Using available Game Engines?

[zerotacg]

As halkun suggested it might be a better to use a complete game engine over just Ogre and writing our own.

In it's current state it shouldn't be to hard to switch to a game engine. We should think about stuff that's hard in general to implement and stuff that is a must and than see which options we have that make them easier to implement (with sticking to current self written engine being an option to).
Thinks that come to my mind are:

• Cross Platform
• ability to read native data PC/PSX, including Sound/Midi formats
• Mini Games
• Tutorials (only for menus? didn't play ff7 in a long time)
• Field Levels, 2D background with multiple layers, animations or video and 3D Data
• transitions between Field Levels, Battle stage, World map
• Vehicles

where as I think sound and video playback with overlayed 3D would be hardest

[Akari]

Using the existing game engine is harder than writing QGears. Writing engine is easy and don't take much time, but gives you flexibility. Most time are spent on reversing and understanding algorythms. you cant use native algorythms with existing engine. You will need to use their algorythms or write them in their script language (it's the same as using C++ but in LUA or Python)

Some times ago I thought about using blender to do ffvii battle. It's cool because you can see models, animations right in 3d editor and edit them as you want. But it's hard to create all battle algorythm in Python.

[Covarr]

zerotacg, you have misunderstood a fundamental purpose of Q-Gears. The goal isn't just to make a new engine, but to make one that's totally compatible with the original datafiles (even if that means conversion in realtime), similar to ScummVM or EDuke32. This simply isn't feasible using an existing engine, such as Unity or Unreal, as the vast majority of them aren't meant to read anything but their own data types. At best, you could write DLLs or something to read/convert all the old data, but if you're going to do that you have to do it separate from the engine you're using anyway, at which point it's once again easier to just write your own engine and build that right in.

where as I think sound and video playback with overlayed 3D would be hardest

This is actually relatively easy if you write the engine to support it. It might be fairly easy with an existing engine, even, but I'm not experienced enough with any of them to know exactly how easy or hard it is.

[zerotacg]

I understand quite well the purpose of Q-Gears an existing game engine might still speed up the project if it fits the requirements

don't get me wrong, I'm not all for it, I'm just open to suggestions and if it speeds up the whole project reasonably one should consider it.
And to get compatibility with original data is a point on the list (how doesn't matter project speed is important)

no matter what, you have to implement gameplay and all, it's the same with an game engine as with writing it your self so that's not a reason
but a game engine might bring stuff that really helps and speeds it up. in general I'd say they bring ai, physics, graphics, user input, entity management (like making
entities react to user input or automagically play animations etc) and, not to underestimate, tools

nothing in there yet where I'd say that would help a lot, since ffvii doesn't use physics and akari already wrote lot's of stuff you get from an game engine, but as I already
stated, open for suggestions

[valentino]

Using the existing game engine is harder than writing QGears. Writing engine is easy and don't take much time, but gives you flexibility. Most time are spent on reversing and understanding algorythms. you cant use native algorythms with existing engine. You will need to use their algorythms or write them in their script language (it's the same as using C++ but in LUA or Python)

Some times ago I thought about using blender to do ffvii battle. It's cool because you can see models, animations right in 3d editor and edit them as you want. But it's hard to create all battle algorythm in Python.

Wow, how i wish i can write engine "easily"... How long did it take?
is it that easy? I'd love to learn about it.

Because if anything qgears could be a custom authoring tools for games with similar pipeline and event designers,
so it'd help lessen the headache on algorythm and whatnot.