Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.soft-sys.math.maple > #946

Re: A hypergeometric formula

Show all headers | View raw

On 10.09.2014 20:32, peter.luschny@gmail.com wrote:
>>> 2*GAMMA(3/2)*hypergeom([1/2,0],[3/2,0,-1/2],-1) /sqrt(Pi) = ?
>> (**) y := 2*GAMMA(3/2)*hypergeom([1/2,0],[3/2,0,-1/2],-1) /sqrt(Pi);
>>                     y := hypergeom([1/2], [-1/2, 3/2], -1)
>> (**) simplify(y);
>>                     sin(2) - cos(2)
>
> Thank you Joe!
>
> Now this was the easy part. Next the question: Why does Wolfram Alpha
> (and presumably Mathematica) gives a different answer?
>
>      2 Gamma[3/2] (HypergeometricPFQ[{1/2, 0}, {3/2, 0, -1/2}, -1]/Sqrt[Pi])
>
> Peter

I suggest to consider only the 2F3. Wolfram Alpha says " = 1"
and writes it as series in Pochhamer symbols. That series is
seen to be sin(2) - cos(2) by Maple (which re-writes the very
task as 1F3, dropping the zeros).

Feeding the series to Alpha gives sin(2) - cos(2)

http://www.wolframalpha.com/input/?i=HypergeometricPFQ[{1%2F2%2C+0}%2C+{3%2F2%2C+0%2C+-1%2F2}%2C+-1]

http://www.wolframalpha.com/input/?i=sum%28%28-1%29^k*pochhammer%281%2F2%2Ck%29%2Fk!%2Fpochhammer%28-1%2F2%2Ck%29%2Fpochhammer%283%2F2%2Ck%29%2Ck+%3D+0+..+infinity%29

Back to comp.soft-sys.math.maple | Previous | Next — Previous in thread | Next in thread | Find similar

Thread

A hypergeometric formula peter.luschny@gmail.com - 2014-09-10 06:44 -0700
  Re: A hypergeometric formula Joe Riel <joer@san.rr.com> - 2014-09-10 09:05 -0700
    Re: A hypergeometric formula peter.luschny@gmail.com - 2014-09-10 11:32 -0700
      Re: A hypergeometric formula Axel Vogt <&noreply@axelvogt.de> - 2014-09-10 20:54 +0200
        Re: A hypergeometric formula peter.luschny@gmail.com - 2014-09-10 15:08 -0700
          Re: A hypergeometric formula acer <maple@rogers.com> - 2014-09-10 21:12 -0700
            Re: A hypergeometric formula Axel Vogt <&noreply@axelvogt.de> - 2014-09-11 19:29 +0200
              Re: A hypergeometric formula peter.luschny@gmail.com - 2014-09-11 10:42 -0700

csiph-web