Share this project


Share this project

Neal Stephenson wants to revolutionize sword fighting video games.
Neal Stephenson wants to revolutionize sword fighting video games.
9,023 backers pledged $526,125 to help bring this project to life.

Technical Update #4

Sword Sync and Force Feedback

In this, which will likely be the last technical update we post during the Kickstarter campaign, we’ll try to roll together some of the material from the earlier technical updates and address the single most frequently asked question we get: how to handle force feedback, or the lack thereof, in the hardware controller.

This is going to be somewhat in-depth, but we think we owe it to our donors to address the question as fully as possible.

By way of introduction, I (Neal) offer the following extended analogy to the development of first person shooters over the years.

In some early video games, you had an infinite amount of ammunition and so could just hold the trigger down the whole time. I can remember the first time I played a game that was actually sophisticated enough to keep track of how many rounds had been fired from the magazine and that had the bad manners to stop firing when I allowed myself to run out of ammunition. Many “lives” were lost. I might even have stopped playing the game in disgust and gone back to an old-school shooter where I didn't have to worry about such annoying details. But after a little adjustment I saw it wasn't that difficult to keep an eye on the user interface widget that told me how many rounds were left. I started making allowances for this and reloading when I had an opportunity. The game became more challenging and more engaging as a result. I ended up feeling pleased that the designers of the game had given me credit for having some brains and adaptability. Going back now to an old-school "infinite ammo" game would feel like a step backwards.

If we are going to advance beyond the current doldrums in the swordfighting video game world, we need to make similar reasonable demands on the players. We're confident that they'll step up and handle it, and never look back.

Just as real gunfighters don't lose track of how much ammo is left in their magazine, real swordfighters don't make wild, uncontrolled swings without regard for what their opponents are doing. Both of these are bad ideas in a real fight. A video game that lets you get away with making those errors isn't being honest with you. At some level, you know this.

While swinging your blade into an attack, you'll have time--not much, but enough--to anticipate whether your opponent's sword is going to block yours. Recognizing that, you are, of course, still free to "swing for the fences," but the predictable and fair result of making such an overcommitted attack will be that you'll lose sword sync. That is, your controller's position will get completely out of whack with what's depicted on the screen. Like a shooter player who has just run out of ammo, you'll lose control of the situation.

With the benefit of some game play experience and other user interface features such as tachypsychia (mild slowing down of time) and simple visual, auditory, and haptic feedback, players will learn the knack of anticipating blocks, and other sword-on-sword contact, and then controlling their movements so as to maintain sword sync. This in turn will give them access to relevant gameplay options such as the ability to flow around a block and parry a possible counterattack, or to change the attack into some other technique on the fly.

To translate that into specific user interface terms: all of the moves you can use in a given situation should be represented on the screen by an array of user interface widgets that come and go dynamically based on the game situation. We’ve depicted one possible set of widgets in Technical Update #2 and its associated video clip. Losing sword sync, or letting your character get into a bad position vis-a-vis the opponent, is going to cause a lot of those widgets to gray out or disappear altogether, narrowing your options.

So much for the abstract description of our game design intent.

In the second half of this Technical Update, we now want to get into the weeds a little more, and show why all of this makes sense from the point of view of hard-core swordfighting geeks, or, as our donor Jackalgirl has dubbed them, HSGs.

We’ve had vigorous internal debate over whether and how to release this material, because if we just throw it out there as “proof” that we have got the force feedback problem solved, donors who are coming at this from a more code-centric, software engineering standpoint may feel as though we’re begging the question by changing the subject to medieval swordfighting mumbo jumbo. “Your sad devotion to that ancient religion has not helped you conjure up the stolen data tapes!” But it’s our view that any solution to the software problem has to be rooted in the physical reality of swordfighting. We have to start from the original sources and then look for ways to embody that reality in code.

In that spirit, we recently connected with old friends at the School of European Swordsmanship in Helsinki and Academie Duello in Vancouver--two of the leading organizations in the historical swordsmanship revival that has inspired CLANG. The result is a series of short videos that help to introduce the force feedback “problem” and elucidate why we think it’s solvable from a game design point of view.

First, an illustration of what it is that people are concerned about, filmed at Academie Duello.

Next, a discussion of overcommitment in the fight, from Guy Windsor at the School of European Swordsmanship. The book that Guy is referring to here is, of course, Il Fior di Battaglia, by Fiore dei Liberi (1410).

Finally, a followup from Devon at Academie Duello, discussing some practical approaches to the overcommitment problem from a swordfighting point of view with application to gameplay mechanics.

If you’re still with us at this point, I think you can see that what we’re calling the sword sync problem--loss of registration between the position of the controller in the gamer’s hands vs. position of the virtual sword in the game world--looks like a whole different issue from the point of view of professional longsword fighting instructors. When we come at this from an engineering mindset, it’s natural to get hung up on the fully general problem of what happens when a big swing slams into a strong block. And that is a difficult problem to solve, when you put it that way. But people who play CLANG are going to be inhabiting the characters of trained swordfighters, who don’t fight that way--instead they are thinking a couple of moves ahead, adapting to what the adversary is doing, getting ready to flow around it into the next move. We’ve seen some examples of that “flowing around” in the videos from Guy and Devon, and we’re hoping that experienced gamers--who’ve had to master equally challenging moves when playing other types of games--will see that this style of gameplay is learnable, it’s workable, and that it relegates the force feedback issue into the category of “things that we can figure out given some time and some resources.”


Only backers can post comments. Log In
    1. Tim McClennen on December 11, 2012

      I'm sure that someone smarter than I has already suggested this, but what about self-disarmament? By this I mean that you hit so hard that your sward leaps out of your grasp and lands on the ground. Fight over. In my (limited but not non-existent) sword fighting experience, the same people who regularly over-commit, also grip their swords to tightly, and it is the excessively tight grip, not the excessively loose one, which the sword will break out of. So, when the controller leaves Sword-Synch, the character drops his weapon. This will RAPIDLY train players to use prudent attacks, and to pay attention to maintaining Sword-Synch. Another thought is that "rumble packs" as I think that they were once called, would tell a player when they are supposed to stop moving their controller, although it will not guarantee that they will. A single sharp shock from such a device might coincide with all incidents of hard-on-hard contact, just to give players a heads-up that they should change the direction of their motion.

      This is probably stuff which you already thought of, but here it is again in case it helps.

    2. Peter on July 7, 2012

      In a video titled "Sword Sync and Force Feedback" there was plenty of mention of Sword Sync and over-commit but little to no mention of Force Feedback in the videos. As far as Force Feedback is concerned I am curious what type of hardware you had in mind. A sword game such as this could be aided by force feedback that stops the movement of your hand and arm where it is, although the trick is to do that in a cost effective way. I look forward to the progress you make in the coming days.

    3. Subutai Corporation Creator on July 5, 2012

      As 2/3 of us here at Subutai are left-handed, I hope we will! I consider it likely.

    4. Missing avatar

      Ting Hsu on July 5, 2012

      I noticed that the demonstrators were all right handed hitters. Will the controls work properly for left handed hitters? Obviously, I'm asking as a left handed hitter, but if you look at baseball and hockey statistics, you'll notice many of the players are left handed hitters (over 25%).

    5. Subutai Corporation Creator on July 5, 2012

      @Douglas -we have only considered controllers that are already available on the market.

    6. Douglas Summers Stay on July 5, 2012

      Has the virtual chanbara design featured at SIGGRAPH in 2002 (Daijiro Koga from U. of Tokyo) been considered and rejected? This uses flywheels at different orientations that spin up, and on contact are suddenly braked. Perhaps this would require too much battery, too many moving parts, too expensive, too many patents, or be unsuitable for other reasons, but I would like to know that it has at least been considered.

    7. Missing avatar

      Michael Frilot on July 5, 2012

      Very well said, plus I like the idea that at some point the player is responsible for raising to the occasion and will undoubtedly have a repository of appropriate defensive moves to get back into the swing of things. [This kind of good thought and design will continue to drive this in the right directions and testing will fine tune the game play to make it a fun, but challenging experience. :-) ] (the videos certainly illustrate the correctly design drive very clearly)

    8. Missing avatar

      Zarlan on July 5, 2012

      @Jackalgirl Mwahahahaha >:D

    9. Missing avatar

      Justin Milatz on July 5, 2012

      Beautiful explanations of the subject matter. I agree that this problem is probably smaller than it would seem. If a player allowed his controller to get so far out of sync that he could not recover it, that would mean his character had struck with a force that would make the bind (the swords point of intersection) so unbalanced, that the opponent could easily work agains the player from that.

      A Longsword is both the weapon and the shield. In other words overcommitment gets you killed because it opens your stance and takes you out of balance.
      So allowing this to happen should most likely be the players fault, this is not something that just happens to him.
      The situation is much like not taking cover or not reloading in one of those Tom Clancy Spec Ops Shooters. Players deserve to die for that.

    10. Jackalgirl on July 4, 2012

      Dangit, @Zarlan -- did you /have/ to post those incredibly useful links? I can't afford another obsession right now - LOL!

    11. Will Petillo on July 4, 2012

      Question: suppose I am an unskilled swordfighter--or have a human tendency to screw up once and a while despite training--and I get out of do I get back in? Does the digitial sword just stop until I bring the controller back in sync, does the digital sword move itself back to where the controller is as soon as all digital barriers are removed, does the sword "jump" back into position after a set amount of time, a combination of several of the above, or what?

    12. Missing avatar

      Zarlan on July 4, 2012

      The problem of Sword Sync is just one part of the issue of Kinetic Disconnect. Still, I hope things go relatively well, with the motion controller. This problem may even have the benefit of forcing people to learn to swing more properly, so...

      Oh and...
      @Jackalgirl, You know, there are many HEMA groups, in many places in the world.
      You can try looking at these sites, to find the closest one to you:…

    13. Jackalgirl on July 3, 2012

      I agree that it's true that when using small arms (can't tell you about long arms), the comments are absolutely true about not counting ammo. I always have a hunch, though, as to when roughly I expect to run dry (although if I were swapping out for half-used magazines, I grant you that would be harder) and I get a lot of visual and haptic feedback from the weapon (I like autoloaders) when it does.

      I would liken the smooth, speedy reload to someone with a sword "flowing" around a bind and back into the attack. In both cases, you have encountered something that stops your weapon, and by not overcommitting (in the case of the gun, knowing how to reload and not freaking out), you can get back into play. That video game guns do not require the same kinds of motions to reload as real guns, and also don't provide the same kind of feedback as real guns isn't a problem (or, at least, it doesn't seem to hurt their popularity or playability for most people) - you just learn how to "flow" with the game's virtual interface, which is easy to adapt to as long as it isn't annoying cludgy. I think that Clang will be exactly the same, especially after seeing these awesome videos.

      Grumble. I can see even MORE money coming out of my wallet if I ever move within effective driving range of any of these schools. ; )

    14. Chris Aardappel on July 3, 2012

      I am positive you folks have discussed this but I believe you have an excellent opportunity to deal with a loss of sword sync in the "default" state of the character. You've mentioned that the system will try its best to match actual movements when the player is not in a known stance. Adding some intelligence to this, where the avatar naturally attempts to find the next logical stance based on input, or back away defensively, or displays a thought icon of the next most appropriate stance or whatever should allow for more than enough reaction time for the player to "reload" his weapon into a legitimate fighting stance.

      TL;DR - I have faith that you guys can tackle this issue effectively!

    15. J on July 3, 2012

      Yeah, Erik has it right. I've never had a trainer express any interest in how many rounds you've fired, or have left. Being able to know what you've got left presupposes you know how many you started with, and that isn't reasonable in an environment where you might be reusing partial magazines which you preemptively changed out.

    16. Joshua R on July 3, 2012

      If you'll forgive me ignoring the subject matter and going full-meta: These are without question the best KS updates I've ever seen. I'm bummed there won't be any more like this!

    17. Mark Brown on July 3, 2012

      Kind of reminds me of when Tesshu Yamaoka asked a street fighter nicknamed ‘The terror of Edo’ about how he won so many duels...
      “As soon as the fight began I would get close enough to touch the tip of my enemy’s sword with my own. If he held his sword stiffly I knew I could win easily, but if he held his sword in a flexible grip with a strong projection of ki, I didn’t take the risk of a fight. If I meet such a man I throw my sword at him and run away, and thus remain undefeated.”

    18. Nicholas Russell on July 3, 2012

      Nice to see another who took up the way of the samurai, though I admit it has been a while since last I wielded my boken. I look forward to facing you in the game and seeing how you'll adapt that art to the European longsword.
      Still, speaking to the update, I like what I see. Your answer to the sync problem is to respect the player's intelligence and give them not just full control over the sword but the freedom to screw up royally during a fight. I think not holding the player's hand is precisely what the sub genre needs and I can't wait to give it a try.

    19. Erik Berls on July 3, 2012

      Actually, "real gunfighters" lose track of how many rounds are in their magazine all the time. There are reports on post-incident interviews where police officers were unaware of how many rounds they actually discharged. Magazine changes are drilled such that they are fast and automatic for exactly this reason.

    20. Apotheosis Studios on July 3, 2012

      We can do it! Spread the word! We need to get other swordfighting enthusiasts interested in this project! I practice Iaito, and while I see no Katana anywhere here, I know this project has the ability to pave the way for other groups who are passionate about the sword art I love. So please, spread the word and lets make this project happen!

      If you are needing something to show other to get the word out about CLANG, you can use this blog posting.

      Please help bring new members to the table. Subutai is close, but they need YOUR help to make this happen!