RPG Maker MZ (Patreon)

Greetings! I'm still at work on 0.9's long quest, but I do have some other stuff to share. I wrote a post (with some pictures) to both inform you and work through my own thoughts, but it ended up being over 3,000 words long so I've decided to break it up into two posts. I'll post the second in a few days, likely Sunday. Unfortunately that's the one with all the pictures :/ The first post will be about RPG Maker MZ, the upcoming successor to RPG Maker MV (the engine used to make this game.)

RPG Maker MZ doesn't come out for another 8 days, but the Japanese site has already released some sample games, which means I can both evaluate the engine's performance and take a look at the code. I'll have more to say, but the tl;dr is that MZ seems to perform much better than MV, and the data structures it uses are almost identical. This latter part means it should be very easy to transfer the game and all the code I wrote over to MZ.

RPG Maker MZ

I invite you to check out the sample games if you're interested - the second one down (Louise and the Secret of the Cellar?) seems like more like the kind of game I'm making. (The first one, Battle Senki 3D Maiden Wars, has a cool name but looks pretty bizarre, IMO.) Of course, they're all in Japanese, so if you (like me) have a very limited understanding of that language, you'll probably just wander around clicking random things and getting into fights. But it'll still give you an idea of what I'm talking about!

Performance was the main thing I was keeping an eye out for. Minigames have always been a feature of The Proteus Effect, and they'll continue to be. But I've always felt a little guilty about making so many of them in an engine that - because of its crappy performance - clearly wasn't suited for it. Would MZ change that? Could it allow me to make minigames fun, or at least responsive?

I believe the answer is yes. Performance was very good, far superior to MV. Everything was very smooth. I got a consistent 60FPS (the cap) except during map changes, and there was none of that weird sprite stuttering you see with MV. It looks like the weird audio lag some people get was gone, too. Animations didn't seem to slow it down at all. The differences aren't night-and-day, but they are significant. It finally feels like a real game engine, rather than something a couple of guys somehow duct-taped together from Node.js and a web browser in their basement.

It's worth saying more about the animations, since MZ uses a completely different system. MV used spritesheets, which was both expensive in resources and didn't look that great unless it had a ton of frames (thus increasing the resources required.) Well, MZ uses files from a particle/effect generating program called Effekseer, so the animations look like what you'd find in any other modern game. The ones in the sample games look very nice - better than spritesheets, IMO - but they're nothing amazing. I know what this sort of software can do, though, and I fully expect incredible things.  There are some very talented artists who make spritesheet animations for MV using a tool very similar to Effekseer, and I fully expect them to switch over and start making animations for MZ. And they will look amazing. If only I hadn't bought so many of their spritesheet animations!

MZ's Codebase

Much to my (pleasant) surprise, a huge portion of the codebase didn't seem to have changed much at all. Almost all of the objects/methods I use in my own code haven't changed their interfaces. This may sound like a bad thing - "What a ripoff! They barely rewrote anything!" - but it's actually very good. MV is a great engine in a lot of ways, it just has really crappy performance. So, they improved the parts that needed improving and kept everything else the same. This makes it very easy to port projects over from MV, and will even make it easy to port many plugins over from MV! Plugins that mess with a lot of stuff deep in the engine may require a bit of work, but I expect mine will be very easy.

So, the long and short of it is that I'll definitely be switching the project over to MZ. The performance gains alone are worth it, and the other stuff - improved animations, better editor, etc - are just gravy.

Switching Over

I use a lot of plugins for this game. For those who don't know, plugins are bits of JavaScript code that modify or extend the functionality of the engine in various ways. The current version of 0.9 uses 8 plugins that I wrote, but most of the rest - 38! - come from Yanfly, the most talented, famous (in these circles) and prolific plugin maker for MV. Before I can port The Proteus Effect over, I need most of the plugins I use to be ported over as well.

Yanfly retired amid some drama back in October, and the task of porting his plugins to the next RPG Maker (MZ) was supposed to fall to a sort of trust he'd set up. Now, many of Yanfly's plugins are quite complex, and are among the sort that wouldn't be trivial to rewrite. I was skeptical about how quickly this porting process would happen - or if it would happen at all - but it looks like his designees have *already* ported over a huge amount of his codebase. And the engine isn't even out yet!

I compared what I use with what they've already converted, and as it stands today I'm only waiting on 3 more plugins to get ported over. And I could probably port one of them myself if need be. This has happened much, much faster than I thought it would, so I'm seriously considering moving up the timetable for switching this project over to MZ. Don't worry, this won't delay 0.9, but the upshot is that I may do the port/partial revamp in 0.10. More on this in the upcoming 2,000 word post! (With pictures!) Thank you for your support!