It finally happened. I backed a Kickstarter project that spent the money and didn’t manage to create a product. It was bound to happen sooner or later, but it really is a shame it happened to this project in particular.
As I’ve said before, the 1994 classic System Shock was a really important game, both for the industry and on a personal level. It was the first of the immersive simThis genre name is wonky and confusing now, but back in 1994 “sim” wasn’t so strongly associated with Will Wright-style simulations. games, making it the progenitor of Thief, Deus Ex, BioShock, Prey, and (to a lesser extent) the Dishonored series. It’s a game I loved so much I novelized it. The sequel, System Shock 2, is often considered one of the greatest games of all timeParticularly for PC gamers above a certain age..
So when I saw that Nightdive Studios was crowdfunding a remake of System Shock, I didn’t have much choice in the matter. Of course I was going to support it. But what really made me glad to put my money in was this blurb from the Kickstarter:
A modern take on System Shock, a faithful reboot; it’s not Citadel Station as it was, but as you remember it. Many improvements, overhauls and changes are being implemented to capture the spirit of what the original game was trying to convey, and bring it to contemporary gamers.
(Emphasis mine.) This was exactly what I was looking for: The game as I remember it.
The tricky thing about nostalgia titles is just how much we forget their faults. I remember the game as looking cool. I remember the sounds being spooky. I remember the gameplay feeling frantic. But then I launch the game twenty years later and discover the visuals are so blocky I can’t tell what things are, the font is illegible, the gameplay is awkward, and the interface is an abomination. I didn’t notice those problems at the time because all of the technology was new and nobody knew how to do better. What the Nightdive team seemed to be offering was a way to revisit the game with all those problems fixed. And maybe this would give us a way to share this classic with the younger generation without having to explain “You can’t use the mouse to look around. You have to use the keyboard. I’m so sorry. We didn’t know any better.”
As a backer I read and watched their development updates. As time went on, the news made me more and more uneasy and their creative decisions became increasingly inexplicable. Every update announced a disastrous idea with optimism and enthusiasm. I wish I’d written about it at the time so I could seem prescient now, but I was actually second guessing myself. “Oh, I’m just worrying because the project means so much to me. If I speak up now I’ll come off like an entitled, overprotective fan. They probably know what they’re doing.”
But no. The project has now failed, and it failed in exactly the way I feared. They ran out of money and quit. My gut instinct was right.
So let’s go back in time and look at how the project evolved.
The initial Kickstarter asks for 900k. That’s not a lot of money. Keep in mind that this isn’t a 2D sidescroller. This is an RPG shooter with very large environments. That means complex visuals, complex mechanics, and complex animations. This was one of the earliest examples of audiologs, which means you also need a bunch of voice acting. There are dozens of named characters. Sure, everyone is dead by the time the game starts so you don’t need mo-capped in-game cutscenes, but you still need character designs and portraits to go with the audiologs. The game also has a hacking minigame, upgrades to your abilities, inventory management, and a fly-through cyberspace minigame. There are sixteen different weapons of all different types: Melee, tranquilizer, energy weapons, and traditional firearms. On top of the sixteen weapons are an additional eight different types of explosives. Also, there are multiple types of ammo for some weapons. There are different movement modes that including flying, crouching, lying prone, leaning left and right, moving in low gravity, and cyber-skates that let you glide off the walls. The environments need to be dynamic and feature elevators, lighting changes, destructible objects, moving walls, forcefields, and several different types of doors. The game is non-linear and has a little Metroidvania in its DNA, so the player needs to be able to backtrack across the levels and have their changes to the environment maintained. And on top of all that, there are opening and ending cutscenes that have aged far better than many other games of the same time period, and doing them justice will be difficult without spending a lot of money.
There’s a reason this game is considered a classic. This is a big game with a lot of content. I have no idea how they planned to get all that done for just 900k. Heck, I wouldn’t be surprised if that was less than the budget of the original game in 1994.
According to the business types and executives I’ve worked with, it costs almost double someone’s salary to employ them. If you pay someone $50k a year, then it probably costs you close to $100k once you’re done paying for health insurance, taxes, equipment, unemployment insurance, and whatever else it costs you to keep them around. So 900k is enough to employ 10 people for just one year at 45k each. 45k sounds low for industry veterans, ten people sounds like a very small team for a project this complex, and a year sounds like a really tight schedule. And that’s assuming you spend everything on salaries and don’t have any operating expenses and you’re not planning on doing any marketing.
It’s June now, and they’re promising a delivery date of December of 2017. At best that gives them a year and a half to get this done. It’s probably less than that, since if you want to release in December you need to go gold a few weeks earlier.
But fine. Maybe the right people could get this done for 900k. Maybe these people are willing to take small salaries and manage their budget very, very carefully. They have a playable demo, after all.
The end of July rolls around. The Kickstarter ends, and the team gets 1.3 million. That gives them some breathing room. Maybe I’m just worrying.
I’ve been ignoring the game since the Kickstarter campaign ended. But when the team makes a big announcement I take notice.
“We have officially switched our engine for System Shock to Unreal Engine 4.”
When the campaign was going you had a working demo made in Unity. It had scenery, models, animations. You’re throwing that all away? Why? You already had a tight budget and tight schedule.
Worse, the new stuff looks terrible. The original Unity demo looked like remastered System Shock. For context, here is what the original game looked like:
And here is what that same hallway looked like in the demo featured in the Kickstarter pitch video:
That’s so perfect. It captures the feel of the original while using modern rendering techniques. This is exactly what you want if you’re trying to make the game look, “The way you remember it”.
And now here is a screenshot from the game after the switch to the new engine:
This looks like a generic corridor shooter. It looks flat and sterile and boring. The earlier screenshot is unmistakably System Shock, and this one could be from anything.
The team defends the change by saying that it’s not fair to compare the two. See, that original demo represented six months of work, and this new footage is just their first rough draft.
Yeah. I get that. That’s why we’re upset. I know games can look good in Unreal Engine. It’s a good piece of technology. The point is that you’re not there yet. We’re now nine months into this project. Fully half of the time budget is gone, and you’ve actually made negative progress. You haven’t even worked your way back to where the game was during the Kickstarter campaign.
To explain this seemingly insane move, Game Director Jason Fader said:
After listening to everyone during the Kickstarter campaign, it became clear that console support was very important to a lot of you. We took a hard look at what Unity could do on consoles, and what we wanted to achieve for both visual quality and performance.
This is really annoying because it’s highly unlikely but impossible to disprove. This is a remake of a 24 year old game in a niche genre that was exclusive to the PC. Are these aging, PC-focused, immersive-sim fans really clamoring for console support in significant numbers? Are there really that many people demanding to take this first-person game with a complicated interface and play it with thumbsticks? Or did you change engines for other reasons and are now trying to justify it by claiming it’s the “will of the backers”, since that’s impossible to disprove?
Regardless of what the community wanted, what this looks like is they took that 1.3 million from old-time PC fans and used it to pivot to the console market.
Moreover, even if you really thought that “visual quality” was lacking, was jumping engines really your only option? What is it that Unreal Engine could do that Unity couldn’t? Are you seriously suggesting you couldn’t get that demo running smoothly on the Xbox OneThe initial campaign promised Windows and Xbone versions.? I get that UE is the bigger, more impressive engine and is favored by AAA developers, but you’ve got a budget of 1.3M. You’re not making a AAA game. The backers were evidently fine with the visuals. So who are you trying to please?
Moreover, this jump took the project away from the grid-based design structure of the original. See, System Shock used this odd engine that was reminiscent of Wolfenstein 3D or Rise of the Triad. The environment layouts could be depicted on graph paper. Shock had the small improvement that you could optionally have floors and ceilings slope at exactly 45 degrees, which helped break up the boxy look and make the levels look more interesting.
Yes, things like this constrain the artists, but they also simplify design. If you’ve ever opened up a modern level editor you know there’s a lot of fussing around required, even if you just want to make a simple box building. That same building might only take a few seconds to throw together in something simplistic like Minecraft.
What I’m getting at is that this change didn’t just create a massive setback, it also made the game harder to develop. It takes more time to construct a freeform environment than to build one based on a simple grid. Not only was the original budget and schedule very tight, but now they’ve thrown everything away and started over with an even more ambitious design. Madness!
I must be missing something. Nobody is this crazy. Certainly not these industry veterans. I’m just being an obsessive fan and second-guessing them because my emotional investment is too high. I should probably just ignore the updates. The game will either be a success or it won’t. Speculating won’t change that. Neither will arguing with a press release.
The Nightdive team releases a video showing their progress. Unable to contain my curiosity, I watch it.
The good news is that the art style has recovered. The game has shed the “everything is made of dull plastic” look and is starting to develop a style of its own. It still doesn’t look anything like the original game or “System Shock as I remember it”, but it no longer looks boring. I guess it looks contemporary in terms of art design. I think looking like a modern game seems like a terrible mistake from a cost perspective, but maybe they’ve found a way to streamline it somehow? Maybe they’re going to use prefab rooms, like the Snap Map system they had in Doom 2016. Because if they’re trying to make the entire game at this level of detail, then I have no idea how they’re going to pay for it all.
The video is narrated by art director Kevin Manning, who says:
“I started on the project about a year ago as the lead artist, and just as the switch to Unreal [Engine] was happening. It was an interesting time for sure, as our small team learned the new toolset[…]”
You… you guys didn’t already know the Unreal toolset!? This is beyond absurd. I sort of assumed you chose UE because it was familiar. But not only did you throw away six months of work and and commit to a more ambitious design, but you decided you were going to do this with unfamiliar tools?
A bit later in the video:
“The team, the public, and our Kickstarter backers were all split between the retro look of the demo and reimagining it similar to what Doom and Wolfenstein had done.”
I’m sure you can find a handful of people who were asking for modern visuals, but are the backers really “split” over this? I certainly never heard any of these alleged demands for more advanced visuals. More importantly, this isn’t a democracy. Even if there is a paradoxical “vocal yet silent majority” pushing you in the direction of Doom 2016 and Wolfenstein: The New Order, that simply isn’t an option. Those are AAA games with multi-year development cycles and tens of millions of dollars to spend. You’ve got 18 months and $1.3 million.
The rest of his talk follows the same lines. It discusses spending months learning tools, doing tests, and working on concept art. Those are all things you do at the start of a project. It’s clear they’re not going to hit their release date, since that’s next month. But I feel like progress should be further along than these static environments and motionless placeholder foes.
Later in the video Game Director Jason Fader shows off “shatter tech”. He blasts a cryogenic barrel to freeze a zombie. Then he blasts it to show that the body shatters. You can even cut off specific limbs. Cut off a leg and it’ll topple over, then shatter.
I guess that’s impressive. But how is this moving you closer to your goal of shipping the game you promised? This gun is not a weapon from System Shock. Cryo barrels were not a part of System Shock. Shattering wasn’t part of the game either. With so much to do, you’re working on complicated technology-heavy features that weren’t even part of the game you’re trying to remake?
On top of all the other work you have to do, you’re now going to add balancing new weapons, creating new layouts, integrating new mechanics. You’re not “remaking” System Shock. You’re just making a new AAA shooter and slapping the System Shock name on it.
They release another video. This time they’re focusing on concept art. Considering how the game was supposed to release last month, it seems like we’re really late in the project to still be mucking around with concept art. This should have been done ages ago. You should be in full-tilt content creation by this point, and you’re still trying to figure out what the game should look like?
Concept art is important. But it’s not that important. You can make a game without concept art. But you can’t make a game without content. You can’t make a game without code. If you’re behind schedule and budgets are tight, then concept art is one place you can cut some corners. On a small team you don’t hire laser-focused specialist workers, you hire generalists and have them work on may different parts of the game.
I want to stress that this concept art isn’t just rough sketches to establish architectural cues and color palette. This is very elaborate work. It’s the kind of stuff I’d expect to see from a huge studio when they’re trying to keep their two dozen environmental artists organized.
In the video, Environmental Concept Artist Robert Simon talks about making 3D models so he can turn them into Photoshop images to be used as concept art. This seems like a decadent expenditure of resources, and that’s before you get to the part of the video where it’s revealed that he’s not the only concept artist on the team.
It’s like they think they’re a AAA studio. Maybe their project lead has only managed large-budget games and he doesn’t know how to scale down to something this size? Maybe they have other money besides the Kickstarter funds? I don’t know.
I can’t make sense of this. This behavior is so divorced from reality that I feel I must be missing something. The only theory I can come up with is that maybe their real goal is to have a major publisher take them under their wing and give it proper AAA funding. Maybe they’re trying to imitate Doom and Wolfenstein because they’re hoping to follow in their footsteps. I don’t have a shred of proof, but I think this theory makes more sense than “The project lead doesn’t understand the concept of money”.
That game is now on “hiatus”. They’re calling it a hiatus, but I would wager the odds of them resuming work on this project are extraordinarily low. I guess “hiatus” sounds better than “we laid off the team because we ran out of money”. The closure was announced in a post hilariously titled Sometimes You Need To Take a Step Back In Order To Take Two Steps Forward.
Yes, you can put a good spin on bad news. But there’s only so far you can get with corporate-speak. Even if we’re willing to call shutting down the whole team “one step back” instead of “falling over dead”, there’s nothing in the announcement to hint at what those “two steps forward” could possibly be. This announcement is like the doctor calling you a week after the funeral to tell you he thinks grandma is gonna pull through.
From the announcement:
“Maybe we were too successful. Maybe we lost our focus. The vision began to change. We moved from a Remaster to a completely new game. We shifted engines from Unity to Unreal, a choice that we don’t regret and one that has worked out for us. With the switch we began envisioning doing more, but straying from the core concepts of the original title.”
Your project was “too successful”? You blew through all your money and you never even escaped pre-production! Maybe you personally had fun working on it, but from the standpoint of everyone else on the planet this result is identical to taking the Kickstarter money, putting it in a big pile, and setting it on fire. From our point of view it’s the same: Money went in, and nothing came out.
You don’t regret changing engines? You say it “worked out” for you? How so? The state of the game at the end of the project is not as robust as what we saw during the pitch video, when you had combat and more than one working weapon. After all that work, you never even made it back to your original position at the starting line.
More from the final announcement from Nightdive:
As our concept grew and as our team changed, so did the scope of what we were doing and with that the budget for the game. As the budget grew, we began a long series of conversations with potential publishing partners. The more that we worked on the game, the more that we wanted to do, and the further we got from the original concepts that made System Shock so great.
(Emphasis mine.) So that seems to confirm my suspicion that the real goal was to get adopted by a big publisher. Now the development decisions make more sense.
- Moving to Unreal Engine was a ludicrous waste of resources if they were trying to ship the game they promised us, but it makes sense if you’re trying to court publishers who prefer projects with cutting-edge technology. Same goes for that shatter physics stuff.
- The move towards a more grounded, conventional, and expensive art style made no sense if you’re trying to please the backers who fell in love with that original demo, but it makes sense if your real goal is to impress publishers who don’t know or care what System Shock is or was.
- It’s wasteful to make such high-quality concept art for a small team, but it makes sense if you’re trying to show a publisher your vision.
- Making consoles a priority is somewhat questionable if you’re looking to fund a game exclusively through the support of older PC-focused backers, but it makes complete sense if your goal is to spin the project up to AAA scale.
- Prioritizing visuals over gameplay and financial viability seems crazy unless you’re betting everything on the hope that you’ll get a huge infusion of cash, far larger than the initial Kickstarter fund.
I wouldn’t mind if they were shopping the project around while also working to deliver the game they promised. But several of these moves made it so that the game could only reach the market with the help of a publisher. I can forgive a project that fumbles due to gradual scope creep, changing marketplace needs, or unforeseen technology problems. But these were deliberate, willfull decisions that ran against the promises made to the Kickstarter backers. I can’t prove they intended to use the KS money to tart themselves up and look for a publisher to be their sugar daddy, but throwing the original demo away was inexcusable in the face of their original promises, regardless of their motivation.
I realize I’m accusing a group of total strangers of being dishonest with the backers. I don’t know these people and I don’t know any facts beyond what they’ve publicly shared. I want to believe this is just the result of ambition, naivety, and recklessness. But if we assume they used the Kickstarter to build a tech demo so they could sell themselves to a publisher then everything they did makes sense. If we assume they just wanted to ship the game they promised, then nothing they did was rational.
The tragically ironic thing is, this is exactly the problem that led people to turn to crowdfunding in the first place. We couldn’t get the big publishers to fund these mid-budget projects because they’re so obsessed with photorealism and high production values. All they care about is big-budget blockbusters. And then Nightdive comes in, gets funded, and decides to make a big-budget game.
I’ve said before that I think it would be good if project leads would release the source code and art assets if their crowdfunded project failsI realize that the license itself can’t come with it. If a crowdfunded Batman game were to fail, the released art and code would still be bound by their ties to the trademarked character. It would need to be stripped of the Batman stuff before it could be released.. The code could have educational value for aspiring developers. Maybe the public will finish the project for you and something good can come of it. But most importantly, it would be a sign of good faith and a demonstration that you were really working on a game and not playing ping-pong for the last N months. (I’m not suggesting this was the case at Nightdive, since it’s pretty clear they were working hard. I’m just saying it would be a good thing to make standard practice.)
Of course, I’m not expecting the Nightdive team to release the source for the their half-finished gameMore like one-tenth finished, but who’s counting?. The final announcement makes it pretty clear they’re still dreaming of finding a publisher willing to bankroll their ever-expanding ambitions.
Where can the franchise go now? Even if the license somehow made it into the hands of a new team, they couldn’t very well crowdfund it again after the failure of this crowdfunding effort. Private investment? Good luck trying to raise a proper AAA budget in a pitch that begins with “This is a remake of a game that invented a genre that’s really expensive to produce and that has never sold particularly well.” Publisher support? I think it’s pretty clear they’re not interested. If they weren’t interested in adopting the project when production was underway, then I think they’re going to be even less eager to acquire it now that it’s been closed down and parts of the team have moved on.
As much as I love this series, I don’t think it’s viable as a AAA product. The audience simply isn’t there. The modest $1.3 million Kickstarter proves this is a niche game. (For contrast, Pillars of Eternity – a text-driven top-down party RPG like the kind the AAA industry stopped making a decade and a half ago – got four million bucks. The System Shock Kickstarter had the best possible proof-of-concept demo and they only got a third of that.) If you spend fifty million dollars making System Shock then you’ll never get a return on your investment. This game is only viable as a low / mid budget title, and Nightdive has made it clear they’re not interested in making that sort of game, even if they somehow got another infusion of cash.
Nightdive CEO Stephen Kick is still insisting that he “doesn’t regret” changing engines and he’s not talking about going back to the original plan. This means there’s no route forward for the game. System Shock is now deader than it was three years ago.
What a sad, needless waste. They should have just made the game they promised us.
 This genre name is wonky and confusing now, but back in 1994 “sim” wasn’t so strongly associated with Will Wright-style simulations.
 Particularly for PC gamers above a certain age.
 The initial campaign promised Windows and Xbone versions.
 I realize that the license itself can’t come with it. If a crowdfunded Batman game were to fail, the released art and code would still be bound by their ties to the trademarked character. It would need to be stripped of the Batman stuff before it could be released.
 More like one-tenth finished, but who’s counting?
Two minutes of fun at the expense of a badly-run theme park.
Internet News is All Wrong
Why is internet news so bad, why do people prefer celebrity fluff, and how could it be made better?
C++ is a wonderful language for making horrible code.
Starcraft 2: Rush Analysis
I write a program to simulate different strategies in Starcraft 2, to see how they compare.
Even allegedly smart people can make life-changing blunders that seem very, very obvious in retrospect.