Split Fiction is published by EA.

Even just split-screen multiplayer has value. Replication is handled by the engine. User accounts are handled by your storefront. Anti-cheat is something you’re thinking about if you’re designing an e-sport, but if you’re just making a fun video game that you might play with friends, it’s a nice-to-have. Why are we even collecting data such that GDPR is a problem? I know these are all things that multiplayer devs tell you they’re thinking about as to why this is so complicated, but we’ve lost the plot here so much that they’re building a game that they’re already expecting is going to reach millions of people without even being sure that they’re going to hit thousands. Which is how we get to an article like this one.

We’ve come a long way from the days when one programmer added multiplayer into Goldeneye at the very end of development, that could never happen today.

Why? I can’t name a reason why this couldn’t be. Even extrapolating out for added complexity of network multiplayer, maybe it wouldn’t be feasible to add in just a handful of weeks, but if you’re already developing with client-server in mind, the same thing can still be whipped up today in a reasonable amount of time.

Even the rest of your comment makes it seem like if there aren’t thousands of concurrent players weeks after launch that it’s somehow failed as a multiplayer game. The industry has broken all of our brains so thoroughly that most of us can’t remember a time where that wasn’t a goal, and I’m arguing that it’s better if we didn’t make it the goal. If you make a multiplayer mode that you can play with friends, that has bots to fall back on when you don’t, and is designed to scale to very few players in a match, that multiplayer mode offers just as much value in week 1 as it does 20 years later. It’s not falling back on a single player mode, nor is it a failure as a multiplayer game in a competitive market if you build something that can withstand reaching a small audience, like the industry used to. That we used to get both modes in tons of games back in the day is what made these games “the full package” rather than only a single player game or only a multiplayer game, and I reject the idea that one of those two things has to suffer for the other to be good.

Halo didn’t have Xbox Live until the sequel because Xbox Live didn’t exist yet when Halo 1 was built, but it did still have network multiplayer. And that was still very much serving multiple masters, just like its predecessor.

I seem to recall the reason they dropped the first one being that the tech stack it was built on couldn’t support the number of players trying to play it at once.

Halo was a game with a single player campaign, that could be played co-op, and also had versus multiplayer. It served many masters. This game only serves the latter. Halo’s multiplayer was played for years by a core group, but probably the most common use case was that it was played only a handful of times with friends, everyone had a great time, and it didn’t matter that people didn’t keep playing it after those handful of times. What would make FPS games great again, to me, is if we remembered all of that stuff about Halo rather than trying to be the one viral success out of tens of thousands of game releases every year, where failure results in tons of job losses because your company has no Plan B.

Thanks. A lot of rally games have come up over the years, but to my eye, surprisingly few of them have any multiplayer to speak of, let alone local, so I have yet to try one.

There were three edits in the past few days. Feel free to look at the diff, but I’m not making a Fandom account to do so. It would stand to reason that it was this list, and those three edits probably account for Palworld’s number being lower in the graphic and why the wiki has two more games on it, if they pulled the data more than a few days ago.

https://vgsales.fandom.com/wiki/2020s

Its source is a wiki, noted in the bottom right corner. I’ve seen plenty of these numbers reported publicly before. By its very nature, it’s going to lag behind real time.

Our demographic is dwarfed by the type of person who only plays 4 or fewer games per year. These people play Assassin’s Creed, Call of Duty, EA’s soccer game, GTA, etc. Call of Duty is one of the highest selling games each year because it sells to people whose only video game for the year is Call of Duty.

I played it at release, and I had a negligible number of technical issues on PC. Not everyone had a bad time with it back then.

This is number of copies sold. It really did sell that high. And if that blows your mind, wait until you find out Human Fall Flat sold like 40M copies.

Well, I know at least one part of this is pretty out of date. Baldur’s Gate 3 just got confirmed a few weeks ago as having sold over 20M copies. We have so many round numbers here because companies generally only share milestones.

That might be a solid recommendation for others, but speaking for myself, licensed cars and tracks do nothing for me and in most cases will probably put some drag on my enjoyment, because real racing asks you to do things like “not checking the car next to you” that would put real people in harm’s way; and damaging licensed cars in video games is generally frowned upon by the licensors. And also speaking for myself, the store page says it has no local multiplayer, which is my primary use case for a racing game, so its omission is a deal-breaker. Most of the genre has gone this way in recent years, catering to the crowd that likes licensed cars and real tracks, and that’s why I haven’t had as many racing games to play of late. There’s still some stuff for me, though.

The thing is, for a game like Clair Obscur or Elden Ring, I’d echo those same complaints, but I still enjoyed them; in Elden Ring’s case, despite those complaints, I’d still call it one of the best games ever made. You might share those criticisms but still find plenty to love about it.

These probably mostly are the consensus most anticipated games of 2026, but I’ll throw a few of the ones I’m most excited for in here.

If you like fighting games, this is looking to be a great year. We’ve got Marvel Tokon: Fighting Souls, Avatar Legends: The Fighting Game (which may still be a working title?), and the one I’m personally most excited for, Invincible Vs.

I love Batman Arkham combat, and if you do too, you should keep your eye on Dead as Disco.

The FPS genre has largely disappointed me in the past decade, but despite the absence of any multiplayer modes, Mouse: P.I. for Hire looks to be delivering what I haven’t been getting from this genre for years. We should also, finally, presumably, maybe, see a release for Judas.

Similar disappointment has followed racing games, but the indie scene has been trying to pick up the slack, and we’ve got a AA endeavor from racing game veterans that looks cool, complete with a story mode, called Screamer.

In the survival space, both Palworld and Enshrouded are set to leave early access in 2026.

For metroidvanias, I’ve got Bloodstained: The Scarlet Engagement and the beautifully animated The Eternal Life of Goldman on my radar.

And in the RPG space, I’ve got my eye on Warhammer 40,000: Dark Heresy and The Expanse: Osiris Reborn coming up, both from Owlcat. Like The Expanse, Exodus is also planning to fill the Mass Effect void, because it’s unlikely that a new game called “Mass Effect” will do so.

I know how it works. Do you know of a game on GOG with dedicated servers that the company is paying for that also uses GOG’s matchmaking to find those dedicated servers? Because at that point, they may as well run the matchmaking themselves and open up the possibility for cross play, and I can’t imagine what value they’d get from GOG’s services. For instance, I’m pretty sure I’m hitting GOG’s matchmaking servers for the likes of Star Wars Battlefront II, but all that’s doing is registering player-run servers that it then connects me to.

If they’re using GOG matchmaking to find dedicated servers, then those binaries are in our hands already, as far as I know. Feel free to provide a counter example if you know of one. The whole point of using the store’s infrastructure is that the developer doesn’t have to pay for it, and I’ve never heard of a store that offers hosting for bespoke dedicated servers for different games.

I played Clair Obscur, liked it, and it’s in my top 10 for the year out of about 18 games. But man, the reception of that game in the Game Awards and such is wildly out of sync with what I thought of it, and it makes perfect sense to me that it wouldn’t end up in an outlet’s top 20.

And that likely stopped making financial sense once online multiplayer operated at larger scales. On PC, GameSpy servers came with ads. Even downloading patches for games meant going to an ad-supported third party web site.