12-12-2017, 09:15 AM
Hi stefano!
Everything you wrote is clear and understandable :-) I may or may not agree with some of the opinions, but I won't ever get angry with someone that is evaluating and suggesting things to make a better project. So feel free to talk openly about it!
1. Text effects are "high-level" constructions, that can be implemented with tiles, sprites or background bitmap. So they must be done in the application level, not the engine level. The C# "SuperMarioClone" sister project has some basic text for the HUD using tiles. What do you mean by "shaders"? In the graphic level, 2D-style SNES games are already possible with tilengine: it has all the basic elements (tilesets, sprites, palettes. raster effects...) plus some snes-specific features, like affine-transformed layers (mode 7), transparencies/blending and mosaic pixelation.
2, 3, 4. I agree, C# is a fantastic language, it's cross-platform and has powerful free editors with great autocompletion. The C# binding is structurally very similar to the python one (objects, properties...) but with the syntax conventions of C#. It needs a bit of update to be in sync with library v1.19 and the documentation strings. However tilengine doesn't offer sound, so the C# binding won't either. That's why the python platformer example uses SDL2/pysdl. There are also C# bindings for SDL2.
5. I disagree here. The included examples should be as simple as possible to show the graphic features in a clean way. The barrel (castlevania) and shooter (thunder force) samples are cool but way too complex. The sister projects "SuperMarioClone" and "TilenginePythonPlatformer" are intended to show full game mechanics as you say: user input, sound, gameplay... That's why they're separate projects, to keep tilengine examples clean and simple.
You can use any library of choice to provide windowing, sound, post-processing shaders to tilengine. It is designed in a way that makes easy to integrate your own components. In the examples I use SDL2 because it's nearly universal and easy to work with, but there isn't anything in tilengine that forces you to use it, you can use your preferred framework.
I haven't ever worked with Lua, so I don't know its syntax or its way to interface with external libraries in native code.
The future C# binding documentation will be very similar to the current python one, as both bindings have the same structure. I didn't use pygame because pygame is also a binding for SDL2, but it's python-only, and has many features that I won't use and others that are already handled by tilengine itself. In other words: it's non portable and overkill for just providing sound effects. I always chose the most simple and standard/portable option, in this case plain SDL2. I expect to update the C# binding soon, for the documentation part I still have to research which tools are used (C uses Doxygen, Python uses Sphinx...)
Of course you can contribute with anything you think it's interesting! I'm a single individual doing all this in my free time, so there's always more things to do than time to do them...
Everything you wrote is clear and understandable :-) I may or may not agree with some of the opinions, but I won't ever get angry with someone that is evaluating and suggesting things to make a better project. So feel free to talk openly about it!
1. Text effects are "high-level" constructions, that can be implemented with tiles, sprites or background bitmap. So they must be done in the application level, not the engine level. The C# "SuperMarioClone" sister project has some basic text for the HUD using tiles. What do you mean by "shaders"? In the graphic level, 2D-style SNES games are already possible with tilengine: it has all the basic elements (tilesets, sprites, palettes. raster effects...) plus some snes-specific features, like affine-transformed layers (mode 7), transparencies/blending and mosaic pixelation.
2, 3, 4. I agree, C# is a fantastic language, it's cross-platform and has powerful free editors with great autocompletion. The C# binding is structurally very similar to the python one (objects, properties...) but with the syntax conventions of C#. It needs a bit of update to be in sync with library v1.19 and the documentation strings. However tilengine doesn't offer sound, so the C# binding won't either. That's why the python platformer example uses SDL2/pysdl. There are also C# bindings for SDL2.
5. I disagree here. The included examples should be as simple as possible to show the graphic features in a clean way. The barrel (castlevania) and shooter (thunder force) samples are cool but way too complex. The sister projects "SuperMarioClone" and "TilenginePythonPlatformer" are intended to show full game mechanics as you say: user input, sound, gameplay... That's why they're separate projects, to keep tilengine examples clean and simple.
You can use any library of choice to provide windowing, sound, post-processing shaders to tilengine. It is designed in a way that makes easy to integrate your own components. In the examples I use SDL2 because it's nearly universal and easy to work with, but there isn't anything in tilengine that forces you to use it, you can use your preferred framework.
I haven't ever worked with Lua, so I don't know its syntax or its way to interface with external libraries in native code.
The future C# binding documentation will be very similar to the current python one, as both bindings have the same structure. I didn't use pygame because pygame is also a binding for SDL2, but it's python-only, and has many features that I won't use and others that are already handled by tilengine itself. In other words: it's non portable and overkill for just providing sound effects. I always chose the most simple and standard/portable option, in this case plain SDL2. I expect to update the C# binding soon, for the documentation part I still have to research which tools are used (C uses Doxygen, Python uses Sphinx...)
Of course you can contribute with anything you think it's interesting! I'm a single individual doing all this in my free time, so there's always more things to do than time to do them...