Some *years* ago, I asked in MSE [a question][1] about the Chudnovsky brothers pi formula. Later, I asked in MO a [related question][2]. The former was unanswered *until a few days ago* when L. Miller gave me a clue to find a missing piece of the puzzle. (It seems it may take years for a question to be resolved.)

>**I. Question**

Given $A,B,C$ and binomial $\binom{n}{k}$. If the series below converges, is it true that,

$$\frac{1}{C^{3/2}}\sum_{k=0}^\infty\tbinom{2k}{k} \tbinom{pk}{k}\tbinom{2pk}{pk} \frac{\color{blue}pA\,k+B}{C^{\color{blue}p\,k}}=\\ \frac{1}{(C+4r)^{3/2}}\sum_{k=0}^\infty\tbinom{2k}{k}\sum_{j=0}^k r^{k-pj}\tbinom{k}{pj}\tbinom{2j}{j}\tbinom{pj}{j} \frac{A\,k+B-Dr}{(C+4r)^k}$$

where $D = \frac{2A-4B}C$, and generic $r$ such that $C(C+4r)>0$?

>**II. Ramanujan** 

Ramanujan's formula is then the case $p=2$,

$$\frac{1}{\pi} =\frac{192 \sqrt 2}{(396^2)^{3/2}} \sum_{k=0}^\infty \tbinom{2k}{k}\tbinom{2k}{k}\tbinom{4k}{2k}\frac{\color{blue}2Ak+B}{(396^2)^{\color{blue}2k}}$$

yielding

$$\frac{1}{\pi}=\frac{192\sqrt{2}}{(396^2+4r)^{3/2}}\sum_{k=0}^\infty\tbinom{2k}{k}\sum_{j=0}^k r^{k-2j} \tbinom{k}{2j}\tbinom{2j}{j}\tbinom{2j}{j} \frac{Ak+B-37r/4}{(396^2+4r)^k}$$

where $A=58\cdot15015$ and $B=72798$ and $C=396^2$. 

>**III. Chudnovsky**

The Chudnovsky formula is the case $p=3$,

$$\frac{1}{\pi}=\frac{12\sqrt{-1}}{(-640320)^{3/2}} \sum_{k=0}^\infty \tbinom{2k}{k}\tbinom{3k}{k}\tbinom{6k}{3k}\frac{\color{blue}3A\,k+B}{(-640320)^{\color{blue}3k}}$$

yielding

$$\frac{1}{\pi}=\frac{12\sqrt{-1}}{(-640320+4r)^{3/2}}\sum_{k=0}^\infty\tbinom{2k}{k}\sum_{j=0}^k r^{k-3j} \tbinom{k}{3j}\tbinom{2j}{j}\tbinom{3j}{j} \frac{A\,k+B+1448r/3}{(-640320+4r)^k}$$

where $A=163\cdot1114806$ and $B=13591409$ and $C=-640320$.

---

I am not aware of a formula that uses $p=4$. **So is the answer to the question a "yes"?**


  [1]: https://math.stackexchange.com/questions/1952204/the-chudnovsky-pi-formula-1-pi-revisited
  [2]: https://mathoverflow.net/questions/265528/numerology-with-ramanujans-pi-formula