Monday, November 11, 2019

Killer Queen Black Interview with Liquid Bit



An interview between Darrel of The Nintendo Jump Podcast and Liquid Bit’s Adam Walters and Mike Chorak.  Liquid Bit is the developer of Killer Queen Black.


Darrel: Today, I am super excited to be joined with two members from Liquid Bit, Adam and Mike.  How are you guys?

Adam: Good!

Mike: Doing well, thanks.

Darrel: For anybody who’s not aware at this point, Liquid Bit is the developer of Killer Queen Black which is, as you guys know, Nintendo Jump Podcast’s Game of the Month for October.  So, all this month we’ve been talking about it and this is going to be linked to our full review of the game itself and its final breakdown.  To jump in quickly, can you guys talk about your involvement with the game, like your different roles?

Adam: Sure.  So I’m the CTO, Mike is the Chief Architect, but we’re a studio of three.  Matt Tesch is our CEO and does all the production and coordinating between all the different people, like contractors, ourselves, and features.  Mike and I are primarily in the code developing.

Darrel:  So you guys are going to be the ones that will be able to get into all the balancing and how we actually got this game.  That’s great.  Can you tell us a little bit about Liquid Bit, like how did you guys come to be?  What did you work on, that kind of thing?

Mike: Yeah, that’s an interesting story, actually.  It kind of goes back quite a ways, but to give you the skinny of it, the three of us (me, Matt, and Adam), we used to work together at a company called ÄKTA.  I actually worked with Matt at a company before that called Fueled.  We merged with ÄKTA, and Adam came on.  Our company got acquired by Salesforce so we’ve been working together for a long time.  We used to build mobile apps, actually. We were building iPhone apps, and one of our clients from Texas was like “hey, I’ve got some capital, I want to build a video game.  I don’t know what you’re doing at Salesforce, but if you want, let’s start a company and see what we can make of it.”  So that was three years ago.  And it’s like a really weird story.  This never happens, right?  But yeah, we did.  We put in our two weeks notice and started this.  Been doing it ever since.  I went to school for game programming, Adam didn’t have any game development experience, but we worked on a lot of stuff together and we were confident that we could do something that wasn’t a full AAA shooter or something, you know?


Darrel: Which by the way, thank you!

Mike: Right!  We’ve seen enough of those!

Darrel:  That’s really cool.  It’s not every day somebody just approaches you with “hey, I’ve got lots of money - do something!”

Mike: I know, right?  Yeah, it was kind of crazy… it’s been a wild ride!  A lot of learning, and it took us about three years to build this.  We knew we wanted it to be some kind of networked multiplayer game, some kind of team-based thing.  We didn’t know what that was going to be, but we were building the tech around that to try and get it going.  We did that for a long time, we didn’t even have a game, we just had a prototype.  But we went to Logan Arcade here in Chicago, and they have Killer Queen Arcade.  A ten player, two teams of five awesome game that we played and fell in love with.  “Wow, wouldn’t it be great if we could somehow do something like that?”  We built a prototype pretty quickly and we got in contact with BumbleBear Games, guys in New York that built the Arcade.  We sent them a demo of what we had, played it, and they were like “wow, this is actually better than we thought it would be!”  Shortly after that, we flew out to New York and we did a little game jam with them and kicked around some ideas.  Then we started a formal partnership and we’ve been working with them ever since.

Darrel:  That’s awesome, so you actually had the tech built and were just looking for a game to throw it on and came across one of the coolest arcade games there is.  

Adam: Like I said, we fell in love with it.  The partnership we formed with them has been really great because, obviously Mike and I are very much on the technical side of things. We like video games and whatnot, but don’t really know that much about designing them.  And Josh (DeBonis) and Nik (Nikita Mikros) (of BumbleBear Games) are both game designers who obviously put a lot of care in that game and are really great at what they do.  We complemented each other really well in that sense.  That was a big missing thing, from our perspective.  And I know they’ve enjoyed having us hashing out the technical stuff with them, so it’s been a really great partnership and great experience.  

Darrel: So you took this arcade game that is huge, and you went through the route of putting this on a console.  What would you say were the biggest challenges to actually getting it on the console?  Was it more making sure it was stable online, or process-oriented challenges?  Can you talk through some of that?

Adam: Sure. There were definitely technical challenges.  Not necessarily just console, but when we were developing the tech beyond wanting it to be an online game.  It was really important for us to make it cross-platform.  That’s something Mike, he’s kind of been our network guy, really cranked out and built from the ground up knowing. So it’s never been like we were locked into a PC-only networking option or some proprietary service that one of the big console makers provides.  That maybe would have been more challenging to integrate between the two.  Having that from the get-go, networking is always hard, but I think we kind of mitigated a lot of that risk by knowing it up-front and always keeping it in the back of our minds.  Cross-platform was super important to us.  And so, there’s definitely challenges, especially with UI between platforms, and different hardware specs.  All of those things have always been the case.  [To Mike] Does anything come to mind?

Mike: Maybe to take a step back.  We’re talking about a ten player arcade game, right?  We want to bring that to console.  We knew that multiplayer was going to be the cornerstone of that.  We knew it needed to work well over the internet and the experience had to be just about the same as you standing next to someone at the arcade.  Some of the design choices we had to make to accommodate that are what you see in the differences between Killer Queen Black and Killer Queen Arcade.  Namely, we have weapons like a Joust mechanic where the highest player wins.  We found out very early on that basing kills on height (who is the highest) when they contacted wasn’t going to work in a network environment.  Because due to latency, I might see you higher on my screen, but you see yourself lower.  There’s always paradoxes, anybody that’s played multiplayer games knows it’s a very familiar paradox.  We knew that we needed more wiggle room, so we started toying around with the idea of area of attack weapons and things like this to give a little bit more of a buffer.  It was more based on positioning and where you were strategically, not necessarily precise pixels the “I was two pixels above you” kind of thing.

Adam: Or when you hit the button, you just know that if you get a miss, there’s times where “hey, you were just too slow doing it.”  Not something automatic and triggered.

Darrel: Right, so you basically have it checking when each person hits the button as opposed to just a consistent height basis.

Mike: Yeah, it’s still doing collision checks and things like that but the important thing is it’s more of a volume so there’s a little bit of like… when you play the game, it does sort of feel like you did get the kill if you got it, right?  Honestly, we knew that region based match-making was going to be really important because in order for you to have a good experience, you need to have low latency.  We wanted to make sure people in the same region were playing together.  A lot of different factors went into the decisions that we made and the design choices that we made that became what you see today in Black.

Darrel: Actually, as the game was coming out, there was this overall thought of the uninformed masses of “please don’t lag, please don’t lag, please don’t lag.”  But honestly, our group has been playing on Nintendo Switch which is not exactly known for its really solid online play… I think I can fire a little bit of shots there.  But, any lag at all is reduced to one person kind of skipping around a little bit.  I’ve actually been really impressed with how the game handles lag, even in cross-nation play.  A lot of us are kind of spread out, so that’s been really cool to see.

Mike: That’s really great to hear.  We worked really hard on that, it was something that was very important from the get-go.  We were targeting about 100ms.  If you’ve got about 100ms of latency, you can still be pretty competitive.  Anything beyond that, it starts to break down a little bit.  But we still have a few more tricks up our sleeve in terms of how we’re trying to reduce latency.  We’re working with a partner, a company called Network Next.  They provide premium routes through the internet, like a private network so you pay a premium, but you have lower latency and lower jitter.  So even people that aren’t playing in the same region should get a better experience.  And through our testing, we’ve seen improvements of up to 20ms in some cases.

Darrel: Oh wow, okay.

Mike: Yeah, it’s significant.  We’re really excited… we’re coming out with that pretty soon.  It’s going to be in our next patch most certainly.

Darrel: I mean, that trends right into what I was going to get into next.  If you can, can you talk about some of the updates you’re planning for the game?  Are we looking to see any more maps, modes, or weapons?

Adam:  Yes to pretty much all of those, we group all that together as "content."  The things I can talk about - having new maps.  There’s new weapons that we’ve prototyped and want to continue to experiment with that will hopefully roll out at sometime.  Cosmetic things like new skins for the workers or queens.  Those are the main things I can think of from the content side.  The thing I’ve been working on right now is making good progression on cross-party friends.  We try to stay active with the community on our Discord, or through reddit, or whatever. To be honest, we ran out of time to launch.  So it’s cross-platform right now in the sense that everyone is put in the same pool and can compete against one another.  So if you’re on the Switch, you can go to Ranked Match, and be playing against a PC person as well as a Switch on the same team.  What you can’t do is enter a friend code and become friends with someone on a different platform which kind of prevents you from being able to party up.  It’s always been something we’ve wanted to do and we’re focusing on trying to get that done now.  That’s one thing we’re all pretty excited about.

Darrel:  That sounds awesome.  We do have some players who are playing the game currently on PC as well as Switch and we’ve had some of those conversations, too.  So that’s great to hear.  Can you talk a little bit about some of the changes you made from Killer Queen Arcade? We’ve kind of gotten into it.  Obviously there’s new weapons, it’s down to four versus four, you can throw berries and such.  Can you talk about how some of that came to be and what the design methods were?

Mike: A lot of it was that Nintendo Switch was on our radar.  We weren’t quite sure that we were going to be on that platform but we knew that we wanted to be.  So we knew that we needed to design for a smaller screen size and it was important that you be able to make out certain details in that smaller form factor.  If you look, I believe one of the major differences between Arcade and Black is the size of the characters with respect to the size of the platform.  I think we’re the opposite of Arcade, actually.  The characters are really tiny in the Arcade, the platforms are big.  In Black, it’s kind of the opposite and I think part of the reason for that is because a lot of our weapons are very dashy, you’ve moving around a lot.  It’s zany, especially with the queen, right?  We didn’t want the platforms to get in the way of that so we’re trying to create a lot of negative space to allow you to move around and not get frustrated or clip corners or things like that.

Adam: The other thing, and this is kind of one of the first things we came to a conclusion on when we went to New York.  You know in the arcade, you’re inches away from a giant screen in front of you, so those smaller characters are easier to make out.  But whether it’s on the Switch like Mike mentioned, or you have it in docked mode, or you’re sitting back on your couch, those things become a lot harder to see when they’re not right in your face like they are in the arcade.  So that’s why the map is a little bit smaller, characters are a bit bigger, tweaking between those two things.  Other than that, I don’t remember how, I think it was just through lots of game testing, we made things really easy to change.  We just have a config file that you can live update so you can tweak things like the speed of flight, speed of the runners, tune the jump perimeters.  The kind of pace of the game, I think, largely came through trial and error and iterating over and over again.

Darrel: Oh, nice.  I was going to ask, with all these changes, how was that balancing act that you guys did?  The arcade game is extremely well balanced, but I would say the console game is, too.  How did you accomplish that?

Mike: You know, that’s truly a testament to BumbleBear Games, to be honest.  We worked everyday with those guys, not side-by-side but telecomming, right?  They were heavily involved in the design process, we did a ton of play testing.  We knew, through the beta, we did a lot honestly, we put it in people’s hands, we watched replays of the game.  We saw what was working and what wasn’t.  I remember one thing in particular, I think as a gag, I added berry tossing.  I went “hey guys, check this out, this is kind of cool!”  And it made it in, right?  Like, the repercussions of that trickle.  Okay well, runners can just toss berries now.  If you can get really good at that, you can just rush through an Econ win. Then it became overpowered which de-balances the whole game, so we were kicking around almost removing it because of that.  We said “do we need to take berry throwing out?  This is getting out of hand.”  And then Josh DeBonis worked with Nik on it, and they decided to nerf the speed of the throw, I believe.

Adam: We used to have a longer cool down, I think.

Mike: Yeah.

Adam: And I don’t think anyone really likes the feel of the cool down, so we wanted to minimize that as much as possible but not make it zero so people can just stand on berry piles and throw them out.  I mean, they still kind of do that in a way today, but it’s not as bad as it was at times.  That and I think he moved a lot around, that’s why there’s berry piles hanging vertically, hanging from ceilings…

Darrel:  …so you have to jump.

Adam: Yeah, so you have to do that jump to get it.  So that was kind of a cool way to not have as bad of a cool down feel but still delay that ability to clear out a pile and just throw them all on the ground.

Mike: And what’s funny about it is it’s one more concern for the game designer when they’re creating a level.  I can’t place a berry pile in the direct lane to the hive because people can just spam throw them in really quickly.

Darrel:  Except for that one map where you put that very alluring pile right between the two hives.

Adam: Yeah, Tally!

Darrel: Yeah, that one always is a lot of fun.  And one thing that we have seen from berry throwing is teams get really good a throwing to each other.  You’ll have one person stationed in the hive making sure all the berries actually make it in.  It’s really fun.  So I gotta ask.  You have bots in the game, so you can fill in spots with hyBotenuse and other awesomely named bots, how did you go about making the AI because they seem really good.  Was this a big trial and error process, did you have any funny moments where they were doing something unintended?  

Adam: Oh yeah, tons of those.  It still happens.  I started on a version of the AI a while back and then Josh and Nik knew a student at NYU, Ahmed (Khalifa) who kind of jumped in and helped us out and then got busy being in school.  So he got the ball rolling and it’s one of those things, you know, when you’re doing some of the pathing and it’s hard to do with line map changes as the maps are evolving.  It might expose some sort of edge case or an unsolvable issue that the AI runs into.  We’ve had all sorts of runners just standing still at times.  It’s a little bit involved.  We kind of knew certain approaches we wanted to do, but there was also a lot of trial and error to see what situations they would get in.  What might be going on inside their AI brains, and trying to figure out how to get around that.

Mike:  You know what’s funny?  So Ahmed Khalifa, shout out to him at NYU, he helped us a ton there.  I kind of took over- he had to pull off at one point.  I adopted all of that code, tried to bring it home.  One of the challenges with some of our maps - in order to deposit berries you have to do some kind of tricky moves. There’s one where you have to jump on a tiny little ledge in the middle of the hive and then you have to jump out and there’s three holes.  The only way you can get in is by doing that, and I think I had to write specific code just to handle that.  If we release a new level, our designers are working on one right now, I’m sure I’m going to have to load the level up.  Who knows what’s going to break?

Darrel: You just have workers running around in a circle and you’re like “oh no!”
Mike:  Yeah, throwing out question marks.

Darrel:  I will say especially as workers, man they seem to be doing something useful most of the time.  It’s nice to see.  I saw some of the footage of your guys and your team and BumbleBear Games at Nintendo Treehouse.  You guys briefly spoke on what it was like working with Nintendo.  Based on that conversation, you got it running on Switch really quickly, at least as an initial build, is that true?

Mike: Yeah!

Adam:  Yeah, we had to.  By the time we figured out we were going to E3 and by the time we had to be ready for it, there wasn’t much time.  So thankfully, it was a really smooth process to get up and running.

Mike:  For the most part.  We use Unity for anybody who’s in game dev or who’s heard of Unity.  So, you know, they shipped us a few dev kits pretty quickly and I just got rolling on it.  Fixed errors as they came up, but I think within a day I got it at least compiling.  By the end of the week, you could at least play the game.  But then we had a bunch of performance issues.  The Switch is amazing, I love it, but the hardware on it leaves much to be desired and even though we’re a little pixel game and the camera doesn’t move and such, you’d be surprised.  There were some issues there, but we were able to get it going and it ran pretty great at E3.  We didn’t have any issues, I was really surprised.  Save for a few minor bugs, but they were actually just game play bugs that would’ve been there anyway.

Adam:  Yeah.

Darrel:  Yeah, I bet that was a relieving feeling when you didn’t have any big crash moments or anything like that.

Adam: I know!  Especially when Nintendo came down to our booth.  Mike and I were playing and it was just like “please don’t crash now, of all the times!  Don’t crash now!”
Mike: I don’t think we got one crash…

Adam: No.

Mike: We were really lucky there.

Darrel: Can you talk about the responses that you’ve had to the game both on the Switch and other platforms?  Has it met the expectations that you‘ve set forward, has it been better?  What’s it been like?

Mike: You know, this whole thing has been really surreal.  We’ve never built a game before, and we worked with a contractor that helped us out a ton especially during the last parts of development.  He’s been working in the industry for a long time in some capacity and he’s like “first of all, I can’t believe this is your first game, this is a lot to take on.  All this networking stuff and everything, it’s crazy.”  He seemed pretty confident “you guys are going to do okay.”  But we had no idea, right?  We had no idea what to expect so we were internally like “if we could get an 8 out of 10 in reviews, that would be awesome!”  And then, October 10th, the day before we released the game, all the review sites put up their reviews and it started at 8.5, 8.6 and lowered down a little bit.  The reviews we got were phenomenal and I was almost in tears.  I was biting my teeth, so stressed out about it, but when I saw the positive response that people are giving it, that was awesome.  And we have a pretty active Discord community and the people in there are great, and we’ve been getting a lot of great feedback from them.  A lot of feature requests, quality of life things.  For the most part, yeah, we’ve been really pleased with the response we got.

Adam: And even beyond the Discord community, we’ve gone to Logan quite a few times, we’ve met a good number of the people in the Killer Queen Arcade community.  They travel all over the country going to tournaments, a very tight-knit group, and you never know how a game that they spend so much time and energy and love on… how they’re going to respond to it.  They have their own Killer Queen Black room in Slack and I’ve talked to some of them and they’ve all just been great.  You know there’s clearly differences, I always tell people this is an evolving project for us.  We’re not done, we’re going to keep adding stuff and keep trying to listen and provide our feedback to others’ feedback.  But the Arcade community’s been great, so that was really nice to see as well.

Darrel: Yeah, I bet that was relieving, too.  That they’re not like "aw man, this is not my Killer Queen.”

Adam: Yeah, right?  That’s always what we wanted.. to kind of feed off one another and cross-pollinate.  We really hope that people maybe get exposed to Killer Queen Black and then decide to check out the Arcade because they’ve never done so before.  And you should, because it’s an awesome game.

Darrel:  It definitely, definitely is.  One of the differences from the Arcade is some of this background art in the stages which is just jaw dropping.  It’s really pretty in this game.  Was that kind of something that you guys started with or was it just something that was added later on in development?  How did that art come to be?

Mike: I think that’s another follow up answer to a previous question of what some changes are when you go to console versus arcade.  The arcade, you get away with very little.  You don’t really need flashy, fancy graphics and the gameplay speaks for itself.  We knew to be appealing and maybe I’ll say “competitive” in the industry we’re trying to break out into, we knew we needed to up the game a little bit in terms of presentation, graphics, and things like that.  We were very fortunate to come across this amazing artist whose name is Blake Reynolds and he did the majority of the background art and stuff.  He just did a phenomenal job, and is continuously impressive in the work he’s put out.  I mean, most of what you see there in terms of gameplay art is his doing.

Adam: And he did a lot of the concept stuff, too.  He would look at pictures of bugs over and over and get ideas from that.  It was almost too technical, but it really paid off.  Even the gates.  A lot of people don’t realize when it closes, it gets this mantis or insect head that it forms into.  That was all him coming up with some concepts.  Mike and I weren’t always involved in those calls, but bringing Nik and Josh and Matt, they would have an art review that would tweak stuff.  He would live edit stuff in Photoshop as he’s going… yeah, he’s absolutely incredible. 

Darrel:  Similarly, you also have an awesome soundtrack.  The one thing I’ve really grown to like about the game is how the soundtrack is developed in the sound effects of the game.  So as the workers promote or get berries, it all kind of forms this musical soundtrack.  Were you guys ever tempted to actually have music playing in the background of the matches, or was this always the design?

Mike: Great question.  To be honest, I’m not entire sure why we decided not to do music during gameplay.  Adam?

Adam:  Yeah, I’m pretty sure that was another thing, similar to the art calls they would have.  They would also have audio reviews.  That was definitely something that people had different opinions on.  Whether they should or shouldn’t, but I don’t know a definite reason why or who “won.”  Well “won” is the wrong word, but something that was debated for quite a while.  We decided to go with this and sometimes background music can get a little distracting and with so much going on in the game, picking up on cues really helps you, I think.  If you turn down the sound effects and turn up the background music, you might miss some of the clues you’re trying to pay attention to.

Darrel:  Right, I’ve noticed that as you play the game, you can tell what’s going on without seeing it, which is neat.  I like the way it’s done, personally.  I’m going to get into some fun questions, here.  What would you say is your favorite win condition in this game?

Mike: Oh, good question.  I mean, how could you not say Snail, right?  I mean, it’s so rewarding.  Especially when you sneak in and they have no idea.  I’ll even get a little cocky, I guess, and start throwing “HAHA’ emojis before I get there.  But I think I’m more military.  I always played first-person shooters and stuff growing up.  I was more action/Rambo based.  Go in and do some carnage.  I’d probably gravitate more towards military.

Adam: I guess I would say Econ.  Three quarters of military victories are because someone put the queen in a situation they shouldn’t be in because they added pressure from the berries or from snail.  But I like the teamwork aspect, and I love some of the battles that happen at the entrance of a hive.  Or even blocking that as a warrior is such a good feeling.  You’re just stopping them from getting that last berry.

Mike: Yeah, what’s interesting about military victories is if you ever watch BumbleBear Game’s BumbleBash, a yearly tournament they hold for Arcade, there's gameplay of their top-tier players and those games can go on for fifteen minutes, no joke.  Our games are what, usually?  A minute if we’re lucky?  Usually really short, and I think a lot of that is because you give the queen such a powerful weapon.  She can dash and dive, and is all over the place, but players don’t quite realize how precious she is and they kind of just go all out.  I’m interested to see how the meta evolves there.  If people start playing more of a conservative queen style, see if the games start getting a little bit longer.

Darrel:  For sure, and we actually had a little poll on Nintendo Jump Podcast’s Twitter.  It was up for about an hour.  Snail took 86% for top win condition, it was crazy.  Okay, so what is your favorite skin and dance for the workers?

Adam: Checkers.

Mike: Same.

Darrel:  The robot?

Adam:  Yeah!  I mean, robot aside, I’ve always liked his outfit.  We didn’t add that until kind of late in the dev process.  It’s one of those nice to have things.  I’m so glad we did add it, but it’s my favorite dance, too.

Mike:  Totally.  Weren’t we just joking around and saying “wouldn’t it be hilarious if we had an outtake of Checkers doing the robot dance post-match and trips and falls off the platform?”

Adam:  Spins and gets too wound up.

Darrel:  Most people have to learn that you can make them dance, but yeah.  Checkers has been my favorite from day one.  That leads me to… do you have internal names for the snail, the workers, or the queen?

Adam: I can answer the snail one because someone asked that in Discord and I jokingly gave the first name that came off the top of my head.  Everyone kind of jumped on it so I guess it’s canon, now.  I said Winston in the snail’s name.

Mike: I don’t know if we could get in trouble for that because it’s really BumbleBear Games’ lore.

Adam:  Well, I was talking to one of the game designers Chris Wallace, and we decided they’re different snails between the games.

Mike: Ah, okay!  Yeah, that works.

Darrel:  Fair enough.  I’m going to start using that.  Do you have a favorite snail pun or joke you keep coming back to with this game?  Because it can get kind of meme-y, which is great.

Adam: Huh, snail ones.  I’m not sure.  

Mike: You know, I usually don’t hear snail ones.  Do you have any?  What are some you’ve heard?

Darrel:  Oh I don’t know.  This is one of our listeners being weird.  We actually started gathering these questions for you and inevitably somebody said the same question as somebody else and we all jumped on that as “you just snailed them!”  Because it’s the opposite of ninja’d.

Adam:  Haha!  “Snailed it” I’ve said before.  That reminds me of the BumbleBash before last, we went up to Portland for it, and we brought some Switches and had the first Killer Queen Black tournament.  Me, Matt, and Mike formed a team and our name was Chuck Berry.  We liked that one.

Mike: Berry White!

Darrel:  That’s a deep one, I like that.  Okay, how many different end scenes are there, such as Headshot?

Adam:  I think they finally found them all by playing Custom Match, let me see if I can count them out.  Seven, I want to say?  I guess I could just look at the code… but I wanna say it’s seven.

Darrel:  Do you have a favorite?  Mine is probably That’s Amore!

Adam:  Yeah, that’s definitely it.

Mike: Oh, Amore, that’s a great one!  That one was actually found pretty quickly, I think it was like maybe a day or two after we launched.

Darrel: We found it on the first day by complete accident.

Adam: Yeah, we see that at PAX or something, people who had never played the game, it would just happen and it was pretty cool.  When our artist made that one, she did all the UI, all the non-pixel stuff we were talking about from our other artist.  She actually had listed out what all those chocolates were in great detail.  What’s sprinkled on them, what’s inside of them…

Mike: Caramel swizzle.

Adam: So who knows, maybe come Valentine’s Day, we’ll have those on the merch site.

Darrel: I hope so.  And just a couple more.  Are there any other games that you guys are playing in the office that you really like now? Any other influential games that are important to you?  What’s your favorite game?

Adam: Games that we play in the office… TowerFall is probably the biggest one.

Darrel: I was going to ask about that.

Adam: Yeah, I know.  Any time I see our game compared to that, it just blows my mind.  Because that’s one of my favorite games in the last couple of years.  We have an Xbox sitting at the office, so we would just fire it up and play all the time.

Mike:  Funny thing about TowerFall, like obviously our game wraps around the screen, and to be honest, that was one of the things that gave me the biggest headache developing.  I got that wrong so many times, so many bugs, people getting lost in the wrap…

Adam: …or fall off the screen.

Mike: I remember playing TowerFall late at night, and if you get one of the upgrades, the whole screen moves, like everything’s wrapping.  How did they do that?

Darrel:  Yeah, as I started playing your game, I kind of remembered back to TowerFall saying “we don’t want this game to be online.”  Because it is kind of a twitchy game and lag is actually important.  Real quick, have you guys reach out to Matt Thorson and said “hey, can we put your game online?”... you don’t have to answer that.

Adam: No, no, haven’t done that.

Mike: Wait, let me write that down.

Darrel: I think you should definitely reach out to him and say “look what we did!” Because I love TowerFall too, I just wish it were online so I could play with more people.

Mike: Yeah, for sure.  I mean, I love that game so much just because of the way it feels, the movement and mechanics.  So on point, and that carried to Celeste, obviously.

Adam:  That was probably my favorite game last year, Celeste.

Mike: Yeah.

Darrel:  That was my game of the year last year for sure.  Two more, and this is an odd question.  I apologize in advance.  What is your favorite chess piece, if you play chess?

Mike: Oh wow, um…

Darrel:  I didn’t know why this one was asked.

Adam: Mine’s probably knight.

Mike: I never played a lot of chess.  I didn’t like the knight because when I was a kid, I always had a hard time figuring out what the right moves were.  But yeah, I don’t know.  I guess maybe the rook.

Darrel: I’d probably go bishop there.

Mike: Bishop?  Yeah, bishop - diagonal moves are pretty all right.

Darrel: And one final question for you guys.  Are you reaching out to Nintendo to get queen or snail or somebody into Smash Bros?  Because I think there might be potential.  

Adam:  We haven’t done it yet, but that’s been brought up before.  It would be a dream come true to see that.

Darrel: Or at worst, a Killer Queen stage which would actually be kind of awesome.

Mike: So cool.

Adam:  No.  Who knows, maybe one day.  We’ve definitely brought that up before, how cool that would be.  That would be a dream.

Darrel:  It really would be awesome.  Is there anything else you guys want to talk about here?  Anything else you’re working on that you want to let us know about? 

Mike:  Honestly, I think all I really want to say is “thank you” to everybody who’s checked out the game.  We put a ton of work into it and to see the feedback from the community has really made it worth it, so just thank you to everybody at this point.

Adam: Yeah, the feature requests, everything.  Especially during our beta.  There were some rough builds that we put out.  Everyone that played on the beta, the game wouldn’t be what it is without you guys helping, so thank you!

Mike: 100%.

Darrel:  That’s awesome.  And honestly, thank you guys for coming on here.  I had a lot of fun hearing some of the background behind some of the decisions made.  Thank you guys so much, I really appreciate it.  I love your game.

Adam: Awesome, thanks!
Darrel:  With that, Mike, Adam… thank you so much for coming on the show.  We look forward to more updates in the future.

Adam; Great, thanks for having us!

Mike: Thanks!

No comments:

Post a Comment