#launch problems

Of course a studio can delay the launch, but a launch delay will have its own set of consequences. Remember, no game launches in a vacuum. Timing is the key factor.

The most obvious consequence goes back to the adage "Time is money". The developers working on the game need to pay rent and buy food. The studio needs to pay those developers during that time to keep them employed. Those costs don't go away when the game's launch is delayed - they will still accumulate. If the studio doesn't have the money they need to keep the lights on during that delay, it won't be able to make payroll for its employees or pay its bills/debt. In such a situation, a delay is impossible - the company will die without money from the game sales.

The second consequence is about the state of the gaming world over time. Imagine that we delay our game to dodge Hades 2. What happens after Hades 2? Will we need to dodge another big release? Will we need to dodge another competitor in our specific game genre? Will the market still be good for our game? Will the platform still be good? Remember, Facebook was a major gaming platform for years and then it wasn't. Imagine spending two years and millions of dollars to develop a major Facebook game only to have the platform die out from under you before launch. Things always change over time.

These two consequences feed into each other - every day we delay, more things change and new problems arise. Every day we delay, costs build up that we need to pay for somehow. Every day we delay release, our income on the game remains zero. At some point, the game will need to launch or the company will assuredly die.

[Join us on Discord] and/or [Support us on Patreon]

Got a burning question you want answered?

Short questions: Ask a Game Dev on Twitter

Long questions: Ask a Game Dev on Tumblr

It has certainly happened in the past, though not necessarily specifically the narrative part of the game. Many games are pushed to launch without development being as far as they want it to be due to reasons like hitting their budgetary limit and needing to recoup some of the investment. Our estimates are only estimates after all, sometimes we run into unforeseen problems and things take longer than expected. We can't stop paying the developers when we hit snags like that, so certain features end up more costly than others, which eats into the budget that was earmarked for other stuff instead. Most games in this situation have a lot of other launch issues too for the same reason - when you're pushed out the door to make the deadline due to running out of budget, things that should have been fixed are often not.

When World of Warcraft launched in 2004, there were several entire world zones that were incomplete and (mostly) locked off from players. Some players were able to sneak in through various exploits and take screenshots of those areas. Most notable were that the zones were primarily unpopulated by anything - no mobs, no quests, empty towns and buildings, just environment geometry that had been built out. This accompanied other incomplete bits of the game like quests that still had XML code in them. It would take years before players would finally see the incomplete-at-launch zones in some form or other.

Cyberpunk 2077 famously launched after multiple delays with numerous bugs and weird issues. Notably, the dev team also completely cut the multiplayer mode of the game that they had been building in order to consolidate resources to ship the single player game. The game came in super hot and had a huge number of launch issues that were eventually (mostly) ironed out, but the multiplayer mode was never resurrected.

The most famous example of this is probably Knights of the Old Republic 2. The publisher famously moved the deadline up and Obsidian scrapped the in-development ending since they didn't have the time to finish it. Instead, the story was wrapped up super quickly to ship the game. Notably, the partially-finished original ending was left on the disc and modders eventually discovered (and later restored) it.

[Join us on Discord] and/or [Support us on Patreon]

Got a burning question you want answered?

Short questions: Ask a Game Dev on Twitter

Long questions: Ask a Game Dev on Tumblr

It entirely depends on how rocky the launch is and what the extent of the damage is. While there are the super rare unicorn lottery-winner exception stories like how Final Fantasy 14 managed to turn itself around after rebuilding the entire game, most of the time the results are pretty cut and dry when it comes to whether a game sinks or swims. What typically determines a game's survival is whether the overall player base that we manage to hang on to is sufficient for critical mass.

Community rage and blowback are indicators, but the real important figure is whether enough players continue to play. Critical mass is necessary for the game to feel like it is firing on all cylinders. Critical mass means the auction house is stocked. Critical mass means that looking for a group doesn't have extremely long queue times. Critical mass means that you can find players to play with, trade with, fight against, and so on. All of the multiplayer aspects of the game need enough players to fill those community roles.

If a live service game can reach and sustain critical mass, it will likely continue on even if the launch was rocky and the initial blowback was severe. If the game cannot reach critical mass or the player churn rate becomes too great, the game is doomed. We can stave off the inevitable by spending more on marketing and user acquisition, but burning money like that is not sustainable. Without a game that is engaging enough to sustain a critical mass of players, the publishers will pull the plug.

[Join us on Discord] and/or [Support us on Patreon]

Got a burning question you want answered?

Short questions: Ask a Game Dev on Twitter

Long questions: Ask a Game Dev on Tumblr

The writing was on the wall when they announced the delay after going gold. The warning signs were already there before then - CDPR is fairly well-known in the industry for their sweatshop working conditions and pay, the game had been in development since the PS3/X360 console generation, and there had been multiple delays - but the post-gold delay really confirmed that it was going to be a hot, buggy mess. So let’s talk about it for a moment.

First, Cyberpunk’s extremely long development cycle basically meant that the leadership couldn’t figure out what they wanted the game to be for a very long time. Nobody plans to take eight years to build a game - the video game market changes and evolves way too quickly for that to happen. When Cyberpunk was announced, League of Legends hadn’t hit its peak yet and everybody was starting to fall off of the big AAA MMOG boom. Hearthstone didn’t exist. Fortnite didn’t exist. Players were all super excited to play COD: BLOPS 2. Having a dev cycle take longer than 4 years generally means the dev team wasted a lot of time building things, discovering they didn’t work, and then throwing them out. The finished Cyberpunk isn’t the sum of eight years of work, it’s the sum of maybe three years of work and five additional years of cut systems and content. 

Second, CDPR is already known in the industry for their brutal working conditions. They’re a crunch culture studio and there have been confirmations from one of the devs on Cyberpunk that they’ve been crunching on it since at least June of 2019 despite public assurances that it wasn’t mandatory. These kind of working conditions are nothing new; they did it for the Witcher 3, they did it for the Witcher 2, and there’s no reason to think they won’t continue in the future since it clearly continues to work for them. Unfortunately, the effect of the exhausted dev team gets multiplied by the systemic effects of delaying the game as well.

Finally, Cyberpunk was also delayed several times - in January, then in June, then again in November. Delays are never planned into the schedule - they are responses to shit hitting the fan. They only occur when the producers realize that there is no way that they’ll be able to hit their target that they delay the game. There are several reasons for this, primarily budgetary - changing release dates greatly affects the marketing campaign, business deals, as well as the overall development cost of the game. Thus, delays are usually a last resort - the added time is the minimum estimated time to reach the goal by crunching super hard to get past the finish line. The fact that they had to delay three separate times meant that they blew their estimates and schedules three times and had to course-correct each time. This is why the effect of the death march crunch mentioned previously gets multiplied - the devs have already been on a year-long deathmarch crunch and are exhausted, causing their work to suffer. Delaying on top of that means that their work will continue to suffer.

Cyberpunk is the result of multiple contributing factors - years in development hell, a brutal sweatshop studio culture, and all exacerbated by pivoting due to delays changing the schedule. It is the result of bad planning multiplied by developer exhaustion, compounded by sudden scope and schedule changes. All of the warning signs were already there if you knew where to look. I actually would have been very surprised if this wasn’t the resultant state of Cyberpunk at launch.

[Join us on Discord] and/or [Support us on Patreon]

The FANTa Project is being rebooted. [What is the FANTa project?]

Got a burning question you want answered?

Short questions: Ask a Game Dev on Twitter

Long questions: Ask a Game Dev on Tumblr

I doubt it. You can generally tell what’s in a patch by its relative size. Big patches include asset changes. Small patches are only code changes. Cyberpunk probably has a lot of asset changes in it for various bug fixes.

Code is just text that’s been compiled and linked down to machine language, so it tends to be very small - just take a look at the executable size of any PC game you’ve got installed and you’ll see it is usually only a megabyte or two at most. We usually change code for bug fixes on a live game, since we trust the engineers more than any other discipline to make changes to the game while maintaining stability. Code changes comprise the rules and behavior of the game, so things that may have behaved one way will now behave differently. This also generally includes numerical data - stats and abilities are generally represented as text in the files, so changes like “this ability does X more damage” and “that character now has Y ability instead of X ability” are also likely to be in the small-file camp unless it involves visual asset changes or additions.

Assets like models, textures, and audio can’t be compressed down as much as code or text can, so they take the lion’s share of any patch’s file size. Each of those types of files can be multiple megabytes on its own (or even bigger, in audio’s case), so any changes to assets (e.g. adjusting a seizure-inducing visual effect so it doesn’t induce seizures so much anymore) will greatly increase the size of a patch. Assets change the way the game looks and sounds and are usually included alongside changes to the rules.

The general rule of thumb is this: 

If the patch is only a few megabytes or less, it’s probably just code changes. There are always code changes.

If it’s in the tens or hundreds of megabytes, it’s changing or adding some assets in addition to the code changes.

If it’s in the gigabytes, it’s changing or adding a lot of assets in addition to the code changes.

[Join us on Discord] and/or [Support us on Patreon]

The FANTa Project is being rebooted. [What is the FANTa project?]

Got a burning question you want answered?

Short questions: Ask a Game Dev on Twitter

Long questions: Ask a Game Dev on Tumblr

Any game with a development cycle of over four years has almost certainly spent a significant amount of that time in development hell. Cyberpunk is no exception. Most might not realize this, but Cyberpunk 2077 (May 2012) was announced about six months before Star Citizen (October 2012). It’s been in development for over eight years by now, twice as long as the Witcher 3. Cyberpunk was announced a year and a half before the PS4 launched and will likely hit store shelves around when the PS5 launches, meaning that they came close to skipping an entire console generation.

Nobody sets out to build a game with a development cycle that long. Between then and now, here are some of the notable gaming trends that have come and gone:

MMOGs were still in fashion (SWTOR launched six months before Cyberpunk was announced)

MOBAs hadn’t hit their peak yet (League of Legends didn’t hit their peak of 100 million unique monthly players until four years later)

Fortnite and Battle Royale wouldn’t be a thing for another five years

Doublefine had just run their first kickstarter campaign three months before Cyberpunk’s announcement

Hearthstone and online card games wouldn’t be a thing for another two years

It’s no secret that the Witcher 3 (released May 2015) hoovered up practically all of the Cyberpunk devs during their final push to launch, but the fact it’s taken another five years on top of that to get to (near) release is also not a good look for the game’s development process. These signs point to meandering development that had a hard time nailing down what the game was supposed to be about, building a lot of things, and then scrapping them. This also gels with the stories I’ve heard about their internal development process. 

Additionally, the game has had announced delays along the way. A delay signals two things - that the delay wasn’t planned, and that the project fell so far short of their expected scheduled progress that the dev team could not make it up by crunching. Cyberpunk delayed not once but twice. If they had initially thought that they needed the additional two months from September to November for their release, they would have delayed the game from April to November, rather than April to September, and then again to November. The fact that there are two delays instead of one (in addition to their eight year dev cycle) means that their leadership thought they could get the needed work done in six months and realized that they were so wrong that they they needed two more on top of that. 

Scope is a major concern with a game this size. The game’s multiplayer mode, for example, is one of the features they promised for Cyberpunk 2077, but will not be in the game at launch. They have promised they will patch it in later, which means they had to cut multiplayer from the launch scope in order to make their release date. Overall, all of these signs point to a troubled development process - starting with the abnormally long development cycle, and then factoring in other signs like delays and scope changes in addition to CDPR’s baseline crunch culture. So yeah, I feel very comfortable saying that I believe Cyberpunk 2077 was in development hell for quite some time. I can’t think of a game that took so long to build that wasn’t.

[Join us on Discord] and/or [Support us on Patreon]

The FANTa Project is being rebooted. [What is the FANTa project?]

Got a burning question you want answered?

Short questions: Ask a Game Dev on Twitter

Long questions: Ask a Game Dev on Tumblr

Baldur’s Gate 3 just released to Early Access. It costs $60USD and it launched with only the first of three planned acts. It looks beautiful but is very buggy. If they called this a full release instead of early access with more to come, the reception would have been terrible because players expect a game that can be played from start to finish and doesn’t crash or break when they spend that kind of money. Early Access, at least for now, convinces audiences to pay money for buggy and unfinished work without necessarily triggering an ensuing backlash. The general understanding is that more will be coming. That may not necessarily be true, but it is the expectation.

[Join us on Discord] and/or [Support us on Patreon]

The FANTa Project is being rebooted. [What is the FANTa project?]

Got a burning question you want answered?

Short questions: Ask a Game Dev on Twitter

Long questions: Ask a Game Dev on Tumblr

Not all situations are created equal and most come with tradeoffs. When a title goes to Early Access it is, for all intents and purposes, a (soft) launch. The game is live, the players are real, and any money spent by those players will be real too. This is extremely important to note because many of these projects don’t have sufficient funding to get from Early Access to the “full” launch with accompanying marketing push and the like. The team may be [running out of runway] and in desperate need of cash to keep the lights on. 

In these situations, Early Access effectively is the game’s actual launch. These teams don’t have the resources to get to the “full” launch, so they are banking on revenue from Early Access to keep the lights on. If the Early Access launch can’t hit its revenue targets, the company will likely take the usual cost-cutting measures - layoffs, scope reduction, or even shut down. It isn’t so much a question of “Should we go Early Access or should we do the full launch later?” so much as “If we don’t go Early Access by this date, we’re going to run out of funding and we’ll all be out of work”.

Going to Early Access like this is also a double-edged sword because the vast majority of a game’s marketing and user acquisition comes from when the game is new. Early Access followed by a “real” launch dulls the newness effect because the “real” launch is typically more like expansion content to the Early Access game than it is a whole new product. It’s much harder to pick up new players when taking an Early Access game to “launch” than it is at the initial launch. This is why Early Access means something very different for AAA games than it does for smaller projects that stay in Early Access for a long time. For AAA games, it is typically part of the marketing push up to launch as a reward for preorders - it allows preorders to play a week or so earlier than others. Small and indie teams don’t always have the funding they need to take this route, which is why they stay in Early Access for a really long time. It certainly isn’t the ideal situation, but it’s a lot better than layoffs.

[Join us on Discord] and/or [Support us on Patreon]

The FANTa Project is being rebooted. [What is the FANTa project?]

Got a burning question you want answered?

Short questions: Ask a Game Dev on Twitter

Long questions: Ask a Game Dev on Tumblr