Pixel Sizes

[1magus]

I have been focusing on these consoles lately and I just realized I don't know what these terms mean

8x8 Pixel
16x16 Pixel?

Does 8 x 8 mean a certain size of the pixel and the bigger the sharper they look when stretched on screen?

I'm also trying to figure out why exactly the Saturn is better at 2D stuff than the PlayStation does anybody have any specs for why?

[Joe Redifer]

Sounds like a tile size. A pixel can only ever be 1x1.

[1magus]

Sounds like a tile size. A pixel can only ever be 1x1.

mm Maybe thats what they mean in tech specs of lets say Sega Genesis it says it can support:
32 x 32 Pixel Sizes and I was wondering what that meant lol.

[retrospiel]

I'm also trying to figure out why exactly the Saturn is better at 2D stuff than the PlayStation?

It was designed as a 2D machine so it got special hardware features for 2D graphics.


http://en.wikipedia.org/wiki/Sega_sa...specifications
http://en.wikipedia.org/wiki/PlaySta...specifications

[dragonboy]

they are either refering to the size of the sprites (in lamen's terms: size of the video game puppets) or the size of the tiles (in lamen's terms: size of the Mario blocks)

[tomaitheous]

I have been focusing on these consoles lately and I just realized I don't know what these terms mean

8x8 Pixel
16x16 Pixel?

Does 8 x 8 mean a certain size of the pixel and the bigger the sharper they look when stretched on screen?

Those are just a matrix of pixels, a block of pixels. Doing single pixel manipulation is pretty cpu intensive back, so moving around objects or building out backgrounds were done in blocks of pixels. In the case of objects, the larger the block of pixels, the less cpu time had to be spent lumping smaller blocks together. Using blocks of pixels also allowed for a type of compression. If a block was reused more than once in a BG pattern, then you save space on not having to include more tiles. Same with sprites. Flipping horizontal/vertically also helped in lessening the required blocks in exploiting a BG or sprite pattern.

By the time the Genesis generation came around, this wasn't as big an issue. The CPUs were fast enough, but the issue of how many blocks of pixels could be shown on a single scanline was still important. If the Genesis was limited to 8x8 sprites, you would only get half the screens worth of sprites before it would stop drawing them, resorting to flicker. So having up to 32x32 sprites helped to show more sprite pixels onscreen until you hit the limit of 320 pixels or 256 pixels depending on the screen mode. Other systems of the time had the same or similar limitation.

The down side of larger blocks for tiles or sprites was the you limited the number of colors over the range of pixels in the block. For 32x32 pixel sprite you were limited to only 15 colors, were as a 32x32 pseudo sprite made up from four 16x16 pixel sprites could have different set of 15 colors assigned to each 16x16 group.

Saturn and PS1 could show more sprites then the screen width and I believe saturn can use 256 color palettized sprites/textures. I know the PS1 was limited to 16 colors if the object was a flat 2D plane textured object.

[dragonboy]

fun fact: The Nintendo Entertainment System was made to use 8x8 tiles but Nintendo always used 16x16 tiles instead.

[tomaitheous]

That's because a single palette index was for a group of four 8x8 tiles (16x16 pixels) for BG layer of the NES.