Defense Grid Interview

07/28/2011 by Brad Cook | Source: Virtual Programming Blog
Defense Grid: The Next Level in Tower Defense28.07.2011 Written by Brad Cook

(This article originally appeared on MacGamer last November.)

When it debuted for PC and Xbox 360 in December 2008, Defense Grid: The Awakening offered a unique twist on the tower defense sub-genre of strategy games. Virtual Programming published Defense Grid for the Mac this past July.

Set in the distant future, the game puts players in the role of a nameless character tasked with reactivating a long-dormant defense system and stopping invading aliens who want to take the power cores that keep the system going. Only one power core needs to remain to progress to the next level; if the aliens take all the cores, the game ends.

Many of the levels feature ruined military bases and other decaying structures, and the storyline suggests that the aliens have been to the dead planet before. The computer controlling the defense grid talks to the player as the levels progress, revealing more and more about the story in the process.

It burns! It burns!

We sat down for a chat with Jeff Pobst, co-founder and CEO of Defense Grid developer Hidden Path Entertainment, to learn more about the game’s development.

Q: There are many tower defense games out there. How did you seek to differentiate Defense Grid from the others?

A: We started on Defense Grid in May of 2007 and at the time, we did a study of tower defense games. We found 84 different flash games and mods to great products like StarCraft and WarCraft III and felt that while tower defense games were pretty well covered in the domain of the amateur game developer or mod maker, we didn’t see anyone bringing this addictive and fun gameplay to the professional market at that time.

We basically asked ourselves what would the definitive tower defense game look like, and as we talked about it more and more we saw a vision of something that we felt should be made and that we thought we could do well. We focused on having just the right balance of features and options that we had seen in other tower defense games (often less was better in our opinion), but then we asked a lot of questions about the emotional curve that one experiences as they go through a tower defense level or as one progresses from level to level.

We pulled from our past experiences making AAA games and realized that we didn’t see anyone out there spending time on strong intentional support for the emotional arc locally or globally, the balance for many different kinds of players, progression, immersion, visualizations, etc. that are a part of making a large retail game, and we decided that we should make an effort to fill that void. Also, the new channel for downloadable games on the consoles and the increase in popularity of PC download channels like Steam opened up new opportunities where we felt that we could really deliver something different than what others were doing.

We felt there were a couple of opportunities as well for some new changes in the tower defense experience. Prior to Defense Grid we found tower defense games that were either all open-map experiences or all fixed-path experiences. As we looked deeper we found that the learning curve for fixed-path was much easier and opened up the gameplay to a larger audience. We agreed, however, that open-map presented a more creative challenge and allowed one to really get deeper into the gameplay. We decided to mix both map types in our game and actually have fixed paths connected to open areas in interesting level designs, and over time we allow the player to learn some of the nuances of playing in both types of systems. We found that this hybrid-map approach seemed to bring something new to the table.

I think the newest thing we brought to the table, though, was the idea that the incoming enemies were there to “steal” something rather than just trying to get from point A to point B. This had several positive impacts on the game experience. For one, the emotional curve was now multi-peaked. There was an emotional increase as the aliens got closer and closer to your cores (similar to when the aliens would get close to the exit in other tower defense experiences), but then there was a second emotional experience from the time they actually got a core (and the music swelled up more), to the time where they would or wouldn’t be prevented from escaping with that core.

Kill the thief!

There now was a new mid-game experience of stress and response where the player knew a bad thing had happened, but it wasn’t the end of the world, and there were steps they could still take to prevent it from hurting them further.

In addition, the “handoff” mechanic of the aliens being able to pick up dropped power cores (dropped when other aliens were killed) turned out be a great gameplay experience in that it helped keep the map fresh, preventing easy “front-loaded” or “back-loaded” solutions. This was because the main hot spot on the map where you need to prevent enemies from escaping could move around to different parts of the level as the cores were stolen and then passed off to other aliens when an enemy died.

One other great thing about the power cores being part of the game is that by providing 24 possible things to be stolen, and setting up our rule that you could progress to the next level as long as you kept one of them, the levels could be self-balancing for different players of different skill sets.

Novice players saw the cores as “lives” and worked hard to keep that last life in order to be able to continue through the game. We could balance this to be difficult to do appropriately for novice players, but not hugely difficult or impossible as many tower defense games seemed to be at the time. Meanwhile, we could keep the more serious players engaged by awarding achievements or medals for retaining all of the cores at the end of a level, which was much more difficult to accomplish. In this way the same level had different objectives for players of different skills.

Q: Defense Grid’s storyline is more detailed than what many other tower defense games feature. Why did you decide to use a story in the game?

A: As part of our goal in making the experience feel more polished, balanced and professional, we felt that giving the player a bit of motivation and immersion into the world was important. We had all played Portal, which had come out the previous year, and saw a situation where GlaDOS was hugely engaging even though you were busy doing something else most of the time she was talking. Even the comments made by the turrets was cool and interesting. So, we decided that we wanted to take the next step and provide a story to give the player some context and a bit of meaning. Paramount to the story design, though, was that it couldn’t get in the way of gameplay, and that turned out to be a challenge.

Our initial design was to create a story between the AI and the player. The actual player wouldn’t really converse, but what we’d do is simulate as if you were “typing” on the screen asking questions to which the AI would respond. In this way we kept you anonymous allowing it to really be “you” in the story, but we effectively put words in your mouth and had the AI character respond to those typewritten words.

The original script was great. The design team at Hidden Path put together the general story arc and progression and brought a lot of ideas to the table, and then worked closely with a couple of exceptional writers, Sam Ernst and Jim Dunn, who brought new ideas as well; they helped bring the story to life and got the dialogue just right.

At the time, they had written a “West Wing” episode that was well known, had worked on some other games, and had done some very good writing, so we were excited to collaborate with them. As we were working together they told us about a TV series they were pitching based on Stephen King’s “Colorado Kid” story, and of course we now know a couple years later that they were successful in getting it made, as it debuted this year: “Haven” on SyFy.

So we had a great script, everyone loved it, the writers were very happy, we were happy, and we started implementing the dialogue and story in the game. It became quickly clear, though, that waiting for the player to type something out and then have the AI respond created a lot of undesirable waiting time. In addition, if we sped it up, or played around with different presentations, your eyes kept moving from the playing field to the text and back and it was easy to feel uncomfortable and unfocused. We’d be really messing with the timing of the game.

I’ll never forget the call we made to Sam and Jim when we said: ”Hey guys, you know that two-character interaction that you nailed so well? Well, what would you think about converting it to a one-character monologue where we just ‘imply’ what the player may be asking or wanting to know?” If you ever want to break a writer’s heart, you would walk down such a path and boy did we feel awful about it. It was clear, though, that our original idea of the player and the AI going back and forth was getting in the way of the pace and the play of the game. We could give a voice to the player, but we didn’t like that idea of it then not being or feeling like it is “you.”

To their credit, Sam and Jim simply said, “We don’t know if this is going to work, but we’re up for giving it a try,” and after a few iterations, we all stood back and were amazed. It was actually a better script with just one character. I think everyone was stunned by that. It of course went into the game in a very straightforward manner, and as we played, it all fit together perfectly, and this was with the stand-in voice acting. Later when we got the actual actor to perform the lines, we all got very excited with the realization of the entire story process and how it was performed.

Q: How did you ensure there was the right balance between available towers and the various types of aliens?

A: Our background in working on many past AAA games allowed us to take the lessons from those games in balancing systems and apply them to this game. The balancing was one of the toughest things to get right and something we spent a tremendous amount of time doing. For us that was probably the key differentiating factor of what we thought would make a top tower defense game: so many of them aren’t balanced, have only one way to solve them, or aren’t true puzzles where you can solve them many different ways. For us, Defense Grid levels had to be solvable by lots of different people in many different ways and the only way to really accomplish that was to balance the towers, balance the levels, balance the alien waves, and make it all work tightly.

The aliens become harder to kill as you progress through the levels.

It turns out, for example, that the difference between a boringly easy level and a too hard to complete level could be as little as three percent of the hit points the aliens have. So, it couldn’t be “close” and be successful the way we wanted to be successful, it really had to be much more precise than that.

With respect to the different tower types, each was designed to give the player a different play style option, and each alien was designed to prevent certain simple strategies from being all you need to do. There are of course towers that target single aliens, or those that target groups. There are those that pack a strong impact less often, or those that continuously hit for small damage over a longer period of time. There are aliens that come in packs that can overwhelm towers that just target single aliens, or those that move quickly to escape short range towers.

At the end of the day the goal was to create different towers that could be more or less effective where they were placed on the level, and alien waves that would challenge the player without causing such difficulty that players would feel that the level wasn’t solvable.

Q: What are your favorite strategies for success in Defense Grid and why?

A: I tend to prefer longer-range towers when I can use them, but they do have a downside: decoys. Just like most aliens in Defense Grid, decoys are designed to prevent the “one strategy solves everything solution” and their special power is that they can’t be seen by towers that are more than a grid or two away from them. Meteor towers pretty much never see them, and cannons often aren’t very effective against them since both those tower types have an inner radius of fire as well as an outer radius (so they can’t shoot at an alien when it is too close to them – and in this case they can’t see it unless it is close to them).

There are in my opinion a couple good anti-decoy strategies. The obvious one is to have more gun, inferno, Tesla, concussion and laser towers around, which will see them and shoot at them when they are near. Also, combining those or other short-range towers with a temporal tower to slow down the decoys is good so that they can get a lot of damage done to them while they are near a set of short-range towers. Second, command towers not only help provide more resources in an area, but they also illuminate decoys so that other towers can see the decoys – even from far away – when they’re near the command tower. So, a small killing zone with some short range towers, a temporal tower, and a command tower not only do a lot of damage to a decoy, but they also allow cannons and meteors in the other parts of the map to see the decoys and attack when they’re near the command tower as well. This typically can do the trick.

Q: How do you feel about Hidden Path’s success in finding a niche as an independent developer?

A: It’s kind of funny to us, actually. The staff at Hidden Path — there are about 25 of us right now in the studio — mostly have experience shipping retail games. Our team averages 11 to 12 years in the industry and averages something like 7 or 8 games shipped prior to working at HPE. We’ve also worked at HPE on some games you haven’t seen yet that are larger than Defense Grid and were made for publishers that have put those projects on hold or have moved on to other projects — typically when a key person we were working with has left. Someday you may get to play these games, or games that evolve from them — I hope you do, because I think they’re hugely fun for where they are at in development. But that said, we loved the opportunity to be able to put our own money into the development of Defense Grid and see it work out so well for game players and for us.

Recently we’ve been working with the folks at Valve and in conjunction with the folks there we released the first major update to Counter-Strike in six years, fixing a lot of past issues, adding achievements and stats, and features that people are coming to expect in games they play on Steam, and working to update the technology and features of the game.

At the end of the day, we love making and playing great games, and Counter-Strike is a great game. We’re proud to be working in that franchise and proud to be the creators of Defense Grid.

Q: How do you feel about the current state of indie gaming, and where do you think it will go in the future?

A: The term “indie gaming” is a funny one to me. Valve is an “indie,” Epic is an “indie,” Bungie is an “indie,” but of course so are 2D Boy, Uber Entertainment, Twisted Pixel, Frozenbyte, Introversion, etc. Indie is a studio that isn’t affiliated directly with a publisher and has some more freedom on charting their own course. At the end of the day whether we’re an indie or not, we all have a customer to satisfy — sometimes that customer is internal management, sometimes it is just the folks who will buy the games and need to buy a certain amount so the studio can keep going.

I think the thing that’s really new in gaming these days is that there are many more channels available to release your game, and depending on your budget, you can make different channels work out well for you. I think Steam, Xbox LIVE Arcade, PSN, Direct2Drive, Impulse, OnLive, WiiWare, Deliver2Mac all provide new and exciting opportunities for game developers to reach an audience without having to have the kind of financing required to make a retail game product. The prices are different there too, so consumers are more willing to give something a try, more willing to take a chance, and have different expectations than they do for the next big retail game.

As long as I’ve been in the game industry, and I suspect for much longer than that, there has been a cycle of large publishers locking in game franchises and talent by buying up game studios. As big publishers get larger and larger, they make more product, some of which works out well, and some of which doesn’t work out. To try to keep the balance sheets working for them, they then lay off a portion of their staff every so often or people leave because the culture has changed and doesn’t work for them any more. Some of those folks often form new studios, and some of them have success, and there is this constant cycle of studios being bought, publishers getting too large, new studios being formed, studios being bought up, and so on.

I think the thing that is different this time, is that the small studios can potentially make it longer on their own and don’t necessarily need to do the publisher buy-out as the main way to make their independent studio work out and survive. I credit the new channels online to reach customers and find ways to make it all work.