advantage of floating point unit

Status
Not open for further replies.

dizgah

Member level 5
Joined
Nov 8, 2009
Messages
91
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,286
Location
iran-8par
Visit site
Activity points
2,049
hi every body
i saw very article & notes in benefeits floating point unit processing but why & where & when really we need this?
we can do floating point calculation in software form altough i dont know application of this!
WBR
mahmoud
 

hi,
Using a 'hardware' mathematical calculation processor is faster than using maths software .

You load the maths processor with the calculation parameters and start the calculation process, when the result is available, the maths processor will create an interrupt to the main processor.

While the maths processor is doing the calculations the main processor is running the software program.
Basically 'parallel' processing.

E
 
Reactions: dizgah

    dizgah

    Points: 2
    Helpful Answer Positive Rating
really thanks for good answer
but when or where we need this float point calculations?do you have some good example ?
thanks again:-D
 

hi,
I have used FP maths for GPS Lat/Long conversion to Range/Azimuth and Elevation.
The 'trig' math functions requires FP in order to get the accuracy required.

Its possible to do the FP maths in software, but depending how the compiler optimises the coding, it can take some time to process the maths.

E
 
Reactions: dizgah

    dizgah

    Points: 2
    Helpful Answer Positive Rating
You will especially need floating point maths when you multiply a very large number with a very small number and still retain accuracy.
i.e. say you want to multiply two variables where one or both have a possible range of from 0.001 to 100000 and you want to retain 4 digit accuracy. If you do it in integer, you will need at least 40 bit wide multiplication. With floating point you should get away with 16 bits. Sometimes the range is much bigger and you'll need hundreds of bits such as calculations for solar positions etc.
The disadvantage of floating point is that it takes longer to calculate, especially because the result has to be normalized after a calculation.
 
Last edited:
Reactions: dizgah

    dizgah

    Points: 2
    Helpful Answer Positive Rating
Status
Not open for further replies.

Similar threads

Menu
Log in
Register
Cookies are required to use this site. You must accept them to continue using the site. Learn more…