I can’t say a lot about the “project that should not be named” of course. But I can give you sort of a high-level run through of the type of work required to move Battleborn to UE4, should we ever need to do that.
But, Battleborn was built in UE3 after YEARS of customizing tools. Most of what let us build the game as designers were custom additions to the UE3 framework, shaped around the features we use regularly. Many of those systems were inherited from previous projects at Gearbox.
When you Upgrade to UE4, they changed enough paradigms about how the game works that most of those tools wouldn’t easily port. We would have to make smart calls about what the most important ones would be, and then move/adapt them one-by-one to a new system.
For example, the editor for Battleborn has a wizard for making a Title. Simple thing. That Wizard creates a couple of definitions in Unreal, links them, sets up naming and package structure, and then jumps out to our backend system, creates an object there, links it to the in-editor definition. At which point, the designer has to open all of those objects and definitions and input text, link the icon file, etc etc. Without the wizard, that’s about a 20 minute process. With the wizard it’s about 5. (Not including the time to graphically create the icon, which is the most time consuming, but completely engine indepedent). Multiply that times 400 titles, and you can justify the day it took us to code up the wizard.
Now, apply that very very straightforward simple component to something like a Battleborn Character. We have a wizard for that, too. It creates a framework where pieces may be built - a web of somewhere near 100 or so elements that comprise the character. when we have wizards for mutations. Wizards for skills. So on so forth. Each of those things designed to save time and tedium and maintain standards which allow us to enhance and balance the game after launch. THEN, consider that each one of those skills is hand-assembled in a logic-based visual design language we implented custom in Gearbox (called “Construct”). That framework was years in the making.
All of those things, as needed to make component pieces, would need to be re-created in UE4 for us to pull Battleborn forward into the editor.
When you hear things like “Nintendo/Microsoft/Sony” has made it easier for a developer to move their games to our platform, what that generally means is that if I already have a game in a particular engine on one platform, there are tools for porting that to the same engine on another platform (usually PC to one of the other platforms). Moving from one version of an engine to a newer version (like, 10 years newer, right?) is a much much larger undertaking. That’s why we don’t do it all the time, even when it can bring significant benefits.
Hope this helps understand a bit more about the process, and what most developers have to go through when “changing an engine”. It’s always a big deal.