# Exploring Minibits and Experimental Fracbits for Graphics Representation

#### ThetaSigmaPhi

Registered Member
Hello I am Theta Sigma I have been enamored with computers and science, I have been exploring something. I am trying to create computers based on floating points in between 0 and 1 that are fracbits that work like analog I’m trying to
Here's an explanation of minibits and fracbits, along with their potential use in graphics representation. A minibit is a small unit of information, representing a single bit or a fraction of a bit. In the context of graphics, minibits can be used to represent subtle color variations or texture details. By using multiple minibits, you can achieve a higher color depth or texture resolution than traditional bit depths. A fracbit is a fractional bit, representing a value between 0 and 1. Fracbits can be used to represent fractional values in graphics

Potential use in graphics representation:
- Minibits and fracbits can be used to create a more detailed and nuanced representation of graphics data.
- This can lead to improved image quality, especially in areas like texture mapping, shading, and color gradients.

And:
- Imagine a pixel represented by 4 minibits (2^4 = 16 possible values).
- Each minibit represents a fractional value (fracbit) between 0 and 1.
- Combining these minibits, you can achieve 16 different shades of a color, with smooth transitions between them.

We already talk about 16 bit or 24 bit colour, for example, which is how many data bits are assigned to each colour pixel in an image. For instance, a pixel that has 4 bits assigned to it can display one of 16 possible colours.

I don't see how your "minibits" change anything about the total amount of information that needs to be stored to get a particular colour depth. If you want to represent values between 0 and 1 digitally then you'll need to assign a certain number of bits to do that, same as with any digital value stored.

Conceptually, I can't see any difference between storing, say, a grey-scale range of pixel brightnesses as a number between 0 and 255 (8 bits) and storing it as 256 discrete values between 0 and 1.

Can you please explain what you think the advantage of your "minibits" is supposed to be, because I don't understand so far. Am I missing something?