I miss the day when Fallout 4 announced and then release in just a few months. Ever since FO76, they’ve been announcing stuff waaaaaay ahead of schedule. Kinda kill the excitement.
I would love for a big franchise or studio to just drop their next big hit some day with zero leadup. HL3 would be perfect for it - just pop it up on the steam homepage and see what happens.
From the official press release:
Our teams are now developing The Elder Scrolls VI and Fallout 5 on Creation Engine 3…
Fuck!
Well… At least it’s going to be moddable I guess, so we’ll wait a year after launch for the community to fix the game’s issues.
So new games are still ten years away? Five years away?
With no insider information, Elder Scrolls VI probably comes 2028 or 2029, as that’s about what Bethesda’s cadence is. That would make Fallout 5 somewhere around 2033 or 2034.
Whatever Fallout Obsidian is doing is happening now, not after the next Elder Scrolls game is done.
Might not be Fallout 5 — Obsidian’s last Fallout game was Fallout: New Vegas, which I’d consider mainline, even if it fit between Fallout 3 and Fallout 4 and didn’t get a number.
I bet it’ll be a live service 🙄
They’ve already got one of those, Fallout 76. They’ve kept it going, and it’s probably even profitable at this point thanks to the TV show, but I doubt they’ve got the appetite for another one of those right now.
I doubt that they’re going to do a live service thing either, but at least part of the problem was, as I recall from postmortems, that they didn’t have people with experience doing network games going into it and figured that multiplayer capability would be easy to retrofit in. That generally isn’t the case — if you want to permit client-side prediction, to keep the game responsive, which you probably do, then your engine has to support rolling back all state in the event of a misprediction. Like, say my client predicts that another player character keeps walking straight, doesn’t push any keys. Given that, it calculates that the rifle grenade I just fired sails past them and hits a monster and kills it, preventing it from afflicting me with a poisoned status.
Then it gets a new update. Actually, it turns out that a while back, that player veered to the right, right into the path of that rifle grenade, and it hit him and killed him. The player is no longer alive. The monster that my rifle grenade would have hit is not actually dead, and so it survived and afflicted me with poison. All of that world state needs to be rolled back and recalculated (and this rollback done efficiently). That means that you can’t just leave any state about the world anywhere that doesn’t get rolled back.
Plus, Fallout has a scripting engine, Papyrus, and if it has any state stored, that has to be rolled back too.
That’s not so hard to do if you think about it from the get-go, and are careful about how you store state. But if you’ve been writing a purely-single-player game for decades and not thinking about any of this, it’s easy to stick state somewhere that doesn’t get rolled back. And that’s gonna manifest in all kinds of unpleasant ways during runtime, when you occasionally hit corner cases.
IIRC, they had to bring in network game people to help out.
But my guess is that that’s also something of a one-off effort to fix most of that. Like, I doubt that doing Fallout 76 2 would be as bad technically as Fallout 76.
That being said, I’m not really looking for a live service game myself, and I doubt that it’s what they’re doing. Plus, from a business standpoint, if you already have a live service game, I suspect that you’re probably better off just extending the existing game and leveraging the existing player base than starting from scratch.
