An interview on Full Metal Furies’ reception and where they went wrong
Rogue Legacy took a lot of people, myself included, by surprise. It wasn’t perfect, but as a debut title in the roguelike genre, it was a fantastic experience. Naturally, I was excited when the developers at Cellar Door Games announced they were working on another title, Full Metal Furies, even if it was not like Rogue Legacy at all.
I was also excited when I learned that I was going to be the one reviewing the game for the site. I played it, I enjoyed it, and felt that it earned an 8.5 on our review scale. As per usual, I like to see what other critics have thought of the games I reviewed. Turns out, it didn’t even have the four required reviews for a Metacritic score. Currently, it has nine. Two months after release, and the follow-up to Rogue Legacy has less than ten reviews.
How could that possibly be?
I’ve been following Cellar Door on Twitter since Rogue Legacy, and a particular tweet stuck out to me:
Yeah, the puzzles are almost half the game. We wanted to keep it a secret, and we succeeded! But I think in hindsight that was a really bad idea. Like… really bad.
— CellarDoorGames (@CellarDoorGames) January 24, 2018
This seemed like a very genuine tweet, which made that last line stick out even more. For those unaware, Full Metal Furies is a hybrid of brawler games like Castle Crashers and puzzle games like The Witness. The puzzles come in the form of riddle-esque monoliths hidden within the brawler levels and usually require some outside-the-box thinking. For example, you may have to push a specific button (or use a specific skill) in a certain location, or even go in a direction that looks impassable until you try. Many, myself included, were very caught up in the brawler aspect and saw the puzzles as sidequests of sorts. Turns out, this is not what Cellar Door Games had intended.
We got in contact with the team back in January for an interview about their thoughts on the game’s reception. Specifically, I spoke with Teddy, Ryan, and Kenny Lee, the designer, producer, and programmer (respectively) behind Full Metal Furies.
Below is the raw interview, somewhat truncated for length, and with some notes from me for clarification on specific items. My parts are in bold.
First of all, thanks for making such a stellar game. These days, it’s too often that I drop a game after a review is done and don’t think about it much after that. With Full Metal Furies, I’ve been back to it almost daily, attempting to solve more monolith riddles and whatnot. As a huge fan of Fez and The Witness, I’m getting that same joy. I currently have this on my desk at all times:
I even recruited my fiancée to help me out, and both of us are still scratching our heads!
Anyway, I saw that tweet from earlier and it caught me a little off guard. So I guess my jumping off point is what, in particular, makes you feel like your decision to keep the secrets a secret was a “really bad” one?
Thanks again man, I really do hope you enjoy the puzzles! As for why we thought keeping it a secret was a bad idea, it’s a long answer.
Do you know about the story with Terminator 2? James Cameron originally wanted the reveal that the Terminator was a good guy to be a surprise, but marketing revealed that in the trailer. It removed all of the surprise from that scene. What should have been monumental reveal became more of an “OK, finally, we got past the intro filler.”
We were really proud of our puzzles (especially the ones in Khomli*), so we kept that whole section of the game a secret. Like really secret. Let people have the T2 surprise. And if you haven’t solved it yet, there’s one puzzle in particular which will explain why we felt obligated to do this.
*Editor’s Note: Khomli is a late-game puzzle-heavy area that is very difficult but equally rewarding. Many players may never encounter it due to it being available only after the “big boss” is defeated. Overall, it makes up a ton of content.
This meant that a huge portion of our game (probably ~40%) was kept under wraps. This includes additional bosses, additional stages, and the true ending. This is all supposition, but I think if we were forthcoming with what we planned to do with the game, there’d be more critic and public interest in the game. I mean, you just have to look at Metacritic, we’ve only had three reviews (btw thank you for the review!). And even in those reviews, none of the puzzles were beaten, because – and rightfully so – it was all seen as optional content.
I think NieR: Automata best explains the differences. After you beat Nier, a system message from Square pops up saying. The game is not over, please keep playing. It’s straight up explicit.
We hid that stuff in marketing, we alluded in the game, and it just bit us in the butt. I think we wanted to do something too different with the assumption that when the game came out, people would give it a full once-over. But honestly, with the amount of big-name games like Monster Hunter World coming out in January, nobody has the time. We sat on this stuff for two years, so it was a mistake we kept on making. But once again, this is all supposition, and we’ve been living in our heads for these past few days.
Sorry for this super long-winded answer.
First of all, I am very shocked at the lack of reviews myself. I went to check the other day and saw just two, and was taken aback.
I admit on my part that I did indeed see everything after the first “final boss” to be the “enjoy at your leisure” kind of content. I knew there were deeper puzzles, plenty more levels, and who knows what else. I hope you don’t hate me for reviewing it when I did.
What would you have changed in the pre-release period, looking back? Would you have led with a promise of a buttload of content? Emphasized puzzle-solving more?
We don’t hate you trust us, if it wasn’t for your review we would have launched with zero reviews. But when we read the review (and an 8.5 is amazing, thank you), when we got to the part regarding the story caps at 50% bit was heavy. But we don’t blame you, it was more years of a having a sinking feeling that we made a terrible mistake, and then having it confirmed. It’s like a ten-ton weight with a sad face drawn on it, hitting your stomach.
As for what we would have done differently? It’s hard to say what exactly we’d change since we sat on it for two years. But just having that arsenal in our hands would have changed how we did interviews, press releases, even our store page (which hides this stuff). But if you wanted specifics, I would have been straight blatant with the hidden ending. Probably done something along the lines of saying Full Metal Furies is so large, that getting the first ending, is only halfway through the game. And maybe go with something like saying it’s an action-RPG with puzzles harder than most puzzle games. This isn’t what I’d actually say, it’s too cocky, but things along those lines to just get the narrative going.
Johnathan Blow did something similar with The Witness. Every single interview and mini article he did, he just constantly dropped that there’s a whole other half to the game. Everyone looked out for it on day one. Blizzard also did something similar with Diablo III, where they kept pressing the whole, you haven’t played Diablo III until you’ve played Infernal.
These examples I’m giving you aren’t the best. Honestly, we have a pretty good idea of what we did wrong, but we lost our launch window. So these past few days in between patches and such, we’ve been thinking more along the lines of what we can do now to save the game, instead of what we could’ve done back then.
To clear up my intentions, I in no way intended for that to be taken as the story “caps” at 50%, just that when I thought I was done with the story, I was only at 50% and immediately knew I had way more to uncover. Perhaps I should have worded that better. Even now I sit at 80% story completion. I guess I didn’t completely realize just how much that other 50% was, or what it really entailed. As I now get into the more meta puzzles (Khomli stuff) I feel like I do let you guys (and the game) down a bit in order to hit a deadline. I’m sure you’ll put the onus back on you somewhat, but it is part of my job as well. These puzzles are so awesomely insane, and I hope that this interview piece can help do them justice to the readers/potential buyers.
Hell, I think just likening the game to Terminator 2, NieR, and The Witness would have people rushing to the Steam store.
I know the monoliths and those puzzles are strewn throughout, but did you ever think about changing the ratio of combat to puzzles in the early stages of the game during development? I think most people probably look for (but not hunt for) the monoliths until they get the first ending, then realize that they’re needed to progress further, so they go on the hunt. I know that personally, it was the constellation monoliths that got me super intrigued towards the “other” half of the game. Then I stalled out there and went back to the first couple of worlds.
Thank you for being so honest with me during these questions, I really appreciate it.
Sorry, I didn’t mean you implied the story ended at 50%. We knew what you meant, and I think most of your readers did too. It was 100% our fault. If we dropped our ego, and just told critics once about the secrets, I think things might’ve gone differently, but we didn’t. If someone told me back then that we could either make a game that really surprised a few people, or a game that kinda surprised a lot of people, we would have changed our entire philosophy to marketing.
As for changing the pacing of the puzzles, yeah, I think if I made the game again, we probably would have made one or two of them mandatory. Pacing-wise, probably make it closer to La-Mulana than The Witness (if that makes sense). A few mandatory, with the majority being optional, but at least forcing them to engage in it. It’s tough though because even though I say that, I wouldn’t want to actually do it. My reasons tie into the narrative, and I can’t really say more without spoiling things (and I wouldn’t want egg on my face if it didn’t pan out), but there was a reason we wanted to get so puzzle-heavy with an action game, and it just ties in with us wanting to do something different. I think we tried to do too many things at once, and it blew up in our face.
It’s pretty clear that you guys aren’t happy with the state of the game on release. I’ve noticed you’re continuing to address the online stuff*, which is great, but has your (evidently) diminished disposition hurt your drive to work on the game at all? In these early stages, is FMF a “failure” for you?
*Editor’s Note: The “online stuff” that was an issue on release has since been fixed.
For the state of the game, I think that’s two different sides to this answer. Game-wise, I think we’re really happy with it. We know it’s not for everyone, we did contentious design decisions, but we are still really happy and proud of the end result. This was the first game we’ve ever made that had so many people rooting for us, outside of our direct family and friends. Like, when we outsourced the networking and porting to Blitworks, the developers were psyched for the game. They really thought it was gonna be something special, and were even asking for bonus keys so that they could give it to their friends, and play the game in-studio. It was like the exact opposite experience of Rogue Legacy. That game, we only had ourselves, and released the game in a bubble. So when this game failed, it felt like we were letting down a lot of people.
But the technical release of the game, yeah, we’re not happy. No one is, especially because of how badly it blindsided us. I mean, there were a few bugs that we couldn’t fix in time but everything else was so solid. I would put my name on it, and I did. I don’t know what happened on release day, the game just exploded technically, and everything fell apart. We delayed the game for almost two years in order to get networking in, so we thought we were well prepared.
Disposition-wise, it hasn’t affected our drive to fix the game. If anything, it’s stronger because now our pride is on the line. But it has greatly reduced the chance of any post-release content (except lobbies which we’re discussing internally right now). This is our game with our name, so every time someone bashes it, it’s very personal. There was no middleman forcing our deadlines, or our design decisions. We made them, so when people bash the networking, we feel bad. We really do, and we really, really, really freakin’ want to fix it. I mean, over the past week*, we’ve been working day and night, and almost everyone’s been averaging less than six hours of sleep.
*Editor’s Note: This interview was conducted in late January/early February.
As for Full Metal Furies being a failure? I mean I’m proud of the game, I really am. I really do think it’s something special, but then again, I’m pretty biased. 🙂
But yeah, the game is still a pretty massive failure. A lot of people had their hopes and dreams tied up in this game, and we failed to connect with the critics, we failed technically with the public, and we failed as a company to break even. In a few weeks, FMF is on trajectory to sell less then Rogue Legacy daily, so I think there’s a point where one’s personal opinions are outweighed by the commercial reality. We’ve got a big patch coming through soon*, but I’m not sure if it’s too late for second chances.
*Editor’s Note: The patch mentioned here has been out for some time now.
I’m glad to see you guys so determined to fix the game. I can only imagine how difficult it can be dealing with internet connectivity issues!
You mention that the game was delayed for almost two years for networking — what did this game look like ~two years ago? Was it really more-or-less content complete at that stage? In other words, if you released the game in 2016 (ish), would it look much different, other than being local only multiplayer?
Yeah, the game was delayed for around two years, but to be fair it was a complete disaster, and we were way out of our league. We first tried to do it internally, hiring new people and whatnot, but after two years, they weren’t able to get the project running. It’s strange that something so awful can be condensed into one sentence.
I think at that point we should have taken it as a sign to call it quits, but instead, we dug deeper and outsourced the networking to another company, Blitworks. Because we decided to do that, we had to get the game to a very nearly complete state before handing it off to them. There was a lot of polish needed and a lot of bugs to fix, but it was nearly content complete at this point. If we had called it quits on the networking, we probably only needed an extra two to three months. But Blitworks needed an extra year to get networking in, so we just kept working on the content we had (there’s nothing else we can do).
We built the game with multiplayer in mind, so the whole game had these “hooks” to make rigging networking easier. For example, we designed the lobby system to allow for drop-in drop-out, reconnection points, etc. without ever having the ability to test it. Just hoping it’d work when it was all rigged together. But you can only go so far, and the things that really bit us in the butt, were things we couldn’t have predicted. Stuff like, much of the collision system had to be redesigned because what we originally had wasn’t conducive to net-play. These change the fundamental feel of the game and also meant we had to test two versions, the one we made, and the one “replicated” for online play. So on top of network bug-testing, we had to go back and do “game-feel” testing. “Are the knockbacks the same distance? Is the wall dampening applying correctly? Is corner collision similar?” Like… Oh man, it was way too large for us.
But yeah, we still saw ourselves with a lot of extra months to improve the game, so we just kept working. For example:
- We implemented a checkpoint system.
- All of the bosses got remakes, including the super final boss. In fact, he got redone three days before the game launched, so the Xbox still has the old version.
- A couple of enemies got remade (like timebombs and hoplites).
- A bunch of puzzles got remade. (Some of the original ones were a touch too difficult, including one of my favorites. But that puzzle played way too much like The Witness than deserved to be in this game.)
- We added the musical instruments.
- We remade a few of the equipment pieces to make them more unique.
- And so on, and so on, and so on.
Honestly, I think in total, I dunno. Maybe the game is like 10-15% better? I’m happy with all the changes, but I’m not sure how much of an effect it had on our Metacritic.
If anything, I think those two extra years actively hurt sales. Aside from the fact we would have launched earlier – before indie games got so insanely competitive – but all the networking issues at launch actively hurt our user review. I mentioned this already, but the first day we started with a ~50% user review, which made Steam knock us off the front page. We got review bombed by a group of four friends who couldn’t connect, and each of them dropped a negative review before refunding the game. We fixed the vast majority of network bugs across three patches, distributed across nine days, and we’ve been able to climb up to an 86% user review. So now over half of our negative reviews are for problems that don’t even exist anymore.
I know you guys are busy busy busy with the game itself, so I’ll start to wrap this all up.
What are your future plans/where do you see yourself dedicating your time? Again, I know you’ve been hard at work patching FMF on Steam, but are you considering an attempt to give the game new life on the new “indie darling” — the Switch? Or are you more interested in getting FMF to a state that you’re comfortable with and moving on?
Our future plans right now is to add lobby support. We’re working on that now, and we plan to release it alongside our first sale to see if it can boost sales. We had other features which we’ve spoken about, and planned on doing depending on the game’s success. But those have all been tabled for the time being.
Our initial, very rough roadmap idea of future features (sort of like a Kickstarter, where the better we did, the more we would’ve probably added) would have been:
1. Lobby system (doing now – Free update).
2. FURY MODE (A new difficulty setting from the outset. Initial idea was no checkpoints, all equipment at start, player level is capped based on max mastery, etc.).
3. Procedurally Generated Dungeon and extra NG++ (raise max mastery, possibly an alternative economy for people who wanted to keep playing post end-end-game content).
4. DLC skins (reskins for the main characters. Since it’s pixel art, it’d mean every frame would need to be redrawn completely).
5. DLC character (tied into a small extended campaign bit, so that she could get her own item drops added as well).
Once again though, most of this is on hold. Probably indefinitely.
We’re under embargo regarding ports.
Guys, thank you so much for sharing your story and thoughts with me. I’ve appreciated how in-depth you went with each question, and that you didn’t shy away from anything.