PC when its done. I'd buy it.
-B
Just an afterthought:
Had you considered not just modelling existing slopes people fly on but coming up with impossible ones? I've played a few very convincing snowboarding games where they created snowboarding runs bigger than anything that could possibly exist in the real world. Slopes 20 miles long full of jumps and swoopy curves that were so realistic they induced delightful vertigo. Your game already does that, I'm just thinking you could tack together ten maps the size of whats in that video back to back for that fantasy flying element. How about flying down the side of Olympus Mons, biggest mountain in the solar system?
That video is the only decent depiction of wingsuit flight I've ever seen. I wasn't born with enough thumbs or I'd give you 3 thumbs up.
How many levels is it gonna have?
Live and learn... or die, and teach by example.

Share this post


Link to post
Share on other sites

Thanks for the feedback, that just put a very big smile on my face. B|

The rigidity of the character is what I'm working on now, so that will definitely start to look better.

As for supported platforms: I'm we're building this using Unity, which supports PC, Mac, Wii, iPhone, and soon als PS3, Xbox360 and Android phones; so a lot is possible. It's going to depend on the business model really, not the tech.

As for levels: If at all possible I'd like to try huge, larger than life runs like they do in some snowboarding games. There's some issues that will need to be dealt with.

You have absolutely huge freedom of movement in a wingsuit, compared to a snowboard. On a board the shape of the mountain limits your options way more than when you're zipping through the air. This means:
- You either limit the player's freedom of movement over the mountain in some specific way, being careful the limitation doesn't feel artificial (invisible walls, 'you have 10 seconds to turn back', etc.), or you have to be prepared to model entire regions of mountain area. Which in turn means:
- You have to generate a huge amount of content, which is not possible for a two-man team. With content I mean anything that is interesting to interact with. You could quite easily do endless randomly generated content, but if it contains not interesting features it is all for nothing.
- You need to have tech that can handle that huge amount of space with that huge amount of content.

That said: A cool thing about bigger, open environments is that you can go exploring and experimenting. You could grab a map and plot your course, setting yourself an easy flight or a heavy challenge. Heck, you could even hike all the way up a mountain for some pre-flight immersion.

A pro for a more linear approach is that it will be easier to chop levels up into chunks that can be loaded as the player gets closer to them in a streaming fashion. This would make it possible to do 10 minute runs, which would be very cool.

So those are some of the options and their consequences. I'm not sure which way we'll go yet.

Anyway, since we're talking larger than life, here's a crazy idea I've been playing with:A mode with multiple gravity sources. Since gravity is the only 'motor' force a glider has available for propulsion, why not have multiple ones? As you learn to use multiple gravitational fields you'd be able to fall/glide forever, and use some tricks that astronauts use to fling yourself through the air very fast. ;)

Out of interest, which snowboarding titles are you referring to? I'm familiar with the SSX titles, and more recently Shaun White snowboarding. Any others I should check out for inspiration?

Lastly, if you have any more ideas on what would be interesting in terms of level design: Keep them coming!B|

Share this post


Link to post
Share on other sites
A few questions:
-any thoughts about tracking as well as wingsuit?
-Have you developed the canopy flight at all yet?
-thoughts about 2 ways(in a 'follow me' sort of challenge seeing if you can stay as close to the wall as much as a computer flyer jumping right before you)?
-Will there be aerials allowed on exit that will take skill to come out of and be stable?
- will it just be proxy flying or will there be some iconic objects of all types for the whole spectrum of BASE?

And the final but very important question: in the D3 game, if you held the track button you were automatically stable and knew exactly how far you could go so there was no excitement or balance to being able to stay on a certain glide ratio. It would be very nice if there was some sort of balance you needed to have to stay flying well and if not then you would stall, or spin out, or something so you never actually knew if you were going to make it over something until you had balanced your way past it much like the way you need to stay balanced during a grind in skateboard video games. Having the chance of at any moment spinning out and hitting the wall or whatever you were over at a given time adds so much to the game.

Also in the aerials for the D3 game it was completely random when you pulled off an aerial and sometimes for no reason of skill with the controller you would be unstable out of an aerial and die. It would be great if this was a balance/player skill as well and had complete control over how you came out of an aerial and better control everything instead of having rotation be automated.


Keep up the great work guys, cant wait to hand you over my money! PS3 would be SICK...

Share this post


Link to post
Share on other sites

Thanks for the questions. :)
Tracking would be a nice addition to the game, but it is not currently my priority. Right now I'm just going to focus on getting the wingsuit right, and then we'll see.
I've done some quick prototypes for canopy flight early on in the process. Back then my knowledge was quite lacking, and I tried a way too complicated and realistic approach. No canopy flight yet, but I'll definitely give it another shot with all the knowledge I've gained since.

The thing with both wingsuit flight, tracking and canopy flight is: I want to simulate it all with a certain amount of depth. Basically, my biggest design guideline is: All the stuff you'll be able to do in this game, you will be able to do because the physics system allows you to do it. As of yet, I have not been willing to compromise on that principle. Expect no pre-animated sequences with scripted/hacked behaviour, that kind of design doesn't really fly with me [/pun], as far as simulation goes anyway.

This means that to perform aerials and aerobatics, you will have to manipulate your body to do it, instead of triggering an animation with a button press. This is cooler, and offers deeper and meaningful play (requiring skill, finesse and timing), but it is also a lot harder to pull off from a technical perspective. So, one feature at a time .;)

Regarding game types: 'Monkey see, monkey do' is definitely a good call. You could do this by following a computer controlled character, or by taking turns with other players in a multiplayer context.

About other BASE objects: Proximity flying is the priority for me, as -- from an outsider's perspective -- this type of flying is by far the most adrenaline fueled. For an insider this is of course entirely untrue, but I'm not just targeting this at insiders. That said, if everything is up and running, and if time, money, and life allow it, I'd love to expand the game to encompass the entire BASE spectrum.

Since I'm building a simulation of wingsuit flight, the whole game will essentially be one huge balancing act. As the game is now it already demands quite a lot of skill to fly efficiently, and this will only increase as the depth of the simulation is increased (by dynamic wind, non-linear distribution of profile-drag, all bodyparts being affected by drag/lift individually, and affecting each other, etc.)

For easy of access I'm working on functionality that kind of resembles the 'tracking button' you mentioned. It's kind of like the fly-by-wire control systems that modern jet fighters have. It will be able to perform basic balancing tasks for you, like canceling out oscillations, correcting drift, and making sure you don't get into a stall or spin. Like the car-racing game Forza 3, I plan to let the player select which assists he would like, and how much they should help in navigation. One type of player will just want to breeze through the air in an arcade fashion, completely ignoring many of the finer points of the simulation, while another will want full control over every tiny detail and all the risks that it all brings.

If, when, and how all of this makes it into the game: I have no idea what the future will bring. But those are some of the things I'm thinking about. :)

Share this post


Link to post
Share on other sites
Ok, further thoughts: I used to build maps for Unreal Tournament so I've got a pretty good handle on the limitations and needs you have.

I'd suggest the linear approach. Maybe make ridges, work em in so they aren't obvious walls but once you're flying you can't fly over them. If I were building it I'd make the flyable area 3 to 5 times wider than the field of view, so you can hop a ridge and find a whole different field to fly down with various crosscut valleys allowing you to go from lane to lane, but go too far to the side and the ridge wall is just too high to hop over. To keep it from getting predictable you can make it so sometimes theres a ridge in between you and another flight lane where you have to wait for a gap to become available, and sometimes the ridges are low enough that you can go across lanes at will. Constant variation from tight confined flying and sudden breaks into breathtaking wide open spaces where you can fly anywhere you want.

Autogenerated content gets stale really fast. After a bit it all looks the same. Like you said, interesting features are necessary. If its at least semi-linear and has SOME kind of limits to how far you can fly to the sides it becomes possible for you to create content without having to model stuff 2 miles to each side.

I honestly don't remember the title of the snowboard game...just that the visuals you generated reminded me of it. SSX rings a vague bell and might be it...it was 4 to 6 years back. Arcade game. Big-ass screen, an actual board mounted on a platform you steered by leaning. Very flashy colorful graphics, lots of places where you could launch off the slope and find yourself airborne with the ground dropping away beneath you for that vertigo effect, large visual areas where you could see a mile ahead of yourself and downward and get a real "whoa!" feel for the tasty slopes awaiting you further ahead.

Heres another idea for limiting the side travel without invisible walls or unhoppable ridges: Have it so that you CAN fly up onto the side of the map, but outside the valleys you're supposed to be flying in, make it so the ground on top of the ridges flattens out flatter than your best glide slope. So if the player insists on trying to escape the map, sooner or later the ground comes up under them and they fail to outfly the slope and biff in. Instinct would tell em to get the hell off the high ground before they inevitably hit it.

Trees, man! How bout some trees? a dozen or two different types- make it so you can do glancing contact and get a spray of leaves off of it, hit it closer to center and it sprays more leaves and knocks you off course, hit it dead center and it smacks you out of the air.
Get too close to the surface and it starts kicking up rocks and gravel and shit?
Arches. Stone arches you can fly under, or fly hard and clear em, fly over.
Tunnels, man. How about some freakin' tunnels? BRANCHING tunnels? That exit onto different lanes? Tunnels with broken roofs so sometimes it IS a tunnel and sometimes just a wicked deep v-cut valley? One can flow naturally into another. You come around a corner in a valley and see a ragged hole in the slope ahead of you, can either stay above the surface and fly above the ground or dive into the hole, fly through tunnel, get glimpses of the terrain ahead through ragged holes and gaps in the ceiling.

I'm not a cliff BASE proxy flyer yet so I do proximity cloud surfing, go after the biggest cosmic scale white puffy clouds in range. I've got god knows how many flight hours worth of chasing after ragged holes and canyons in clouds.
How about combining both? A map where the first half is surfing a big-ass cloud with lots of holes cuts and valley features, arrive at mountain below, then surf the walls?

Surface contact warning: Your smoke effects are cool. How about some snow covered slopes where if you get REAL close to the ground it starts generating a snowblast cloud behind you like the spray a swooper raises when dragging a toe across the pond? You could have a 3 or 4 state blast effect...slight, medium, heavy like you're actually dragging your feet, and superheavy where your chest is one inch off the surface and the blast cloud all but obscures your flyer. Step 5 would be an actual crash, tumbling wipeout because you finally dug into the surface.

Cliffs you can fly off the edge of:
One of my favorite visuals in cloud surfing is where I find a slope or semi-flat cloud surface beneath me. When I reach the edge of the cloud the surface drops off in a sheer wall thousands of feet high, causing a WONDROUS sense of flying off a cliff makes me look down and gasp every time. Breathtaking "flight feeling" effect.
You could work this into the terrain, have the ground suddenly drop away beneath the flyer, wall across the way. If you set up for it you could just clear the drop and fly over the ridge on opposite side, or if you know you aren't going to make it and you're stuck in the canyon, forced to hang a hard left or right until terrain opens up allowing you to fly out of valley and rejoin the main course.

Most of this stuff can be done by linear mapping where you only need to load the segments immediately ahead of the flyer or a single segment to each side whereever theres an intersection allowing you to go from one valley cut to another. And like I said, much of the linear nature of it can be disguised by varying the tightness of the terrain.

Hope some of this stuff helps, its all that comes to mind right now.
-B
Live and learn... or die, and teach by example.

Share this post


Link to post
Share on other sites

Excellent stuff, Lurch!

Some of those things I had considered, others I hadn't thought of yet. Those are all really convincing arguments for a linear approach. Basically, you have full control over *flow* of the level, really fine-tuning the geometry for the most exciting experience possible. Contrasting sections; drops with wildly varying slope angles, different lanes (with different trade-offs and difficulties), close-to-ground followed suddenly by a wide open valley; they should all really help to make the experience more dynamic and engaging, like you said.

Tunnels, overhangs and trees are something that I'll need someone else to model for me. I can do terrain generation by myself for now, but that's about the limit of my 3D modeling skill. ;) Regardless, they are all very interesting terrain features, and should definitely be present in the game.

Also, some good ideas for creating level boundaries there.

Particle effects for various surfaces and objects are planned. I'll probably exaggerate their intensity a little to make it more dramatic.

Thanks again for your many thoughts, they're great help!

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

0