Interview with Lars Doucet
Today we have an interview with Lars Doucet of LevelUp Labs, one of the two core members of the team that brought us the delightful RPG / Tower Defense game Defender’s Quest (a game which, in my opinion, was the first such hybrid to actually get the “RPG” part more or less correct).
Today, LevelUp Labs issued a surprise announcement about Defender’s Quest II’s music. Kevin Penkin, who composed the soundtrack for the first Defender’s Quest, will be returning; no big surprises there. What is a surprise is that the title theme will be composed collaboratively with game music legend Nobuo Uematsu.
Hit the jump for more details on how LevelUp Labs snagged Nobuo Uematsu; their plans for Defender’s Quest 2 design and mechanics; and the promising open source game development framework they’ve been using to make these games.
I’m Lars Doucet, I’m a programmer for Level Up Labs. Our most recent game was Defender’s Quest: Valley of the Forgotten, a tower-defense / RPG hybrid. I write a lot about the design and business of independent game design on my blog and also at Gamasutra. I’m also pretty big into open source development, I’m one of the core contributors to the HaxeFlixel open-source game engine.
How did you get into developing games?
I’ve always wanted to make games, and I got my first taste of it at a part-time high school job in at the local 3D animation shop (you know, the guys who do the 3D talking cartoons for the local car dealership commercials, et al). My actual job was making interactive stuff in Flash (version 4! It was the stone age). I used to make little (terrible) games in Macromedia Flash / Director. Years later in college, I worked on educational game research grants and the like, first doing Super Energy Apocalypse and then CellCraft. The latter project is where I met my current business partner, Anthony [Pecorella], who formed Level Up Labs with me. He’s now the lead designer for the Defender’s Quest series.
What gave you the idea for an RPG / tower defense hybrid?
It was Anthony’s idea. He works for Kongregate as his day job, so he’s played every single tower defense variant you can imagine. I believe he first got the idea for DQ when playing Final Fantasy Crystal Chronicles, an FF tactics spin-off game for iPhone [Most likely, he's referring to Crystal Defenders--Ed]. The idea sounds brilliant — Final Fantasy Tactics + Tower Defense! Except it’s not that great, the Final Fantasy Tactics bit is just a visual skin, and the rest of it is by-the-numbers TD gameplay. So Anthony kept thinking about what it would be like to really merge the two genres. Playing kongregate submissions, he’d seen plenty of games that tried to mash-up the two genres, but never in exactly the way he’d imagined. After we were done with CellCraft, he pitched me the idea, and I thought it sounded great, so we went for it.
Why Defender’s Quest II–why not something brand new? Is it a business decision, or a creative one?
It’s both a creative and a practical decision. Personally, I feel like games are one of the few forms of media where the sequel not only has a chance of being better than the original, but also a good chance. The first time around, you spend so much time building the game before you really know what it will be like when it’s done. The second time, you can already start making decisions that affect rules, mechanics, fun, etc. It gives us the confidence to build on top of it and take it in new directions, such as a completely different setting with an all-new cast of characters, and a battle system that relies more on unique units rather than generic stand-ins. All that said, after DQ2 we plan on trying something completely different.
Other than relying more upon unique characters, what changes are you making to the mechanics in Defender’s Quest II?
We’re overhauling the skill trees. A common complaint from the first game was that upgrading and re-speccing characters could get tedious, and not all the decisions in the upgrade trees were interesting. So this time around, we’re looking to make skill upgrades less frequent and more powerful. We’ll also be building more either-or upgrade paths into the process to emphasize diversity in character builds.
Equipment is getting some changes, too. In the original game, all the equipment just had a single stat boost, until New Game+ where a few special abilities came in. This time we’re going to be building special abilities in from the start so that equipment becomes a more important part of the customization process. At the same time, we’re going to make equipment less linear and disposable than it was in the original game.
Battles are being upgraded in various ways. First of all, the battle maps will be larger. Also, enemies will be able to inflict status effects on defenders, so there will be more to do on the party-maintenance side than simply throwing down some healers and keeping their HP above water. The biggest change is that each character’s top-level boost attack will now be a special “ultimate” skill with a long charge that has the be activated manually, so it’s a bit more strategic and intentional.
The interface is being overhauled, and we’re cleaning up the guts of the stats system so players don’t have to do as much math in their heads. For instance, the “speed” stat just made everything unnecessarily confusing, so we’re dropping it entirely.
A quick word on the unique characters move — in DQ1, our players created a very popular self-imposed challenge called “Hero Mode” where you refused to hire generic units. It was pretty challenging, but also brought out a lot of interesting strategies you never saw in the regular game, because you couldn’t just spam archers (or berserkers, or whatever your favorite was). Only having one of each class really forced you to think about what each one was best at and make the most of it, and it also blunted the effects of natural imbalances that favored one class over another. We’re planning on having a cast around 12, with a little bit of intentional overlap since there will be no generic units this time.
What made you decide to switch the visual style of the series so dramatically? Did you abandon pixel art in the interests of having something that could scale to fit different display resolutions more easily?
Yeah, it’s mostly a technical decision. It’s kind of hard to get pixel art to behave well in multiple resolutions, and we also wanted to have a more coherent overall look for DQ1, as DQ1 had a pretty big mismatch between cutscene art and the rest of the game’s pixel art.
You just announced that Nobuo Uematsu is going to be composing the main theme for Defender’s Quest II. That’s pretty astounding–how did you make that happen?
We’re extremely fortunate. As it turns out, our musician for Defender’s Quest 1, Kevin Penkin, had actually collaborated with Uematsu-san before. This was not the work of clever planning on our end – we had randomly discovered Kevin on youtube several years before, and all of a sudden we find out he had this amazing connection. When I found out, I asked Kevin, “Hey — I have a crazy idea…” Kevin then helped us set up the connections. He introduced us to Hiroaki Yura of Creative Intelligence Arts, who helped us make the final pitch. It’s all been very surreal, but extremely exciting.
Why is he only composing the main theme? Budget constraints?
We didn’t want to be over-ambitious with the pitch or ask too much of his time. We’re honored to have him working on the game’s signature song
Our approach was to solicit professional translations for the major languages we thought had a good chance of earning back the costs of creating them, and then support any crowd-sourced translations that happened to turn up from volunteers. With DQ2, we actually have a real *budget* and some decent numbers to guide these decisions, so we hope to do more professional translations in the sequel. As for the volunteer translations, it’s worked out pretty well. That said, I don’t recommend relying on crowd-sourced translations as a primary solution. That’s because you have *no idea* what you’re going to get from your community — don’t expect a complete French translation to magically land on your doorstep. The main reason we added the volunteer translations is because people just kept sending them to us and begging us to put them in the game. I understand the impulse, as I used to do volunteer translations myself — back in college I translated various comic books in the Knights of Dor and Miranda series by Norwegian comic book artist/author InkaLill Røsberg. If we weren’t basically breaking even on the time spent integrating the volunteer translations from smaller language groups I don’t think I’d be comfortable including them in a commercial title.
On a related note, we’ve overhauled our translation framework for DQ2, which is now publicly available as the Haxe library firetongue.
You’ve mentioned to me that Defender’s Quest has mod support, though I had no idea that that was so until you told me. Is that planned for Defender’s Quest II–and if so, are you planning to make a bigger deal of it?
Yes, DQ2 will have mod support, just like the first one did. I would like to push that a bit more if I can. In DQ1’s case, we released mod support well after initial release, and without any useful tools — all players had to work with was an intimidating stack of files and our official mod guide. I blogged a little about this in our last update post.
Basically, I’m working on open sourcing a lot of important bits of the engine, and this is the first step in making some half-decent tools available to the community. As for the technical backbone of mod support, OpenFL makes things a lot easier, because of the way it handles assets – it’s very easy to override stuff at a low level so you can swap in an alternate asset pack (for images, audio, data files — everything), which is essentially what mod support does. With DQ1 we didn’t take plan anything like that ahead of time and had to create a ridiculous gum-and-duct-tape solution.
Let’s talk about Haxe-Flixel for a moment, if you would. What is the advantage of using that over a more established tool like Unity?
So I already covered why I switched from Flash to Haxe/OpenFL (and thus as3 Flixel to HaxeFlixel) in this article. But why did I choose it over, say, Unity? Two things — freedom vs. security, and not wanting to start from scratch.
Not Starting From Scratch
This one’s totally practical. Haxe is really, really, similar to AS3. Also, HaxeFlixel is really, really, similar to vanilla Flixel. It’s probably the most mature of all of the flixel ports, and in terms of community involvement I’d say it rivals or even surpasses the original in terms of new features and bugfixes. If I picked something like Unity, I’d have to start from scratch with a whole new 2D toolkit and engine. HaxeFlixel is the easiest path for me to get my game to compile natively cross-platform.
Freedom vs. Security
This one’s pretty simple — I just jumped out of one proprietary boat, and I won’t jump into another one without good reason. If I honestly believed porting to Unity would be *faster* for me, I might consider it, but as of right now everything’s pointing me to HaxeFlixel. The only Flixel port for Unity that I’ve heard of hasn’t been updated since 2010.
What are you looking to make once you’re finished with Defender’s Quest II?
I really don’t know. I would like to take another whack at TouretteQuest, but I also have this idea for a cool way to subvert the 4X space empire genre, and I’ve always wanted to make a proper city-building game driven by a real energy economy, as hinted at in Super Energy Apocalypse. Whatever it is, it won’t be Defender’s Quest 3 — I will need a serious break from Tower Defense / RPG hybrids by then. That said, depending on interest and how we feel, there may be a DQ3 sometime in the future, just not immediately after this one.
Thanks for your time.