Re: how to efficiently storing floating point numbers with limited

> I have calculated the bits needed to represent the range of 8 digits +
> decimal point and sign to be 31....

Yes you are right, in fact I mistaken my calculation by a digit (about
3 bits)

> I am attempting a video codec which will be based on least square line
> fitting to generate functions that will approximate the RGB values of
> a pixel through a series of sequential frames. I am attempting to
> "compress" each function's term's multiplier by reducing the 32 bit
> floating point number to a 24bit or 16 bit floating point number.

I'm not sure of how the data you have can be compressed thanks to
assumptions on its input in fact. Probably the best choice is to
reduce precision as needed as discussed in the later posts.

So long,
Stefano