I take full responsibility for how good or bad the game balance is. ;) Oh, and I already know you think that Tafari is too good and that the random number generator is broken. Neither is true and everyone says both of those things when they start.

You can discuss the game on my forums, if you like.
Good luck!
--Sirlin

I forgo several crossup oppertunities in this vid, partly because we were distracted with talking partly because CigarBob has crazy reversal abilities. In later matches I went for the crossups more, but whatever. At some point I did a crazy dragon punch vs his sweep and declared the maneuver "old-school only."
Enjoy.
--Sirlin

1. Monday: Kongai, 1.8 balance patch, closed beta
2. Tuesday: Yomi: Fighting Card Game, semi-closed beta for shoryuken.com members
3. Wednesday: Street Fighter HD Remix, open beta

Yomi: Fighting Card game is a fixed-deck card game using modified poker cards that simulates the mind-games that go on during fighting games. I started this as a side project three years ago. Kongai virtual card game is the main site-wide game at Kongregate.com and is also an exploration of "reading the mind of the opponent." I started the design of this game about 2 years ago. Street Fighter HD Remix is an update to the 14-year-old classic, Super Street Fighter 2 Turbo. It has all sorts of features, but most importantly, I completely rebalanced the game and I hope it will hold up under the intense pressure of tournament play. I started the balancing process about a year ago.

Kongai: 20 characters
Yomi: 18 decks (18 characters)
Street Fighter HD Remix: 17 characters

That's 55 characters total, and tens of thousands of variables that have to be set a certain way to make sure things are balanced for competitive play--and fun!. (Maybe hundreds of thousands if you really count everything in Street Fighter.) It's a lot to balance! Especially all at once.

If you're not in the closed beta for Kongai, you'll get your chance to play it soon anyway. The public release is getting close. Shoryuken.com members can play Yomi right now if they are willing to print out some playtest cards and are invited to bring prototype cards to the Evolution Fighting Game Championships this August in Las Vegas. The real Yomi game should be out next year. And anyone who buys Commando 3 on XBLA can play the Street Fighter beta starting tomorrow. The full game will be out when the art is done.

Enjoy!
--Sirlin

For the record, I'm available but I don't think Blizzard will call. ;) For one, they might say that fighting game balance doesn't apply to their game, and that I don't know their game. Shrug. The same concepts seem to apply to Puzzle Fighter, Street Fighter, Kongai, Yomi card game, and the other projects I can't tell you about. Balance is balance and the same issues show over and over.

And second, I don't think my article two years ago won me any points there. Even though much of what I complained about has indeed been acknowledged and addressed, I don't think my tone went over well.

Anyway, thanks Vir.

--Sirlin

And now, gamasutra interviewed Larry about his company. It sure sounds a lot like what Grassroots Gamemaster talks about. The people in Larry's circle are all contractors who come together for projects but don't all work in the same "factory." I'm getting used to the idea.

Meanwhile Jenova Chen is doing his own thing in this other interview. I will officially state that Jenova Chen has beaten me. He really is making things of his choosing, things that the rest of the industry isn't so good at making. Most of all, he really seems to lead a band of programmers and developers to implement his vision. Bravo, Jenova Chen.

--Sirlin

Maybe I should have a weekly show or something?

--Sirlin

GameCyte also asked a lot of questions about game balancing in general, what it actually means, how it's done, and how it differs in different types of games. Those didn't make it into the final cut, but maybe they'll create a part 2 out of all the footage. In any case, enjoy!
--Sirlin

You can hear some information about this in this video (which incidentally has less than a third of the views of my interview on the same site, sorry for the uncalled-for cheap shot).

Anyway, I just couldn't let this video go. I don't get why World of Warcraft has taken so many steps towards being reasonable and yet can't take the last, logical few. The key point here is that 10-man raids will get worse rewards (one tier lower in WoW-speak) than the 25-man versions of the same dungeons. Kaplan (lead designer) explains that 25-man raids are much harder to coordinate, have more logistics to worry about, and are more work. No argument there, I think we all agree with that. But this is the *reason* that they need to have better rewards, he says. That sounds a bit backwards.

To put this another way, there are two versions of each raid dungeon: the 10-man version and the less fun one. If they had the same rewards, not enough players would play the less fun one. So...why even have the less fun one? Shouldn't players be encouraged to play the content that is the most enjoyable to them, rather than encouraged to play content that is more logistically difficult to coordinate? (10-man versions can be tuned to take just as long, of course.)

Kaplan has moments of clarity in this video where he explains that both sizes of raids should have easy dungeons and hard dungeons. The size of the raid is not a judgment call on your worth, it's simply about how many people you feel socially comfortable hanging out with. Either one can be hard or easy, depending on the specific dungeon design. Yep! (And either can have the same rewards? No, apparently.)

Then in the same video Kaplan mentions that they considered solutions(?) like having the final 25-man raid on Arthas *unlock* the 10-man version. (What??) Or maybe when you get to the end of the 10-man version, Arthas just isn't there. Perhaps he left a note behind: "Hey guys, this is Arthas. I only value players who play in large groups and I'm a little grumpy about whole no-more-40-man-raids thing. I won't even fight second class citizens such as yourselves."

I think Kaplan knows on an intellectual level that 25 is not better than 10. (He flat out said it, in fact). He might also know that 10 is not better than 5, that 5 is not better than 2, and that 2 is not better than 1. They are just all different. Each one of those sizes can have tasks and challenges of extreme difficulty. Each one can have endless time-sinks. And yet the 40-man values of WoW's past still echo today. "Yeah, yeah we'll *have* 10-man raids, but they can't have equal loot!"

One last thing I'd like to point out is the years-old argument that players who enjoy large raids would enjoy them regardless of the loot. For the majority of raiders, this is false. I know it's false, you know it's false, and Blizzard definitely knows it's false. The last place I'd look to find people motivated by intrinsic rewards is a 40-man World of Warcraft raid. (Dear both of you who really do enjoy big raids for the own sake, even with no rewards. You are not like the others.) The actual case is that the vast majority are motivated by gear-rewards to spend time in dungeons that they would otherwise not choose to play in. If I'm right about this, why not let the rewards be equal so that they can play 10-man raids and have more fun? And if I'm wrong about this, why not still make the rewards equal? In that case, 25-man versions will be full to the brim anyway because raiders love the intrinsic rewards of completing a logistically difficult task, after all.

To summarize, challenge should lead to rewards. (A separate gate of time-spent can also lead to rewards since this is an MMO. I'm not ruling that out.) Challenge can come 25-man or 10-man. If you accept all that, the final step is that it's true for 5 man...and 2 man...and 1 man. You can have just as much challenge (and require just as much time spent, if you like) in any of those sizes. Different players will have a sweet spot group-size that they prefer, and no size is really second-class. The value judgment shouldn't be on group size, but rather that we judge inclusive design as better than exclusive design.

If there were challenging 2-person dungeons that I could play with my girlfriend, I'd still be playing World of Warcraft today. I get the feeling that if I made it past a gauntlet of virtually impossible monsters in a 2-person World of Warcraft dungeon that the final boss would disappear and say, "Sorry, but the princess is in another castle."

--Sirlin

Guile is a solid mid-tier character in Super Street Fighter 2 Turbo (ST), but it takes quite a bit of player-skill for him to fulfill that potential. As a mid-tier character, he’s eligible for a few upgrades, especially in his bad matchups. More than that, he’s eligible for some fun.

Crazy New Flash Kick
Guile’s roundhouse flash kick goes straight up very high in ST, and has no use I’m aware of. I’ve never seen a good US or Japanese Guile player use it in a real match, so it’s a ripe place for some new spice. Instead of traveling straight up, it now travels very far forward and diagonally up just a bit.

Guile players salivate at this newfound range, but it comes with a price. Great flash kicks come with great recovery. If blocked, it’s the most vulnerable move in the game. A fierce Dragon Punch is pretty vulnerable if you block it too, but Guile’s roundhouse flash kick leaves him pushed right up against you so that it’s very easy to time a combo as he recovers.

The new flash kick means you can’t jump straight up and down at mid-range versus him anymore. He can use it as an escape if you try to cross him up. He can use it to trade or even clean hit Dhalsim’s Yoga Fire and Sagat’s Low Tiger at mid-range. He can use it easily hit Vega’s slide, while his old flash kicks often missed in that situation.

Vega was a very bad match for Guile before, and a lot of it had to do with how hard it is to counter Vega’s slide. It’s good news that he now has a better answer. Sagat was a very tough mach as well, though not so much anymore due to Sagat’s slightly worse fireball recovery. The change in Sagat’s fireball recovery combined with Guile’s new roundhouse flash kick might give Guile the advantage now, but that remains to be seen. Guile’s worst match was probably versus Dhalsim, so his ability to do *something* at mid-range now combined with Dhalsim’s inability to low punch cleanly under Sonic Booms brings this match from near-hopeless to near-fair. Blanka is still a tough match for Guile. You can’t win ‘em all.

Upside-Down Kick
The upside down kick (hold toward or away + roundhouse while close) always looked like an overhead attack…and now it finally is. The opponent now must block this move high. In addition, you can now also activate it from any range by holding toward + roundhouse (it still has really long range). This kick had 15 frames of startup in ST, which would make it the fastest (and longest range) overhead in the game. I slowed it down to 20 frame startup which puts 2 frames slower than the slowest overhead in the game (Fei Long’s).

Guile is such a defensive character that it’s a breath of fresh air to have a new offensive option. The opponent really should be able to block this overhead most of the time, but Guile can tack it on after making the opponent block a Sonic Boom almost any time he wants, as long as he’s willing to lose his charge. (You have to be pretty close to get the overhead to come out when you’re holding back, remember.)

I also had to monkey with the hitbox below Guile during this move. After the first pass, we realized that he could cleanly upside down kick over Sagat’s Low Tiger Shots every single time on reaction. It was a bit too crazy for me, so now those Low Tigers will hit Guile in this situation. There’s still a very tiny window where Guile can go over the Tiger Shot cleanly, but it’s impractical to do in a real match. Also note that even though Guile appears to be off the ground during this move, he can still be thrown. I didn’t change that; it was true in ST also.

Super
Guile’s super move fails to connect correctly all the time in ST. It took about four tries to fix this, but it finally works as it seems like it always should have. The super move consists of two consecutive flash kicks, and the second one has a bit more range now. All the hits also knock down and juggle now, and I even had to do some subtlety about changing how fast the enemy falls when he’s hit by this move. You won’t notice that, but without it, the super would miss air juggle hits on certain characters only. Vega was one of those characters, but now Vega gets juggled like everyone else.

In equally big news, there’s now an optional new motion to perform the super. You can still do the old motion, but the new one is hell of a lot more practical: charge down/back, then down, down/forward, forward, up/forward. In other words, charge down/back, then do the old Tiger Knee motion + kick. Ironically, I removed all other Tiger Knee motions, but it seemed appropriate here considering how complicated the old motion was. (The old motion could be performed charge db, df, ub OR charge db, df, db, any up.)

You might ask, “Won’t I get the super to come out accidentally?” I have never had this happen even once, so I don’t think so. You might also worry that Guile can now easily combo into this super. For the most part, comboing normal moves into supers is not practical in ST (yes, I know you can do it). We could have made it very easy for everyone to combo normals into supers, but the lack of this feature is, in my opinion, one of the nice and defining qualities of the SF2 series. It’s not about poking with safe attacks and comboing into a super. Supers are, for the most part, an additional move in your arsenal rather than a combo ender.

So doesn’t Guile break this rule by easily comboing low strong into super with his new motion? You’d think so, but some deep mysterious property of ST’s engine saves us. I cannot figure why this is true, but you cannot combo normal attacks into Guile’s super using this new motion. If you try, you will always get a flash kick instead. Even more deeply mysterious is that if you try to combo a normal move into super using the old, difficult motion, you CAN. This really should be true, but somehow it is. So combing into super is no easier or more difficult than it ever was for Guile. We may never know why.

Things That Didn’t Change
One of the most common requests from players was to change around the commands of Guile’s standing kicks. In ST (and HD Remix), the kicks are:

• Back + short = knee (advances Guile while charging)
• Back + medium kick = backward sobat (Guile jumps slightly off the ground and moves back while kicking)
• Back + roundhouse = lunging kick (Guiles moves forward a bit while charging and kicks)

The common request is to find some way to make back + short into Old Guile’s rapid fire kicks. This would allow Guile to rapidly standing short to stop incoming slides while keeping his charge. I admit, this would be useful against Blanka, and Blanka is a bad match for Guile, but we can’t fix everything. The new roundhouse flash kick hits the other slides anyway.

More to the point though, no one could ever agree on which move Guile should lose of the above three. The knee is incredibly useful and basic to his gameplay, so that needs to stay for sure. The back sobat is not used much by most players, but the experts find it indispensable because of a few important, specific things it counters (Dhalsim’s drill at a certain range is one, and there are others). And Guile players were not willing to lose the back + roundhouse lunge kick either. Rather than assign the knee to back + jab + short (strange for a normal move), I just left all those kicks as-is.

I hope Guile players won’t complain about that, because with a huge new flash kick, a long range overhead kick, and an easier, better super, Guile players should thank their lucky stars.

–Sirlin

Now that that's out of my system, check out Kongregate.com. Jim and his crew are highly intelligent, highly motivated, and foster an environment where everyone contributes ideas and the best ones rise to the top. Kongregate is worth your look. The site-wide metagame called Kongai is where I come in, and I hope that's worth your look too, once it's out of beta.

--Sirlin