RPGM U2U

RPG Maker Questions
1

What is your opinion?

It is build on unity, so hopefully it will be optimized better allowing bigger maps and so on. But I worry about performance on mobiles/android since it will run heavy Unity architecture.

Hopefully porting will work just fine to WIN/Android/Linux

AND I HOPE, it will support webm/webp/ogg formats and many other.

Playing workarounds to make webp to work with RPGM MV was annoying. Reduce size by x3-5 so it is a must I think. Especially for mobiles with their limited size capacity (probably relevent only for older devices, is 2-4gb MAX still a thing?).

1 Like
2

It’s hard to make any useful statements without seeing how it’s coded, but this is my experience. Optimization and flexibility are often in opposition, and unfortunately RPGmaker must be geared towards flexibility.

3

How did RPGM Unite performed? I never really dig into it as I switched the engine after completing a massive commercial project. Wasn’t happy with java performance.

4

Unite ran incredibly slow. I was not able to research much into its code architecture but one of the key mistakes was it tried to build an engine that did not use tilesets. Instead, it attempted a system where every image was its own game object. It had the side effect of every tiny icon being prepackaged with tons of metadata and it negatively impacted loading times.

Another key mistake: It also built a system where NPC’s are defined in a global database where RPGmaker traditionally ties NPC’s to the map. This choice is fine if you were building a Stardew Valley where NPC’s have only one location in the game world and travel between maps. However most RPG’s don’t need this feature and work better with the event system that most of us use.

Also, javascript runs just fine but many plugins out there have hidden performance hazards that aren’t common knowledge. If you’re a dev who writes javascript in notetags for plugins to run, it runs thousands of times slower than javascript written in a normal plugin. Javascript must be translated into machine code for the computer to do anything with. Plugins are fine because it’s all translated into machine code when the game boots, but any code you write in a database, or in a script call in your events has to be translated and matched with memory addresses post hoc which is a very slow process.

Ultimately, game performance is on the onus of the dev.

2 Likes
5

oh also the fact that unite ENTIRE game was built in a UI CANVAS and not normal 2D plane. this one was wild to me. canvas is not optimized for that in unity. However as mention if you follow unity workflow you can achieve incredible level of flexibility

but the way unite was written was basically : lets just make a BOX built on top of Unity and make it virtually incompatible with any Unity assets.

Its the only rpg maker I have actively refunded

2 Likes
6

Not super excited about it, I think they went wrong when they try to use other game engines instead of making their own/doing their own.

7

I think my biggest issue was how terribly MV would scale with resolution. Changing to 1080p would immediately have a striking performance effect.

8

I might be a bit blunt but I have Unite, and I was quite disappointed. Many things were left when compared to previous versions of the engine, while it was also not great for using Unity’s own capabilities.

Things like not being able to create looping maps is always something that feels weird, when most engines (not RMXP) have those features. I also felt that the curve of learning was way steeper, while it didn’t feel like it was trying to do much different to MZ, for example.

Since we’re talking about Unity:

  • Unity’s runtime footprint is small compared to MV/MZ. Chromium is giant. An empty project in Unity builds into about 20Mb or so, while a complete project (of course depends on scale) can very well fall around the 100Mb of just binaries. Compare that to 200Mb that just the Chromium binaries use.
  • Formats shouldn’t be an issue. Unity and most game engines convert all assets into their preferred format, so usually what you need is some sort of importer for the assets you’re using. Unity by default supports WAV, MP3, OGG. And will convert those internally into OGGs. What it does with textures and video I’m not 100% sure but they seem to do something similar. I do know that it also converts models from Blender/etc into FBX, which is what the engine supports internally.

Using canvas for gameplay was a weird choice indeed!

But I think using Unity as base for your engine is a smart move. You don’t need to build things from zero all the time. Also Unity has lots of tooling that should be useful, and would cost a lot to develop in-house. Even things like getting console approval, having tooling available early and access for giving it support. Those things don’t come for free, and by leveraging from Unity you kinda get those things done for you.

I do think that what was done for Unite was not really good. I understand some of the decisions taken, but it also felt overengineered in parts. The tileset setup was especially awkward for me.

2 Likes
9

Broadly speaking, JavaScript isn’t actually translated into machine code. It gets compiled into bytecode that is then consumed by a virtual machine that is written in a language that does get compiled into machine code. When running in browser-like platforms, these virtual machines have JIT compilers that generate machine code from the bytecode, not from the JavaScript language source text.

Embeddable JavaScript engines such as QuickJS also execute bytecode, but don’t carry JIT compilers.

What you’re describing is the cost of launching an eval (compilation into bytecode, cold path, still not JIT’ed) for the strings of code found in the database or event commands, and running it in the plain bytecode interpreter mode.

It’s a terrible idea, both from a technical point of view and from a “social” perspective. It’s like sailing an oil tanker for fishing, when all of your fishermen are small-boat experts.

I understand you find it enticing, because you think it gives you things for free. But that kind of thought process leads to the current state of software: a bloated mess.

I encourage everyone interested in programming to watch the “Preventing the Collapse of Civilization” talk by Jonathan Blow:

That said, I’d like RPG Maker U2U to do better than Unite because I wish GGG well. But I hope they’ll come up with a proper “mainline” or “standalone” RPG Maker version after it, leaving Unity behind.

In the meantime, I’ll be keeping an eye on the engine revamp that @niokasgami is working on, and probably publish some things to extend the life of MZ too.

2 Likes
10

the problem with how they built unite is they USE none of the tooling and basically make it impossible to use them either. or was very cumbersome.

1 Like
11

Jonathan Blow has millions to spare. Most businesses try to make a profit.

12

I have two dumb questions, if Unite is made in unity does that mean each time Unity gets updated does Unite need to be updated? Also do games made with Unite follow Unity’s guidelines like the tiered revenue pay structure?

13

Unite and Unity are separate. You install one version of Unity, and then create a project using the Unite Wizard application, which then imports all relevant packages and opens your project in Unity.

When setting up, Unite will give you multiple suggested versions to work with. I think last time I was doing something in Unite I tried 6.0 and it worked, I don’t remember right now if Unite itself says that it supports it.

1 Like
14

Thanks for the response, you’ll have to forgive me I’m pretty not very up to date on how unity works :sweat_smile:

So if I have it right, Unite generates a RPG Maker like project in Unity then it’s just a Unity project after that. And if a newer version of Unity maybe doesn’t work with Unite properly, you can set it to a previous Unity version that does?

I ask because what I like about the stand alone RPG Makers is that they’re just a one time purchase and then there’s no strings attached after that to using it. In some ways Unity almost felt like it’s never a done deal, that makes me have a bit of apprehension to using RPG Makers that use Unity.

15

Hey all, thanks for the interest in what’s coming with U2U!

We had the first hands-on trial of the new map maker at BitSummit last month, and got a lot of great feedback from everyone. We also looked at all the comments from people who weren’t able to attend: the worry about Unite being repeated, concerns about Unity as the backend again, all of it.

So while we’re not ready to release some new information at this time, know that the voices have been heard, and I hope the next information release will help people to better understand just how different U2U is from Unite in all the ways that matter.

So please be patient, and do keep us updated on your wants and desires! Our ears are still very open!

9 Likes
16

u2u looks promising, but the file size will probably be larger. This engine is going to really test the artists.

I’m optimistic about this engine, but let’s hope I don’t get slapped in the face later.

1 Like
17

Personally I’m excited about a non unite engine using the Unity framework because it doesn’t require a wrapper to publish unity projects to consoles. MV and MZ both used HTML5 which Sony and Microsoft won’t host unwrapped due to security concerns, so this should make console publishing more accessible without having to go through additional steps!

18

Unite was globally a failure - too expensive, and unfortunately Unity. Things would have gotten interesting with Unreal Engine, but I fear unity remains lacking. I personally liked the nodejs version.

Also, U2U2 is taking its first steps at 2D-3D, and late compared to RPG Developer Bakin on that topic. It has to be extremely good, at MV/MZ level, for people to want to use it.

We’ll see, but we need a limited time demo version too - I personnaly didn’t try Unite as I had no way to test it when it came out without paying a price I found excessive. And I missed the opportunity later on when it was freely accessible for a short while.

Edit : It seem unite is less than 20€ right now on the store - perhaps time to still try it out

19
Offtopic rant on Unreal lack of actual 2D

From my experience, Unreal is kinda poor for 2D. It can do it but it doesn’t really offer anything special for it.

Unity at least tries to pretend it has a 2D renderer.

Both do the same approach: a 3D space with orthogonal projection and a camera set to face directly forward. But Unreal doesn’t even offer special 2D classes, instead what you do is grab all the 3D ones, lock them in the Z axis, and pray.

If we go even deeper, none do proper 2D. For that you want either Godot or GameMaker.

U2U would do fine on Unreal but I’d say only if it focused on the “perspective 2D”, as I think Unreal is kinda poor for pure 2D. It’s just that technically it can do it but even the example project is kinda barren.

I think the main concern I have with U2U is if it really brings Unite’s base features to MV/MZ level. And by that I mean specifically the editor experience. I think the games themselves worked okay, but editing the games felt miserable because it either lacked common features that were staples forever and were common sense stuff. Or the ones it had just had very obtuse implementations.

I know Unite is meant to be treated as a new engine. As in, it doesn’t need to carry over things from previous ones. It doesn’t need to limit itself by accomodating existing users necessarily. I do think that doing so would present it as welcoming to existing customers, but it is not something you have to do by design to make it work. However I do think that the current design and the current experience is lackluster.

I’m sorry if I am way too blunt on this point onwards, but I have to explain myself. I was revisiting Unite yesterday and it literally pisses me off every time I open it because I find even more things that are just plain wrong in terms of UX. And you’d think are super minor nitpicks but they pile up.

What do you need to do to delete an event in MV? Press delete. Boom, it’s gone. You don’t need a confirmation window. It just slows down the process. Imagine if I create a whole map full of events, do I just delete the map instead and redo it from scratch instead of deleting 300 events?

image
image486×281 40.4 KB

I had a couple interface bugs but I can’t say a thing about that because I was using Unity 6.3 since that was what I had at hand, so that’s on me, recommended is still v2022. v6.3 generally works but yeah with issues. I’ll install 2022 later for other testing.

I post this with big hopes that U2U can learn from Unite’s shortcomings and become something better.

20

Honestly, I probably won’t get U2U, solely because I sort of dislike Unity. If they made something built on Godot or Unreal, I might give it a shot. (I haven’t actually tried Godot, but I’ve heard good things about it.) And if they made another JavaScript-based engine in the vein of MV and MZ, I would strongly consider buying it.

Response to off-topic Unreal comment

I don’t think this is really true about Unreal? I’ve only done 3D in Unreal myself, so I’m not sure how good the 2D stuff is, but there is a whole Paper2D module dedicated to this stuff. I know it can handle extracting sprites from a tilesheet, and I think it might have support for tilemaps as well. (I even heard a rumour that it can import Tiled map files.)

You can certainly also do 2D in Unreal using 3D assets and a fixed perspective though.