Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.
BCD floating point B12345678 (DFP085, B) 8-byte BCD encoded number. These are obsolete and should no longer be used.
The first byte of an B12345678 contains the sign and the base-10 exponent. The remaining seven bytes contain the 14 BCD digit mantissa. B12345678 values are transmitted in big-endian order.
The sign bit is the high order bit of the first byte; 0 means the value is positive, 1 means the value is negative.
The base-10 exponent is determined by subtracting 64 from the value of the low order seven bits in the first byte; the exponent ranges from -64 through 63.
B12345678 numbers are normalized so that the high order digit of the mantissa is non-zero unless the value is zero.
B12345678 values range from -9.9999999999999e63 through -1.0000000000000e-64, 0, 1.0000000000000e-64 to 9.9999999999999e63.
O.K.
If you want to make calculations with 10 digit accuracy with microcontrollers,which format can be used ?
It doesn't seem that 32bit floating point be suitable .
Do you have any idea ?
To cover 10-digit accuracy you will need at least 34-bit binary floating point calculations, but with microcontroller almost everything is possible: 40 bits, 48 bits etc ..
Hi AVR_Luciano
The URLs you mentioned are about 32 bit (and less) floating point .
Our subject was more than 32 bit routines ( 40 bit , 48 bit , 64 bit , ...) .
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.
