Problems & Puzzles: Puzzles

Problem 90. Find the first 8 in...

On March 29, 2025 my friend G. L. Honaker, Jr. has published the following Curio/Challenge-Reward and problem:

G. L. Honaker, Jr. offers $8 to the person or team that finds the first occurrence of digit 8 in the constant

 1/2 - 1/3 + 1/5 - 1/7 + 1/11 - 1/13 + ... = 0.269...
 
.

(See the Curio here).

"...(this) alternating series converges (Robinson and Potter 1971)". See A078437 and this.

Q. Please send your decimal expression of this alternating sum, up to the first asked decimal (8), indicating the largest prime needed to get it.

Important Note: Regarding this challenge, I will be just an intermediate between Honaker and the participants. While I will annotate the day and the hour I receive your email/answers, he alone will decide who is the winner.


On May-1-2025, 2:08 am (México, central time), Simon Cavegn sent his solution to this Problem:

Continued h(k) from https://oeis.org/A078437

                         h(k) =
            k  (f(k-2) + 2*f(k-1) + f(k))/4
   ==========  ============================
            2     0.29166666666666666...
            4     0.28095238095238095...
            8     0.26875529011751921...
           16     0.27058892362329746...
           32     0.27009944617052797...
           64     0.26963971020080367...
          128     0.26959147218377685...
          256     0.26959653902072193...
          512     0.26960402179695026...
         1024     0.26960568606633210...
         2048     0.26960649673621509...
         4096     0.26960645080540929...
         8192     0.26960627432070023...
        16384     0.26960633643086948...
        32768     0.26960634835658329...
        65536     0.26960635083481533...
       131072     0.26960635144743392...
       262144     0.26960635199009778...
       524288     0.26960635199971603...
      1048576     0.26960635195886861...
      2097152     0.26960635197214933...
      4194304     0.26960635197019215...
      8388608     0.26960635197186919...
     16777216     0.26960635197171149...
     33554432     0.26960635197146884...
     67108864     0.26960635197167534...
    134217728     0.26960635197167145...
    268435456     0.26960635197166927...
    536870912     0.26960635197167200...
   1073741824     0.26960635197167416...
   2147483648     0.26960635197167454...
   4294967296     0.26960635197167462...
   8589934592     0.2696063519716746115951753723391177045452704199650018570665...
  17179869184     0.2696063519716745911151644809251657437243751753378500233783...
  34359738368     0.2696063519716745933571121943551630117783437944488045026659...
  68719476736     0.2696063519716745948464031246903256452110028349813594133690...
 137438953472     0.2696063519716745947477102160199100133802487290017570637917...
 274877906944     0.2696063519716745949231122858949180938393568249345518460973...
 549755813888     0.2696063519716745948577548318546254752594315752839685372454...
1099511627776     0.2696063519716745948490506428349040904990465420119594007519...
2199023255552     0.2696063519716745948520206964438658721432622448204777318986...
4398046511104     0.2696063519716745948516189025839897789324530848126518723457...

I calculated 3 different averages:
Average of last 3 values weighted by pascal triangle entry 1,2,1, this is the h(k) function used by https://oeis.org/A078437
Average of last 8 values weighted by pascal triangle entry 1,7,21,35,35,21,7,1, named r(k)
Average of last 8 values weighted equally, named a(k)

h(k)=0.26960635197167459485161890258398977893245308481265187234570, k=4398046511104
r(k)=0.26960635197167459485161890281395580110416961656337841996337, k=4398046511104
a(k)=0.26960635197167459485161890353310749587938919254573665846864, k=4398046511104

The constant's value : 0.2696063519716745948
Next digit is likely : 0.26960635197167459485

The calculation took 11 days and 10 hours on a Notebook CPU "AI 9 HX 370", implemented in C# using https://gmplib.org/

***

and this is the comment by Mr. Honaker, Jr.:
Phi = 0.2696063519716745948... [Cavegn]
Simon,
Thank you very much for this result! I have submitted it to OEIS for approval. If you email me your mailing address, I will be happy to send you the promised reward money ($8US).

G. L. Honaker, Jr.
Content Editor: Prime Curios!

P.S., If OEIS does not respond within a week, you may wish to submit the result yourself: https://oeis.org/A078437

***

Records   |  Conjectures  |  Problems  |  Puzzles