Hi JGO,
My first post here. I am not sure if this is the right board to post this to, but for me this is pretty much
a newbie question. If it should be in the API & Engines board i’m sorry and would appreciate it if someone
could move it. I don’t wanna doublepost it.
I am absolutely new when it comes to anything gfx related other then doing UIs with AWT/Swing.
I also have no experience in working with any modelling software.
I started to get my feet wet with OpenGL ( using LWJGL ) by making a game as a small hobby project.
I got started by looking up tutorials and just playing around with it, but now i need a little help.
My game is going to be in an isometric perspective. The x/z plane is my ground plane and y the height.
In isometric that gives me 4 possible camera directions by rotating around the y-axis by 90° each.
I haven’t messed around with anything but building simple things and getting a texture to appear on them.
I now need a very simple human looking model with a walking animation and because i lack any knowledge in this area
i am writing this post in the hopes that someone can point me in the right direction.
The possible ways of doing this that i came up with:
-
Just using a 2d texture on a quad and change the texture every frame to animate it.
I would need to draw the human figure frame by frame in multiple directions. -
Make a basic model out of cuboids( is that the correct terminology? ) and try to animate them cuboid by cuboid.
I guess this is how the minecraft playermodel works, at least it looks like 6 basic cuboids. -
Bother to “learn” to use a 3D modelling software.
Well i looked at blender and basic .obj exports are just textfiles that contain all the data. but anything with a
skeleton for animation seems to be another beast. And i also searched the lwjgl javadoc and i would have to write
code to handle all that myself.
The model size is only in the range of 64 pixels ( at least so far ) and there would be a maximum of around 100 of those
visible at a time.
I have no idea about performance, but option 2 seems to be the one i am favoring because it seems to need the least
amount of work and considering the size of it in a 1024*768 resolution ( or bigger resolution ) doesn’t require a too good
looking animation ;D. But as i mentioned i have no idea if that would give me a decent performance.
So far i am doing everything by using glBegin(GL_QUADS/TRIANGLES/LINES/POINTS)…glEnd(); which seems to be a deprecated
way of drawing in opengl, but i would like to keep it really simple for starters so i would also use this to draw my models if i am
going by option 2.
Can you tell me if i may run into any performance issues going down the road of option 2? Is there another possibility of doing this
i haven’t thought about?
I appreciate any input on this, thank you.