Stop Expecting Smooth Day 1 Game Launches
Why? The short answer is that you’ll be continually disappointed.
The long answer starts with four examples of companies who have each, in their own way, failed to smoothly launch their games: Blizzard with Diablo 3, EA with SimCity, Square Enix with FFXIV and now Rockstar with GTA Online (these are just immediately off the top of my head, but I’m sure I’m forgetting some). All of these games’ launches created massive user backlash, and all sorts of scrambling drama for the devs. It’s all too tiring to recount here, but I’m guessing you remember at least one of those.
At Game Front, we’re usually staunch consumer advocates, ready to call out anyone trying to rip you off. And when you can’t play a game on Day 1, it’s understandable to feel swindled. How is it that Rockstar, who made $1 billion in two days, can’t figure out how to get GTA Online to run smoothly? There are smart, creative people working there.
I’m starting to think we’re banging our collective heads against a wall that isn’t going to break anytime soon. Moreover, the wall may not have been built by bricks of developer or publisher incompetence, but rather an economic reality. In all likelihood, it doesn’t make economic sense for, say, Blizzard to buy enough servers to put a small country online, just for Day 1. For any online game launch, Day 1 is the most difficult.
For developers, the problem seems like it must come down to one of investment. How much do you invest in servers, and not just in the short term, but the long term? To use an analogy, in order to accommodate a horde of scampering hamsters, a tunnel must be built, which costs money. The size of the tunnel should be commensurate with the size of the hamster horde. But how many hamsters will there be? Let’s say you have some sort of ballpark. It’ll be about 10,000. You have two choices: A). Build a large, expensive tunnel big enough to allow 10,000 hamsters to flow freely, at any given time. B). Build a small, budget tunnel, one that will be clogged with hamsters for days, as they push through in anguish…until they don’t, several days later, when the traffic dies down. The large tunnel gains you hamster-happiness, and loses you money. The small tunnel saves you money, but causes hamster-anxiety for a few short days.
NOTE: this analogy was inspired by this Diablo 3 meme.
I’ll give you one guess as to which of these options most developers choose. And can you blame them? If you ran a business, and those were your only two options, which would you choose? How this decision is communicated to your customers is another matter entirely (more on that later).
What do developers have to say about this?
Usually they’re silent, but back in March we spoke with Cryptic’s Jack Emmert about the then-forthcoming launch of the MMO Neverwinter, and asked Emmert to chime in on online launches. In the words of a man who’s done this before, here’s what he said about building a hamster tunnel that’s too big:
“Inevitably, the population decreases over time, and you end up shutting down servers and merging them. That alienates players, and it gives you a bad taste in your mouth as a company.”
“As soon as you start shutting servers down, it’s a snowball, because there’s nothing more important in MMOs than the belief that there’s a future. It’s a persistent world. You’re committing your time into the game, and if you don’t see that there’s new stuff on the horizon, you’re not going to want to play anymore.”
If you’re looking for answer to the question “why can’t they just have enough servers at launch,” that’s one answer right there — over-compensating is a possibility, and shutting down servers post-launch presents its own problems for game developers. Not to mention, over-buying servers sounds expensive. Then what happens when you don’t need them anymore? You have to tear down your hamster tunnel, and you’ve wasted money.
So What the Hell do we Do?
Like I previously said, this doesn’t seem a problem that’s going to just vanish any time soon. It’s a messy collision between IT limitations, business realities, bad PR/community management and serious user backlash. Consumers aren’t necessarily powerless, but this won’t change without developers taking more active risk. Along both of those lines, here are some thoughts on how to move forward.
1. Someone start a 3rd party server company specializing in helping online games have smooth Day 1 launches. They’d focus on competitive server rental prices, and would be contracted for a number of days or weeks until the game is well on its feet. Developers shell out some cash for a smooth launch, the community notices, and a 3rd party company employing people takes a cut. Everyone wins.
2. Developers should try Jack Emmert’s idea of a “shardless” design, which they used for Neverwinter:
“One of the things that we do to avoid that is that we have a shardless design,” he replied. “We’re doing that instead of the typical MMO structure where a shard supports maybe 10,000 to 20,000 concurrent connected users, and you end up with five, seven, maybe 10 servers at launch. But what we have is a single server. Everybody is playing simultaneously in the same exact shard. How this is modular is that we can add or subtract CPUs on that shard to make it more robust, or to shrink it as demand changes. That way players are never inconvenienced about changing shards.”
Sounds like a step in the right direction, but it’s worth noting that even Neverwinter launched with some issues.
3. What about actually having digital versions of a game “sell out” for a while, when you’ve reached the number of preorders equal to your ability to serve those players smoothly on Day 1. This could even create a little buzz-building scarcity. Hey, it works for iPhones, PS4s and hard copies of games, doesn’t it? Square Enix kinda/sorta did this with FFXIV, but only after the game’s server woes kicked into high gear. It’d be nice to seem some experimentation on this with actual planning.
4. Consumers (and I think this is a big one), stop buying online games on Day 1. Whatever the actual release date of the game is, choose a date at least one week from that date, then buy the game. When you pay a developer $60 up front for a game that hasn’t even been tested yet, you’re sending a pretty big message about your willingness as a consumer.
5. Developers, stop acting like your game is going to work on Day 1. Be transparent with the community, admit you may have some problems, but be very communicative about what those problems are, and when solutions will be exacted. Rockstar took this road with GTA Online, outright admitting they wouldn’t have a smooth launch. That’s at least something and I applaud them for their candor. It helps that GTA Online is free, too.
6. Consumers, understand that just because you’re angry, and you passively expect a game to work on Day 1, it still may not. I don’t blame you for that, but instead I’d encourage you to refer to #4. Just vote with your wallet. Stop giving developers your money until you’ve given the game time to get on its feet, and you’ve verified it works. At some point, it isn’t just developer greed/incompetence causing you pain — you chose to spend money on a product that past precedent shows is very, very unlikely to work right at launch.
Once more, this article is in no way meant to belittle the anger or injustice you feel when a game doesn’t work right at launch. Rather, understand that merely expecting something to change (‘it’ll be different this time’) doesn’t mean it will. Additionally, understand there’s more going on here than outright greed and/or incompetence. These launch issues aren’t a fluke, and it’s an actual problem that’s probably here to stay for a while.
How you react to that is up to you, and unfortunately how much developers care is up to them. We can hope they get the messages we send (refund requests, not buying games on day 1, etc.) but passively expecting a change won’t work. Furthermore, you’ll just end up wasting a ton of anger and energy.
What do you think? We certainly don’t have all the answers, so we invite any ideas/suggestions you may have in the comments.
Follow Mark Burnham on Twitter: @MarkBurnham