The Road To Puzzle Quest 2

This generation, one of the standout surprise hits was Puzzle Quest, which blended casual puzzle game mechanics with an RPG theme and in doing so came up with something that appealed to a wide audience -- addicted them, in fact.

However, its sequel, Puzzle Quest: Galactrix wasn't quite as well loved. The game had a complicated design and, on the Nintendo DS, major interface issues.

Here, Steve Fawkner, president of Puzzle Quest developers Infinite Interactive, discusses the lessons the team learned from the contrast between the successes of the first two games, and how that has played out in the decisions made with the upcoming Puzzle Quest 2 -- in terms of how gameplay design, UI, and technical improvements are considered.

I just played a bit of Puzzle Quest 2, and was reminded of the balance of luck versus skill in these games. For example, I got a crazy chain off a random jewel drop, but so could the enemy.

When the player gets that kind of bonus, they feel great. It's awesome. But when it starts to happen too much on the enemy side, they feel like the game is stacked against them -- even if it really is random or luck-based.

Steve Fawkner: And it is indeed random and luck-based. There's no cheating going on in the game. But it wasn't in the first one, or Galactrix either, although it quite often feels that way.

I think one of the challenges as a designer, and something you want to actually try to do, is you want to have the games just frustrating enough the player wants to keep playing it. I feel we actually hit that with Puzzle Quest. People kind of feel cheated, but not cheated so badly they want to ragequit. They feel cheated, and they want to get back at it. If you can get that, you've got a very, very addictive game.

So you actually don't mind the potential for a bit of frustration there?

SF: No. I think a little bit of frustration is actually a good thing, provided it doesn't make the player want to quit. It makes the player want to keep playing and win. I feel that in Puzzle Quest the players generally right from the start have enough hit points and enough life points that, even though they might have a string of bad luck, with a little bit of good play following that, they get back in to the game and ultimately win.

Puzzle Quest 2

Have you played Might & Magic: Clash of Heroes?

SF: I haven't played that yet. No, unfortunately.

It's a player versus player, or player versus AI player puzzle game. It's similar to PQ, except it's got a different mechanic. What they do for randomization is they will drop a random number of units into your well, but they will never set it up for an instant combo. What is the appeal of doing one versus the other?

SF: Well, we actually experimented a little bit early in development with this game, with what I'd call anti-cheating. Which was actually rigging the drops so that those large cascades didn't happen. The game actually loses a little bit of its charm, we felt, when we looked at that. We could turn it on just for the AI, or we could turn it off for both the AI and the human.

It's obviously something we want to leave on for the human. Because when the human gets that, it's awesome. It's one of those things that makes your day. We felt it detracted a little bit from the game when the AI couldn't do it too. You remember those fights where you've had a bad run of luck, and then you come back from behind? That's a really exciting, memorable fight. So we wanted to keep that.

When there's player versus player, how does that change how you have to design the battles? Because it's one thing for a huge chain to come down versus an AI opponent, but it's even more horrifying if it happens with a person you know is going to beat you.

SF: Yes, it is. And there's a number of things to think about with player versus player. The first one is that adding in balance to a PVE -- player versus environment -- is great. Because as the player discovers that imbalance, they enjoy exploiting it. If you put in too much imbalance into the environment, like in PVP, nobody enjoys getting their butt whipped by someone who's bet on the imbalances.

So the imbalances still have to be there, because you need a player to reach where he can't beat the other player, whether you're talking about a turn-based strategy RTS or one of these puzzle games.

I believe that some imbalances just need to be a lot smaller and there have to be good counters for the imbalances. And so it just means a lot more design time that we need to put in, that we know we have to put in just to balance things.

But there's that second issue where we have to cater to the fact that a player just may want to sit down and play with a friend. We've got something like that on Xbox, the XBLA version of Puzzle Quest 2.

We've got a tournament mode we can put in anywhere, where the players can just sit down with a friend and play with a bunch of monsters versus their friend. Just whose goblin is better -- "Is my goblin better than your goblin?" And you can show who's the better player.

We tried to build that into Galactrix as well, where one player could choose a ship and the next player could choose any enemy ship to fight it with. That meant that a player could sit down who was brand new to the game and could play against his friend and they could build their own handicapping in there.

It's important to give players choices, I think. Set that stuff up. But if you've got a ranked online system like XBLA, it's really important to have some kind of matching system going on there. So it's not the guy with the level 50, unbeatable ship, that goes out there and wins every battle, or the level 50, unbeatable hero with the broken spell system. The only way to combat that is throwing a ton of design work at the game.

How far do spells in the game (which change the playing field, or damage enemies) go toward achieving the balance that's disrupted by, say, the randomness?

SF: I really feel that achieving that balance is partly about the channels that damage is dealt through in this game, and that having skills as your only [method of] dealing damage would allow for pretty large imbalances.

So in the first Puzzle Quest, you had skills plus spells to deal damage. That allowed two ways of dealing damage, and of course you can deal more damage that way.

So a good player can manipulate the game; using spells and the board will deal more damage. That will kind of bring him back, if he gets behind. Some good players will bring themselves back.

We've added a third channel of dealing damage in Puzzle Quest 2, which is the action points and weapon system. We feel that it makes it easier for us to balance the game.

Does the new weapon system also add some a loot element? Do the monsters have item drops, or are there conditions to meet to get certain weapons, et cetera?

SF: Well, there's a number of things about the weapons. To start with, you've got exactly which kinds of weapons you're going to equip. So am I going to use a two-handed weapon and use up both my weapon slots? It'll do more damage, sure, but it will take more action points to activate, and that might be the way I'm gearing my character, for a build that uses a lot of action points to do a lot of damage.

Do I want to use two weapons, a bigger one and a small one? A bigger one to do the big hits, and a smaller one to finish of an opponent, perhaps. Do I want to use a weapon and a shield to improve my defense? Do I want to walk out there with a potion in my hand, that I can increase my mana with?

An interesting [character] build I saw the other week was a two potion-wielding wizard. One yellow potion, one blue one. He wasn't about dealing damage with his weapons, he was about pumping up his mana to cast more spells. So it's left us with a lot of variety in how we can build the characters, and that's really nice.

The monsters don't actually drop items. They're not actually dropping swords, but they are dropping what we call trait items and components, and those I things I can use to upgrade my weapons in the game. And monsters will drop bits of metal and wood, or amber, or pearls, or emeralds. I can take these to a merchant in town and use them to upgrade certain weapons.

It's a bit like an item crafting system, really. I can take my sword and turn it into a fine sword, then turn it into a masterwork sword. Each time I upgrade it, it's getting more bonuses, and as I finally upgrade it through the last levels, epic up to legendary, it's picking up special bonuses, and doing things like increasing my critical chance and increasing my spell resistance.

So there aren't multiple swords that you would have to collect necessarily. It's all about gearing your one weapon set.

SF: Well, there are 33 different types of weapons in the game, each with slightly different stats and different abilities. For instance, if I pick up a missile weapon, missile weapons increase my initial action points, as well as doing damage like a regular weapon. Two-handed weapons just do a lot of damage.

We have other variations like the larger two-handed weapons only the barbarian can use. The bigger shields and armor, only the templar can use. The big poisons, only the assassin can use. There's certain weapon types tied to certain classes.

So it's quite a complex system, even down to having what we call racial types on items, which is like elven swords and dwarven swords, and providing slightly different bonuses to each other.

Regarding Galactrix, I loved Puzzle Quest, but Galactrix frustrated me, which was interesting because I'm actually more of a hardcore puzzle player type. I really enjoy Puyo Puyo and Super Puzzle Fighter Turbo, but Galactrix...

SF: Galactrix is a bit of an interesting beast. It appealed to some people and not to others. That appeal wasn't based on how hardcore you were or how casual you were. It wasn't based on how smart or how dumb you were, or anything like that. To some people it struck a chord with them, for some people it didn't.

There's no doubt that the original Puzzle Quest probably struck a chord with more people than Galactrix did. We went through after Galactrix, and we really picked it apart and tried to identify exactly why, more people liked Puzzle Quest than Galactrix. When we came to do Puzzle Quest 2 we did some things, like setting it back in the fantasy universe. Because we felt that sci-fi just maybe didn't quite lend itself as well to this kind of game.

We made the game more accessible in a number of ways too. Galactrix had a lot of mechanics in there that were difficult to convey to the player. If you got them, it was great. If you learnt them, you played the board, it was great. If you didn't, it just seemed that we never gave you an opportunity to learn that stuff.

The timing-based puzzles in Galactrix were another interesting one. Some people really hate time-based puzzles, in a turn-based puzzle game. You'll have noticed in Puzzle Quest 2, that we removed those time-based levels entirely. Having said that, if you enjoyed those three elements, if they felt right to you, you really had a good time with Galactrix. Myself, I loved it.

What was the thinking behind the travel and its UI? I played it on the DS primarily, and I found it was very difficult to select what I wanted, like my ship, versus the planet.

SF: Yeah, on the DS, there are a few technical issues on the DS and stuff being very close together. We learnt a lot of lessons from that game about how to use UI space very well, use the interface on the DS much better to keep things apart. Don't put things too close together because as you hit the screen on the DS, you can get a bit of an echo on parts of the screen. We've also improved the tech behind that as well; we can pick up much better where the stylus is hitting.

So they were all things we took on board and learnt, because Galactrix was actually our first DS game, done in-house. [With] Puzzle Quest, we oversaw the development of it, the DS version was done by some other people.

With the DS [Puzzle Quest 2], we treat is as a separate platform. It's all the same, dungeons, and quests and monsters, everything is identical from a content point of view. But the interface is really, really a good DS interface. It's really slick.

Have you taken any lessons from the quest system? Galactrix would tell me to go somewhere, to a specific planet, and that planet would be within some system, and I would have no idea where it was.

SF: Yes. That was particularly a problem with the DS. I believe Xbox and PS3 skewed that one. We were actually quite good at pointing the way to the planet. But we really didn't have the computing power to do that on the DS.

Yeah, we've learned that the player should never be confused. At any time in the game the player should always know absolutely what they have to do next. And you'll see all sorts of little tricks in Puzzle Quest 2. There's a little sparkly path. It always leads you from where you are to where you should be going next.

Everything you click on, there's always this little bouncing animation, and question marks that let you know exactly what you have to do next. That even takes into account some core stuff like your portal system, where players can portal around quickly to move around the world fast. Something that Galactrix lacked was you'd quite often need to travel right across the galaxy, but you'd have no way to do it apart from moving all the way across the map.

So an amazing number of lessons we learned from that game that have all been rolled into Puzzle Quest 2.

Puzzle Quest: Galactrix

It's interesting that it seems like you have the ability, through having branded Galactrix as not simply "Puzzle Quest", to make Puzzle Quest 2 and continue rolling on the goodwill of Puzzle Quest without any kind of feelings one way or the other about Galactrix.

SF: Yeah. That's certainly the case. But it wasn't our intention when we did Galactrix, interestingly. We did it because I'd been doing fantasy games for about 22 years at that stage. I really just desperately wanted to make one with spaceships in it. So I think that was a lot of the reasoning behind Galactrix. We believed in it and really wanted to do it, get the sci-fi game done.

And then we were like, "Yeah, let's get back to the knights and the dragons and wizards again. That'll be cool."

There have been games somewhat similar to Puzzle Quest before -- dungeon-based RPGs that had a puzzle battle mechanic. But Puzzle Quest really brought it to popular gaming culture. Most of those other games hadn't made it out of Japan.

Since then there have been a lot of other games hopping on the bandwagon. Have you played many? What do you think of this phenomenon? Does it surprise you?

SF: I'm not surprised some of the original ones didn't break out. There's a fundamental difference between them and us. It's that we approached them as fundamentally a puzzle game and secondly an RPG, with Puzzle Quest 1.

So the RPG just became a delivery mechanism for puzzles. I think some of the others first were an RPG and second a puzzle game. Ultimately the game has stand on its own as an RPG and stand on its own as a puzzle game. And I don't feel some of the early ones did that.

Now it's interesting with the ones that have come out since. A few people have been taking a shot at it. And I think people might have underestimated the amount of work, because it really is a whole amount of work that we would put into a large RTS game we worked on in the past.

It's almost like creating a full RPG. There's so much in these things. And it's really easy to look at them and go, "It's just a puzzle game with some RPG stuff and we could churn one out pretty quickly." And I think having been through Puzzle Quest 1, we knew how much work was involved, and that's why we can come into Puzzle Quest 2, and we can come in there strong.

In terms of the sort of streamlining that you've done for Puzzle Quest 2, is that based primarily on the intuition of the team, or was there also usability testing?

SF: We do a lot of usability testing. We really need to know what hardcore people think and what casual people think -- so a lot of people testing all the time. We're constantly streaming people into the studio to have a look at the game. And we have the design team sitting around and watching them play, and watching the facial expressions, because the face never lies. So it's a constant process. And it's very important.

Intuition from the teams? It's an important one too. We'll feel stuff's right or wrong. And we'll also listen to the fan base very, very strongly. The people who played the first one had a lot of comments for us.

And my job, interestingly, is to kind of sift and take the usability stuff, and take the team's intuition, and my own intuition, and take the fan requests, and kind of consolidate it all together and come up with something that can maintain those contributions and is a lot of fun to play.

How do you balance adding depth against adding excessive complexity for the audience when you're layering new mechanics in Puzzle Quest 2?

SF: It's very important with almost every game to keep it very accessible and keep a lot of layers of depth there, so people find it replayable and so that hardcore people can enjoy the game also. And that's definitely something we've kept in mind here.

I think if you go into pre-production the very first day with in mind that every system you design and think, "Okay, we're doing the system, how do we make it accessible and how do we make it deep?" If we think about every single system like that from day one, you make them like that. That's going to be your mission statement.

But if you ever try to do anything quick, or put more accessibility over depth, or depth over accessibility, that's an easy trap to fall into. It's always got to be given 50/50 equal balance.

Is it a difficult as a creator of more hardcore games to keep in mind a casual player -- the "my player's not me" situation?

SF: Yeah, I think it's something every designer battles with. I suspect most designers are very good game players and we have to occasionally step back and watch the usability testing and not be frustrated that people don't understand what we're trying to convey.

We just have to look at it and go, "Okay, they didn't get it. We've got to fix it." You've got to be very objective about that. Over a number of years you develop a very thick skin, so that you don't get insulted. "Stupid players, we know what's good for them!" That attitude has got to be wiped out. I'm very, very strict with the design team, that we never ever exhibit that behavior at all.

The good thing is, we've got a great design team, four or five designers on staff, and we have kind of a hive mind in many ways. Everybody contributes and everybody keeps everybody real. And it seems to work out.

Have you ever felt there were instances where you felt there was something you were pretty sure people would enjoy but they just weren't figuring out to get to it, and maybe you should reposition to guide them on how to enjoy it?

SF: Yeah, I think that happens all the time. We have mechanics that we intuitively know they're good and we just sometimes need to figure out how to get the player to them and to understand them.

The classic case was the way the gems fell in in Galactrix. I still think it's a very good mechanic, I just think we needed to convey it better to the player, and we didn't do that. There are some games in Puzzle Quest 2 that we iterated a lot, some of the mini-games.

The unlocking of magical seals in particular, that we know is a very good game, because the hardcore people enjoyed it very, very much. What we needed to do with that was iterate through it enough times to get the mechanic in a way that could be explained to the newer players and introduce them with their first view of this game in a very easy, non-confrontational manner.

And we did that. It took us a lot of times to do it, but we did it. And it's one of the favorite games on there.

You've moved from kind of having nodes on the map to a map that has people and events on it that are not marked by a specific point. Obviously, as you mentioned, there's the golden path. Does it change the design? And does it change user perception? Is it perceived, perhaps, by the players, as less linear that way?

SF: Yeah. The fact that a player can walk into a room with three doors just gives a perception of non-linearity. And in a sense it is non-linear, because I can go through any door and I can explore bits of the dungeon in any order that I want. And that's great for people who are explorers and don't want to follow the path.

But I think it's really important that the player is never confused and has that little golden marker on the path that tells him where to go.

If you're in dungeons, will there be additional random battles because of that?

SF: The dungeons are big; some of the levels are really quite large. Certainly there areas players can go to where they're grind; where they'll find there'll be random battles, and they'll find some grind spots where they can travel around the rooms meeting some random encounters and fighting some stuff.

There will be areas where there are set encounters and that can respawn constantly which they can also track down. And there's large areas where they'll just defeat the monsters and the dungeon will be nice and empty afterwards so they can freely walk around. So we tried to give a little bit of everything for everybody in there. Certainly the levels are big enough to do it.

There's a slight shift in terms of the character art. Also, the game is now isometric. That's a relatively big shift.

SF: Yeah. Definitely the isometric shift is a big one. But in shifting from anime to a Western style of art had been another quite large shift for the franchise.

One of the reasons for that was now that we're down and we're in the dungeon and we really want to give the players a sense that they're in this damp, dark place, that there's danger around every corner and it's very tense.

The clean anime look wasn't really going to fly with that. To get that in there, we really needed a grittier look to the dungeon.

It's not horrifically gritty, or really disturbingly gritty, or anything like that, but it is a much more earthy appearance. And the Western art style lends itself nicely to that, so we shifted.

Talking of casual mechanics and things like that, have you considered bringing some of these ideas to the Facebook platform?

SF: Yeah, I think everybody's interested in Facebook right now. It's certainly crossed our path. It's something we're just evaluating. But we just want to get Puzzle Quest 2 polished up and done and out the door right now.

What about the older style RTS stuff that you guys used to do? I'm sure you get this question a lot, but is that something that you would return to at some point?

SF: We are constantly looking do something like that again. We love the old RTS games and we love turn-based strategy. We love those grand strategy type of games. We've got the Warlords universe there to play in still. And we're very certain we'll get back there and revisit that one day.

At the moment, we're having fun with the puzzle game. So we'll keep pitching those in the background and I'm sure that eventually one day the nostalgia will kick in at the appropriate level and we'll get that going.

It seems like that nowadays, that kind of thing has to be more of a labor of love if you want to get it out. Because it's not exactly the kind of thing that people are chomping at the bit to fund nowadays.

SF: Doing a PC RTS game is -- I think the days of that have passed for us. Particularly because the Battlecry RTS games are pretty big. But getting an RTS on console platforms or onto handheld platforms, there's a lot of interest there for us to do that because [only] a few games have done it.

Halo Wars is a pretty damn fine attempt; I've enjoyed that a lot. Battle for Middle Earth is another one that worked pretty well. And the Kingdom Under Fire series played pretty well. But I still think there's something there that could be done to make it even more accessible. I'd love to go exploring that one day.

I actually just heard a fellow, he's from Backflip Studios -- they do iPhone games -- talking about how he thinks 2010 is the big year for turn-based. I don't actually know where he got that idea from, but it's an interesting sentiment.

SF: The interesting thing I think, here, is that every year since about 1985, has been the big year for turn-based. I think turn-based games never die out. They come in many guises, from a Civilization-type game to a Bejeweled-type game to Puzzle Quest, which is really turn-based.

Turn-based is always going to be around. I think it's a very good way of playing a game. Some games you cannot play -- take 10-pin bowling for instance, you can't play that real-time, it has to be turn-based.

Have you had to scale up your team at all? Puzzle Quest looked like a small game that made it big. Puzzle Quest 2 looks more like there was some serious visual and UI design polish put into it. Did you have to grow at all for that?

SF: From the time we did Puzzle Quest 1, I think we were about 10 or 12 people, and we're a company upwards of 40 to 50 now. Not all those are on Puzzle Quest. But I can say as of today every single of those people is working on Puzzle Quest at this moment.

So, yeah, obviously the company has grown. It's grown and we've been able to throw a lot of that experience we've picked up into Puzzle Quest 2. It's just been awesome to work with a bigger team and being able to polish stuff so much more, and adding so much art.

Did you expect Puzzle Quest to be such a bolstering thing for the company?

SF: We didn't expect that when we decided to do the first prototype. When the first prototype was up and running and we played it, we knew we had something really, really good. It only took us one month to put that together and we were like, "This is really going to be great."

That's a really good feeling when you do a game like that, and you look at the prototype and know it's going to be awesome. And it's just a matter of polish, polish, polish to get there.

Yeah, it was good because the very first day that we decided, I was the only person that believed it was going to be the game, and fortunately I'm the boss of the company so I get to make the final call. As it turns out it was a lucky one. If not, I'd have a lot of people chasing me down with pitchforks and torches for making a bad call. But I guess I just got lucky. [laughs]