Ohmnivore

The actual reason for Titanfall 2's perceived low player population

Preamble

I bought Titanfall 2 for PC at a discounted price a few weeks ago. I had wanted to get it since launch, but I was deterred by the players' claims of a low player population. I was more than ready to believe them after witnessing EA's Titanfail of a launch. Competing with oneself, what an innovative business strategy! The game's absence on Steam certainly didn't help either. Thankfully, in my experience, the game is yet far from dead. I don't want to go into detail but it's safe to say there are over a thousand concurrent players at peak daily on PC. For any FPS one would expect that's more than enough to guarantee its livelihood. What's the problem then?

The problem

Titanfall 2, like its predecessor, doesn't have dedicated servers. Instead, players queue up for a certain gamemode, and a matchmaking system places them into a server which only lasts for the duration of the match. Each gamemode is a separate queue in the system. The most popular gamemodes are Attrition and Mixtape. The rest of the queues sit empty. This both creates and sustains a feedback loop that drains all the players towards a very limited set of gamemodes. Nobody joins the empty queues because it takes too long to wait for 11 other players to queue up and start a match, and everyone joins the popular queues instead because the matchmaking there is near-instantaneous. In short, good luck finding matches other than Attrition and Mixtape on a regular basis. That's the real shame, and what I believe people mean when they say that Titanfall 2 on PC is "dead".

How dead is dead?

Of course, the definition of a dead game varies from player to player. In a classic user-hosted server environment, I would argue that if at daily peak two servers are well populated, then the game is still plenty alive. Map rotation would allow to vary the gamemodes from time to time. In fact, Battlefield 2: Modern Combat for the PS2 sustained itself with only 40 concurrent players all the way until 2014 when it was brutally taken offline by EA. In my opinion, an online FPS partially dies when at the daily peak you can't find a single saturated server anymore. It fully dies when you can't even play the game at all, like when the servers are taken offline.

Stubborn design decisions

Well we're in luck, Titanfall 2 has a massive amount of players. Unfortunately, it doesn't use a classic user-hosted server model, as explained earlier. The kind of model it uses instead is infinitely less resilient to low player counts than the classic one because servers aren't persistent and are only launched when the necessary amount of player slots are filled. One would think that the developers would have taken note of the issues it creates from the first entry in the series, but nope.

When the concurrent amount of players in a queue is less than a gamemode's expected player count (usually 12), the players can wait forever to no avail. After a minute or so of waiting they would leave the queue and try a more popular one, thus shrinking the queue. Rinse and repeat. Had the servers been persistent, the players could have joined one instantly and waited for it to fill up while already playing. Of course, that means that servers wouldn't be as saturated as they are at the moment, but anything is better than the alternative of never launching a server for the gamemode in the first place and leaving the players hanging.

Additionally, the current system gives the players no control over which map they want to play. The players are already locked into two main gamemodes, so this is adding insult to injury. It feels like we're regressing here. The FPS genre nailed it with server browsing early on, why can't we just leave good things as they are?

Solutions

All of these issues could be solved by adding persistent servers with map and gamemode voting, ensuring there is a running server for each gamemode, and a proper server list browser. Notice that I'm not even asking for user-hosted servers. It would also have been a reasonable request but the current AAA industry seems to be allergic to the very thought of it. Just the above is good enough.

Verdict

I don't actually expect the developers to do anything. I suspect that their hands are tied by the publisher, who doesn't really care what happens to the game once the initial wave of sales is finished. Financially speaking the earlier the game is shut down, the better - maintenance costs can rack up pretty high. Forgive me if I'm being cynical, but I suspect EA shuts servers down as soon as it can do so without incurring too much backlash. If the server hosting model remains the same, I expect the game to die fairly quickly (I give it 6 years) as it requires hundreds of players to sustain it properly, unlike Quake, Doom, and other FPS games that stay alive for decades thanks to their sane server hosting model.

I'm pissed at the short life expectancy of most online games. I especially don't think that this issue should be allowed to exist in PC FPS games as that genre works strongly in favor of a user-hosted server model. Oh did I mention the worst part of it all? Nobody even has access to Titanfall 2's server binaries, so when the official servers go down there will be no way to work around it. It won't be a matter of just reverse-engineering the server browser and/or account management services, we would need to reverse engineer the game server logic - it's almost simpler to just write a new game instead. Studios should really have a contingency plan to let the community host the servers and services once they don't want to do so themselves, but the AAA industry would rather see their games sink like the Titanic.