|
|
| Author |
Message |
amcwill417
science forum addict
Joined: 04 May 2005
Posts: 65
|
 Posted: Sun Jul 03, 2005 2:31 am Post subject:
Representation of Integers
|
|
|
Eah positive integer can be uniquely represented by three ordered smaller
integers per the following examples: 11 = (2,1,2), 18 = (2,3,1), 31 =
(3,2,0). What is the rule?
Alex |
|
| Back to top |
|
 |
Timothy Little
science forum Guru Wannabe
Joined: 30 May 2005
Posts: 295
|
| Posted: Sun Jul 03, 2005 4:58 am Post subject:
Re: Representation of Integers
|
|
|
amcwill417 wrote:
| Quote: | Eah positive integer can be uniquely represented by three ordered
smaller integers per the following examples: 11 = (2,1,2), 18 =
(2,3,1), 31 = (3,2,0). What is the rule?
|
Although there are infinitely many such rules, the first one that
sprang to mind worked:
N is represented by (a, b, c), where
a = floor(N ^ (1/3)),
b = floor((N - a^3) ^ (1/2)),
c = N - a^3 - b^2.
Or more colloquially, a is the cube root, b is the square root of
what's left, and c is what's left after that.
The reason why it sprang to mind is that I thought to myself that the
most compact such representation would asymptotically have
max(a,b,c) ~= N^(1/3).
If you take out the largest possible a^3, you'd have a remainder on
the order of 3a^2, so taking out a largest possible square would be
useful. Then you'd be left with a linear remainder.
- Tim |
|
| Back to top |
|
|
amcwill417
science forum addict
Joined: 04 May 2005
Posts: 65
|
| Posted: Mon Jul 04, 2005 1:25 am Post subject:
Re: Representation of Integers
|
|
|
"Timothy Little" <tim-usenet@little-possums.net> wrote in message
news:slrndces37.soe.tim-usenet@soprano.little-possums.net...
| Quote: | amcwill417 wrote:
Eah positive integer can be uniquely represented by three ordered
smaller integers per the following examples: 11 = (2,1,2), 18 =
(2,3,1), 31 = (3,2,0). What is the rule?
Although there are infinitely many such rules, the first one that
sprang to mind worked:
N is represented by (a, b, c), where
a = floor(N ^ (1/3)),
b = floor((N - a^3) ^ (1/2)),
c = N - a^3 - b^2.
Or more colloquially, a is the cube root, b is the square root of
what's left, and c is what's left after that.
The reason why it sprang to mind is that I thought to myself that the
most compact such representation would asymptotically have
max(a,b,c) ~= N^(1/3).
If you take out the largest possible a^3, you'd have a remainder on
the order of 3a^2, so taking out a largest possible square would be
useful. Then you'd be left with a linear remainder.
- Tim
|
Yes. How do primes behave in this representation?
Alex |
|
| Back to top |
|
|
amcwill417
science forum addict
Joined: 04 May 2005
Posts: 65
|
| Posted: Mon Jul 04, 2005 4:39 pm Post subject:
Re: Representation of Integers
|
|
|
"amcwill417" <amcwill417@email.msn.com> wrote in message
news:b22ye.128$4H2.1394@eagle.america.net...
| Quote: |
"Timothy Little" <tim-usenet@little-possums.net> wrote in message
news:slrndces37.soe.tim-usenet@soprano.little-possums.net...
amcwill417 wrote:
Eah positive integer can be uniquely represented by three ordered
smaller integers per the following examples: 11 = (2,1,2), 18 =
(2,3,1), 31 = (3,2,0). What is the rule?
Although there are infinitely many such rules, the first one that
sprang to mind worked:
N is represented by (a, b, c), where
a = floor(N ^ (1/3)),
b = floor((N - a^3) ^ (1/2)),
c = N - a^3 - b^2.
Or more colloquially, a is the cube root, b is the square root of
what's left, and c is what's left after that.
The reason why it sprang to mind is that I thought to myself that the
most compact such representation would asymptotically have
max(a,b,c) ~= N^(1/3).
If you take out the largest possible a^3, you'd have a remainder on
the order of 3a^2, so taking out a largest possible square would be
useful. Then you'd be left with a linear remainder.
- Tim
Yes. How do primes behave in this representation?
Alex
What I have in mind here is this: 7 = (1,2,2), 11 = (2,1,2) and 13 = |
(2,2,1) so that all permutations here yield primes. This is by no means
general but one can ask the question thus: Are there any other primes p =
(a,b,c) such that all permutations are also primes?
Alex |
|
| Back to top |
|
|
Google
|
|
| Back to top |
|
|
|
|
The time now is Thu Jan 08, 2009 3:28 pm | All times are GMT
|
|
Personal Loans | MPAA | Car Finance | eBay | Guitar Lessons
|
|
Copyright © 2004-2005 DeniX Solutions SRL
|
|
Other DeniX Solutions sites:
Electronics forum |
Medicine forum |
Unix/Linux blog |
Unix/Linux documentation |
Unix/Linux forums
|
Powered by phpBB © 2001, 2005 phpBB Group
|
|
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
Forum index
»
Science and Technology
»
Math
»
Recreational
