7
$\begingroup$

Here is a question about Riemann Hypothesis:

Is $1/2$ of critical line same as the $1/2$ of square-root accurately of error term of prime number theorem $?$

In other words, (just for some brain exercise), let's assume that critical line is at $x=3/4$, does this mean that the error term in prime number theorem will be proportional to the power of $3/4$ $?$ So $|\pi(x) - Li(x) |$ would be bounded by $x^{(3/4)}$ $?$

Which book has a simple proof of this $?$

Thank you.

$\endgroup$
  • 1
    $\begingroup$ Yes, this is standard. This is probably somewhere in Montgomery and Vaughan, or in Iwaniec and Kowalski. Note that if $\zeta(s)$ has no zeroes to the right of the line $\Re(s) = \Theta$, then $|\pi(x) - \mathrm{Li}(x)| \ll_{\varepsilon} x^{\Theta + \varepsilon}$ for all $\varepsilon > 0$, but one cannot remove the $\varepsilon$. $\endgroup$ – Peter Humphries Sep 25 '15 at 19:58
3
$\begingroup$

This is correct and the relationship is most visible in Riemann's explicit formula for the weighted prime counting function. See for instance Garrett's write-up http://www.math.umn.edu/~garrett/m/mfms/notes_c/mfms_notes_02.pdf .

$\endgroup$
3
$\begingroup$

Yes.

Ingham, The Distribution of Prime Numbers, contains a proof that the lower bound $\Theta_1$ of numbers $\alpha$ for which it is true that

$$\psi(x)-x = O(x^{\alpha}) $$

is equal to the upper bound $\Theta_2$ of the real parts of the zeros of $\zeta(s).$

This is his Theorem 31 (with a proof) on page 84.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.