Halfbrick's Luke Muscat talks Tank Turn Tactics: The prototype the studio was forced to ban
The title, which - in Muscat's words - almost tore the studio apart, was inspired in part by Neptune's Pride, a multiplayer PC strategy game described as "either the ultimate office game or an exercise in social devastation."
Using a similar approach, Tank Turn Tactics handed players action points that they could use to move their tanks around a grid in order to shoot at other players and knock them out of the game.
Removed players then formed a jury, inspired by a similar system in the reality show Survivor, to award additional action points to players who were still in the game.
In the end, the prototype testers connived, schemed, spied, backstabbed, plotted with the jury, complained, and eventually brought the studio to a screeching halt. Formally strong studio friendships were broken down and, in the end, Muscat put the project on hold.
But what did he - and others at Halfbrick - learn from the experience? We sat down with Muscat after his GDC talk to get more details about the aborted prototype.
Pocket Gamer: Halfbrick's games really fit the mold of successful mobile games. Casual play, short sessions did the desire to break out of that design scheme drive the creation of Tank Turn Tactics?
Luke Muscat: We weren't deliberately trying to break out of that during the prototype phase. We really like to put no limits on it at all.
We can go crazy during prototyping and it's not necessarily going to be for a particular platform or a particular kind of genre.
Prototyping is as much about learning as much as we can about something as quickly as possible, so we weren't really deliberately trying to break out of the mold. We just wanted to explore an interesting space, and try out something that was much more social and much more multiplayer.
We did quite a few other little games like that, but that one happened to be the most interesting one.
In your talk you said that you had worked on three different prototypes, which you abandoned, before you made the prototype for Tank Turn Tactics. Did any of your three abandoned prototypes contain any of the seeds for Tank Turn Tactics?
There was one - it was a game called Tank Turns, and that's what [Tank Turn Tactics] started off as. That was an asynchronous game, and what we were trying to do was make Frozen Synapse for Words with Friends players.
It was an extremely simple version of Frozen Synapse where you're shooting at each other with these really slow moving shells and every move you could only do one of two things: move or shoot.
We actually got that working on a device and it didn't really kind of work out. It was good, it just wasn't great.
Then we were thinking about how we could we make that massively multiplayer and leverage all the social stuff. Its why [the prototype] was called Tank Turn Tactics.
So basically Tank Turns sowed the seeds of disaster within Halfbrick...
Yes. [Laughs]. That's right.
Did you only play one round, one really long round of Tank Turn Tactics, or multiple rounds?
Just the one and we didn't even finish it.
It was really long because you only get in one action point per day. We were expecting it to take two or three weeks to play out, because we didn't want it to be that intense, short thing, we wanted it to like play out slowly with time and kind of fit in with peoples' lives, but it turns out it destroys them instead. [Laughs].
I tend to think of Halfbrick as a playful company. I've met several people from the studio now and they all seem like pretty mellow, pretty happy people. Did you realise that your colleagues had this in them?
[Laughs]. I think everyone has this in them. I think what happens is rules are so powerful that they bring out particular characteristics in people and amplify particular aspects.
I was just as shocked as anyone else.
Luke on stage at GDC
When this was all going down I was like "Oh my God, you know, we've never, everyone's always so friendly and so happy and so nice, there's something about this game, it's just insane."
After the experiment was terminated, did you ever think back and spot behaviour the Halfbrick team had demonstrating playing other games and think, "Oh, that's where that came from!"
I mean, we would be playing poker and stuff and we'll get really into it and people will be standing on chairs and yelling and excited.
I guess in that way I can see how everyone can get so into what they're doing, but in all of those examples it's just an "all in good fun" kind of thing, whereas [Tank Turn Tactics] took that all in good fun edge off and just made it intense.
It just made them evil.
[Laughs] I mean, I definitely don't think it made them specifically evil, but they were definitely responding to what the game was incentivising them to do.
How much had you played of Neptune's Pride before developing your Tank Turn Tactics prototype?
You know, it's kind of weird.
I started playing and that was where the inspiration for the action points came from, but then we made Tank Turn Tactics and then after I created that and that had started to take off, all of the heavy stuff in Neptune's Pride started to hit me.
I guess that was when I started to think, "What have we done here?"
You didn't feel any of the heaviness of Neptune's Pride before you created the Tank Turn Tactics prototype?
No. I thought "Oh, this is fun! I actually have to talk to people to play the game!" That's the state I was at, but none of that heavy paranoia or those feelings of anxiety that the game creates.
You had no warning.
No. I mean, there's a lot more similarities in the games now that I look back at it retroactively.
When I was [creating Tank Turn Tactics] I was like "They're similar because they're going to take a long time to play out, and we're doing a similar thing with action points, but that was really it. It wasn't until after, when I looked back, that I realised how much there was some common DNA there.
Speaking of common DNA, you'd seen Survivor before you designed this game. You had no idea how the jury system was going to turn out in Tank Turn Tactics?
I didn't think it would be the way it was.
I love the game mechanics in Survivor. For all of the reality TV part of it, and the fact that they're not actually, they get provided a lot of food so no people are always annoyed about that, but the actual game design of Survivor I really love.
I think it's really interesting, but you know, we were doing little things like the fact that the jury is there every day, and are present and always exerting an influence rather than at the end, so no, I didn't predict exactly how that was going to change the way that that jury worked.
But wasn't the idea of the jury really about politicking and backstabbing in the end, though, just like on Survivor?
It definitely was in the end.
I didn't think so much about how the jury was going to be influenced once someone was within the jury. I was thinking a lot about the juror's main interaction ending at the point that they exit the game, and not the fact that there'd be this whole other metagame that would exist within the jury system.
I was really thinking more about "Oh, this person killed me. I'm not going to vote for them." And that was kind of it.
We designed this prototype in a day and a half. We spent about ten hours on it and obviously I hadn't considered even almost all the possibilities.
Why didn't you try changing the length of the game once things started to go awry?
We thought about that, but there was this interesting other issue.
Timing became an interesting thing in the game. I was originally giving everyone their action points, but then this thing happened where I would walk around the office and give out the action points all at once, and people would complain about the order that they got the action points in because they were trying to pull off some very time-sensitive maneuvers.
So then I had to do this thing where I'd randomly generate the order that people would receive [action points] in, and different time blocks, and it just kept getting crazier and more out of control.
Everything I tried to do to keep a lid on things wasn't working.
I wanted to try a time-accelerated version where we'd play it and you'd get an action point every two hours, and I would think the whole game would play out in two days, but I think that two days would be literally 100 percent of the time [of the people] playing the game. I was really hoping that it would fit in and around work but...
Not so much.
No, not so much. [Laughs].
Did you ever think about just running it as a game where you sit down and everyone takes their moves until it was over?
We thought about that. I guess the thing that I was worried about was that the most interesting stuff that happens is the politics, and it's difficult to do it if everyone's in the same room.
It probably would work in a fast version if it was on a device, if everyone was on their iPhone. I would love to get a basic version of [Tank Turn Tactics] running on an iPhone to find out. I think that'd be really interesting.
So it sounds like, to a point, you were kind of asking for it.
[Laughs]. I definitely feel like I went into it quite naively, but getting a prototype done is all about getting it done fast. "Okay, lets do it and see what happens." And then that's what happened.
I definitely feel like I was probably a little naïve going into it and, you know, like I said in the talk, the best thing about it was how much I learned by looking at myself and understanding what happened and understanding my own motivations in making games. Its definitely been a cool learning experience for me.
Why did you wait so long to shut the game down? You knew by the fifth day it was disruptive, but the game went on for four more days...
Every day I was like "Oh, this is the peak. We've hit the climax and its going to drop off," and then every day it would accelerate further.
We did talk about shutting it down. "But it feels like its going to end and well learn so much through it ending," we'd say.
We did eventually pull the trigger but yeah, we could have at the same time we may not have learned as much if we didnt let it run as long as we did, but hindsight's 20/20, right?
If you're doing the damage anyway you may as well get as much as you can out of it, you mean?
Are there any long-term animosities in the office as a result of prototyping Tank Turn Tactics?
Not severe in any way, but I think there probably are a few relationships that will never quite get back 100 percent, they're probably at like 95 percent, but the game really was upsetting and really a powerful experience for a lot of people.
It's infamous within the office. Sometimes well be talking about it at the lunch table and someone who's a new employee who wasn't there for it, it's like "What is he talking about?" and everyone's like "Oh my god, man. This is what happened."
It's definitely an important event that happened in the office. Everybody knows about it. Even the CEO Shainiel is like "Oh, that game just tore us apart. What was going on? Its just so crazy."
Were there any long term positive effects in the office among the employees as a result of this experience with the prototype?
I think we all learned a lot about the power of social dynamics. We're doing casual games, we're doing games on mobile devices. Social mechanics are important, right?
They're very powerful and very prevalent in a lot of mobile games, and we learned a lot through those two weeks about the power of [social mechanics], and how responsible you need to be with that power. The fact that this whole thing happened is a net positive for us.
Those are all long term positive effects in terms of learning about design, but have you noticed any long term positive effects in terms of personal relationships between players?
I haven't noticed any. That's not to say that they don't exist, but no, I haven't noticed any.
Like I said, this game definitely was remarkable for the intensity that it brought out in people, and that was mostly manifested in a reasonably negative way. [Laughs].
Thanks to Luke for his time.
Dennis Scimeca is a freelance writer from Boston, MA. You can follow him on Twitter at @DennisScimeca.