Archive:

State Of Unreal

Posted: 22 June, 2026

The latest State of Unreal happened at Unreal Fest and, as I had a feeling would happen, the other shoe of Epic’s grand plan has finally dropped. Unreal Engine is about to change. Significantly.

Before I harp about Epic’s changes, I should probably set my stall out. In my opinion, Unreal Engine is the best game engine that I’ve ever used. By far. No question. It’s not without its problems, but in terms of workflow, speed of development, shipping, scalability and stability, it is years ahead of Unity and Godot. It’s not even a fair comparison. The reason for this is simple: Unreal Engine has been built, over decades, with the singular focus of shipping big commercial games, and, by doing so, learning the lessons of the countless projects Epic and their licensed third-party studios have shipped, to simplify the life of ALL developers using future versions of the engine. Everything, from getting assets into the build, to iterating on them where it counts, in-engine, is more refined than anything else publicly available. And, when it comes time to ship, UE is replete with all the tools you need to debug, tune and optimize at the lowest level, for all of today’s platforms.

I’ve shipped on Unity, twice, and both times it was a shit-show. Its workflow is badly thought through in comparison and shows no signs of catching up. Godot, as interesting as it is, remains behind Unity, although I don’t doubt that it’s capable of “doing a Blender”, if (and only if) enough experienced devs chip in to help steer it. That’s more likely to happen now, IMO, but it’s a long path…

With all that said, UE has always followed Epic’s whims. When we looked at it, back in the XBox 360 days, it was very much the Gears of War engine; absolutely not fit-for-purpose for the types of things we were looking at making¸and nose-bleedingly expensive if you hadn’t already chummied up to Mark Rein. More recently, it’s very much been the Fornite engine. But, for a good long time before Fuck You money landed in Tim Sweeny’s lap, Epic were bending over backward to make an engine that anyone could use, for near-as-dammit anything, as that was the side their bread was buttered. It was easier to make money with licensing, than it was to make a hit game. That, and the pragmatism of making the engine’s source code open and available to all (so anyone could fix it), helped to get Unreal Engine so far ahead, all while providing a modicum of security; if the worst happens, I have the source code. I can fix it up, or pay someone to fix it up for me (like, support the latest console SDKs, say…)

As we’ve seen with Unity (and now Epic), if you’re making games on someone else’s engine, that’s the minimum security blanket that you need.

—

Epic’s direction of travel has been clear since UEFN was first announced: make breaking changes to the engine “in a safe space”, and slowly push all the smaller (non AAA) developers toward launching their games in Fornite, A) to do a bit of a Roblox, and B) to try and keep hold of the fire-hose of Fuck You money, by taking all of ours.

The problem with this (other than the obvious), and the thing that’s been giving me the heebie jeebies for quite a while, is that everything surrounding the engine – the store, its backend, and the surrounding platform services – are fucking shit. Getting 2umo onto EGS was a monumental ball ache, like actual days of work, all for it to sell fuck all because people simply don’t like EGS. And despite near infinite money, Epic have been incapable of making it into a true competitor to Steam. Fortnite as a platform, regardless of whatever “metaverse” wrapping paper is in vogue “this year”, is a straight-up, bet the company, rent extraction play.

Worse, Epic’s now committed to

• Deprecating Blueprints
• Getting rid of the Actor+Component system
• Going all-in on Verse

Now, I’m no fan of Blueprints and I use them sparingly, but for what they’re good at – anything latent, like UI or Animations – they’re the best thing that I’ve ever used. I can’t imagine going back to a life where these don’t exist. And for bringing people on-board, as a learning tool, or as an enabler for a tech designer / tech artist, they’re fucking exceptional. Removing Blueprints seems, on the face of it, insane. But I can guess why it’s happening; what they get compiled down to is nightmare fuel, they’re a twat to merge / maintain, and they’re never going to sit on top of the new Verse / Scene Graph stuff in a way that’s going to make sense. Oh, and you’re not going to get an AI to easily (and reliably) make them. I’d wager that last point tipped the balance.

Removing the Actor+Component system I’m slightly more sanguine about. Its most important feature is that it’s bullet-proof. It's been battle hardened over decades. It’s shipped again and again and again. But, it’s essentially single threaded and there’s a lot of OOP bloat inherent with it. It comes up frequently in profiling, so yeah, It’s far from perfect, and I can see the attraction in wanting to modernise it, but… to wholesale throw it out?! I mean, maaaybe? I guess it’s not impossible that Scene Graph ends up being great, but I’ll bet money that it won’t end up being stable for a very long time, and that matters, no matter how big your studio is.

Verse though, that’s the deal breaker.

The older I get, the less inclined I am to learn a new language. I’ve lost count of how many I’ve used at this point, and most of those were good. Verse, with its metaverse contract bollocks, Haskell influences, and Python-esque whitespace requirements comes across as an academic circle-jerk. The syntax makes me want to boak. And I have no fucking interest in making server authoritative multi-player games. But… that’s the point, it’s a programming language with the singular purpose of making you reliant on Epic’s servers, so they can take a cut.

—

It’s a weird time to be an Unreal Engine “user", but it’s not an unexpected one. UEFN was a clear sign of intent, I just hoped it would take Epic a little longer to get to this point. That the grand reveal comes right at the point where no one will buy your game is just comically ironic.

For me, personally, it doesn’t change much in the short term. I’d already wrestled with the fact that I’m not going to be able to sustain full-time solo game dev reliably, and one year on, it's clear that I made the right call. Publishers aren’t signing. Amazon is taking over what was left of physical sales, killing it in the process, and nothing’s going to change any time soon. More developers will be laid off, publishers will fold, and Epic will continue to offer Fortnite / UE6 as a life raft.

But it’s not.

—

UE6 is a few years away. It’s entirely possible that some things might get watered down, a new noodle graph visual scripting system might get mooted. Hell, Scene Graph might end up being amazing. But what’s not going to change, at least, not until it fails and takes everyone with it, is Epic trying to build Fortnite into a “platform”.

So now’s the time to start on the exit plan.

I have two unfinished projects on UE, one of which – Gilby’s World – I’m confident will ship before UE6 lands. Maenhir, on the other hand, is gonna be a bit of a problem. This might actually be the final nail in its coffin, unless I restrict myself to 5.8, and forego any easy path for it to land on console. Maybe doable? Unlikely. I suspect there won’t be a stampede of developers racing to jump onto UE6, and Epic will start strong-arming people away from 4.x and 5.x. Lets see.

Regardless, whatever I do from now on will be self-published. I have the EH500 for 2D games, and I’m going to start doing more on devices like the Play Date, and possibly, maybe, (but don’t hold me to it) I might finally get to do something that I’ve always wanted to do: make a game for the Megadrive, or maybe even the Neo Geo.

One thing’s for certain: I am NOT going to make Fortnite islands.

The rest is down to Epic. I enjoy working in Unreal Engine, and I’m incredibly productive when using it, but I’m not wedded to it. It’s far from the only option I have. Lets see what the next couple of years bring, but I think this is the beginning of the end. At least for me and UE.

Friends: