[Rolken]

I venture that the primary threat to this project's success right now is a lack of clear direction and progress in that direction. We can talk grand vision all day, but at the end of the day there's no code committed. A lot of ideas have been thrown out there but few have been settled. So I'd like to push things along the path to clarity.


The first step is to establish, in a single sentence, what the guiding purpose of this project is. For example:

Quote

The Mobius Engine will enable the community to create the games they dream about in a specialized environment built around fluid Sonic-like platforming.

Then, in a paragraph of a few sentences, explain what the vision for fulfilling that purpose will look like from the outside. Another example:

Quote

Building a game in the Mobius Engine will be as easy as possible. You can download collections of standard Sonic art or create your own. Build a level with blocks you place from the control panel. Create obstacles, enemies, and gadgets by defining their behavior in Lua, assigning them art, and dragging them into the level. Hit Go and watch your creation come to life.

And then figure which parts of that vision you'll pursue first. For example (this is something I threw out there on IRC):

Quote

For the Spring 2012 engine build, let's focus on scripting badnik behavior, so we'll have an infinite featureless flat surface on which you've got a single badnik. You can script how that badnik responds to the character, whether it tries to eat you or run away, on which surfaces or when it'll damage you or otherwise interact with you, and so forth. Polish this, put it out, and then on to the Summer 2012 build, focusing on [...?].

Establish a master vision for the project, get consensus, make decisions, chisel them in stone. And then you can focus on the technical details for how to achieve those goals without getting lost in the technical details. Get these few early wins in, start demonstrating progress, start drawing in more developers, and you get a virtuous cycle of compounding success going.

Any thoughts or suggestions along these lines?

edit: also worth noting that I have no idea what the plans are for an IDE, or if an IDE is even the plan, so my examples are definitely not intended to be canonical.

This post has been edited by Rolken: 11 April 2012 - 01:02 PM

[Aerosol]

Remove all instances of the words "Sonic-like" and "badnik" and replace them with "high speed, physics based platformer" and "enemies/generic enemies" and you're golden, in my opinion.

[Gen]

Let's be fairly progressive with the scope of the project here.

That said, let's start with "A game engine with the intended purpose of producing a variety of similar Sonic-like fan games".

We're on Sonic Retro. Calling it anything but a Sonic fan game engine would be kind of stupid at this point.

Let's focus on two things regarding project scope right now: what kind of Sonic fan games do people want to produce, and how general purpose should the engine's internals be designed to suit the needs of the people.

[Kharen]

Quote

Let's focus on two things regarding project scope right now: what kind of Sonic fan games do people want to produce, and how general purpose should the engine's internals be designed to suit the needs of the people.

Yeah, I've been kind of curious about this engine, myself. I don't have any programming experience, so I wouldn't be too useful to help create the engine, but I have been interested in using it when it's done. However, I don't really have any idea about what it's supposed to be.

For one thing, exactly what kind of game would you be able to make with it? Would it be 2D Classic-style side-scrolling, Sonic Adventure-style 3D open-exploration, or Modern-style 3D racetrack-type levels? Would you be able to create cutscenes with it, or just strictly levels? (No, I don't mean full-out CGI cutscenes, more like in-engine cutscenes.) I was thinking that it would be fun to do something like a Sonic 2006 remake, showing if the game would have been any good if it was fully completed, and all the major complaints about it were fixed (bad character design in cutscenes, glitches and massive load times, Mach-Speed segments, etc.). However, unless I know what the engine would be useful for, I can't really anticipate what I would want to make.

Now, if you would like my opinion, and I apologize in advance if this is the wrong topic to put this in, I would recommend something like three individual sets of files (I think they'd be DLL files?) that would allow the creator to make either a Classic-style, Dreamcast-style, or Modern-style game. I mean, you make the initial choice when starting up a new project, and then it automatically sets the basic things like camera system and stuff to allow the appropriate style of level. You could set which setup would be used on a per-level basis, if the designer wanted something like Generations. Things like the enemy creation system would be identical between setups, so it seems like it wouldn't take too much extra work.

Again, I don't know much about programming, especially on something as major as this, so feel free to dismiss my thoughts. Just wanted to get my two bits out there, and try to figure out what the engine would actually be.

[winterhell]

Kharen, you'll be able to remake any kind of console Sonic game, including 2.5D like Sonic Fan Remix /Classic Sonic in Generations, racing like Sonic R and open world 3D.
Technically you'll also be able to make any kind of game with the engine( RTS, Vertical Shooter, etc) but there is going to be a codebase to easen you in making Sonic.

[Candescence]

I say we focus on replicating classic-style momentum physics in 3D FIRST. Once we've done that, that's a solid framework for creating any kind of Sonic game, whether it be Classic, Adventure, or Modern.

[Sofox]

Sounds like a good plan to me Candescence.

[Candescence]

And, also, part of what Kharen said struck a cord with me - the part where he mentioned a possible remake of Sonic 2006.

Actually, scratch that. A 'remake' wouldn't cut it, in my opinion. An outright reimagining of the game would be a much better idea. Take the whole thing back to its beginnings with that E3 trailer, the one that blew everyone away, and the amazing screenshots that came with it. Expansive fields. Hundreds of robots to wreck. There was SO. MUCH. POTENTIAL. And then Sonic Team ended up creating the train-wreck that we know and hate.

This could be a sub-project on its own. Go back to what blew us away. Keep the absolute basic concepts of the game - the island of Soleanna, a princess who needs to be saved, and time travel. And then we go nuts from there - would it be a linear game like SA2, an adventure-style game with hub areas, or a completely open-world game? What kind of plot or art style would be used? Who is playable? And etc etc. Take a few basic concepts, and expand from there. A finished product of this nature could be completely different from the original in nearly every way. Who knows.

At the very least, it would help test out the engine in a 'proper game environment', and could be a fun exercise in design.

[Andeed]

I think it could be an interesting feature, given a 2D side-scrolling engine is implemented that closely follows the original physics and structure of the Mega Drive games, would be a set of tools that would allow for easy porting of the game to the Mega Drive platform. This would have to be in the developers' mind from the start of the project if it were to be a success, but I think that would be such a killer feature to be able to use this as a universal tool to port graphics, music, altered physics, and the like, back into 68000 assembly code! Particularly for those that are daunted by standard ROM hacking techniques...

Any views on this?

EDIT: I'd guess, for this to work, there would need to be very strong cohesion, with the original engine, from the design phase onwards over all aspects like coordinate systems, button handlers, etc...

This post has been edited by Andeed: 16 April 2012 - 05:49 PM

[Retroman]

First of all, is this a custom engine?

Is this built using Unity? C+? C++?

Is it also a level editer like Multimedia Fusion 2 and 3d-editer/exporter? Will it be able to take my models, or levels?

Second, well this is going to be used for an open-source environment for gaming. It has to be open-sourced fully, so focusing on the system and how the community can contribute to code is equally as important as the codes and engine itself.

Since it's a community contribution, you got people who may or may not contribute stuff, depending on interest. So it's a hit or miss. But I'm watching this project, and supporting it.

This post has been edited by Retroman: 16 April 2012 - 09:39 PM

[Kharen]

Quote

I say we focus on replicating classic-style momentum physics in 3D FIRST.

I've never been able to understand why it's so supposedly difficult to do this, and why SEGA doesn't ever have the physics right. I mean, to me it seems as though the process would be as simple as copying the basic movement code (or a different programming language's equivalent of identical code), then take all the x-axis information and copy it to a z-axis. I'm probably wrong and sound incredibly stupid right now, but wouldn't this end up with the correct physics in 3D?

Quote

Actually, scratch that. A 'remake' wouldn't cut it, in my opinion. An outright reimagining of the game would be a much better idea.

Does it cound as a reimagining if you've never actually played through the original? I've only actually seen the original game by watching Pokecapn's LP of it. I kind of figured on changing cutscenes to play out differently (remove all dialog so they play like more elaborate S&K-style cutscenes, use Sonic Unleashed-style character models for human characters, change Mach Speed segments to play like post-Unleashed games, and edit the basic levels to fix glitches, forced Homing Attack against enemies, and remove unenjoyable things such as the billiards puzzle, etc.)

Yeah, I better stop now so I don't end up going too far off-topic. Maybe this should be something like a sub-project. I plan on getting a new computer soon, so it would be something I can actually contribute to!

[Gen]

I've added some frequently asked questions that seem to be popping up here and there in the project information thread.