Thoughts on the LSL4 babes

[Kawa]

The format of an image in a Picture resource is identical to a View's cel. Which means it uses RLE.

In EGA, the RLE format is pretty simple: the least significant nibble specifies the color and the most significant nibble specifies the amount. So you can have up to sixteen colors, in spans of of up to sixteen pixels. And that means any dithering means a new run for every dot. So I'm not surprised the mountain of corpses is 3K+.

Sources:
https://www.tumblr.com/kawaoneechan/706335362813935616
https://github.com/Kawa-oneechan/SCICompanion/blob/da8067cd1c1497f57d943a87d1f786057abff905/SCICompanionLib/Src/Resources/View.cpp#L94

[lskovlun]

The format of an image in a Picture resource is identical to a View's cel. Which means it uses RLE.

Yup, but this is not very efficient for a (dithered) gradient. 1 pixel of color x, 1 pixel of color y, etc.

[Kawa]

That's the second part of what I posted, yeah :3

And that means any dithering means a new run for every dot.