MathGroup Archive 2003

[Date Index] [Thread Index] [Author Index]

Search the Archive

Re: goldbach prime partitions for arbitrary integer n => 4

  • To: mathgroup at smc.vnet.net
  • Subject: [mg43018] Re: [mg42986] goldbach prime partitions for arbitrary integer n => 4
  • From: Daniel Lichtblau <danl at wolfram.com>
  • Date: Fri, 8 Aug 2003 00:26:31 -0400 (EDT)
  • References: <200308070453.AAA24088@smc.vnet.net>
  • Sender: owner-wri-mathgroup at wolfram.com

Gilmar Rodríguez Pierluissi wrote:
> 
> If one wishes to compute:
> eqn={p+q==200}; constraints={2<=p<=100, p<=q, p,q \[Element]Primes};
> wouldn't it be nice that if you evaluate:
> Solve[eqn,constraints,{p,q}]
> you would get:
> {{97,103},{73,127},{61,139},{43,157},{37,163},{19,181},{7,193},{3,197}} ?
> A module (or program) that could solve:
> eqn={p+q==n}; constraints={2<=p<=n/2, p<=q, p,q \[Element]Primes};
> Solve[eqn,constraints,{p,q}]
> for a specified n, (n=>4, n \[Element]Integer), would be even better!
> Thank you!

You can use Reduce in version 5. 

eqn = {p+q==200};
constraints = {2<=p<=100, p<=q, Element[{p,q},Primes]};

InputForm[Reduce[Join[eqn,constraints], {p,q}]]
Out[3]//InputForm= 
(p | q) \[Element] Primes && C[1] \[Element] Integers && 
 Inequality[2, LessEqual, C[1], LessEqual, 100] && p == C[1] && 
 q == 200 - C[1]

The trick is to realize that there is a relatively low bound on how many
discrete solutions are allowed by default (it is 10). If you do

Developer`SetSystemOptions["ReduceOptions" ->
  {"DiscreteSolutionBound" ->100}];

then we get something more along the lines desired.

InputForm[Reduce[Join[eqn,constraints], {p,q}]]
Out[5]//InputForm= 
(p == 3 && q == 197) || (p == 7 && q == 193) || (p == 19 && q == 181) || 
 (p == 37 && q == 163) || (p == 43 && q == 157) || (p == 61 && q == 139)
|| 
 (p == 73 && q == 127) || (p == 97 && q == 103)


Daniel lichtblau
Wolfram Research


  • Prev by Date: Numerical precision problem
  • Next by Date: RE: RE: Dealing with large data sets
  • Previous by thread: goldbach prime partitions for arbitrary integer n => 4
  • Next by thread: Re: Re: goldbach prime partitions for arbitrary integer n => 4