Subject:
|
Re: FW: Something else is needed, I think...
|
Newsgroups:
|
lugnet.robotics
|
Date:
|
Thu, 6 May 1999 07:40:30 GMT
|
Viewed:
|
1069 times
|
| |
|
|
In lugnet.robotics, mario.ferrari@edis.it (Mario Ferrari) writes:
> > It is also not clear what Mario meant by four decimals. Maybe he meant
> > four fractional places?
>
> Sorry I was not clear (it happens when you write in an idiom that's not your
> main one...) Yes I meant four fractional digits. Any long int represent a
> radix 10 number multiplied by 1e4.
28.4 is a pretty good general-purpose fixed-point representation. It gives
a domain of -134,217,728 to 134,217,727.9375, and you can square numbers up
to 11,585 without overflow. 28.4 is also well-suited to vector graphics on
relatively low-resolution devices, for example, where you have oodles of dot
products and sub-pixel coordinates.
A couple of other popular representations are 24.8, 16.16, 8.24, and 2.30,
but these are progressively limited in the magnitude of domains they can
safely handle.
In choosing where to put the radix point, it's important to consider what
sorts of operations can occur in the problem domain. Will there be lots of
adding or lots of mulitiplying? Multiplying large numbers by small numbers
only, or multiplying large numbers by large numbers? Summing small
quantities of large numbers or large quantities of small numbers?
For example, 16.16 is probably excellent for neural networks, but 8.24 --
though it would provide much more detail to the all-important numbers inside
the unit interval -- would likely fail from overflow when summing complex
columns of weight vectors.
--Todd
|
|
Message has 1 Reply:
 | | Re: FW: Something else is needed, I think...
|
| (...) Thanks Todd for your suggestions. 16.16 would probably meet my requirements for the particular task I am working at present moment. Anyway I suppose I can write general-purpose code to be used with different radix points. (...) I made all (...) (25 years ago, 6-May-99, to lugnet.robotics)
|
Message is in Reply To:
| | Re: FW: Something else is needed, I think...
|
| (...) I must admit I didn't think at radix 2 for fixed point math. It is obviously the best choice to implement. I used radix 10 fixed point math because it came more natural to me. (...) Sorry I was not clear (it happens when you write in an idiom (...) (25 years ago, 5-May-99, to lugnet.robotics)
|
32 Messages in This Thread:
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
This Message and its Replies on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
