The third part of the Final Fantasy VII Remake trilogy is already “playable” and currently being polished by the development team at Square Enix. Part of the speed of that turnaround time from Rebirth‘s launch just under two years ago is the team’s continued use of an older version of Unreal Engine. The upcoming sequel’s director recently confirmed why the developers have opted to stick with the older tech for the final game.
“We’ve been using Unreal Engine 4, but we’ve made a lot of modifications to fit our needs,” Naoki Hamaguchi explained in a new interview with GameSpot (via VGC). “It’s more beneficial to have something we’re already familiar with and have customized to fit our needs.” The director said back in 2024 that the team was exploring shifting to the newer Unreal Engine 5 for the third game. Clearly, they wound up opting against it.
Sticking with the more familiar version of the tools could help Square Enix get the third FF7 Remake game out faster. Plenty of other game studios have lost precious development time when shifting to the newest version of an Unreal Engine toolset, though not everyone has faced the same level of trouble with switching over. Sticking with UE 4 could leave fewer unknowns and help as the publisher ports the trilogy to Switch 2, but it also comes with some potential graphical drawbacks.
What role will gen AI play in FF7 Remake Part Three?
Elsewhere in the interview, Hamaguchi was asked about his evolving views on gen AI. Last fall, he said there weren’t any set rules within the company about its use even as Square Enix was pushing to automate entire areas of quality assurance testing work over the next couple of years. “ My stance hasn’t changed on depending on AI to produce something or to create something,” Hamaguchi told GameSpot. “ I don’t rely on AI to come up with ideas and things like that.”
He continued, “That said, AI is a tool that I and other colleagues would utilize so that our tasks would become much smoother or some of the menial tasks are taken care of by AI. I think it’s becoming more and more necessary. I think the purpose of utilizing these tools is so that creators will have a better environment so that they can work more efficiently and they can explore more creative options.”
He said QA was one of those areas, citing the example of testing collision detection, which is a repetitive series of “tedious tasks.” “By automating that using AI, we reduce the stress on the human that is required to check on it,” Hamaguchi said, before adding, “And so we’re hoping that that would lower the bandwidth that is required of our people.”
