> From:         Daniel Serpell[SMTP:daniel@ATMLAB.UTFSM.CL]
> Sent:         Monday, February 02, 2004 4:06 PM
> To:   PICLIST@MITVMA.MIT.EDU
> Subject:      Re: [PIC:] divide by 10 question

> El Fri, Jan 30, 2004 at 05:31:39PM -0500, John N. Power escribio:
>>
>>       Should be:
>>
>> B = N + (N >> 1);
>> for ( i = 0, quotient = 0; i < MAX_ITERS; i++ ) {
>>      quotient = quotient + (B >> 4);
>>            B = (B >> 4 );
>> }
>>

> Note that your routine, as writen, gives bad output for
> a number of input data, ie: 10
. . . .

> That's the problem with many algorithms that treat
> division as multiplication by constants (you are
> multiplying by 0.001100110011....), roundolf errors.

My original post labelled the code as "pseudo code" and
referred the reader to a following paragraph which identified
the need for carrying a fractional part. The above post
was just a correction to the original program.

John Power

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email listserv@mitvma.mit.edu with SET PICList DIGEST in the body