Hello All

I wondered if i could get some pointers for  books for details of
Fixed Point and Floating Point arithmetic.

Thanks in advance.

JO

thunder wrote:

> Hello All
> 
> I wondered if i could get some pointers for  books for details of
> Fixed Point and Floating Point arithmetic.

How about math. textbook for elementary school?

VLV

>Hello All
>
>I wondered if i could get some pointers for  books for details of
>Fixed Point and Floating Point arithmetic.

Well, I would start with something more readable than the book I'm about
to suggest, like this essay (which is all over the place on the
Internet...just an example) :

http://www.eason.com/library/math/floatingmath.pdf

However, that's just floating point.  If you have a high tolerance for
pain, and really want to understand both floating and fixed point, you
might get a copy (inexpensive, but sometimes out of print) of the classic,
J.H. Wilkinson, "Rounding Errors in Algebraic Processes".  You should just
set aside a day to read it, rather than trying to do it in multiple
sessions.  This is the same guy the Wilkinson prize is named after.  You
might also look at the various floating point specs.

It's not clear what you want, but Randy Yates has previously linked to his
intro to fixed point:

http://www.digitalsignallabs.com/fp.pdf

What exactly are you looking for?

>> Well, I would start with something more readable than the book I'm 
>> about to suggest, like this essay (which is all over the place on the 
>> Internet...just an example) : 
>> 
>> http://www.eason.com/library/math/floatingmath.pdf 


Sorry, just realized the version I linked to looks terrible.  This is a
better copy:

http://dlc.sun.com/pdf/800-7895/800-7895.pdf

On Dec 10, 9:33=A0pm, thunder <ja...@hotmail.com> wrote:
> Hello All
>
> I wondered if i could get some pointers for =A0books for details of
> Fixed Point and Floating Point arithmetic.
>
> Thanks in advance.
>
> JO

http://www.amazon.com/exec/obidos/ASIN/081764704X/itbookscatalo-20
Handbook of Floating-Point Arithmetic

Divya Rathore

"Michael Plante" <michael.plante@gmail.com> writes:

>>Hello All
>>
>>I wondered if i could get some pointers for  books for details of
>>Fixed Point and Floating Point arithmetic.
>
> Well, I would start with something more readable than the book I'm about
> to suggest, like this essay (which is all over the place on the
> Internet...just an example) :
>
> http://www.eason.com/library/math/floatingmath.pdf
>
> However, that's just floating point.  If you have a high tolerance for
> pain, and really want to understand both floating and fixed point, you
> might get a copy (inexpensive, but sometimes out of print) of the classic,
> J.H. Wilkinson, "Rounding Errors in Algebraic Processes".  You should just
> set aside a day to read it, rather than trying to do it in multiple
> sessions.  This is the same guy the Wilkinson prize is named after.  You
> might also look at the various floating point specs.
>
> It's not clear what you want, but Randy Yates has previously linked to his
> intro to fixed point:
>
> http://www.digitalsignallabs.com/fp.pdf
>
> What exactly are you looking for?

Thanks Michael. And good question. Depending on the answer, the OP might
also want to investigate a course in numerical analysis at his local
university.
-- 
Randy Yates                      % "She has an IQ of 1001, she has a jumpsuit
Digital Signal Labs              %            on, and she's also a telephone."
mailto://yates@ieee.org          % 
http://www.digitalsignallabs.com %        'Yours Truly, 2095', *Time*, ELO   

On 10 Dec, 17:50, "Michael Plante" <michael.pla...@gmail.com> wrote:
> >Hello All
>
> >I wondered if i could get some pointers for =A0books for details of
> >Fixed Point and Floating Point arithmetic.
>
> Well, I would start with something more readable than the book I'm about
> to suggest, like this essay (which is all over the place on the
> Internet...just an example) :
>
> http://www.eason.com/library/math/floatingmath.pdf
>
> However, that's just floating point. =A0If you have a high tolerance for
> pain, and really want to understand both floating and fixed point, you
> might get a copy (inexpensive, but sometimes out of print) of the classic=
,
> J.H. Wilkinson, "Rounding Errors in Algebraic Processes". =A0You should j=
ust
> set aside a day to read it, rather than trying to do it in multiple
> sessions. =A0This is the same guy the Wilkinson prize is named after. =A0=
You
> might also look at the various floating point specs.
>
> It's not clear what you want, but Randy Yates has previously linked to hi=
s
> intro to fixed point:
>
> http://www.digitalsignallabs.com/fp.pdf
>
> What exactly are you looking for?

Hello

Thank you for the pointer.

I am currently working on a project which is building a compute Engine
for DVB-T2 DeMapper.
Pretty much most of the computation is in Fixed Point arithmetic and
the fixed point format is changing from one step to the next.

Also throw in are some FP computations (not IEEE 754, but custom FP
format).

Thus i wanted to learn as much as possible about Fixed Point and FP
arithmetic, how rounding, quantisation, saturation etc are handled.


Thanks and regards

JO

thunder <jao16@hotmail.com> wrote:
(snip)
 
> Thank you for the pointer.
 
> I am currently working on a project which is building a compute Engine
> for DVB-T2 DeMapper.
> Pretty much most of the computation is in Fixed Point arithmetic and
> the fixed point format is changing from one step to the next.
 
> Also throw in are some FP computations (not IEEE 754, but custom FP
> format).
 
> Thus i wanted to learn as much as possible about Fixed Point and FP
> arithmetic, how rounding, quantisation, saturation etc are handled.

There are books on computer arithmetic, sometimes with names
like "Computer Arithmetic."

For fixed point binary there is twos complement, ones complement
and sign magnitude.  Fixed point can also be done in bases
other than two, decimal (usually BCD) being somewhat popular.

There are many more possibilities in the case of floating point.

-- glen