fencing as a video game

[essell]

hi peeps, i'm new

i don't fence, but i'm very much into martial arts generally and i'm a games design student trying to develop an idea for a realistic fencing game.

when i say realistic, my ideal is that the video game would recreate the essence of the tactical depth of real fencing, to a point where players with a working knowledge of the sport [eg you people] would be at an advantage over those who don't. moreover, i'd like at least some of the long term depth of fencing to translate over to my game.

bear in mind that this whole thing is purely in the ideas stage at the moment, so don't expect any kind of demos or working games at all. also, please consider that i'm not asking you how to design the game or how the control system should work etc - i'm asking you about real life fencing only.

so if you're interested in maybe helping me out with a brief education on what i need to know to accurately recreate the essence of fencing, i've got a few questions:

1. could you sum up what you think is the essence of knowing how to fence well?

2. in order to give the player the necessary freedom of movement and action in the game that you have in real life, what should you be able to do in the game? for example:

is control of the weaponless arm relevent?
does a fencer point their weapon with the wrist, the arm, or both?
what kinds of leg movement are used? [comprehensive list would be good!]

3. i'm thinking it's important to have each of the three weapons - the epee, foil and sabre - included and represented as they actually work in real life. could you explain how the different characteristics of the weapons change the flow and nature of fencing in relation to each other?

those are the main questions that spring to mind now - i might post more some other time if people respond etc. thanks in advance

[Epeecurean]

Hi essell,

That's a pretty tall order & also a huge undertaking for yourself! I would recommend you do some 'field study' by trying out fencing at a club near you. Here is a list of clubs in NE England:

http://www.leonpaul.com/links/uk_clubs/north_east.htm

Who knows, you may find learning fencing more enjoyable than programming it, and in any case doing the former will help you with the latter.

Cheers,
Epeecurean

[Phoenix]

I would defiently go to some fencing classes and obseve or try it yourself. Here is a good intro into the sport of fencing http://www.nbcolympics.com/fencing/about.html?ch=2

your questions:

1. practice. you can never, ever fence well without practice. Just like in anyother sport.

2. the back hand is used as a balance, especially during a lunged, where you kick your front foot (depends on weapon hand) forward while keeping the back leg stationary.
Fencing has a specific footwork style that you can look up at a club near you.

3. well, each weapon has a different target area and some rule differences. A coach can help you with this. I hope this helps you out.

[Epee_Pox]

Some quick thoughts:

1) Hit the other guy without getting hit yourself. Repeat.

2)
a) Need to be able to position your blade in all of the various numbered lines used to parry and attack. These lines are identified by numbers (first, second, etc.) and there are several books and websites that describe them. A mouse is probably the best interface, just move it around to move the point around, and have the rest of the arm adjust into the appropriate line for that position.

b) Need to be able to control distance between yourself and your opponent. As in martial arts, distance control is the key to striking and to faking out the opponent. Distance is managed in real life through footwork -- advances, retreats, lunges, and variations on the three. Fortunately for your design, there is negligible side-to-side movement -- it's 99.9% forward and backward.

c) Need to be able to combine various movements very rapidly (e.g., in one fluid movement you might do the following: advance and straighten arm into 4; opponent counterattacks; half retreat then instantly lunge into 6 and disengage into 4 for touch).

d) Need to be able to do arm and foot movements very slowly, normal speed, and very fast.

e) Need to be able to control extension of the arm, from all the way by the ribcage to all the way out straight.

f) Don't need to be able to control form (posture, forearm covered by the bell, non-weapon arm movements, etc.). Just presume perfect form, it'll be less to think about.

3) This question is better answered by more experienced folks here.

If I may make a recommendation, have the action be from first-person perspective. A side-on view is useless if you want to have it resemble actual fencing. You and your design team should definitely take some beginner classes and bout a little to get a feel for what it should look like.

[mrbiggs]

Fencing is VERY complicated, and is nearly impossible to design a control system for.

I thought of a system that I think COULD work, but I don't know a programming language powerful enough to do 3D graphics. (It's a first person design.) So I'll just share it with you, and you can do whatever you want with it. (Even if I did make it, I would just give it away for free.)

Left hand controls footwork. W is advance, S is retreat. Space is lunge. Several other buttons could be included for things like fleche, duck, jump, etc. I would have to experiment. It would also contain things such as point in line.

Right hand controls bladework. The different keys on the numpad are for where the blade would go-4 for left, 6 for right, etc. The computer would decide whether it was a parrying action based on footwork direction-unless you're going forward, you need to hit their blade. When you ARE going forwards, there has to be a key to designate whether you are trying to beat or disengage, that would have to be experimented with as well. The numpad would also decide where you would finish, high, low, etc.

Whether or not you miss would depend on a random number thing. Possibly, this factor could also be affected by the position of the blade at the time of the lunge.

It's just an idea, but hopefully, it'll give you a head start.

And you should learn the sport before you create a video game. One of the things about fencing is that just about ANY talent can translate into an advantage in fencing-speed, strength, intelligence. Putting that into a video game would be very difficult, and without a knowledge of fencing, it'd be just about impossible.

But good luck!

[prototoast]

See this thread: http://www.fencing101.com/vb/showthread.php?t=11255

It offers an example of a fencing game, as well as constructive criticism.

[JackSparrow]

One of the aspects of fencing which doesn't translate very well in computer games is the use of footwork, in particular the ability to wrong foot an opponent. The only game that has come close to this is in the original Prince of Persia game. In combat if your opponent tried to close the distance between you, he left himself open to an attack into his preparation if it was timed correctly. These kind of stop hits made you think quite carefully about your characters use of distance and footwork.

However, in the original Prince of Persia you could only attack directly and parry in a single line. I would suggest that you can only make a realistic fencing simulator if you switch to first person perspective. Overwise your game will be to simple and arcady to have any tactical relevance to the real sport.

[Black Jeebus]

Hell I would be happy with an arcady-style game that just used upper and lower lines. You might think about an isometric camera angle. something hovering off to the side and above the fencers. That might allow for attacks and parries in the four lines. I would really suggest just trying to epee first, cause foil and sabre have objective parts to them that just wouldnt go over too well.

[jBirch]

I thought about doing this as well: the controls are the key.

Using a PS2 style keypad you have 1 button for attack, one for lateral parry and one for contre parry (circular parry). You have footwork being the four directions + the distance. So a lunge is attack from far away to whatever target you pick. Infighting is to attack/parry at close distance (cool animations).

To attack into high inside, you'd press up and right (against a righty attacker) and then the attack button. To defend against the same, you'd press up and left and the lateral defense button or up and right and the circular defense button (4, contre 6). Beat attack would be lateral parry at distance and attack. Disengage would be like a SF2 fireball: start in one line, traverse all the way to the other line while holding the attack button. You could use the fourth button to emphasise the footwork.

So forward alone becomes step, forward with emphasis becomes ballestra. Tap forward becomes an appel. Tap backwards becomes a half-retreat. Backwards becomes a full retreat. Backwards with emphasis becomes a jump back. Backwards + emphasis + attack + high inside = jump back stop hit the advancing arm on the inside. Lateral Parry + high inside + attack = attack with opposition in 4. Lots of tactical variation.

Hope this helps!

[mrbiggs]

jbirch, how do you move and parry at the same time?

[jBirch]

Originally Posted by mrbiggs

jbirch, how do you move and parry at the same time?

Don't let your coach hear you say that, mrbiggs! You parry, then you move, or you move, then you parry. Never the two together.

Take it easy.

[Epee_Pox]

Originally Posted by jBirch

Don't let your coach hear you say that, mrbiggs! You parry, then you move, or you move, then you parry. Never the two together.

Take it easy.

Are you serious? Sounds like the height of foolishness not to retreat during most parries. And in the rare instances when not retreating simultaneously with a parry, you're doing something else, like a ceding parry with a step forward; or crouching low with a high 9 parry, which melds into a strike at the knee; etc. etc.

I think telling someone to parry first, and only after that to move, is very bad advice.

[mrbiggs]

How about disengage?

You have to disengage while moving forwards....

And how can you defend against marching attacks when you don't parry while moving? If I'm faster than you in footwork, I'll win the point.

[Black Jeebus]

Originally Posted by jBirch

Don't let your coach hear you say that, mrbiggs! You parry, then you move, or you move, then you parry. Never the two together.

Take it easy.

lol...

[essell]

cheers for replies, people - most of this is good stuff, so feel free to keep at it i'm in a rush now but i'm working on a proper post full of questions in reply to certain things people have said...

[jBirch]

Originally Posted by Epee_Pox

Are you serious? Sounds like the height of foolishness not to retreat during most parries. And in the rare instances when not retreating simultaneously with a parry, you're doing something else, like a ceding parry with a step forward; or crouching low with a high 9 parry, which melds into a strike at the knee; etc. etc.

I think telling someone to parry first, and only after that to move, is very bad advice.

We should move this to another thread but...

Opponent attacks, you parry (ie// stop their attack from arriving) then you decide whether to move forward (attack w/ opposition) or backward (retreat out of distance). If you parry and move at the same time, your distance for one or the other is hooped. If you retreat, then parry, what's the point of the parry? Aren't you just batting at their non-threatening blade? Didn't you just get out of distance? Same if you parry, then step forward. Now what? You're too close to attack, so what do you do? Knee your opponent in the groin? *grin*

If your coach is telling you to move and parry simultaneously then I suggest you ask them why since it sounds like sketchy advice. They could be talking about attacks with opposition (in epee you see that a lot), in which case you parry, then move and attack while holding the line closed.

Hope this helps.

[Indy4ever]

To get back on the subject...

I think that having too intricate a control of what your fencer does would simply kill the game - the only people who would have any idea of how to string anything half decent together or have the discipline to keep working at it would be fencers, and the amount of time it would take to work out the order in which to press keys would mean that you may as well just go out and do some real fencing. Think about it - how long did it take you to learn to stick your arm out before lunging? I still can't do it properly. Now think about some of the other moves you normally do on the piste. Doing them on a keyboard will probably take too long to learn to do.

Now take the example of Street Fighter II, or any other beat em up since then. Although you couldn't control the exact movements of each guy you could press a quick sequence of buttons to make him do stuff i.e. throw a fireball or leap in the air and spin his legs around - special moves. In the same way, you could have different set attacks controlled by the combination of keys you press. e.g. you press three keys in a sequence and the guy beats 4, steps forward and disengages, then lunges in 6 - a pretty complicated sequence if you have individual key presses, but straightforward in this format.

I dunno, its only an idea. But I would be willing to give up stringent control of the fencer if it could make the game more playable than it would be with the systems suggested here. I don't have to know anything about kung fu to play fighting video games, and I am pretty sure it is impossible to throw fireballs with my hands: but it is this blissful simplicity that makes Street Fighter a cool game. That's not to say I want the fencers to have six arms and fly, or anything cartooney like that; just that I would be willing to give up some of the more complicated tactical possibilities afforded by detailed controls in order to make this fencing game more playable for others, as opposed to just us weirdos who fence three times a week, give up our weekends for competitions, hand over our money for pricey equipment and log on to web forums to communicate with other equalyl insane idiots. No offence intended you guys.

[ls14evar]

I think the best way to go about this is think exactly what the moves are you want to represent

-Footwork
Short Advance
Long Advance
Foreward Crossover
Short Retreat
Long Retreat
Backward Crossover
Shift left
Shift right
duck
Jump forward
Jump back
Fleche
Appel
Balestra (combo move off lunge?)

-Types of attack
Extension
Lunge
flick
Opposition
beat
bind

-parries
All eight parries.....
Lean back (Yah, not quite move, but I do it all the time to avoid touches)
Lift toe
disenguage

Im sure Im missing alot of stuff here though.

I see it working best as a side view sort of thing (only way to show the distance well.) Parries would have to be simplified to high med, low attacks. Mostly a distance and time game.

BTW I work in game development too. But Im an animator doing ps2 and xbox stuff.

[Epee_Pox]

Use the mouse to control the point. Let the computer move the arm and hand to get the point where you put it.

Left click to thrust.
Right click to lunge. Release to recover.

W to move forward
S to move backward

Done.

[Mar98]

there is a simple solution for blade- (and arm-) work: a dataglove.
but then .. you better stand up for moving your arm right. so we can add data-fencingshoes for perfect footwork control. but then .. well, you can do real fencing then.

i've thought about that for some time too, but i think fencing is too complex for getting a model that covers the thrill and is controllable in the same time. you have to consider the speed changes (wich can differ for arm and foot at the same time).
and then complex maneuvers .. imagine a pil with derobement(?). you can implement this as a combo - but when there`s no search for the blade? in rl you then do something different (or stupid). and how to implemet reflexes?

sounds very destructive .. there must be a compromise, which elements should be taken and which not. not to simple, not to complex ..

mar98