The Sexy Brutale ultrawide and wider solution, lessons learned

The Sexy Brutale was released in 2017 and even got covered in a Detailed Report on WSGF.

Although the game defaulted to the user's desktop resolution, it did not allow to select anything wider than 16:9. Ignoring the lack of ultrawide resolutions and attempting to proceed resulted in the common issues seen in Unity: the UI got bigger and was partially cropped at 21:9, while the buttons allowing to start the game were fully out of view at 32:9 and wider, making the game unplayable at those resolutions. You couldn't even back out of a menu to close the game. Until now!

A friend recommended that I play through the game, so regardless of whether I actually would, there was something to fix as it was on my mind. I tried my old patterns to no avail but managed to use my past knowledge (as covered in the earlier posts here!) to partially navigate through the issues. I had managed to boil down my prior Unity UI solutions to very minor changes or functions removal but this one was tricky in comparison and resembling Milky Way Prince.

Switching the scaling was not sufficient. Remember the YFOV trick in the old Unreal Engine 4 titles, which just converted each horizontal FOV value to vertical, resulting in a drastic vertical increase compared to 16:9? These UI issues were quite like that. I had to introduce my own scaling formula, which took a lot of thinking and calculating. It relies on the idea of the game being meant for 16:9 (although Unity would disagree) and then adjusts accordingly, now preserving the vertical view at all times.

Hours later and with the primary issue out of the way, I also had to allow the user to select their resolution in the game. As if it weren't enough, I quickly found that there are horizontal black bars that otherwise would have been cropped and not visible at 21:9, but here they were, limited to the 16:9 space, just like in Young Souls. The memory of that was very fresh, so I tackled it the same way, by disabling the function that creates them. The code had zero in common with that of the other game and still had to be found though.

Then were the issues of the patcher struggling to properly replace the bytes. I don't imagine that I'd be able to inject in those three non-executable files, so I had to invest time in getting that addressed.  It's the worst part, because once the solution is found and the new code is prepared and made into a smart pattern for longevity, all you want to do is publish and move on. Oh well!

Despite the struggles, going over the code and thinking up the new calculation made me realize that the knowledge would probably be useful for Milky Way Prince and Not For Broadcast, as well as other Unity games. I'm planning to revisit those two in the nearest future.