Unlimited precision in APL? 
Author Message
 Unlimited precision in APL?

Is there any APL interpreter available for the PC which
permits unlimited numerical precision? (i.e. many digits
more than the 16 or-so one gets with most languages which
store double-precision numbers in the IEEE 8-byte format)

I was taught APL at college about 14 years ago, so forgive
me if I am out of date. I now need to perform some matrix
calculations which are very easy to write down in APL, but
require at least 40 digits of precision. I have seen libraries
for other languages (e.g. C) which provide one with functions
for doing this, but writing the program in C or Pascal would
take weeks (more likely months) longer.

Any suggestions?

-----------------------------------------------

National Physical Laboratory, UK
                  FAX: +44-(0)81-943-6453
-----------------------------------------------



Fri, 21 Mar 1997 17:08:30 GMT  
 Unlimited precision in APL?

: Is there any APL interpreter available for the PC which
: permits unlimited numerical precision? (i.e. many digits
: more than the 16 or-so one gets with most languages which
: store double-precision numbers in the IEEE 8-byte format)
...
: -----------------------------------------------

: National Physical Laboratory, UK
:                   FAX: +44-(0)81-943-6453
: -----------------------------------------------

I don't know of any.  There was at one point a remark on the net that this
might not be too hard to add to J.  I hope it is added to J someday.
I am in the slightly embarrasing position of having worked on a math
recreation problem (that will be published, I think) where the editor/poser
asked for code.  I provided the "biggest known" example using some trival
J and the example was 14 digits.  It became clear others were going to soon
beat my record so I recoded in Mathematica - not so hard really, and soon had
24 digit examples.  I suspect that my "solution" will be the 24 digit example
with the J code that would not work because of precision limits.
  I realize that check writing programs don't need 40 digits, but these
things are of real interest to me.
  Cliff
--
Clifford A. Reiter
Mathematics Department, Lafayette College
Easton, PA 18042 USA,   610-250-5277



Fri, 21 Mar 1997 19:47:04 GMT  
 Unlimited precision in APL?
I know of one readily available language that permits selection of precision
up to 1000 decimal digits, if you aren't concerned about speed.  It's the
one grade schools often use to introduce children to computers -- LOGO.


Sun, 23 Mar 1997 18:53:29 GMT  
 Unlimited precision in APL?
Years ago there was a book from IBM that gave APL code for unlimited
precision arithmetic -- only of integers, though, as I recall.

--

  Mathematics & Statistics Dept.            Voice:  413-545-2859 (W)
  University of Massachusetts                       413-549-1020 (H)
  Amherst, MA 01003                           Fax:  413-545-180



Tue, 25 Mar 1997 09:59:54 GMT  
 Unlimited precision in APL?

Quote:

>Years ago there was a book from IBM that gave APL code for unlimited
>precision arithmetic -- only of integers, though, as I recall.

Can someone post this code?  It should be short.
(Does anyone have this book but no longer needs it?)




Wed, 26 Mar 1997 02:35:06 GMT  
 Unlimited precision in APL?

Quote:

>Years ago there was a book from IBM that gave APL code for unlimited
>precision arithmetic -- only of integers, though, as I recall.
>--

>  Mathematics & Statistics Dept.            Voice:  413-545-2859 (W)
>  University of Massachusetts                       413-549-1020 (H)
>  Amherst, MA 01003                           Fax:  413-545-180

That would be "The APL Handbook of Techniques" from IBM. The copy I
have is 1st edition(1978), and it defines functions to do *both*
multi-precision integer and float MDAS, as well as MP float
exponentiation and square root, and conversion, format and setup
functions. They're all pretty short, too.




Wed, 26 Mar 1997 01:57:03 GMT  
 
 [ 6 post ] 

 Relevant Pages 

1. Unlimited Precision in APL

2. Unlimited precision in APL?

3. How Are Unlimited Precision Integers Accomplished?

4. multiple precision arithmetic in APL

5. I-APL, Vanguard APL, and APL.68000

6. Convert single precision to double precision

7. Constant precision & type/precision promotion

8. Internal Precision Vs. Unformatted Precision

9. HP Fortran question: precision of double-precision?

10. double precision vs. single precision

11. single precision or double precision?

12. Double precision runs fast tan single precision in MS Fortran

 

 
Powered by phpBB® Forum Software