[GA2] Making a Game Out of Tower Defense

I have issues with many of today’s tower defense games.  They don’t have a lot of depth to me: figure out your strategy to handle all the waves, upgrade as you get money, and watch the game go by  for a long, long time.  It seems the basic strategies that work on one map will work on all of them:  zig-zag the towers to get the maximum path around each one, put your heaviest hitters and the ones that attack flying enemies in the center path to cut off fliers, the list goes on… When I play, I usually lose health because I’ve gotten distracted and don’t upgrade or build for 4 waves, and not because there was a weakness in my path.

In case you aren’t familiar with tower defense, the game usually involves the player being able to place stationary weapons on the board which will automatically shoot at enemies that walk by them.  The weapons are called towers or turrets and the enemies are commonly referred to as creeps.  Creeps travel across the board in groups called waves, and between the waves players get a quick breather to add more towers or upgrade them.  At the start of the game the player gets some time to put down a tower or two, and then the creeps start walking across the board from point A to point B.  As you kill creeps you get more money to purchase more towers for the board to kill more creeps.  If they get past your defenses and make it to point B, you lose lives.  If you defeat all the creeps before you run out of lives, you win!  The last bit of jargon you’ll read in here is ‘pathing’, which is the idea of placing towers on an open board in such a way that the creeps have to adjust their path to get around them.  They’ll always take the minimum distance to get to point B, so by putting towers in their path you keep them on the board and in range of your towers for more time.

Galactic Alliance is a four player tower defense game created by Vectorform.  It is the only four player co-op game we’ve seen in the genre.  We originally made the game for the Microsoft Surface and later saw the potential for it to be an iPad game as a direct port.  So we put the game out in the iTunes Marketplace, and it did well!  We took the number 4 spot for free app’s in the marketplace, made the new and noteworthy list, and topped the charts in Spain for some time.

When we were given the opportunity to revamp Galactic Alliance with new towers and new enemies, we jumped at it.  It was finally a chance to address some of those glaring issues that we have found with so many of the other games in the genre.  Mind you, it would have been really easy for us to make a Field Runners clone.  We had our different creep types that matched up with the ones in most other tower defense games, had our towers that matched up with most other tower defense games, and it is a known and well accepted experience.  Tie that with our stunning graphics and four player co-op, and we’re good to go.

We wanted to do better; we wanted to create an engaging, immersive game play experience that people will want to come back to.  As we started to get knee deep in the project, the team took a step back and tried to take the game in from varying perspectives.  We asked ourselves the basic questions again:

• What kind of experiences do mobile gamers like to play?
• How long can a game play session really be?
• How can we add variety and interest into the game play?
• How far can we push the new mechanics we’ve discovered?

There were some epic debates, (yeah we get like that, especially Kevin and I), but we came out of it with some solid agreements.

I’d like to start by discussing some of the new tower types that we’ve added, because they are the key to adding variety in our game play.  The goal at the outset was to come up with completely distinct functionality in each of the other towers.   For example, instead of the typical one shot per enemy per recharge, we’ve added in a tower that does area of effect damage (AOE), a tower which will boost other friendly towers in its range, and some really sweet looking towers that work like an electric fence.  We’ve also added in another tower that does nothing.  That’s right, it just floats there.  It’s my favorite.

Each of these towers has its own advantage and disadvantage against certain types of creeps.  For example, the AOE towers are phenomenal when it comes to creeps that travel in clusters, but fast creeps skate by them with barely any damage.  The Arc towers (the ones that act like electrical fences) can do extreme damage, but they’re tricky to place optimally and thus potentially costly to set up.  And the asteroid that chills and does nothing?  It is a super cheap way to setup a path to maximize the damage that the big towers are going to dish out.  The asteroids are easily replaced for real weapons as more money is generated, but you can get as many as 3 additional passes by your strongest towers if the asteroids are utilized properly.  Combine that with some well-placed boost towers, and you can decimate certain levels with just a handful of actual weapons.  Of course, we’ve implemented some creeps that make that strategy fail in certain scenarios, but it sure is a fun way to win a level if you can pull it off.

We also dictate which towers are available per level in the campaign; usually players just get 5 out of the 8 possible types.  Although it would be ideal for the user to have all their tools accessible as they get them, that much freedom makes it difficult to guide players through the experience we’d like them to have.  More so, in my opinion, it makes it more difficult to teach players new game play strategies.  Taking away some of the towers that a player has been relying on will force them to at least see the value in other tower types, even if they don’t incorporate it as part of their ideal strategy.  There are some levels and modes that provide players their full arsenal as well, so the biggest, baddest levels won’t hold players back.

As I mentioned above, we do have new creep types as well.  I won’t get into detail on them in this post, but we’re definitely going beyond the grunt/speed/tank/flyer combos.  Instead of having waves of enemies try to brute force their way through your defenses with speed or crazy HP, these are designed to outsmart the most common strategies and make players think about the moves they’re making.  One pathing strategy won’t get you all the way through the GA2 campaign.

I’m going to wrap up here for this blog, but in my next post I’ll talk more about some of the different modes that we’re adding, and crucial player feedback we discovered in testing.  I challenge you to find the similarities to our Global Killer post.  Can’t wait to hear your thoughts!