A while ago, I answered an inquiry into the representation of floating point numbers in AMSTRAD CPC BASIC (Locomotive BASIC, I think) in a paste somewhere. Since I can’t find it, I thought I’d replicate it here, for posterity and somesuch.
Question: How are floating point numbers encoded in Amstrad CPC’s BASIC?
First things first. A floating point number takes up 5 octets. Octets 03 contain the mantissa in littleendian byte order. Octet 4 is the exponent. The first bit of the mantissa is the sign bit: 0 for positive, 1 for negative. The mantissa is encoded so as the most significant bit is always 1. This allows it to not be stored, and the sign bit to be specified in its place.
The exponent is biased, +128.
Floating point numbers in Locomotive BASIC are displayed to 9 decimal points, unless there is no fractional part.
Let’s conjure up an example:
> PRINT @a!
326
> FOR I=0 TO 4:PRINT HEX$(PEEK(326+I),2);:NEXT I
A2DA0F4982
The mantissa is equal to 0x490fdaa2
(littleendian).
The mantissa MSB is 0, therefore it is a positive value.
The exponent is equal to 0x82.
Let’s calculate the floating point number:

Delete the sign bit, and add the implied 1 bit. In other words, do
mantissa  0x80000000
:0xC90FDAA2 = 3373259426

Calculate the decimal representation of the mantissa:
3373259426 / 2^32 = 0.7853981633670628

Multiply the value above by 2^exponent, which is biased by 0x80:
0.7853981633670628 * 2^(0x82  bias) = 3.1415926534682512
Yes, it was π all along! Note the accuracy up to the 9th decimal.
For more information on CPC internals, I suggest the excellent CPC Wiki.
« Past Future »