MaxCoderz

Unfortunately thats due to the low resolution of the calcscreen and possibly a loss of precision due to the integer math involved. Such things arent likely to change.

That said ive come up with a few ideas to improve efficiency ... unfortunately its only an 8-bit world atm, ill need to rewrite all the projection related stuff if i want things in 16-bits but that would be a massive performance hit i suspect .

_________________
"My world is Black & White. But if I blink fast enough, I see it in Grayscale."

Noooooooooooooooooooooooooo!...

*Water glass*

...Ooooo!


Well, it could be pretty cool to have for game cinematics/title screens. Just like that 2by2/Noice animation stuff.

_________________
My diet consists of nails, code-stealers, and HP fans.
Projects:
Robot War [TI-82, TI Flash App]
Sonic the Hedgehog [Multiplatform]

Very impressive demo! Then again, my love of spinning 3D shapes is hardly a secret...

Extremely low detail starwars models :

_________________
"My world is Black & White. But if I blink fast enough, I see it in Grayscale."

Nice! Is that a game idea brewing, or just experimentation?

At the moment its just experimenting . Right now it runs at about 45-50fps just rendering a single low detail model (well model of lines ). Not sure what the performance hit would be for multiple objects. Plus there is no object handling and subsequently there is no 'camera' yet, which is going to introduce some decent overhead. Not to mention the fact that the world is 8-bits ... which is also what many of my optimisations take advantage of (ill have to reintroduce that pesky division if i expand ).

Oh and there isnt any clipping either. 2D is easy enough, but i was wondering if i could get away with just near-plane clipping as far as the 3D aspect of it is concerned? Think that will suffice or is it going to have to be against a frustrum? :upset:.

_________________
"My world is Black & White. But if I blink fast enough, I see it in Grayscale."

You wouldn't need to expand all coordinates to enlarge the world; you could use 8-bit coordinates for the model, and 16-bit coordinates for its position within the world.

Yeah what i might end up experimenting with is translating everything into camera space and then only running calculations against anything thats within 8-bits of the camera (maybe even only 7-bits). Anything further out would be too small anyway. Of course the camera is going to initially be unrotated so im not sure if thats going to represent a problem.

_________________
"My world is Black & White. But if I blink fast enough, I see it in Grayscale."