No, Bethesda Should Not Stop Using Their Engine

Fallout 76 is Bethesda's latest release, powered by their proprietary Creation Engine. The reviews thus far have been...not great. And I'm not here to convince anyone to like or dislike the game. There is, however, one comment that I've seen many times across reviews, videos, and discussions that I want to address: "Bethesda's engine is outdated and terrible, and they should change engines or rebuild the current engine." They really, really shouldn't, and we should explore why not.

What's In An Engine?

If we're going to have a discussion, I like to clarify terms first. What's an engine? In game development, the engine is the backbone of the game. It connects the graphics, sound, physics, networking, animations and more together. This lets the developers focus on making a fun game, instead of how to make a sound file play on a thousand different hardware configurations. Someone else solved that problem so you don't have to. When you make a new game, you have a solid, tested base to start from so you can get right to the gameplay.

Make or Buy

Okay, so you need an engine. Lucky you. Since you have to pay people to make the engine anyways, why not simply buy one and start building your game faster? Here are a few major reasons, but there can be many more depending on the needs of the company and the game.

1. Cost. Licensing an engine can cut deep into a company's wallet. Many licenses are per-person, so licensing teams with 100+ people gets very expensive. The cost upfront to develop an engine can be significant, but can taper off as it matures.
   
2. Control. If the engine doesn't support something you need to do, you might be out of luck. For very specific things, it's possible nobody provides what you want.
   
3. Speed. Your own engine can do whatever you need as quickly as your developers can get them done. If you're using another company's engine, you might have to wait for them to create or fix the feature holding you back.

Bethesda has at least one key requirement that only their own engine can support -- mods. Their games enjoy massive lifespans due to the amount of control the community has over the experience. There is untold value in releasing games with a mod system that long-term modders are familiar with, letting them jump in immediately and augment the game in countless creative ways.

Skyrim augmented

This means it's a non-starter for Bethesda to use someone else's engine as it would simply be too much work (or, more likely, impossible) to retrofit their modding systems in Unreal, Crytek, etc. So why not start over?

Never Restart From Scratch

There is a blog post from Joel Spolsky that is required reading for anyone that deals with software management. It is about a Thing You Never Do. That thing is rewriting your code from the ground up. He demonstrates all the reasons this is true better than I can, but suffice it to say that it's a great way to not have a company any more.

This company rewrote their software from scratch. Don't be like this company.

Generally if a new engine is underway, it takes several years of development before anything meaningful can come from it. If it's a particularly large engine meant to power the studio's games for a long time, it may take 3 or more years before any real progress can start on a new game. For a company with an existing engine that isn't dramatically slowing down development (there might be a Telltale tale to be told here), putting your game releases on hold or splitting your dev team for that long makes little sense.

There's still a better way.

Engine of Theseus

Over time, an engine's parts are enhanced, replaced, or removed. Even Bethesda's previous engine, Gamebryo, formed the basis of their current Creation Engine. Each time a new game needs to be released, the engine must keep pace with graphics improvements, new technology patterns, and design requirements for the next title. Windows has had massive revisions over its lifetime, but we still call it Windows since it fulfills the same function (even if 10 tries to act like malware whenever possible).

The fact that the "Creation Engine" was used for both Skyrim and Fallout 76, and will be used for future titles is technically true, but misses the amount of changes that happen between each game.

Don't Blame the Hammer for Bad Furniture

I've not made any claims about whether Fallout 76 is a good game or not. By all accounts it seems to have fallen short. But blaming the Creation Engine for producing a bad game is silly. Epic Games shouldn't throw away the Unreal Engine because a subpar game was made with it. Their own Paragon tremendously underperformed, but then they turned around and made Fortnite a massive hit.

Perhaps the engine's multiplayer component wasn't given enough time to mature before Fallout 76 was launched. It's possible some of the game's odd design restrictions were a result of engine shortcomings they had to work around. These are arguments to spend extra time on the engine, or alter the design to fit the engine. You don't try to drive a nail with a screwdriver if you have any sense. It's not that something's wrong with the screwdriver, it's just not the right tool.

Future Games

Game engines can be a tricky business. Chosen or managed poorly, it can cause serious development headaches. And once everyone in your company is ramped up on a particular workflow, it can be costly to switch. I believe Fallout 76's issues stem from a number of sources, and maybe part of it is from the Creation Engine, but I think a majority of the blame lies elsewhere -- in the game design, the presentation, and seemingly not enough development time. All said, I hope Bethesda pushes its engine further than they have before. They just better not replace it.