Here is the final entry in this series. It’s a bit of a downer. Sorry. I edited it quite a bit trying to make it less depressing. But then I realized that while it’s a downer, it’s also honest, and honesty is the whole point of a postmortem. Let’s start with Arvind’s final thoughts:
I've already written about the need to promote your game in an earlier blog post. Now, it is time to discuss the result.
Releasing during a busy period.
There is no “ideal release date” in today's market, considering that hundreds of games are released on Steam every day. We released the game about one month after finishing it, and we picked what we thought was the best release date in that immediate time period. Our finances simply didn't allow delaying the game for any longer, so it is not like we could pick the best day out of the calendar year and wait until then.
Unfortunately, it seemed we were not the only ones who thought the release date was ideal. We also had the misfortune of releasing alongside Enter the Gungeon, also a roguelike shoot â€˜em up, which became the “indie game people talk about for the next week”. Even though (a) Good Robot doesn't share any similarities with EtG beyond the genre, and (b) no website or YouTube channel shares any data with us, I feel that it probably had some effect on our hurt our promotional chances.
Not distinctive enough at first glance.
Even though your game might be a really engaging experience unlike anything else on the market, most people will spend approximately one second looking at it before making a decision on whether to play it or not. This is especially true of game journalists and YouTubers, who get hundreds of games in their email inbox every day.
This is why you need a unique selling point – a pitch in one sentence or less that is supposed to hook the people and convey what your game is all about. For example: Unrest is a story based RPG set in ancient India, or Will Fight for Food is comedy mix of beat â€˜em up, RPG, pro wrestling and whatever the hell I was smoking when I came up with the idea. Good Robot, even with the funny stuff that Rutskarn wrote for it, feels downright bland in comparison.
At first glance, all the components of Good Robot are things you have seen before. Robots, Shoot â€˜em ups, Indie Roguelikes, our art style – Steam has a lot of those right now. Once you start playing Good Robot, the game is really fun and does a good job of differentiating itself. (Case in point: Our Steam reviews currently sit at 96% positive in 114 reviews.) However, to get to that point, you need to stand out in approximately one second or less that they will spend looking at your email or store page.
Sometimes a successful hook is as simple as “next game from famous game designer”, or “reboot of beloved classic”. Sometimes it is something that you can't plan for, like “at the center of this week's games twitter controversy”. This is where marketing budgets come into play – they help you achieve so much saturation that people will eventually look at the game you're making.
For Good Robot, we used the humorous robot storyline to spice up our roguelike shoot â€˜em up pitch. I suppose it didn't stand out enough despite our best efforts, unless:
We were just unlucky.
If you are an independent creator who is self-publishing your book, film, game or youtube channel, you need a certain amount of luck to succeed. For example, my previous point about not standing out enough might have no basis in reality, simply because we weren't the one email that was read out of the thousands in TotalBiscuit's or Pewdiepie's inboxOR any of the other Youtubers â€" trust me, we emailed a ton of them..
The weird thing is that now that the game is old news, the chances of it getting covered by any famous personality are even less, which means the ship might have sailed on Good Robot. That makes me quite sad, though I will try to make another game because I'm an idiot who really should have picked a more stable career by now.
It is also worth noting that there is a certain amount of guesswork involved here, I may very well be completely wrong about these things. However, I feel like getting this out of my system will spur some discussion and help me introspect.
Some of you might have read an article or two about how running your own company can negatively affect your health. Indie Game: The Movie showed it a little bit, except they were like “and then in the end they made a bunch of money and now everything's okay!” Unfortunately, it doesn't always work like that. Good Robot took a great amount of emotional toll on me while I was making it, and after it launched.
Once we started putting the final touches on the game, it was time to begin marketing it. I knew the need for an indie game to stand out, so in order to help us amplify our reach, I decided to also pay a PR company out of my own pocket to help promote the game. Unfortunately, we were unable to secure enough coverage to earn back the money. I knew that it was a risk, but I decided that it was worth taking here.
We needed to sell ~4500 copies to break even; a goal I thought was achievable given our previous titles and Shamus' audience. Unfortunately, I made a huge mistake in that assessment, because the game has sold about half as much as of now. There is still time in the long term for Good Robot to make the money back, but for now the game was a net loss on my finances. That does not feel good.
On top of this, launching a game has a serious emotional toll attached to it. Even though the majority of the feedback was nice (thanks to everyone who took the time to do that, it means a lot), slowly the few insulting and trollish messages creep up on your mind. One such message individually is not that big of a deal, but collectively it is death by a thousand cuts for the mind. If you're doing this for as long as I have, every negative post starts hurting and seriously ruins your day.
This means that every time I release a game, I have a “post release depression” period. Usually I'm okay after a week or two, but after the launch of Good Robot and the month and half of post launch support, I don't think I have recovered as of now. I think I'll be fine again eventually, but right now it feels like my love for making games has been beaten out of me.
The reason I shared this is that the mental part of game development is often overlooked in post mortems. As in most customer facing industries, game developers need to always appear cheerful and grateful when somebody asks them about their job. I'm using this post to get this out of my system as *this* is the biggest obstacle while making games for me.
I hope this post mortem was useful for you, and I hope to see you the next time I'm making a game. Oh god I hate this business but I also love it.
The Silver Lining
I have a bit of an advantage over Arvind, since I don’t have the same creative vulnerability. When Good Robot didn’t meet sales expectations, I could comfort myself with the realization that I had other projects in the works. It sucks when a creative effort doesn’t pay off – particularly when that effort represents months of work – but I don’t depend on game development to earn a living and I’ll have many other chances to entertain people this year.
Having said that, the final sales numbers are pretty disappointing. When Arvind said what we needed to break even, I thought it would be an easy number to hit. 4,500? I nearly hit that with the Witch Watch, which was a novel (old and busted) and not a videogame (new hotness) and which had no marketing at all. And it cost about 50% more than Good Robot. I couldn’t imagine that it would be difficult to move 4,000 units. But here we are.
Earlier Arvind said that he spent some money on a PR firm, and it turned out to not be worth it. I should add that I think our PR firm did their job. Because of them, quite a few YouTubers played our game, including people like Jesse Cox. We’d never have been able to make that happen on our own.
I’m in a strange place, being both developer and member of the gaming press. I’ve seen my share of press releases and marketing emails over the years. I know what a bad marketing firm looks like, and I know what a good one looks like. As a public service, I’d like to offer the following guide:
Shamus Young’s Guide to Marketing a Game via Email
So you’re in the marketing business. People give you money, and then it’s your job to go out and make sure their game is noticed, talked about, reviewed, streamed, and eventually (one hopes) purchased by the public. A big part of that job involves sending out emails to lots of members of the gaming press.
A good marketer knows that reviewers are busy people with a lot of games to sort through. (For our purposes, a “review” is a really broad concept. It might mean a formal review with a metacritic score, but it also might mean ending up on a schtick show like Zero Punctuation, or being played by a popular streamer, or used as fodder for a “screenshot gallery of the day”. Whatever. The goal is exposure.) When offering a game for review, the pitch shouldn’t be the same thing you use on the buying public, because the two have different needs.
Here are some guidelines for avoiding the common mistakes that make me press the “delete” button before I’m done reading:
- Send it to the right address. A good firm knows the private business email of a gaming site. A bad firm just uses the public-facing “contact” email and dumps their promotional message into the email box with all the spam, SEO hustlers, and random jabbering internet crazy people.
- The message should start with the title of game, followed by what genre it’s in and what platform(s) it’s available on. This email needs to be sorted to the proper reviewer, so that info needs to go first. Don’t make the editor read six paragraphs before you tell them that this game should go to the dude that reviews mobile games and not the chick that reviews online RPGs.
- There should be a SHORT pitch that explains what makes this game unique. This should not be the opening crawl! Bad is something like, “A thousand years ago, the kingdom of Backstoria was beset by the dragon Antahgonast. Now one brave hero, a young woman named Mayne Karachter must travel to the lands of Ohpan World to discover…” Ugh. Even the best story in the world would sound tedious when distilled down to a list of proper names. Moreover, reviewers read a lot of those and they tend to blend together in the mind. A good pitch is something like “A tactical turn-based RPG with dating sim elements”. (Or whatever.)
- Include some screenshots. DON’T send them a dozen jumbo 1080p images. Stuff like that goes in the press kit. (See below.) Remember, this isn’t a sales pitch. You just need the potential reviewer to understand what this game is like. Too many images makes this sorting process inconvenient. And reviewers are less likely to be dazzled by screenshots. They have played a lot of crappy games with exquisite screenshots. They’re busy and jaded and just want to know if this art style is something that would interest them / their audience.
- Include convenient instructions for getting a review copy. Some places just dump a Steam key at the end of the email. Usually they say “review copies available on request”. Maybe once in a long time they offer a direct download link, but I haven’t seen that in ages. It’s all fine, as long as it’s clear and simple. But if you don’t include instructions, don’t expect the reviewer to email back and ask, because that’s more trouble than just hitting the down arrow to look at the next game in their inbox.
I’ve passed on a lot of games because of this. This isn’t some official policy on my part. It’s just the default behavior I fall into. If I see a marketing email that looks interesting but it doesn’t make it clear how to get a review copy, then I put the game into my ever-growing Steam wishlist. I’ll end up playing it after a sale months later. In short, if you don’t offer a easy review copy then I’ll behave like a customer instead of a reviewer. And as a customer I’m extremely picky, patient, fickle, and cheap. I’m going to guess I’m not unique in this.
- A clear, obvious link to the press kit. This is a zip of a bunch of image assets that someone will need in order to cover the game: The company logo, the developer logo, the stylized title of the game, along with a generous collection of HD screenshots. The screenshots are often needless: Any reviewer worth their pageview numbers is going to take their own screenshots when they play through the game themselves. But these shots also serve as a “back of the box” style enticement to the prospective reviewer. Also, these shots are often used by the news mills so they can slap their stupid watermarks on everything and post the images without needing to fire up the game themselves. Yes, that’s lame, but it’s exposure. SOME game is going to end up in the “screenshot gallery of the day”, and it might as well be the one you’re promoting.
- This really ought to go without saying, but the promotional email should be tightly proofread. No spelling blunders. No bad grammar. No extraneous fluff. Keep it simple, short, and clear. You’d be surprised how many firms get this wrong. (Or perhaps, how many developers write bad copy and the marketing firm doesn’t intervene.)
After we hired our marketing firm, I went around and checked up on them. I asked some of my game journalist colleagues to let me know how the Good Robot message looked and if it got where it was supposed to go. And it did. They did their job.
I’ve worked with this firm before on the other end of things. They provide me with review copies of games from time to time. They’ve always been prompt, intelligent, and professional. These people are not screwups or cheats.
This means we weren’t ripped off. The firm we used made solid messages (which they showed to us for approval before sending) and even crafted different messages for different review sites. (For example: My name was more prominent if they were contacting people who might know me, while the Pyrodactyl / Unrest names were leveraged everywhere else.) I saw their work from both sides, and I didn’t see any serious mistakes.
In a strange way, this is kind of discouraging. Even with a good firm with good practices and good contacts, it can still be really hard to get your game noticed. The market is well and truly flooded. Speaking of that, Rutskarn has some things to say about entering this line of work…
Indie development is not a safe investment.
I'd like to preface this by pointing out that nothing I'm going to say is based solely on my experience. This is supported by every developer I've ever met, online or offline–and to gently foreshadow the tone of this section, I'd like to point out that I've personally shook hands with more active, involved indie developers than I'd ever even heard of before I started making games.
Lately this industry has developed a warm appreciation for independent developers, which is very pleasant to see. Even gamers who rarely look beyond a few AAA releases a year seem to enjoy and share indie success stories. But I think this exuberance bruises the truth a little bit, and if you're thinking of making your own game, you need to know that. So let me address any new prospective game developers out there with a brief message:
Game making can be fun and artistically rewarding. And in a very real sense, it is no different than getting into painting or poetry or rap. People who love you will say, “Before you commit to doing it for a living, you better know what you're doing.” Those people are not being nearly tough enough. What you actually need to know is that you can't possibly “know what you're doing” well enough to guarantee success.
For heaven's sake, don't think for a minute–for one MINUTE–that making games will fix any fiscal or practical problem in your life. You know how your heating bills went up and you're looking to make a little extra money? How you're losing hours at work and need a new source of income? How you're hoping to switch careers and leave the craft store behind once and for all?
Betcha fifty bucks your game's not going to fix any of that. Your next one won't, either. The best you can say is that it'll put credits on your resume in case you want to apply to an established studio, get paid peanuts to simulate bucket physics, and be unceremoniously laid off just in time for the Christmas release.
If you're dreaming about dollar signs–and we all do–perform a little experiment. Ask yourself how much money you'd charge for your finished game–how much it'd take to make back your investment. Then go to Steam, search for that price point, and scroll through the listings. See how many polished rad expensive-looking games at that price point you've never even heard of.
Those are the ones that managed to get on Steam.
Imagine a lottery where you need to be incredibly gifted, dedicated, and endowed with free time to buy a ticket at all. That's indie development.
Maybe–even hopefullly–all this sounds obvious to you. But you'll be surprised how much wishful thinking develops, scumlike, after you've invested a lot of time and heart into a work. Release one game with real effort and money put into it and your perspective of the industry changes overnight. I mean, people very rightfully get outraged when indie devs go ballistic on YouTubers. There's no excuse for frivolous takedown notices, lawsuits, or any of the other dirty shenanigans that make the news every now and then–certainly it's never helped any of the developers–but I have to admit, after your first game launches it gets a lot easier to sympathize with people whose often (if not always) well-intentioned hard work is turned into grist for a mill of voyeuristic mockery and negativity. That's particularly true when the mockers are making more money on their insulting video than the dev will ever see in sales.
But of course, that's a fine line–right? There's no clear line between YouTube reviewers pointing out that a game is a broken waste of money and guffawing vultures cashing in on some kid's hopeful project. There's no clear line between gamers and journalists expressing frustration with a game's artistic statements and metajournalists half-assing semiotics to feed a content grinder. There's no clear line between a grubby hack developer who exploits a broken system and a media-savvy self promoter. There's no lines at all, just a lot of wearying negativity and bickering and entitlement and BS to deal with ON TOP OF the ponderous business of making and selling a videogame.
The industry is what it is. People will want to steal your game. People will pay for your game and then take their money back. Antisocial fourteen-year-olds will harass you if your game's too sexist or not sexist enough. You almost certainly won't make much money. You may not make ANY money–literally not a single penny.
But maybe, just maybe, it's worth it anyway.
Final Thoughts from Shamus
I certainly don’t regret making a game. But making a game is like forming a band: Do it because it’s fun, but don’t do it because you’re hoping to be rich and famous.
For perspective: For the time I sunk into Good Robot I could have written about at least two novels like the Witch Watch, each of which would have made me far more money than Good Robot. Which is to say: Making games is riskier and less profitable than being a small-time indie author. It doesn’t get much worse than that. (Although to be fair, maybe I’m just a better author than game designer.)
And as rough as this was at time, I still find myself wanting to make games. Which is probably part of the problem.
Thanks for reading. Thanks for playing.
 OR any of the other Youtubers â€" trust me, we emailed a ton of them.
Why The Christmas Shopping Season is Worse Every Year
Everyone hates Black Friday sales. Even retailers! So why does it exist?
A Lack of Vision and Leadership
People fault EA for being greedy, but their real sin is just how terrible they are at it.
Mass Effect 3 Ending Deconstruction
Did you dislike the ending to the Mass Effect trilogy? Here's my list of where it failed logically, thematically, and tonally.
Joker's Last Laugh
Did you anticipate the big plot twist of Batman: Arkham City? Here's all the ways the game hid that secret from you while also rubbing your nose in it.
Overused Words in Game Titles
I scoured the Steam database to figure out what words were the most commonly used in game titles.
140 thoughts on “Good Robot Postmortem #4: Promotion”
I realise that this isn’t entirely the post for this but “The Witch Watch” was a really good read and I do hope you find the time to write more!
What would a Witch Watch game be like?
Exactly like Dark Souls.
So it would be God Hand?
So it’s the Dark Souls of novels?
Massively lazy post of the day: Is the Witch Watch purchasable anywhere, PDF or hardback, where no registration is required?
It’s on Amazon.com if you can buy things from them …
It’s available on Amazon, of course, but also on The Book Depository, in case you’re not American, since they do free world wide shipping. It’s where I got my copy.
Surprisingly, it’s also on Amazon.ca for Canadians (that’s where I got it myself). I don’t know about other countries.
It’s on Amazon or Smashwords, both of which require a user account, which is sad.
However, I suppose you can just make some throwaway stuff up on Smashwords?
Amazon is probably more comfortable you’re more likely to have an account there but if I remember correctly, Shamus gets more money via Smashwords (Amazon takes an incredible amount of the price…), and at least in electronic format, Amazon is restricted to Kindle (possibly with DRM), while Smashwords gives you a choice of digital formats, DRM-free.
To be honest I think Witch Watch is a much better book than Good Robot is a game. Not that Good Robot is bad… it’s just “okay” in a market where being “okay” isn’t enough to make you stand out from the bajillion other indie titles. I guess Shamus has more experience writing than he does designing games, so maybe it’s not that suprising.
I agree that launching just after Enter the Gungeon was really unfortunate. It’s a much more memorable game purely because of its over-the-top theme – and unlike a lot of YouTube-bait games with silly themes, it’s actually a good game with enough content to hold people’s interest for while.
I feel bad for agreeing with you.
I don’t think this means this team has anything to be ashamed of. They made a very solid and enjoyable game. I don’t expect Shamus to be a great game designer. I expect him to be decent at that and really good at coding it which seems to have borne out here.
But the result doesn’t really stand out. If Shamus hadn’t been the dev, I wouldn’t have given it a second glance. Not because its not worth a second glance but because it really does look like so much other stuff (superficially, I think the art style is better than many).
But then I remember the success of Flappy Bird and it just makes me angry. This game is way better than that one. It sold well essentially because Pew Die Pie played it.
Also, I’ll bet if this game had appeared on Jesse’s channel a year earlier than it did, it would have gotten more sales. He’s been going through a transition.
I didn’t buy or play Good Robot (I’m on a Mac) but I did buy Witch Watch. It really was a great book. More, please! But maybe not a sequel? You’re an inventive world-builder, Shamus, and I’d love to see you build another world as interesting as that one.
I agree with what most of the other commenters are saying, but I think Enter the Gungeon was the biggest problem you all had. I’ve never played it and have no interest in playing it, but the descriptions of the games sounded very similar, at least superficially.
Also: I’m totally buying Witch Watch now.
I am a blog follower and occasional commenter, but the designer diaries for Good Robot didn’t make me that excited to play it. Then when I heard about Enter the Gungeon, I decided that’d be my shoot-em-up roguelike purchase for the month.
Having both read the Witch Watch and played Good Robot, I have to disagree. Of course, different folks and all that jazz…
I really need to get around to reading The Witch Watch (and I just got the pun in the name). Sadly it’s sitting in a pile with a bunch of other books I haven’t gotten around to reading.
I haven’t read it yet, but I can only say that the biggest thing in The Witch Watch’s favour for me is that I can probably get past Chapter 1…
I can clearly put up my hand and say that I would love to create a successful indie game one day, but it’s always going to be a side-thing unless something I make sells.
I work on my game whenever I can, but until it’s finished and i know if it’s good or not, I’ll most likely just go and get a scholarship for a PHD study so I actually have money to live on. I totally get what you are talking about when it comes to not ‘relying’ on indie development for your future career, cause the whole thing looks about as likely as a spinning a wheel of chance.
It’s so old it’s a cliche, but “Don’t quit your day job.” is still very relevant to any creative endeavor. Your game might make money, but it probably won’t. Make it for the enjoyment of making it, and treat everything else as a nice bonus. :)
You might have gotten more of a multiplier effect from this blog for “The Witch Watch” than you got for “Good Robot” as well. A very large percentage of your audience will enjoy a humourous fantasy novel, as one of the things that really put this blog on the map was “DM of the Rings”, which was a humourous parody fantasy webcomic. But even the percentage of those who like games and come here for that might not be interested in a rogue-like shooting game, and of those many might not like the side-scrolling, or the robot focus, or the way the gameplay was done, and so on and so forth. In my opinion, gaming is just much more fragmented in its audiences than novels are, which might also explain the issues with independent game development as well.
I’m a case in point: I bought “The Witch Watch” because it was in a genre and was an idea that I might have liked (and I knew about the financial issues you were having that made it a way for me to help out while still getting something out of it). I had NO interest in “Good Robot” because it really wasn’t the sort of game that interested me, despite my being quite interested in games in general. I can’t imagine that I’m the only one. And while it would work the other way as well, my opinion is, for the above reasons, that when to go to video games the audience fragments more than it does to novels.
Me too. It really wasn’t my bailiwick but I bought it to give it a try. I didn’t have the heart to return it.
I’m in the same boat. Good Robot is a genre I don’t play, would require an additional $60 spent on a controller for best results, and even if I wanted to try a shoot em up, the details turned me off. On the other hand, I already knew I liked Shamus’s writing…it’s why I’m here.
The Steam Controller isn’t the best for this kind of game, the touchpad doesn’t lend itself to precision top-down aiming.
And proper controllers don’t work well with Windows, where drivers crash constantly, so you either have to go M+KB, or use the (very badly designed) X-Box controller. I recommend M+KB.
How is the Xbox controller “very badly designed” exactly?
The Xbox One controller is not great for every type of game, but it is, imo, a great piece of kit.
The Elite controller is even better (but overpriced), if you are willing to part with the cash.
Try to play Castlevania: Symphony of the Night with a 360 controller for a good idea of how shit that controller is for 2D games. There is a jump that requires switching to the wolf, running forward and jumping, but if you press up or down even slightly, you lose all momentum and plummet straight down. I actually had to use the analog stick to make this jump after 10 attempts with the d-pad. On Playstation, I make the jump every time.
Me three; I’m very bad at actiony stuff so I almost never play anything in any actiony genre unless it has a super-easy mode of some kind… and even then probably not since I already have a giant backlog of games to play.
But if Shamus had created his procedural game as a RPG or TBS instead I’d have been there with an entire wardrobe of bells on.
I did buy Good Robot, but I have to be honest – I did it because I was interested to see the end result of your articles, not because it’s my style of game. In the end it proved too difficult and frustrating for me (permadeath and roguelikes really aren’t my thing, so add inexperience in the genre to dislike of that premise…) though I got to level… 4 I think? The one with all the worms.
For me, the game doesn’t quite work on the story level. It has atmosphere rather than story, and whilst I liked the jokes it didn’t really have anywhere to go with the story as released
I’m in the same boat here, and I think a lot of people are in the same boat. I was quite apprehensive about Good Robot from what I heard, but I got it anyway, because I wanted to see how it turned out. I also never finished it.
Damn, only about 2000 copies eh? I would have thought as well that it would have easily sold better.
As far as timing was concerned, remember that it was also the week that Steam brought out a huge push on VR, so giving VR games more prominence than anything else as well. Really unlucky timing.
And as Galad says below – *hugs* to Arvind and everyone else who put so much time, effort and soul into the release. Hope the positive memories win through for the smiles to follow.
Oh yeah! I forgot about that. Steam’s front page was covered almost entirely with VR stuff that week, so it was even more difficult than usual to get your (non-VR) game noticed.
I’m sorry to contribute more to the niggling doubts and negative comments, but, well see above re: post mortems and honesty. I think the 96% positive rating is inflated. I (and a couple other people in the retrospective comments) bought the game, didn’t have a great time with it, but decided that if I couldn’t leave a positive review, I would say nothing at all because I didn’t want to feel like I was hurting Shamus’s chances at success.
In the weeks before release, I remember several commenters saying something to the effect of “This doesn’t look at all like a game I’m interested in, but I plan to buy it to support you Shamus”. I have to imagine that that kind of audience loyalty drove at least a few blog followers to write positive reviews that they wouldn’t have written for a game without Shamus’s name on it.
I don’t mean to say that Good Robot is a garbage game that deserves an ultra-low rating, but 96% is the same score as Half Life 2, and I think I can say without being insulting that Good Robot is not as good as Half Life 2.
Half-Life and Good Robot are trying to do very different things, so much so that I don’t think you can really say that one is somehow better than the other. (Now, if one of the games had been a buggy, unplayable mess that would be different.) You sound a little like the people who used to write to Roger Ebert complaining that he gave the same number of stars to some inconsequential comedy as he did to some very serious drama.
Yeah yeah, apples and oranges, but the general consensus is that Half Life 2 is a fantastic game, the sort that routinely makes “Best N games ever” lists, and I don’t see nearly that level of praise for Good Robot among its players.
Well, sure. But a review that boils down to “I’m basically satisfied with my purchase” is equivalent to a review that claims “This is the best game I’ve ever played!” as far as Steam is concerned when it’s tallying up positive reviews. I’m not arguing that Good Robot is (or even should be) as popular or as widely acclaimed as Half Life 2. I’m arguing that the percentage of favorable reviews on Steam is a really ineffective way to measure these kinds of things.
Come to think of it, I would wager there’s more than one niche game on Steam with a greater proportion of favorable reviews than Half Life 2, simply because they have small but passionate fan communities.
We’ve gotten off on a tangent from my original point of “Good Robot’s 96% positive score is inflated”. If you assume only 4% of Good Robot players* had negative experiences, the retrospective comments section alone contains more negative reviews than that number can support.
*Good Robot has a few thousand owners, but it’s extremely common for people to buy games and never or barely play them. According to Steam Spy, its median playtime is under four minutes. If you set the cutoff for “Good Robot player” at a mere 15 minutes, my statement holds.
I think this is probably true. Then again, it’s not because I didn’t have a great experience with it that I have to give a negative review. I posted a positive one, indicating few bugs, no crashes, good communication by the dev team, nice graphics, funny jokes.
It didn’t quite deliver on what I expected, it’s not my genre, and it’s too difficult in ways I don’t like for me to enjoy it, but the game itself, in itself, is pretty good. Admittedly, if this wasn’t Shamus’ project, I’d probably not have written a review at all, or, if I had, focused more on the fact that the gameplay you see in trailers/screenshots is deceptive.
Ditto. I enjoyed the game, but ultimately found it to be too punishing. Then again, this isn’t a genre I’m usually interested in anyway.
Same here. I bought the game more because of Shamus’s involvement than out of interest in SHMUPs. I had fun, but permadeath puts me off of a game after a few hours. I’ll give the game another try with the new easy mode, and I think I’ll like that a lot more.
I feel really bad for being down on the game, but the truth is I was probably never the target audience for it.
I concur. I also don’t want to be unduly negative BUT purely from the point of post-mortems and future planning of games, I think it would be a mistake for the Good Robot team to assume that the 96% score is entirely representative of the game’s appeal. It is pretty clear that a significant portion of the positive reviews are from blog readers who wanted to help Shamus while the lack of negative reviews is likely to be biased in the same way. The low play times of a lot of the positive reviewers back that up (I’m assuming that play time is the total play time to date, not as at the time of the review?).
The comments in the postmortem posts back this theory and also provide some potentially valuable feedback on why the game wasn’t as well received as it could have been.
Of course I agree that timing and luck are likely to have been the biggest factors, just not the only factors.
I had not thought to look at that, you raise a good point. And yes, Steam reviews show total playtime.
Because I was curious I put all the data into an excel spreadsheet. Good Robot’s average positive reviwer playtime is 8.2, median is 4.3, and over a quarter of the reviews are under two hours.
For comparison, I took a look at a hundred of Enter the Gungeon’s most recent positive reviews. The average reviewer playtime is 36.4, median 20.5, and a quarter of the reviews are under seven hours. Even its negative reviews have much higher playtimes.
I think one of the main things that hurt the game was lack of good word-of-mouth. Looking at the comments from previous Postmortems, the consensus seemed to be that the game was way too difficult, very niche and a lot of people ended up not liking it (for that reason and others) myself included.
I’ve read that some other Youtubers will still cover a game after release if they fellow YT highly recommend it. So like if you get Dodger or Jesse Cox, etc to highly recommend the game, then there’s a better chance say TB will play it (although the game came out when he was rarely posted content).
I wonder how it sold relative to other very niche games. I imagine part of the issue is that not everyone on the site is going to be into this type of game versus a book similar to your other content (as Dambert talks about above). And I wonder if some people did buy it but quickly returned it (due to not being their type of game, etc), especially comparing how the game actually plays versus the images showing a fun, close encounter play style. I do think that generally very punishing games tend to have a much smaller audience, which is one thing that should be considered when you need to hit a certain number of sales to break even.
I hate being so negative here, but it seems worth mentioning (Ninety-Three said it better above).
Speaking as a Not-So-Casual Gamer, punishing games — especially ones that advertise themselves as that — lose the casual market. Any game that crows about how hard it is to play immediately gets dropped from my list of games that I might be interested in because I don’t have the time to grind levels or master detailed mechanics through dying, especially if picking the game up again a week later might mean that I have to relearn those mechanics.
I think the “The market is well and truly flooded. ” and “we weren’t distinct enough” hit the nail on the head with the whys of the financial result. Oh well, nothing much to do now but maybe to update the game a few times, then move on, I guess. Oh and Arvind *hug* I hope you do feel better eventually.
I’m not so sure that “not being distinctive enough” is that important of a factor in Good Robot’s failure to break even. I’ll admit that I haven’t bought a copy of the game, and that’s not because I don’t consider the game distinctive enough.
In fact, I don’t think I have many games that are similar to Good Robot. Furthermore, based on what I have seem of it, I consider it quite likely that I would like the game. It’s just that my Steam library already contains 305 games, and I’d estimate that there are about 75 or so of those that I haven’t played yet and a further 50 or so that I would like to play more of at some point in the future.
Buying even more games seem pointless, unless it’s a game that is really well reviewed, that I’m really looking forward to or that is really cheap (and therefore worth the risk that I’ll never get around to playing it). So for me, at least, the reason not to buy Good Robot was almost purely the state of the marketplace.
There might be some truth in the claims that are occasionally made that sales (in particular the Steam sales) are bad for the games industry. But even if that is the case, there is a huge collective action problem that makes it almost impossible to stop having sales or participating in them.
To be more specific, as a fellow gamer with close to around 300 games on Steam and GoG combined, at this point, when idly browsing the store for new games every once in a while, I’d be usually looking for something distinctive *mechanically* . Not necessarily always, however I’d argue that Good Robot is not mechanically distinctive enough from other similar games.
To me the game actually seems (mechanically) novel enough. If somehow my Steam and GoG libraries would suddenly cease to exist, Good Robot would probably be in my top 10 or 15 or so of games to buy as replacements.
But my Steam and GoG libraries have not (yet) disappeared.
+1 hug to Arvind (and the rest of the team, too!)
As a preface to my comment, I think I’m sitting squarely in the region of the Venn diagram that would be Good Robot’s sweet spot, market-wise. I’ve put a hefty chunk of time into shmups (though I really hate that nomenclature for some reason) and roguelikes through the years.
I bought Good Robot because I wanted to help Shamus out. Since then, the number of games I have and want to play has kept me from even installing it. Heck, even Shamus’ own stuff is working against Good Robot. Not long before the launch, I had given in to my curiosity and started playing Lord of the Rings Online thanks to the Shamus Plays article reposts. Also, watching old Spoiler Warning videos has me playing the old Fallout games.
Short story long — I bought the game, but the size of my gaming backlog has kept me from trying it.
“A tactical turn-based RPG with dating sim elements”
So, like Agarest and Fire Emblem?
I would definitely be interested in a tactical turn-based RPG with dating-sim elements. That sounds absolutely bizarre!
Persona came to mind (plus the new SMT/FE crossover Tokyo Mirage Sessions), as well as a game by an indie dev I knew a while back which is better described as a dating sim with turn-based RPG elements. (Fading Hearts by Sakura River.)
More like Dungeon Travellers 2.
There doesn’t seem to be enough actual DATING in that game, compared to the Agarest games. I think you can increase affection by what items you use on them in the dungeons and what you get them to do, but you don’t get the chance, as far as I recall, to go and find them outside of the dungeons and actually interact (there are cutscenes, but that’s not the same thing). Compare that with Agarest, where you have Vacation Days where you do just that, and where the dialogue options themselves increase or decrease affection among the various dates, even if they aren’t actually present.
That sounds like my shit.
Agarest’s been on my watch list for a while now, it just got on my get list.
I’ve mostly actually played Zero and a bit of 2 (called “Record of Agarest War” on the consoles) but let me warn you that the game, as per all of the reviews, can be pretty difficult. Zero, at least, had lots and lots of things you can mess around with — armour, armour types, skills, enhancements, etc, etc — and all of which can have an impact on how good you are, and for all of them you need lots of money and lots of points in order to use them properly, which means fighting lots and lots of things to get them. I found Zero grindy, and haven’t yet managed to finish it, but it was interesting enough. 2 annoyed me because I went from an area where I cleaned everything up without thinking to an area where I got curbstomped by th first fight. I haven’t tried Agarest yet because I only have it for the PC and have other things to play on the PC right now.
I for one look forward to future Shamus created Videogames.
Ditto! The reason I came to this blog in the first place, before I discovered the let’s plays and podcasts, was Shamus’ programming posts. It’s always really interesting system-y stuff, like procedural level generation. :)
Let’s see…..Without meaning to be negative:
1. I bought the game because of Shamus. Given the videos and the Steam page look, I honestly wouldn’t have looked twice otherwise.
2. The choice to go for a very punishing game play experience probably didn’t help. I know I can’t and don’t suggest it to friends looking for a small in-between game – either you play super conservative and it’s not all that fun or interesting, or you play aggressive and entertaining and you never leave Chapter 1. Neither’s a very rewarding experience.
3. I have to admit I’m surprised at the 2.000-ish copies. I’d have expected more, as well.
4. I do hope you guys keep on making games.
5. I also hope Shamus gets bitten by the writing bug again :-)
I imagine at least some of the lack of sales is due to people who visit not being able to play the game on a PC (either have Mac, Linux) or lacks the time to play, etc. I would imagine Shamus has stats on who visits using whatever OS (although that’s thrown off by visiting on a work PC, which you can’t/shouldn’t game on) so perhaps that was taken into account.
From everything I have heard, Mac and Linux are such a small segment of the market that going for multi-platform compatibility just isn’t worth it for most smaller companies.
Pillars of Eternity is an Obsidian game and their lead producer on that game said in an interview somewhere that Linux is maybe 1.5% of their sales and that he doesn’t think that developing for Linux was worthwhile.
Well, from Googling PoE’s sales figures it looks like they sold at least 500,000 copies by October last year, so that’s at least ~7,500 extra sales they got for releasing on Linux.
From what I’ve read from companies that release multi-platform games (like Paradox Development Studios), it’s really not that hard as long as it’s planned for and developed concurrently. A lot of companies it seems focus only on getting things to work on Windows until they get ready to release, at which point they’re hopelessly mired in design decisions tying them to Windows (using DirectX only instead of developing for OpenGL concurrently, etc.). Thankfully a lot of the game engines out there are making it easier to make multi-platform releases, so hopefully it’ll be easier for small companies to do so and take advantage of all the possible sales there.
Sure….but in the case of GR? that’d have been, what, 30 sales? Even at full price, that’s not worth it in terms of time spent.
I don’t know, it’s possible that Shamus’s audience has more Mac/Linux users than the average or less (although from what I’ve read on the site over the years that seems unlikely).
The more copies are sold, the lower the Linux share will be, considering Linux makes up about 0.82% of Steam users (myself being one of those).
I dunno… -if- it had a Linux version, I’d have written up an article for GamingOnLinux, which could have added some sales (though probably not enough to be worth it).
PoE was developed in Unity, which makes supporting Linux comparatively trivial, relative to supporting it with a game developed on top of your own engine.
Not that I’d ever demand your gaming time – especially for an experience you didn’t enjoy, but we have released a less punishing “easy mode” for GR since the start of this postmortem in an attempt to address things for players with your exact mindset. If you’re feeling particularly saintly, you might want to check it out.
I can’t promise the mode will solve any deep-seated systemic issues, but if your problem is that the game simply smacks you down too hard for dying or taking damage, this might address that (and if you do try it and it doesn’t go far enough in the “easy” direction, leave a comment and I’ll see what I can do).
Anyhow, thanks for supporting the game!
You should see if you can get this information into one of the articles. I doubt all the people who played will see this comment, but they likely WOULD see even a blurb in one of these postmortems.
I’ll give it a try!
I’m not sure I can even notice the changes in “easy mode”. I guess there’s a bit more money?
Still doesn’t encourage playing the game the way I could play the early alpha.
This. I’m afraid the problem isn’t raw difficulty – i.e. take more hits or do more damage – but really design. I don’t get the feeling I can wade in and mow down enemies, nor do I get the feeling I can go on the offensive.
I decided to use short range on Easy, and I still usually died around Chapter 2. I *can* play more offensively if I want to, but the whole design of the game discourages it.
I really wonder if any of the screenshots/gifs were made in a game where the GR survived for more than 2 seconds after it was taken.
The differences currently are:
-Player health and health upgrade amount increased by 50%.
-Repair stacking costs reduced (50%); scaling by amount of damage taken increased (~50%). How much damage the player has taken now factors into repair costs more.
-Warranty stacking costs slightly reduced (25%), added cost per upgrade slightly reduced (25%).
-Price scaling on upgrades reduced by the following amounts:
Shields: 1.25 -> 1.2
Damage: 1.1 -> 1.0
Speed: 1.2 -> 1.12
Fire Rate: 1.2 -> 1.15
Vision Radius: 0.6 -> 0.5
Essentially, player HP is higher and goes up faster with upgrades, repairs are dramatically cheaper, warranties are cheaper and less based on your progression, all upgrades increase in cost at a gentler rate.
Sad to hear so few copies were sold, even with the coverage of a Youtuber like Jesse Cox. It’s also sad to hear how abusive the Internet is towards you guys. I’ve heard that from other game developers as well.
I’m an amateur game developer and one of the things I worry most about releasing a game is the flood of death threaths, insults and so on that seem obligetory for each game release. Game development is already a dicy prospect from a financial POV, but I’m starting to doubt if it’s that much fun either.
There’s lots of ways making a game can be fun – the actual creation itself, the enjoyment when you know you’ve finished, and also from people who enjoy your game when you release it. The internet is full of jerks – just ignore/filter them, and look for the constructive feedback. :)
I’m glad you still want to make games.
Standing out among the indie crowd is almost an insurmountable task. It’s not enough to make a good game, or even a very good game – there are too many of those. It needs to be either brilliant, or the right game at the right time. Good Robot simply wasn’t. What stopped it from standing out enough to attract people’s attention or excitement? That’s more or less impossible to say. It might be that the looting or upgrade system were boring as all get out and the style of “roguelike” balancing disincentivized any kind of experimentation or remotely medium-range or close combat gunplay – but even with all that figured out, who’s to say this postmortem would end up any different?
I don’t see how any of that would have affected sales. Those are the complaints here, after the fact. None of those issues are at all apparent to a prospective buyer before they’ve bought the game.
Good Robot didn’t scratch my particular gaming itch but I still bought it out of support for all the work you folks put into it. I think you, Ruts and Chris need to make a game that focuses more on environmental storytelling. That’s a game I’d play. Something like Stardew Valley but faster-paced and with an underlying story that reveals itself while you play. I’m glad you guys tried with Good Robot and I hope you do again.
Provided there’s solid underlying gameplay, making a game full of Looking Glass Studios-style worldbuilding and environmental storytelling is basically my dream gig.
That, and making a tactical game which blends the mechanical consistency, character/unit focus, and story beats of The Best Fire Emblem* with the tone and class diversity of the original Final Fantasy Tactics. These two have nursed their respective, directly opposite flaws for too long!
*Supplies limited; bestness may vary. I am not responsible for sighs, punches, screams of wrath, or waifu wars that may result from this statement. Please, Fire Emblem responsibly.
In reading the development blog from the start, I noticed many times where design decisions weren’t so much based on “what do I want to create?” as much as “what do I think will appeal?” or “what’s the best way to ease players into X?” or “how do we best meet genre expectations?”
If I am being 100% honest, in following the project I was more excited about the game I was hearing about in the beginning of development than where it wound up in the end, and it got the impression (though I’m not 100% sure from where) that the path from point A to point B was paved with compromises made for sale-ability. Making a game because you love it is one thing, but what was your philosophy for balancing “this is the game I want to make” with “this is the game I think people will buy”?
I dunno. I’ve also been following this game since inception, and I think the decisions* weren’t made to try to make more sales, but to make the game fun to play. Sure, making it more fun to play should make for better reviews and more word-of-mouth sales, but I never got the impression that anybody involved in making this game were compromising the game in the pursuit of sales figures. :)
* Except for stuff like marketing, but those didn’t affect the game that got made.
I think if we were courting public opinion, the “easy, open-ended mine flier” design would probably have been the more financially safe of the two. It’s simple to see the game as a lot of different things the potential buyer would want to play, and has few obvious barriers to their enjoyment (“Permadeath? Count me out!”). Any issues the design has appealing to your common player aren’t really visible from the outside, whereas a “roguelike” is automatically polarizing the moment those two words are typed together.
I’d say this is borne out in the number of “the early stages of GR seemed more my kind of thing” comments we’ve received, if nothing else.
I don’t think I’m communicating my point effectively. I don’t mean to say you were going for mass appeal, because that suggests something more like appealing to the lowest common denominator. Rather, I know from my own experience that when I’m working, I will often come to an intersection where I say to myself, “well, yeah, *I* would like it if I designed that this way, but I’m loony and nobody else would know what to do with it” or “this is ‘unconventional’ for a reason and I should steer clear of it even though I prefer it.” It’s not necessarily an issue of wider market appeal, but more like balancing your desire to make what you like against making sure your weirdness doesn’t become a barrier for others’ enjoyment.
I can’t speak more broadly for the rest of the critiquers, but I can say for myself that I have no problem with the ‘roguelike’ label. I played the hell out of Dungeons of Dredmor. However, I think as the project went on GR pivoted from “a schmup non-schmup-playing Shamus would play” to “a schmup with an interesting art-style and a some RPG elements and procedurally-generated content stirred in.” It makes sense to me that such a pivot would probably engender “the early stages of GR seemed more my kind of thing” comments from regular followers of the blog.
This reminds me a bit of the “iron sight effect”, so allow me to say that I agree with the notion… before going into a long diversion:
Back in the day, while making Firearms: Source, we received an overwhelming number of requests to put gun-top sights into the game, despite the fact that FAS was a fast-paced high-accuracy twitch shooter and had no logical reason to include an option which both slowed you down and replaced the time-tested crosshairs with inferior modeled sights. We claimed this a number of times in interviews, but the crowd simply wouldn’t be placated – people were literally passing up the game simply because it didn’t have iron sights in it. The confounded things mattered that much, despite the dev team’s complete lack of interest in designing around them. The common factor in all the complaints was that it added depth to the shooting mechanics you just couldn’t get any other way.
So… we relented, after a fashion. The guns now had sights bound to RMB, but nothing else changed for using them – not your speed, not your accuracy, not your recoil. The sights were a placebo that never saw use in competitive play and only served as eye candy for the immersion-minded.
All of the complaints and demands disappeared overnight with that patch, never to return. To this day I don’t think any of those players have realized the emperor isn’t wearing his clothes.
In the end, with a little finagling we kept our gameplay and they got their sights. It was a surprisingly good decision.
There has to be some kind of balance there. If you create a game that makes you personally EXTREMELY happy and you like every design decision… but the vast majority of players don’t like it, why was it a commercial project at all?
“There should be a SHORT pitch that explains what makes this game unique. This should not be the opening crawl!”
Yes, yes, a thousand times yes! And do the same thing on every online store where you sell it. I very often skip by Steam games that fill their summary paragraph with made-up names and places; I’m much more likely to look at the rest of their store page if they’re obviously in a genre I’d like.
The problem I had with Good Robot is that it is cheeseburger ice cream.
I love cheeseburgers. If one of the decent fast food burger chains around here comes out with a Super Mega Bacon Blaster Burger, I am probably gonna go try it. I love ice cream. Whenever Dairy Queen advertises a new Caramelicious Cookie Dough Explosion blizzard flavor, I am going to go try it. However, cheeseburger ice cream just is not going to get my business. I’m not even going to consider it.
Good Robot is the same kind of combination for me. I don’t mind perma-death roguelike things. I don’t mind 2d shoot em ups. But I don’t think they mix well at all. Shoot-em-ups (good ones) require quick twitchy reflex play and reward daring and risk. Roguelikes are all about slow methodical play and they reward careful thought and risk avoidance strategies.
Those two game types just don’t mesh well together, and you end up with a game where the dominant strategy is carefully hosing each room down with bullets before you enter it.
“Those two game types just don't mesh well together, and you end up with a game where the dominant strategy is carefully hosing each room down with bullets before you enter it.”
This is solved in Enter the Gungeon and Ziggurat by not spawning enemies until you enter the room, and locking the door behind you until you kill them all.
I can think of several other ways to discourage that approach and encourage more risky play as well:
– Endlessly spawning enemies that don’t give rewards, forcing you to press forward to reach the end. (e.g. Left 4 Dead, Vermintide)
– Directly reward aggressive play, such as staying close to enemies or finishing the level in a short time. (Many shmups reward score for grazing bullets, for example)
– Make short-ranged weapons more powerful.
– Give invulnerability for a couple of seconds on hit, so you’re less likely to just instantly explode if you mess up.
– Give a small amount of shield recovery at regular intervals, so you can recover from minor mistakes as long as you don’t make too many. Arguably hats do this a little bit.
– Give enemies weapons with an “arming time”, which makes staying at a distance more dangerous than being up close (e.g. flak cannons in Guns of Icarus – their shells don’t explode until they’ve traveled a certain distance).
Of course, a lot of these would result in a significantly different game – I think Shamus has already mentioned some of them and why they weren’t used in previous posts.
Er, was that supposed to be impatient?
No. He’s patient. He can wait to play a game until it finally hits his price point (or he’s worked through enough backlog to need a new game). He doesn’t need the game now. It’ll still be there when he gets around to it.
I think it’s patient as in “wait 18 months for a really big price drop”.
Oh, I was thinking “impatient” as in only spending a few seconds looking at a game before deciding on it.
Awww. It’s a shame that Good Robot didn’t perform as well as you may have hoped.
i hope you guys are O.K and get out of whichever creative funks you may find yourselves in.
I’m still holding out for what I believe is the greatest Idea you guys have ever come up with, Mook School. Or Mook Manager. I’d still totally buy that.
Nah. “Patient” here as in “I’ll wait for a big sale.”
As a game dev who recently released a game that sold less than 1000 copies, I can relate all too well to just about everything in this post. Even then, with Shamus’ reputation and coverage from people like Jesse Cox, I’m shocked the game sold only ~2000 copies. Makes me feel grateful that mine even broke 100.
Regardless, I hope to see more Pyrodactyl and/or Twenty Sided games someday. (And, I also don’t blame any of you for taking a break from it if you decide to.)
I do think Jesse Cox exposure can sell a lot of copies – but only if the game on offer looks genuinely interesting to his audience. Exposure alone isn’t enough.
True that. I’m reminded of Dungeonmans, (a solid roguelike I really enjoyed) which had a video on it by PewdiePie of all people, and apparently sold almost no extra copies from the video.
The audience’s interests really matter.
I suspect Shamus’s own website is a good example of this. Obviously he gave his own game plenty of coverage in the run up to its release, but how many people who come here are interested in shmup roguelikes, as opposed to the writing in sprawling RPGs?
I myself actually like tough roguelikes – I’ve played the hell out of Darkest Dungeon, Dungeon of the Endless and BoI over the past year. And yet even then, I don’t think I ever considered getting Good Robot. It seems inconsiderate now, reading once again about the amount of work and agonising that go into these endeavours, and how crushing failure feels when you basically did nothing wrong. But I come here for the writing and the complaining about AAA games’ stories, and I knew from a glance at GR that it was about an entirely different set of Shamus’s skills.
I discovered the game by Frankomatic’s videos. Also, yay Jeff!
A FF: Tactics style game! A man after my own heart.
Although 15 dollars for an unknown game from an unknown developer is quite steep. And if I’m being honest, the screenshots are overly busy and none too pretty. This strikes me as the sort of game I (a person who loves grid-based turn based strategy games) would only buy if it was heavily discounted. Nothing about the Steam page really makes it “stand out”.
We must be after the same heart because I agree with your assessment completely. : )
There are good reasons my game didn’t sell well. My art ended up not enough “nostalgic SNES” and a little too much “RPGMaker knockoff”. That’s what happens when you go solo dev but have no artistic ability. As a result no press wants to touch it, and only the most diehard Tactics fans want to buy it.
Still, I’m pretty proud of it. People actually like it (100% positive reviews so far). But I definitely don’t blame you for waiting for a sale. I know I would be at least as skeptical.
…well, I like turn-based Tactics games definitely, so I guess it’s alright to ignore the main post to dig into it.
“A heavy emphasis on exploration”? How does that work for a tactics game?
Disgaea is on Steam (and may still be broken, I dunno.) How would you compare them?
…as a desperate attempt to tie this to the main post: part of the reason I stopped playing Good Robot was because I was still playing Disgaea.
The non-combat is pretty different from a normal Tactics game. It’s basically Zelda exploration, until you run into an enemy, at which point you go into turn-based combat (with no map change). Instead of using special tools like the hookshot to progress, you use your abilities – for example, if you learn a lightning spell, you can activate electronic devices with it.
Also like Zelda, the plot is a bit open, so you have your choice of areas to explore in the order you want.
I didn’t get too far in Disgaea. I think Voidspire has similar tactical combat but with more abilities to play with, and your characters are more flexible (can change class/subclass at any time out of combat). There’s also very little grinding; you shouldn’t need to grind at all to complete the game on the default difficulty.
…and to be relevant: I stopped playing Good Robot because I was busy developing my game : P
Ooo, a tactical RPG! I dig the couple I’ve played, to the point that the video game I’ve been working on is the same genre. The $15 price tag gave me pause, and I’m not in an impulse buy mood today anyway even if it had been <$5, but I at least signed in to wishlist it!
I have a feeling I'll pick it up one of these days.
Btw, while you may not have moved many copies, $15 sounds reasonable for a sizeable, fairly niche, indie game. Gives you room to put it on sale sometimes, and I doubt it would've gotten noticed as $3, either. I don't think the price is unfair or necessarily even unwise; I'm just not in the market at the moment.
Cool, thanks! Yeah, I don’t regret the price point I chose. It’s low enough for the hardcore fans – but a lower price point would probably not attract many more.
Glad to see another Tactics dev! Do you have a twitter or something I can follow development on? I like keeping up with upcoming tactical RPGs.
I can’t but feel like the final product that is Good Robot could have been made much, much quicker in Unity3d, gained quite a bit in terms of visuals and in turn would not have needed as many copies sold. But then, it’s easy in hindsight to say something could have been done very fast.
Maybe all I’m trying to say is, my personal experience with Unity has led me to believe that making your own engine is a pointless endeavor for any small dev, where you mess with graphics and rendering tech while you should be prototyping and making the game itself. There is no real cost anymore to these engines. They’re trivial to learn. They support fantastic graphics technology from high quality post-processing to lighting and physics, saving countless hours. Iteration rate is lightning fast due to how quickly you can recompile and test. Even AAA devs use these engines to prototype (See: Ubisoft). Even then you get free multiplatform release, expanding your userbase.
This is something we might disagree on, but it’s a common saying in amateur game dev communities – if you’re making an engine, you’re not making a game.
I certainly hope you’ll keep making games, and I’ll keep buying them because that’s how I like to support the blog.
How would it be quicker? Shamus would have to learn a new language (and while C# is quite like C++, it’s different enough that it takes month to get really good with it) and would have to learn Unity. He’s already extensive experience with making game-like applications with his current setup. It’s difficult for a framework to improve on that.
I also don’t believe that using Unity would suddenly make the game look better. The rendering of Good Robot is done nicely and the artwork wouldn’t magically become better because of Unity.
There’s also the issue that if you go the Unity route, your company is basically dependent on another company. Shamus and Pyrodactyl can remain fully independent with their current setup, without worrying if f.e. Unity suddenly changes their pricing scheme.
If it had been a 3D game I’d probably agree. Shamus himself said in part 1 that a 3D game takes too much fiddling with rendering nonsense to make shippable. But making a 2D game engine is absurdly easy if you’re as experienced as Shamus. Also, while it’s gotten a lot better with recent major version releases, Unity’s 2D capabilities are still a bit wonky.
If you skim back through the dev blogs, he only had one really big tech bug that would’ve been fixed with Unity (the bug where he was accidentally rendering occluded objects due to a logic flip). It’s not that there weren’t other tech bugs, but they were more game bugs that would’ve manifested themselves in a Unity game as well.
Absurdly easy or not, it took more than 0 time. When you use a third party engine, you can start implementing gameplay from day 1. Speaking as someone who has shipped a couple of games professionally on Unity, the benefits far outweigh the risks. And as someone who migrated from C++ to Unity, the transition is pretty easy.
It’s a big paradigm shift, but there are dozens of good books & YouTube videos on how to get started with Unity development, an active community that’s great at responding to questions. And even if Unity changes their licensing tomorrow, you can always continue to develop on the version of the platform you’re currently using.
“you can start implementing gameplay from day 1. ”
Assuming you know the engine.
Assuming you know the language it uses.
Assuming it does what you want.
Assuming the docs are clear.
” there are dozens of good books & YouTube videos on how to get started with Unity development, ”
Exactly. I’ve spent a lot of my professional life reading docs and learning things. Sometimes it’s nice to DO rather than learn yet another system that may or may not even do what I want. In the past, I’ve found third-party engines have been TERRIBLE at procedural content. They’re all built around the assumed paradigm of pre-made levels. Maybe that’s changed since the last time I looked. Maybe it hasn’t. The only way to know for sure is to sink a bunch of time into learning the new system and see if I hit a wall.
I’ve spent years having people evangelize some SDK / engine / tool only to have the New Thing eat a bunch of my time. “The docs are great!” (In their native language.) “It does everything!” (That they wanted.) “It’s easy to learn!” (If you already know all this unrelated stuff.)
I downloaded Unity a few weeks ago. I starts with a level editor. Uh… where’s the code? Where’s the game loop? Is there a version of this for PROGRAMMERS? I downloaded some examples. They flaked out because they’d been written in Unity 4 and I had Unity 5. Then I was having trouble opening them because clicking on a project in one window does something different than clicking on the exact same thing in another window. I wrestled with the thing for a good hour just trying to perform simple tasks. Instead of writing code, I was learning the Unity interface. I certainly wasn’t anywhere near “implementing gameplay from day 1”.
You are grossly underestimating the extreme cost of learning a new system.
Thank you for making the game, and thank you for writing about it! I haven’t gone back to try and beat it, but I quite like Good Robot, and I don’t think I’d regret the purchase even if I had never heard of you guys.
TotalBiscuit had a video a while back about an indie company who made a game that didn’t break out, but I can’t find it now. Basically, huge amounts of it are luck.
One thing I noticed that I would say didn’t help promotion; on day 1 ( I think) you had a pinned thread about bugs, but it was locked and seemed to encourage people to make new discussion threads about bugs they’d encountered. So while the review page was positive, the community page was basically a laundry list of problems. I’d imagine at least some people use those thread discussions to inform purchasing decisions as well.
…I would have mentioned it at the time, but I didn’t want to make a thread about it.
EDIT: Oh right:
I think most of it the problem is that the game just doesn’t really give that good of a first impression. The main theme I see in every video is “what does it do? Is it helping me? What is this green platform? Why am I losing my invincibility hat?” This is a roguelike, but it just isn’t one where part of the fun is figuring what you are capable of. Especially because the it is kinda bad to convey what all that stuff do.
The game needlessly hides information from the player. The worst offender being the weapons. 95% of the time, the people playing it gets into a groove with a certain weapon and don’t even realize when a higher tier weapon shows up that is objectively better than their previous weapon in every way… and then as they advance and the robots get stronger, they feel like they are equipped with pea-shooters against tactical nukes and that simply isn’t fun. Not to mention that I still think the upgrade system is hopelessly imbalanced. The shield-upgrade is objectively better than any other upgrade. The damage upgrade is barely noticeable and the rate-of-fire ditto, and still both cost the same as the shield upgrade which is ridiculously, absurdly good.
With two damage upgrades you can increase weapon damage by 15%, which is something that Shamus has always criticized about certain RPGs (barely noticeable incremental damage increases that doesn’t make the slightest difference), while with the same investment in shields not only you get two free shield repairs out of the ordeal, you also MORE THAN DOUBLE your shield bar. Which means that the shield upgrade is almost an order of magnitude objectively better than the damage upgrade (and still much, much better than any single other upgrade). The runs that I got the furthest were the ones where I invested in vision and shield exclusively, ignoring all else (except maybe movement speed to help dodge missiles).
At the very, very least there should be some kind of UI notification of the tier of the weapons. So that the players figure that the reason they are falling behind is because they are just not switching to better weapons. That they are still using a tier 1 weapon in an area where there are tier 3 weapons available. Another thing that would help immensely would be if the upgrade system was more upfront about how much each point is improving you.
Also, one really frustrating quirk is that when the enemies drop a weapon you don’t want (for being the same you already use, being inferior tier, or simply because you prefer the one you already have) it just sits there, uselessly, taunting the player that it is there instead of a drop of money that you desperately need. If there was a way to scrap weapon drops into money it would be a really visible improvement.
Maybe think about getting the game tucked into a humble bundle?
I mean, the problem is audience, not quality.
Have you thought about finishing your space adventure novel? Last I heard you put it on hold and released what you had. I wrote enjoyed what you released.
I bought Good Robot on launch day and promoted it to all my friends, but have only played it a couple of times since it means closing everything I’m doing and rebooting into Windows.
I enjoyed it even though I’ve barely gotten into it.
I’m not a gamer, but I thoroughly enjoyed the Good Robot series, as well as posts about Shamus’s other development projects. The journey of creating a game and writing about the experience, and especially about the technical problems encountered along the way, was enriching for the audience, and for that I am grateful — grateful enough to have purchased Witch Watch, which was a great read in spite of the formatting flaws and apparent reliance on Microsoft Word’s spell/grammar-checking (as a former college textbook copy-editor I’m particularly sensitive to these things). But see, Shamus, your posts about Good Robot did result in a sale to this reader, indirectly.
I didn’t buy “Good Robot” (it’s the kind of game I would require a physical copy of to rage-stomp on after rage-quitting). I’m sorry Arvind, please don’t be sad and do not stop making games. “Unrest” was fantastic; I still catch myself thinking about its world and story quite often.
Fun fact: Unrest did actually print and ship physical copies, with box art and everything. Posters too! Ripe for the stomping!
Less fun fact: My Unrest came pre-stomped by the Canadian (or Indian? Or both?) postal service! Thankfully, it retained the majority of its majesty, and now adorns my side-desk in a place of honour.
Extremely fun fact: At one point during development, I transferred game scripts between locations on a 3.5-inch floppy disk. It remains the superior hard copy version of Unrest to this day.
Luridly fun fact: I kept my 3.5 inch drive in the PC case for the long, long time after I stopped using it. Not long after I took it off (all new insides, additional ventilation, blahdy-blah), my cat barfed into the slot and fried some of them new insides.
I can verify this: my 3.5 inch drive is solidly in place even now, and never once have I had cat vomit enter my PC unprompted. With results like that, I’m amazed more people don’t keep them around.
It also helps that I do not own a cat.
>> “It also helps that I do not own a cat.”
Well, that’s just cheating.
But you have had prompted cat vomit enter your PC?
My cat may well be the champion of barfing (crowning achievement: every single bedroom surface in a single night, plus the litterbox and living room) but he has never yet managed to barf inside any of my PCs. How on earth did yours manage?
I rather liked Good Robot, and even bought a few extra copies to dole out to a few friends, who all largely responded positively, so I’m sorry to hear it wasn’t much of a success. I wonder if the original script you had for the game would have made a difference, so that there was more to talk about in the game than “great shooting mechanics, great game feel, witty snippets”. It makes for a more interesting discussion to go from “You are a good robot! :D” to “You were the bad robot all along! D:”, and a discussion around the game might have helped. As it is, it’s a great little shooter with a delightful aesthetic and it’s pleasing to play, but beyond that there’s not a lot to say about it, even among the friends that have beaten it.
Then there’s that problem, I guess, which is the problem inherent to all roguelikes–it’s hard. The whole genre is built on luck. Having a good run is a lot harder than having a bad run, and with the fail states the genre inspires, a mediocre run is just a longer bad run that you don’t see coming and drop an hour or more into, which can be frustrating. FTL saw this problem pretty distinctly, where you could get 5/8ths of the way through the game before you found out pretty quickly that failing to put any points into medbay upgrades meant your entire run was forfeit, and that you had to start all over. And every time you have to start all over, the chances of someone abandoning the game go up.
I’m almost tempted to say this genre needs to look back at old games and dig up one lost corpse–level codes (or something comprable). You beat a few levels, die, and go back to the beginning. But now, you get the code, and poof, back to level 4 instead of having to drop another hour and a half getting back to that point. Maybe you just get a randomized item assortment and a few skill points (but not as many as if you had gotten there on your own) and continue on with the roguelike, just so that you can stay invested in the game. A full stop is a hard thing to come back from, time and again.
Anyways, that’s just my thoughts. I hope you don’t give up anytime soon, Arvind, because I do really enjoy your games, and want them to succeed.
Leaving this here since I saw you complaining (rightfully!) about Skype on Twitter:
Also, I use XMPP for instant messaging and VOIP, and videocalls. Jitsi is a pretty cool client but there are dozens more, and there are also loads of providers (It used to come with a Google mail account until they decided to lock their users in…). To me, that’s how communications should work. As with e-mail, you can talk to others independent of the client and the provider, as long as you’re on the same protocol. You can even set up your own server if you’re willing and able.
I like the bit where Rutskarn calls himself “incredibly gifted, dedicated, and endowed.”
I never implied we had a chance.
Single manly tear.
I don’t always agree with you, but you’ve got heart, kid.
Thanks for doing this postmortem. That was a rather dispiriting final entry. Thanks for writing it anyway and being honest. Hugs to everyone, especially Arvind.
Thanks for writing the postmortem articles (as well as all the articles during development). It’s an interesting insight into indie game development.
Did you consider releasing a demo? I have no idea whether it would be worth the time spent on it, but it seems to me it would at least encourage people to try the game. Or do Steam refunds make it somewhat unnecessary now?
For what it’s worth, I enjoyed the game, even though I generally don’t like rogue-likes.
From what some devs have said(Puppy Dog? The guys who made Titan Attacks IIRC), demos cost more to make than you get back in sales, reliably. You’re better off making the game DRM-free and posting download links on imageboards and such, at least that is proven to create a spike in sales.
I anxiously awaited the reading of the new games list on the cooptional podcast that week, because they usually do a little mini commentary on the sale page, then I was hugely disappointed when the show ran long and they just read off the list. It’s cool you got Jesse Cox to do a video though.
Good Robot seems to at least be a good portfolio game for Pyrodactyl and Shamus.
And as long as it does not cost anything to keep it available for purchase then let it remain available for purchase to get whatever you can from the long tail.
Though maybe 5 years from now you’ll need to tweak the code a little to make the game run properly on Windows 10 v3.0 or whatever. This will extend the long tail even futher.
The graphics of Good Robot is timeless enough that it will age pretty well, assuming the code holds up on future systems.
Also, why not set up a patreon (or indiegogo or whatever) to see if you can get a Linux and Mac port fully funded.
BTW! What PR firm was it? I don’t plan/need a PR firm myself, but if they did a good job then a link to their website would be nice for those that might need a good PR firm.
The beauty of digital distribution is that we can keep Good Robot available as a title on Steam basically as long as the service exists, for no additional cost. I’m not even sure PC indie teams of our scope and size could properly exist if it wasn’t for the rise of internet distribution systems (though it might be argued that we’d be more discoverable and thus have more money to pay distributors/publishers in a world where Good Robot wasn’t just a mere speck of dust in the industrial-size mega mall of similar products that is the Steam store at any given time).
Well done with the game.
It is a nice achievement to see it through and complete a commercially released game. Obviously it is hard to be that big selling breakout indie game.
I have lots of sympathy this didn’t turn out better.
FWIW, I was really excited by all your descriptions of Good Robot, but when I tried it, I found it hard to get started and figure out what was supposed to be going on. I would like to try it again at some point, but i didn’t actually spend much time on it. Sorry that’s more helpful.
And I realise this is probably not practical advice because it’s just how I feel, but I would really enjoy reading more novels if you ever write any. I also wonder, this is probably a terrible idea, but have you considered trying to write a more story-focused game? Or maybe even something procedurally generated?
Thanks for joining the discussion. Be nice, don't post angry, and enjoy yourself. This is supposed to be fun. Your email address will not be published. Required fields are marked*
You can enclose spoilers in <strike> tags like so:
<strike>Darth Vader is Luke's father!</strike>
You can make things italics like this:
Can you imagine having Darth Vader as your <i>father</i>?
You can make things bold like this:
I'm <b>very</b> glad Darth Vader isn't my father.
You can make links like this:
I'm reading about <a href="http://en.wikipedia.org/wiki/Darth_Vader">Darth Vader</a> on Wikipedia!
You can quote someone like this:
Darth Vader said <blockquote>Luke, I am your father.</blockquote>