Hello, I have a new question - it is a theoretical one.

I know it is possible to, say, load a .3ds file into OpenGL using a custom load function or the like. I decided that I didn't want to have a .3ds file sitting around, so instead of loading from a file, my .3ds model is a part of the code (I initialize an array with these vertices, etc).

Now I am working with texture maps and want to do the same thing. I know how to load a texmap from a file, but now I want to just include it as part of my code - you know, initialize it as a variable without having that extra .raw in with my executable.

So I would end up with a bigger executable, basically.

Now, the 3ds model I am using as well as the texture map both appear throughout the entire run of the program, so I figure I don't care about the overhead, but... I also haven't seen this approach taken very much and I wanted to know: is it a good idea? Or is it better to load from files?

Opinions and ideas? I would like to develop an educated opinion before I get too far in.

Thank you very much for your comments. The more I think about it, the more it seems that loading a model as a part of the code's compilation (or loading a .raw bitmap as the case may be) is not very versatile or flexible. Also, I suppose that if you have a loader, you can reuse that as well instead of having to create a new array of verts. Also, if I assign an array as part of my code, even if I free that array when I am done, all that space must still be taken up in memory since it is literally a part of my program/executable (where I assign it).

Do you think there are any advantages to making a (ie) 3ds file a part of the code rather than loading it from a file? Or would it be totally useless?