This week I'm starting on the game graphics system for Ice Slider. I'm not sure how long this will take as I'll need to learn a bit more OpenGL in order to do this properly - I haven't yet fully looked into using depth buffers, stencil masks, or playing around with the alpha blending. Also from some sprite tests I've done I think my present rendering code using OpenGL immediate mode can only render a few thousand sprites on my computer before getting noticable slowdown. Given I'll be rendering a lot of small sprites I don't think this will be enough, especially because my computer has a semi-decent video card in it. I'm pretty sure it's down to me doing something very naive with the rendering code - doing a few expensive calls or switching on unnecessariy rendering options - but I'll have to spend some time experimenting with options sooner or later.

First I'm getting the basics of displaying the "ice playing field" right before worrying too much about the speed issues. So far I've got a little way into displaying it correctly, but I'll wait until I'm finished before posting a screenshot.

Gameplay-wise I'm still undecided about exactly what I'm going to do. The dilema is about whether I stick to my original plan - having a simple ice maze that the player has to navigate through, where clicking a direction key makes the player avatar slide from their position until they hit a wall - this was based on a mini-puzzle I remember from both one of the Ultima Underworlds (I forget which one) and Tales of Symphonia. Given I've had so much trouble sticking with an idea I decided to just go with this come what may - but that was before I realised that John Hattan's Duck Tiles is based on the same premise. I'd already sketched down a gameplay listing before I had played Duck Tiles so there are some significant differences (most notably I doubt I'll lovingly handcraft several hundred levels for my game [smile]) but I'd feel a bit more comfortable if I could distinguish Ice Slider a bit more from Duck Tiles.