Search All of the Math Forum:

Views expressed in these public forums are not endorsed by NCTM or The Math Forum.

Notice: We are no longer accepting new posts, but the forums will continue to be readable.

Topic: Integer pairs in sum of reciprocals
Replies: 39   Last Post: Jan 22, 2001 6:02 PM

 Messages: [ Previous | Next ]
 David C. Ullrich Posts: 21,553 Registered: 12/6/04
Re: Integer pairs in sum of reciprocals
Posted: Jan 6, 2001 9:20 AM

On Sat, 06 Jan 2001 13:25:21 GMT, George Cantor
<the_great_nathan@my-deja.com> wrote:

>In article <3a55fd38.124617239@news.newsguy.com>,
> randyp@visionplace.com (Randy Poe) wrote:

>> >>First, the sum would be too large if either A or B
>> >>were less than 2002. Second, the sum would be too
>> >>small if both A and B were greater than 4002. So,
>> >>one of the variables must be between 2002 and 4002,
>> >>but an exhaustive search of that range, taking less
>> >>than .00001 seconds, uncovers no solutions.

>>
>> The reasoning appears to be sound. Note that both A and B in
>> the above solution are > 2002, and only one exceeds 4002.
>>
>> Apparently something was wrong in the validation step of the
>> "exhaustive search".

>
>Yes, the validation step was flawwed. I accidently
>confused a Truncation opcode with a Rounding opcode
>when I tried detecting and correcting errors introduced
>by the use of floating point variables. As far as I
>knew, no one else was accounting for rounding errors.

Presumably(?) no one else was using floating-point arithmetic.
Seems like an unreasonably clumsy approach: Although
it's possible to do this search using floating-point if
one accounts for the errors _properly_ I tend to doubt
that you personally are capcable of figuring out exactly
what it means to account for the errors _properly_ here.

>That's why, eventhough I noticed my results disagreed
>with the rest of the group, I assumed I was correct.
>Isn't that ironic?

I don't think "ironic" is quite the right word. I think you
might learn something from the fact that you disagreed
with everyone, and then incredibly it turned out that
everyone else was actually right. But I doubt it will
happen.

>Anyway, if I would have read all the posts I would have
>known better than to claim no solutions, but as soon as
>I read the post claiming an infinite number of solutions
>I went insane!
>
>
>--
>I'm a little crackpot short and stout.
>I've got a new handle cuz of the drought.
>When I get all steamed up then I shout,
>snip a poster and flame the lout.
>
> - George (0+0=oo, for sufficiently large zeros) Canto
>
>
>Sent via Deja.com
>http://www.deja.com/