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


Groups > sci.math > #513607 > unrolled thread

What is the next best simple fraction of pi meeting this criteria?

Started bydjoyce099 <hlauk.h.bogart@gmail.com>
First post2021-03-07 12:35 -0800
Last post2021-03-14 13:58 -0700
Articles 20 on this page of 65 — 13 participants

Back to article view | Back to sci.math


Contents

  What is the next best simple fraction of pi meeting this criteria? djoyce099 <hlauk.h.bogart@gmail.com> - 2021-03-07 12:35 -0800
    Re: What is the next best simple fraction of pi meeting this criteria? Sergio <invalid@invalid.com> - 2021-03-07 14:52 -0600
      Re: What is the next best simple fraction of pi meeting this criteria? "mitchr...@gmail.com" <mitchrae3323@gmail.com> - 2021-03-07 13:23 -0800
        Re: What is the next best simple fraction of pi meeting this criteria? Bob Aya - PhD <bobac42@hotmail.com> - 2021-03-07 21:34 +0000
          Re: What is the next best simple fraction of pi meeting this criteria? "mitchr...@gmail.com" <mitchrae3323@gmail.com> - 2021-03-07 13:42 -0800
            Re: What is the next best simple fraction of pi meeting this criteria? "mitchr...@gmail.com" <mitchrae3323@gmail.com> - 2021-03-07 14:18 -0800
        Re: What is the next best simple fraction of pi meeting this criteria? Sergio <invalid@invalid.com> - 2021-03-07 18:23 -0600
          Re: What is the next best simple fraction of pi meeting this criteria? "mitchr...@gmail.com" <mitchrae3323@gmail.com> - 2021-03-09 10:20 -0800
            Re: What is the next best simple fraction of pi meeting this criteria? Sergio <invalid@invalid.com> - 2021-03-10 08:20 -0600
    Re: What is the next best simple fraction of pi meeting this criteria? "Chris M. Thomasson" <chris.m.thomasson.1@gmail.com> - 2021-03-07 16:33 -0800
      Re: What is the next best simple fraction of pi meeting this criteria? Sergio <invalid@invalid.com> - 2021-03-07 19:12 -0600
        Re: What is the next best simple fraction of pi meeting this criteria? djoyce099 <hlauk.h.bogart@gmail.com> - 2021-03-07 20:57 -0800
          Re: What is the next best simple fraction of pi meeting this criteria? djoyce099 <hlauk.h.bogart@gmail.com> - 2021-03-08 20:10 -0800
            Re: What is the next best simple fraction of pi meeting this criteria? Sergio <invalid@invalid.com> - 2021-03-09 08:25 -0600
              Re: What is the next best simple fraction of pi meeting this criteria? djoyce099 <hlauk.h.bogart@gmail.com> - 2021-03-09 07:59 -0800
                Re: What is the next best simple fraction of pi meeting this criteria? Peter <peterxpercival@hotmail.com> - 2021-03-10 15:09 +0000
                  Re: What is the next best simple fraction of pi meeting this criteria? djoyce099 <hlauk.h.bogart@gmail.com> - 2021-03-10 11:37 -0800
                    Re: What is the next best simple fraction of pi meeting this criteria? Sergio <invalid@invalid.com> - 2021-03-10 19:25 -0600
                      Re: What is the next best simple fraction of pi meeting this criteria? * _Joseph W Smith - PhD_ *
The College of Applied Health Science
602 Eeast Green Street
Urbana, IL 61801
osephsmith@illinois.edu <joseth@illinois.edu> - 2021-03-11 11:41 +0000
            Re: What is the next best simple fraction of pi meeting this criteria? Peter <peterxpercival@hotmail.com> - 2021-03-09 15:06 +0000
              Re: What is the next best simple fraction of pi meeting this criteria? djoyce099 <hlauk.h.bogart@gmail.com> - 2021-03-09 07:52 -0800
                Re: What is the next best simple fraction of pi meeting this criteria? djoyce099 <hlauk.h.bogart@gmail.com> - 2021-03-10 06:11 -0800
      RE: Re: What is the next best simple fraction of pi meeting this criteria? earle <earle.jones@comcast.net> - 2021-03-15 01:52 +0000
        Re: What is the next best simple fraction of pi meeting this criteria? "Chris M. Thomasson" <chris.m.thomasson.1@gmail.com> - 2021-03-14 19:09 -0700
        Re: What is the next best simple fraction of pi meeting this criteria? James Waldby <j-waldby@no.no> - 2021-03-15 03:43 +0000
        Re: What is the next best simple fraction of pi meeting this criteria? FromTheRafters <erratic@nomail.afraid.org> - 2021-03-15 04:05 -0400
          Re: What is the next best simple fraction of pi meeting this criteria? Mostowski Collapse <bursejan@gmail.com> - 2021-03-15 03:19 -0700
            Re: What is the next best simple fraction of pi meeting this criteria? Mostowski Collapse <bursejan@gmail.com> - 2021-03-15 03:24 -0700
              Re: What is the next best simple fraction of pi meeting this criteria? Mostowski Collapse <bursejan@gmail.com> - 2021-03-15 03:40 -0700
                Re: What is the next best simple fraction of pi meeting this criteria? djoyce099 <hlauk.h.bogart@gmail.com> - 2021-03-15 07:30 -0700
                  Re: What is the next best simple fraction of pi meeting this criteria? Peter <peterxpercival@hotmail.com> - 2021-03-15 14:53 +0000
                    Re: What is the next best simple fraction of pi meeting this criteria? djoyce099 <hlauk.h.bogart@gmail.com> - 2021-03-15 11:53 -0700
          RE: Re: What is the next best simple fraction of pi meeting this criteria? earle <earle.jones@comcast.net> - 2021-03-15 20:53 +0000
            Re: What is the next best simple fraction of pi meeting this criteria? FromTheRafters <erratic@nomail.afraid.org> - 2021-03-15 17:11 -0400
    Re: What is the next best simple fraction of pi meeting this criteria? Peter <peterxpercival@hotmail.com> - 2021-03-09 14:56 +0000
    Re: What is the next best simple fraction of pi meeting this criteria? Michael Moroney <moroney@world.std.spaamtrap.com> - 2021-03-11 00:35 -0500
      Re: What is the next best simple fraction of pi meeting this criteria? djoyce099 <hlauk.h.bogart@gmail.com> - 2021-03-10 22:25 -0800
        Re: What is the next best simple fraction of pi meeting this criteria? djoyce099 <hlauk.h.bogart@gmail.com> - 2021-03-10 22:30 -0800
          Re: What is the next best simple fraction of pi meeting this criteria? djoyce099 <hlauk.h.bogart@gmail.com> - 2021-03-10 22:35 -0800
        Re: What is the next best simple fraction of pi meeting this criteria? djoyce099 <hlauk.h.bogart@gmail.com> - 2021-03-10 22:44 -0800
          Re: What is the next best simple fraction of pi meeting this criteria? djoyce099 <hlauk.h.bogart@gmail.com> - 2021-03-11 07:39 -0800
            Re: What is the next best simple fraction of pi meeting this criteria? djoyce099 <hlauk.h.bogart@gmail.com> - 2021-03-13 08:32 -0800
              Re: What is the next best simple fraction of pi meeting this criteria? djoyce099 <hlauk.h.bogart@gmail.com> - 2021-03-13 08:43 -0800
                Re: What is the next best simple fraction of pi meeting this criteria? djoyce099 <hlauk.h.bogart@gmail.com> - 2021-03-13 11:16 -0800
                  Re: What is the next best simple fraction of pi meeting this criteria? djoyce099 <hlauk.h.bogart@gmail.com> - 2021-03-13 11:29 -0800
                  Re: What is the next best simple fraction of pi meeting this criteria? djoyce099 <hlauk.h.bogart@gmail.com> - 2021-03-15 12:33 -0700
                    Re: What is the next best simple fraction of pi meeting this criteria? djoyce099 <hlauk.h.bogart@gmail.com> - 2021-03-15 12:36 -0700
                    Re: What is the next best simple fraction of pi meeting this criteria? Mostowski Collapse <bursejan@gmail.com> - 2021-03-16 00:56 -0700
                      Re: What is the next best simple fraction of pi meeting this criteria? djoyce099 <hlauk.h.bogart@gmail.com> - 2021-03-16 06:58 -0700
                        Re: What is the next best simple fraction of pi meeting this criteria? Mostowski Collapse <bursejan@gmail.com> - 2021-03-16 07:23 -0700
                          Re: What is the next best simple fraction of pi meeting this criteria? djoyce099 <hlauk.h.bogart@gmail.com> - 2021-03-16 07:43 -0700
                Re: What is the next best simple fraction of pi meeting this criteria? Sergio <invalid@invalid.com> - 2021-03-13 14:09 -0600
                  Re: What is the next best simple fraction of pi meeting this criteria? djoyce099 <hlauk.h.bogart@gmail.com> - 2021-03-13 14:46 -0800
                    Re: What is the next best simple fraction of pi meeting this criteria? Sergio <invalid@invalid.com> - 2021-03-13 20:22 -0600
              Re: What is the next best simple fraction of pi meeting this criteria? James Waldby <j-waldby@no.no> - 2021-03-13 23:46 +0000
                Re: What is the next best simple fraction of pi meeting this criteria? djoyce099 <hlauk.h.bogart@gmail.com> - 2021-03-13 16:20 -0800
                  Re: What is the next best simple fraction of pi meeting this criteria? djoyce099 <hlauk.h.bogart@gmail.com> - 2021-03-13 16:22 -0800
                    Re: What is the next best simple fraction of pi meeting this criteria? James Waldby <j-waldby@no.no> - 2021-03-14 04:29 +0000
                      Re: What is the next best simple fraction of pi meeting this criteria? Sergio <invalid@invalid.com> - 2021-03-13 23:05 -0600
                      Re: What is the next best simple fraction of pi meeting this criteria? djoyce099 <hlauk.h.bogart@gmail.com> - 2021-03-14 09:48 -0700
        Re: What is the next best simple fraction of pi meeting this criteria? Michael Moroney <moroney@world.std.spaamtrap.com> - 2021-03-14 03:40 -0400
          Re: What is the next best simple fraction of pi meeting this criteria? djoyce099 <hlauk.h.bogart@gmail.com> - 2021-03-14 09:59 -0700
            Re: What is the next best simple fraction of pi meeting this criteria? Brodie Wiser <bdwi@committee.aero> - 2021-03-14 17:01 +0000
            Re: What is the next best simple fraction of pi meeting this criteria? James Waldby <j-waldby@no.no> - 2021-03-14 19:03 +0000
              Re: What is the next best simple fraction of pi meeting this criteria? djoyce099 <hlauk.h.bogart@gmail.com> - 2021-03-14 13:58 -0700

Page 3 of 4 — ← Prev page 1 2 [3] 4  Next page →


#513931

Fromdjoyce099 <hlauk.h.bogart@gmail.com>
Date2021-03-11 07:39 -0800
Message-ID<2766d605-d3a5-42de-86f0-342a378d6077n@googlegroups.com>
In reply to#513899
On Thursday, March 11, 2021 at 1:44:06 AM UTC-5, djoyce099 wrote:
> On Thursday, March 11, 2021 at 1:25:46 AM UTC-5, djoyce099 wrote:
> > On Thursday, March 11, 2021 at 12:35:08 AM UTC-5, Michael Moroney wrote: 
> > > On 3/7/2021 3:35 PM, djoyce099 wrote: 
> > > > The first is a well-known approximation in one simple fraction. 
> > > > 
> > > > 113/(2^4) = 7.0625 
> > > > 1/7.0625 +3 ~ pi 
> > > > Or 355/113 
> > > > 
> > > > A better one below 
> > > > --------------------------------------------------------------------------------- 
> > > > (((2^12)*113)+1)/(2^16) =n 
> > > > 1/n +3 ~pi 
> > > > Or 462849/(2^16) 
> > > > --------------------------------------------------------------------------------- 
> > > > What is the next one? 
> > > The answers so far all seem to involve 113 somehow. That is unusual. 
> > > 
> > > I will reword the question slightly: What is the smallest n, where k/n 
> > > is an approximation of pi, where k and n are both integers, and k/n is a 
> > > better approximation of pi than 355/113? 
> > > 
> > > ( |pi - k/n| < |pi - 355/113| ) 
> > > 
> > > Just like 113 is the smallest n where k/n [k=355] is a better 
> > > approximation of pi than 22/7, another common approximation of pi. 
> > 113 is used because to start these equations where n = 7.0625 then 1/7.0625 = 0.1415929203... +3 = 355/113. 
> > Here is the next one-- 
> > (((2^328)*113)+116413629295091038246264489256036237717551434878188590657475340766437085830461471938008682720975)/2^332 correct to 103 decimal digits of pi
> Should have added above at the end of the equation = n then 1/n +3 ~ pi too 103 correct decimal digits.

So basically the trick is to find the right integer to add to the smaller power of 2 on the left side of the equation.
It is just that, a trick, that takes seconds to find.
It appears all that matters is the two powers of 2 are (4) apart and even and the larger power of 2 is the divisor.    

[toc] | [prev] | [next] | [standalone]


#514138

Fromdjoyce099 <hlauk.h.bogart@gmail.com>
Date2021-03-13 08:32 -0800
Message-ID<2e7f5dbb-aee2-416d-b430-ec7860953accn@googlegroups.com>
In reply to#513931
On Thursday, March 11, 2021 at 10:39:59 AM UTC-5, djoyce099 wrote:
> On Thursday, March 11, 2021 at 1:44:06 AM UTC-5, djoyce099 wrote: 
> > On Thursday, March 11, 2021 at 1:25:46 AM UTC-5, djoyce099 wrote: 
> > > On Thursday, March 11, 2021 at 12:35:08 AM UTC-5, Michael Moroney wrote: 
> > > > On 3/7/2021 3:35 PM, djoyce099 wrote: 
> > > > > The first is a well-known approximation in one simple fraction. 
> > > > > 
> > > > > 113/(2^4) = 7.0625 
> > > > > 1/7.0625 +3 ~ pi 
> > > > > Or 355/113 
> > > > > 
> > > > > A better one below 
> > > > > --------------------------------------------------------------------------------- 
> > > > > (((2^12)*113)+1)/(2^16) =n 
> > > > > 1/n +3 ~pi 
> > > > > Or 462849/(2^16) 
> > > > > --------------------------------------------------------------------------------- 
> > > > > What is the next one? 
> > > > The answers so far all seem to involve 113 somehow. That is unusual. 
> > > > 
> > > > I will reword the question slightly: What is the smallest n, where k/n 
> > > > is an approximation of pi, where k and n are both integers, and k/n is a 
> > > > better approximation of pi than 355/113? 
> > > > 
> > > > ( |pi - k/n| < |pi - 355/113| ) 
> > > > 
> > > > Just like 113 is the smallest n where k/n [k=355] is a better 
> > > > approximation of pi than 22/7, another common approximation of pi. 
> > > 113 is used because to start these equations where n = 7.0625 then 1/7.0625 = 0.1415929203... +3 = 355/113. 
> > > Here is the next one-- 
> > > (((2^328)*113)+116413629295091038246264489256036237717551434878188590657475340766437085830461471938008682720975)/2^332 correct to 103 decimal digits of pi 
> > Should have added above at the end of the equation = n then 1/n +3 ~ pi too 103 correct decimal digits.
> So basically the trick is to find the right integer to add to the smaller power of 2 on the left side of the equation. 
> It is just that, a trick, that takes seconds to find. 
> It appears all that matters is the two powers of 2 are (4) apart and even and the larger power of 2 is the divisor.

Continuing this as a new question about pi---

pi ~ 4 - 1/n - 1/(2*n) - 1/(3*n) - 1/(4*n) - 1/(5*n) ...

What is the largest value of n  possible where the convergence is always >pi? 
n =?

[toc] | [prev] | [next] | [standalone]


#514139

Fromdjoyce099 <hlauk.h.bogart@gmail.com>
Date2021-03-13 08:43 -0800
Message-ID<0c511bc6-d2e3-4ead-872b-c88f708b6318n@googlegroups.com>
In reply to#514138
On Saturday, March 13, 2021 at 11:32:56 AM UTC-5, djoyce099 wrote:
> On Thursday, March 11, 2021 at 10:39:59 AM UTC-5, djoyce099 wrote: 
> > On Thursday, March 11, 2021 at 1:44:06 AM UTC-5, djoyce099 wrote: 
> > > On Thursday, March 11, 2021 at 1:25:46 AM UTC-5, djoyce099 wrote: 
> > > > On Thursday, March 11, 2021 at 12:35:08 AM UTC-5, Michael Moroney wrote: 
> > > > > On 3/7/2021 3:35 PM, djoyce099 wrote: 
> > > > > > The first is a well-known approximation in one simple fraction. 
> > > > > > 
> > > > > > 113/(2^4) = 7.0625 
> > > > > > 1/7.0625 +3 ~ pi 
> > > > > > Or 355/113 
> > > > > > 
> > > > > > A better one below 
> > > > > > --------------------------------------------------------------------------------- 
> > > > > > (((2^12)*113)+1)/(2^16) =n 
> > > > > > 1/n +3 ~pi 
> > > > > > Or 462849/(2^16) 
> > > > > > --------------------------------------------------------------------------------- 
> > > > > > What is the next one? 
> > > > > The answers so far all seem to involve 113 somehow. That is unusual. 
> > > > > 
> > > > > I will reword the question slightly: What is the smallest n, where k/n 
> > > > > is an approximation of pi, where k and n are both integers, and k/n is a 
> > > > > better approximation of pi than 355/113? 
> > > > > 
> > > > > ( |pi - k/n| < |pi - 355/113| ) 
> > > > > 
> > > > > Just like 113 is the smallest n where k/n [k=355] is a better 
> > > > > approximation of pi than 22/7, another common approximation of pi. 
> > > > 113 is used because to start these equations where n = 7.0625 then 1/7.0625 = 0.1415929203... +3 = 355/113. 
> > > > Here is the next one-- 
> > > > (((2^328)*113)+116413629295091038246264489256036237717551434878188590657475340766437085830461471938008682720975)/2^332 correct to 103 decimal digits of pi 
> > > Should have added above at the end of the equation = n then 1/n +3 ~ pi too 103 correct decimal digits. 
> > So basically the trick is to find the right integer to add to the smaller power of 2 on the left side of the equation. 
> > It is just that, a trick, that takes seconds to find. 
> > It appears all that matters is the two powers of 2 are (4) apart and even and the larger power of 2 is the divisor.
> Continuing this as a new question about pi--- 
> 
> pi ~ 4 - 1/n - 1/(2*n) - 1/(3*n) - 1/(4*n) - 1/(5*n) ... 
> 
> What is the largest value of n possible where the convergence is always >pi? 
> n =?
I should have stated, n as an integer.

[toc] | [prev] | [next] | [standalone]


#514150

Fromdjoyce099 <hlauk.h.bogart@gmail.com>
Date2021-03-13 11:16 -0800
Message-ID<f891ae96-6106-4824-8451-cdbc79adc14en@googlegroups.com>
In reply to#514139
On Saturday, March 13, 2021 at 11:43:59 AM UTC-5, djoyce099 wrote:
> On Saturday, March 13, 2021 at 11:32:56 AM UTC-5, djoyce099 wrote: 
> > On Thursday, March 11, 2021 at 10:39:59 AM UTC-5, djoyce099 wrote: 
> > > On Thursday, March 11, 2021 at 1:44:06 AM UTC-5, djoyce099 wrote: 
> > > > On Thursday, March 11, 2021 at 1:25:46 AM UTC-5, djoyce099 wrote: 
> > > > > On Thursday, March 11, 2021 at 12:35:08 AM UTC-5, Michael Moroney wrote: 
> > > > > > On 3/7/2021 3:35 PM, djoyce099 wrote: 
> > > > > > > The first is a well-known approximation in one simple fraction. 
> > > > > > > 
> > > > > > > 113/(2^4) = 7.0625 
> > > > > > > 1/7.0625 +3 ~ pi 
> > > > > > > Or 355/113 
> > > > > > > 
> > > > > > > A better one below 
> > > > > > > --------------------------------------------------------------------------------- 
> > > > > > > (((2^12)*113)+1)/(2^16) =n 
> > > > > > > 1/n +3 ~pi 
> > > > > > > Or 462849/(2^16) 
> > > > > > > --------------------------------------------------------------------------------- 
> > > > > > > What is the next one? 
> > > > > > The answers so far all seem to involve 113 somehow. That is unusual. 
> > > > > > 
> > > > > > I will reword the question slightly: What is the smallest n, where k/n 
> > > > > > is an approximation of pi, where k and n are both integers, and k/n is a 
> > > > > > better approximation of pi than 355/113? 
> > > > > > 
> > > > > > ( |pi - k/n| < |pi - 355/113| ) 
> > > > > > 
> > > > > > Just like 113 is the smallest n where k/n [k=355] is a better 
> > > > > > approximation of pi than 22/7, another common approximation of pi. 
> > > > > 113 is used because to start these equations where n = 7.0625 then 1/7.0625 = 0.1415929203... +3 = 355/113. 
> > > > > Here is the next one-- 
> > > > > (((2^328)*113)+116413629295091038246264489256036237717551434878188590657475340766437085830461471938008682720975)/2^332 correct to 103 decimal digits of pi 
> > > > Should have added above at the end of the equation = n then 1/n +3 ~ pi too 103 correct decimal digits. 
> > > So basically the trick is to find the right integer to add to the smaller power of 2 on the left side of the equation. 
> > > It is just that, a trick, that takes seconds to find. 
> > > It appears all that matters is the two powers of 2 are (4) apart and even and the larger power of 2 is the divisor. 
> > Continuing this as a new question about pi--- 
> > 
> > pi ~ 4 - 1/n - 1/(2*n) - 1/(3*n) - 1/(4*n) - 1/(5*n) ... 
> > 
> > What is the largest value of n possible where the convergence is always >pi? 
> > n =?
> I should have stated, n as an integer.

My next one --
(((2^952)*113)+8104404765941412889702286114479258869520636215834051707467153534189991882023910364929120556800695084661119334983817134490962909515315408427569855783444429124224573862605853199320109614444831382137591181375499492619928811818429775149930083922377757409905197865478907212143145193701698)/2^956 
pi correct to 290 decimal places..

[toc] | [prev] | [next] | [standalone]


#514152

Fromdjoyce099 <hlauk.h.bogart@gmail.com>
Date2021-03-13 11:29 -0800
Message-ID<9e733d6e-3e65-43d4-9b06-8eee2cf430cen@googlegroups.com>
In reply to#514150
On Saturday, March 13, 2021 at 2:16:44 PM UTC-5, djoyce099 wrote:
> On Saturday, March 13, 2021 at 11:43:59 AM UTC-5, djoyce099 wrote: 
> > On Saturday, March 13, 2021 at 11:32:56 AM UTC-5, djoyce099 wrote: 
> > > On Thursday, March 11, 2021 at 10:39:59 AM UTC-5, djoyce099 wrote: 
> > > > On Thursday, March 11, 2021 at 1:44:06 AM UTC-5, djoyce099 wrote: 
> > > > > On Thursday, March 11, 2021 at 1:25:46 AM UTC-5, djoyce099 wrote: 
> > > > > > On Thursday, March 11, 2021 at 12:35:08 AM UTC-5, Michael Moroney wrote: 
> > > > > > > On 3/7/2021 3:35 PM, djoyce099 wrote: 
> > > > > > > > The first is a well-known approximation in one simple fraction. 
> > > > > > > > 
> > > > > > > > 113/(2^4) = 7.0625 
> > > > > > > > 1/7.0625 +3 ~ pi 
> > > > > > > > Or 355/113 
> > > > > > > > 
> > > > > > > > A better one below 
> > > > > > > > --------------------------------------------------------------------------------- 
> > > > > > > > (((2^12)*113)+1)/(2^16) =n 
> > > > > > > > 1/n +3 ~pi 
> > > > > > > > Or 462849/(2^16) 
> > > > > > > > --------------------------------------------------------------------------------- 
> > > > > > > > What is the next one? 
> > > > > > > The answers so far all seem to involve 113 somehow. That is unusual. 
> > > > > > > 
> > > > > > > I will reword the question slightly: What is the smallest n, where k/n 
> > > > > > > is an approximation of pi, where k and n are both integers, and k/n is a 
> > > > > > > better approximation of pi than 355/113? 
> > > > > > > 
> > > > > > > ( |pi - k/n| < |pi - 355/113| ) 
> > > > > > > 
> > > > > > > Just like 113 is the smallest n where k/n [k=355] is a better 
> > > > > > > approximation of pi than 22/7, another common approximation of pi. 
> > > > > > 113 is used because to start these equations where n = 7.0625 then 1/7.0625 = 0.1415929203... +3 = 355/113. 
> > > > > > Here is the next one-- 
> > > > > > (((2^328)*113)+116413629295091038246264489256036237717551434878188590657475340766437085830461471938008682720975)/2^332 correct to 103 decimal digits of pi 
> > > > > Should have added above at the end of the equation = n then 1/n +3 ~ pi too 103 correct decimal digits. 
> > > > So basically the trick is to find the right integer to add to the smaller power of 2 on the left side of the equation. 
> > > > It is just that, a trick, that takes seconds to find. 
> > > > It appears all that matters is the two powers of 2 are (4) apart and even and the larger power of 2 is the divisor. 
> > > Continuing this as a new question about pi--- 
> > > 
> > > pi ~ 4 - 1/n - 1/(2*n) - 1/(3*n) - 1/(4*n) - 1/(5*n) ... 
> > > 
> > > What is the largest value of n possible where the convergence is always >pi? 
> > > n =? 
> > I should have stated, n as an integer.
> My next one -- 
> (((2^952)*113)+8104404765941412889702286114479258869520636215834051707467153534189991882023910364929120556800695084661119334983817134490962909515315408427569855783444429124224573862605853199320109614444831382137591181375499492619928811818429775149930083922377757409905197865478907212143145193701698)/2^956 
> pi correct to 290 decimal places..
The above again pi = 3+ 1/n results from (n) above

[toc] | [prev] | [next] | [standalone]


#514393

Fromdjoyce099 <hlauk.h.bogart@gmail.com>
Date2021-03-15 12:33 -0700
Message-ID<9c320867-48be-4be2-bc35-5751cd371849n@googlegroups.com>
In reply to#514150
On Saturday, March 13, 2021 at 2:16:44 PM UTC-5, djoyce099 wrote:
> On Saturday, March 13, 2021 at 11:43:59 AM UTC-5, djoyce099 wrote: 
> > On Saturday, March 13, 2021 at 11:32:56 AM UTC-5, djoyce099 wrote: 
> > > On Thursday, March 11, 2021 at 10:39:59 AM UTC-5, djoyce099 wrote: 
> > > > On Thursday, March 11, 2021 at 1:44:06 AM UTC-5, djoyce099 wrote: 
> > > > > On Thursday, March 11, 2021 at 1:25:46 AM UTC-5, djoyce099 wrote: 
> > > > > > On Thursday, March 11, 2021 at 12:35:08 AM UTC-5, Michael Moroney wrote: 
> > > > > > > On 3/7/2021 3:35 PM, djoyce099 wrote: 
> > > > > > > > The first is a well-known approximation in one simple fraction. 
> > > > > > > > 
> > > > > > > > 113/(2^4) = 7.0625 
> > > > > > > > 1/7.0625 +3 ~ pi 
> > > > > > > > Or 355/113 
> > > > > > > > 
> > > > > > > > A better one below 
> > > > > > > > --------------------------------------------------------------------------------- 
> > > > > > > > (((2^12)*113)+1)/(2^16) =n 
> > > > > > > > 1/n +3 ~pi 
> > > > > > > > Or 462849/(2^16) 
> > > > > > > > --------------------------------------------------------------------------------- 
> > > > > > > > What is the next one? 
> > > > > > > The answers so far all seem to involve 113 somehow. That is unusual. 
> > > > > > > 
> > > > > > > I will reword the question slightly: What is the smallest n, where k/n 
> > > > > > > is an approximation of pi, where k and n are both integers, and k/n is a 
> > > > > > > better approximation of pi than 355/113? 
> > > > > > > 
> > > > > > > ( |pi - k/n| < |pi - 355/113| ) 
> > > > > > > 
> > > > > > > Just like 113 is the smallest n where k/n [k=355] is a better 
> > > > > > > approximation of pi than 22/7, another common approximation of pi. 
> > > > > > 113 is used because to start these equations where n = 7.0625 then 1/7.0625 = 0.1415929203... +3 = 355/113. 
> > > > > > Here is the next one-- 
> > > > > > (((2^328)*113)+116413629295091038246264489256036237717551434878188590657475340766437085830461471938008682720975)/2^332 correct to 103 decimal digits of pi 
> > > > > Should have added above at the end of the equation = n then 1/n +3 ~ pi too 103 correct decimal digits. 
> > > > So basically the trick is to find the right integer to add to the smaller power of 2 on the left side of the equation. 
> > > > It is just that, a trick, that takes seconds to find. 
> > > > It appears all that matters is the two powers of 2 are (4) apart and even and the larger power of 2 is the divisor. 
> > > Continuing this as a new question about pi--- 
> > > 
> > > pi ~ 4 - 1/n - 1/(2*n) - 1/(3*n) - 1/(4*n) - 1/(5*n) ... 
> > > 
> > > What is the largest value of n possible where the convergence is always >pi? 
> > > n =? 
> > I should have stated, n as an integer.
> My next one -- 
> (((2^952)*113)+8104404765941412889702286114479258869520636215834051707467153534189991882023910364929120556800695084661119334983817134490962909515315408427569855783444429124224573862605853199320109614444831382137591181375499492619928811818429775149930083922377757409905197865478907212143145193701698)/2^956 
> pi correct to 290 decimal places..

In case anyone is wondering how I came up with the right integer to add
 to the left side of the equation
(2^956) * (1/(pi -3)  -  ((2^252)*113) = k . if the remainder of k is > 0.5 then round up and <0.5 just int(k) .
This k gets added to the final equation --
(((2^952)*113)+k)/(2^956) = n  1/n  +3 ~pi.
The actual correct decimal expansion is 289 decimal places.
The length of the divisor ( 2^956) = 288
I am not sure if any of these powers separated by 4 will produce 
much better than 3 decimal digits greater in length than (2x).

[toc] | [prev] | [next] | [standalone]


#514394

Fromdjoyce099 <hlauk.h.bogart@gmail.com>
Date2021-03-15 12:36 -0700
Message-ID<112c0f80-2326-49dd-b1f0-8c561b4d8961n@googlegroups.com>
In reply to#514393
On Monday, March 15, 2021 at 3:33:55 PM UTC-4, djoyce099 wrote:
> On Saturday, March 13, 2021 at 2:16:44 PM UTC-5, djoyce099 wrote:
> > On Saturday, March 13, 2021 at 11:43:59 AM UTC-5, djoyce099 wrote: 
> > > On Saturday, March 13, 2021 at 11:32:56 AM UTC-5, djoyce099 wrote: 
> > > > On Thursday, March 11, 2021 at 10:39:59 AM UTC-5, djoyce099 wrote: 
> > > > > On Thursday, March 11, 2021 at 1:44:06 AM UTC-5, djoyce099 wrote:
> > > > > > On Thursday, March 11, 2021 at 1:25:46 AM UTC-5, djoyce099 wrote: 
> > > > > > > On Thursday, March 11, 2021 at 12:35:08 AM UTC-5, Michael Moroney wrote:
> > > > > > > > On 3/7/2021 3:35 PM, djoyce099 wrote: 
> > > > > > > > > The first is a well-known approximation in one simple fraction. 
> > > > > > > > > 
> > > > > > > > > 113/(2^4) = 7.0625 
> > > > > > > > > 1/7.0625 +3 ~ pi 
> > > > > > > > > Or 355/113 
> > > > > > > > > 
> > > > > > > > > A better one below 
> > > > > > > > > --------------------------------------------------------------------------------- 
> > > > > > > > > (((2^12)*113)+1)/(2^16) =n 
> > > > > > > > > 1/n +3 ~pi 
> > > > > > > > > Or 462849/(2^16) 
> > > > > > > > > --------------------------------------------------------------------------------- 
> > > > > > > > > What is the next one?
> > > > > > > > The answers so far all seem to involve 113 somehow. That is unusual. 
> > > > > > > > 
> > > > > > > > I will reword the question slightly: What is the smallest n, where k/n 
> > > > > > > > is an approximation of pi, where k and n are both integers, and k/n is a 
> > > > > > > > better approximation of pi than 355/113? 
> > > > > > > > 
> > > > > > > > ( |pi - k/n| < |pi - 355/113| ) 
> > > > > > > > 
> > > > > > > > Just like 113 is the smallest n where k/n [k=355] is a better 
> > > > > > > > approximation of pi than 22/7, another common approximation of pi. 
> > > > > > > 113 is used because to start these equations where n = 7.0625 then 1/7.0625 = 0.1415929203... +3 = 355/113.
> > > > > > > Here is the next one-- 
> > > > > > > (((2^328)*113)+116413629295091038246264489256036237717551434878188590657475340766437085830461471938008682720975)/2^332 correct to 103 decimal digits of pi 
> > > > > > Should have added above at the end of the equation = n then 1/n +3 ~ pi too 103 correct decimal digits. 
> > > > > So basically the trick is to find the right integer to add to the smaller power of 2 on the left side of the equation. 
> > > > > It is just that, a trick, that takes seconds to find. 
> > > > > It appears all that matters is the two powers of 2 are (4) apart and even and the larger power of 2 is the divisor.
> > > > Continuing this as a new question about pi--- 
> > > > 
> > > > pi ~ 4 - 1/n - 1/(2*n) - 1/(3*n) - 1/(4*n) - 1/(5*n) ... 
> > > > 
> > > > What is the largest value of n possible where the convergence is always >pi? 
> > > > n =?
> > > I should have stated, n as an integer.
> > My next one -- 
> > (((2^952)*113)+8104404765941412889702286114479258869520636215834051707467153534189991882023910364929120556800695084661119334983817134490962909515315408427569855783444429124224573862605853199320109614444831382137591181375499492619928811818429775149930083922377757409905197865478907212143145193701698)/2^956 
> > pi correct to 290 decimal places..
> In case anyone is wondering how I came up with the right integer to add 
> to the left side of the equation 
> (2^956) * (1/(pi -3) - ((2^252)*113) = k . if the remainder of k is > 0.5 then round up and <0.5 just int(k) . 
> This k gets added to the final equation -- 
> (((2^952)*113)+k)/(2^956) = n 1/n +3 ~pi. 
> The actual correct decimal expansion is 289 decimal places. 
> The length of the divisor ( 2^956) = 288 
> I am not sure if any of these powers separated by 4 will produce 
> much better than 3 decimal digits greater in length than (2x).

That should be in the above line -- 2^x

[toc] | [prev] | [next] | [standalone]


#514439

FromMostowski Collapse <bursejan@gmail.com>
Date2021-03-16 00:56 -0700
Message-ID<50429283-43cd-4b00-8906-71151b7cccf8n@googlegroups.com>
In reply to#514393
Seems your last numerator is wrong. I get:

?- X is mp(mp(log(2^952*113-pi*2^956)/log(10),600), 6).
X = 0d288.378
?- X is mp(mp(log(2^951*113-pi*2^956)/log(10),600), 6).
X = 0d287.375
?- X is mp(mp(log(abs(2^950*113-pi*2^956))/log(10),600), 6).
X = 0d287.923

So 2^951*113 is closest.

But because:

?- X is pi*2^5.
X = 100.53096491487338

I woud try 2^951*101.

djoyce099 schrieb am Montag, 15. März 2021 um 20:33:55 UTC+1:
> On Saturday, March 13, 2021 at 2:16:44 PM UTC-5, djoyce099 wrote:
> > On Saturday, March 13, 2021 at 11:43:59 AM UTC-5, djoyce099 wrote: 
> > > On Saturday, March 13, 2021 at 11:32:56 AM UTC-5, djoyce099 wrote: 
> > > > On Thursday, March 11, 2021 at 10:39:59 AM UTC-5, djoyce099 wrote: 
> > > > > On Thursday, March 11, 2021 at 1:44:06 AM UTC-5, djoyce099 wrote:
> > > > > > On Thursday, March 11, 2021 at 1:25:46 AM UTC-5, djoyce099 wrote: 
> > > > > > > On Thursday, March 11, 2021 at 12:35:08 AM UTC-5, Michael Moroney wrote:
> > > > > > > > On 3/7/2021 3:35 PM, djoyce099 wrote: 
> > > > > > > > > The first is a well-known approximation in one simple fraction. 
> > > > > > > > > 
> > > > > > > > > 113/(2^4) = 7.0625 
> > > > > > > > > 1/7.0625 +3 ~ pi 
> > > > > > > > > Or 355/113 
> > > > > > > > > 
> > > > > > > > > A better one below 
> > > > > > > > > --------------------------------------------------------------------------------- 
> > > > > > > > > (((2^12)*113)+1)/(2^16) =n 
> > > > > > > > > 1/n +3 ~pi 
> > > > > > > > > Or 462849/(2^16) 
> > > > > > > > > --------------------------------------------------------------------------------- 
> > > > > > > > > What is the next one?
> > > > > > > > The answers so far all seem to involve 113 somehow. That is unusual. 
> > > > > > > > 
> > > > > > > > I will reword the question slightly: What is the smallest n, where k/n 
> > > > > > > > is an approximation of pi, where k and n are both integers, and k/n is a 
> > > > > > > > better approximation of pi than 355/113? 
> > > > > > > > 
> > > > > > > > ( |pi - k/n| < |pi - 355/113| ) 
> > > > > > > > 
> > > > > > > > Just like 113 is the smallest n where k/n [k=355] is a better 
> > > > > > > > approximation of pi than 22/7, another common approximation of pi. 
> > > > > > > 113 is used because to start these equations where n = 7.0625 then 1/7.0625 = 0.1415929203... +3 = 355/113.
> > > > > > > Here is the next one-- 
> > > > > > > (((2^328)*113)+116413629295091038246264489256036237717551434878188590657475340766437085830461471938008682720975)/2^332 correct to 103 decimal digits of pi 
> > > > > > Should have added above at the end of the equation = n then 1/n +3 ~ pi too 103 correct decimal digits. 
> > > > > So basically the trick is to find the right integer to add to the smaller power of 2 on the left side of the equation. 
> > > > > It is just that, a trick, that takes seconds to find. 
> > > > > It appears all that matters is the two powers of 2 are (4) apart and even and the larger power of 2 is the divisor.
> > > > Continuing this as a new question about pi--- 
> > > > 
> > > > pi ~ 4 - 1/n - 1/(2*n) - 1/(3*n) - 1/(4*n) - 1/(5*n) ... 
> > > > 
> > > > What is the largest value of n possible where the convergence is always >pi? 
> > > > n =?
> > > I should have stated, n as an integer.
> > My next one -- 
> > (((2^952)*113)+8104404765941412889702286114479258869520636215834051707467153534189991882023910364929120556800695084661119334983817134490962909515315408427569855783444429124224573862605853199320109614444831382137591181375499492619928811818429775149930083922377757409905197865478907212143145193701698)/2^956 
> > pi correct to 290 decimal places..
> In case anyone is wondering how I came up with the right integer to add 
> to the left side of the equation 
> (2^956) * (1/(pi -3) - ((2^252)*113) = k . if the remainder of k is > 0.5 then round up and <0.5 just int(k) . 
> This k gets added to the final equation -- 
> (((2^952)*113)+k)/(2^956) = n 1/n +3 ~pi. 
> The actual correct decimal expansion is 289 decimal places. 
> The length of the divisor ( 2^956) = 288 
> I am not sure if any of these powers separated by 4 will produce 
> much better than 3 decimal digits greater in length than (2x).

[toc] | [prev] | [next] | [standalone]


#514460

Fromdjoyce099 <hlauk.h.bogart@gmail.com>
Date2021-03-16 06:58 -0700
Message-ID<3212c2e6-611f-43f4-94b0-13d81a7d7041n@googlegroups.com>
In reply to#514439
On Tuesday, March 16, 2021 at 3:56:56 AM UTC-4, Mostowski Collapse wrote:
> Seems your last numerator is wrong. I get: 
> 
> ?- X is mp(mp(log(2^952*113-pi*2^956)/log(10),600), 6). 
> X = 0d288.378 
> ?- X is mp(mp(log(2^951*113-pi*2^956)/log(10),600), 6). 
> X = 0d287.375 
> ?- X is mp(mp(log(abs(2^950*113-pi*2^956))/log(10),600), 6). 
> X = 0d287.923 
> 
> So 2^951*113 is closest. 
> 
> But because: 
> 
> ?- X is pi*2^5. 
> X = 100.53096491487338 
> 
> I woud try 2^951*101.
> djoyce099 schrieb am Montag, 15. März 2021 um 20:33:55 UTC+1: 
> > On Saturday, March 13, 2021 at 2:16:44 PM UTC-5, djoyce099 wrote: 
> > > On Saturday, March 13, 2021 at 11:43:59 AM UTC-5, djoyce099 wrote: 
> > > > On Saturday, March 13, 2021 at 11:32:56 AM UTC-5, djoyce099 wrote: 
> > > > > On Thursday, March 11, 2021 at 10:39:59 AM UTC-5, djoyce099 wrote: 
> > > > > > On Thursday, March 11, 2021 at 1:44:06 AM UTC-5, djoyce099 wrote: 
> > > > > > > On Thursday, March 11, 2021 at 1:25:46 AM UTC-5, djoyce099 wrote: 
> > > > > > > > On Thursday, March 11, 2021 at 12:35:08 AM UTC-5, Michael Moroney wrote: 
> > > > > > > > > On 3/7/2021 3:35 PM, djoyce099 wrote: 
> > > > > > > > > > The first is a well-known approximation in one simple fraction. 
> > > > > > > > > > 
> > > > > > > > > > 113/(2^4) = 7.0625 
> > > > > > > > > > 1/7.0625 +3 ~ pi 
> > > > > > > > > > Or 355/113 
> > > > > > > > > > 
> > > > > > > > > > A better one below 
> > > > > > > > > > --------------------------------------------------------------------------------- 
> > > > > > > > > > (((2^12)*113)+1)/(2^16) =n 
> > > > > > > > > > 1/n +3 ~pi 
> > > > > > > > > > Or 462849/(2^16) 
> > > > > > > > > > --------------------------------------------------------------------------------- 
> > > > > > > > > > What is the next one? 
> > > > > > > > > The answers so far all seem to involve 113 somehow. That is unusual. 
> > > > > > > > > 
> > > > > > > > > I will reword the question slightly: What is the smallest n, where k/n 
> > > > > > > > > is an approximation of pi, where k and n are both integers, and k/n is a 
> > > > > > > > > better approximation of pi than 355/113? 
> > > > > > > > > 
> > > > > > > > > ( |pi - k/n| < |pi - 355/113| ) 
> > > > > > > > > 
> > > > > > > > > Just like 113 is the smallest n where k/n [k=355] is a better 
> > > > > > > > > approximation of pi than 22/7, another common approximation of pi. 
> > > > > > > > 113 is used because to start these equations where n = 7.0625 then 1/7.0625 = 0.1415929203... +3 = 355/113. 
> > > > > > > > Here is the next one-- 
> > > > > > > > (((2^328)*113)+116413629295091038246264489256036237717551434878188590657475340766437085830461471938008682720975)/2^332 correct to 103 decimal digits of pi 
> > > > > > > Should have added above at the end of the equation = n then 1/n +3 ~ pi too 103 correct decimal digits. 
> > > > > > So basically the trick is to find the right integer to add to the smaller power of 2 on the left side of the equation. 
> > > > > > It is just that, a trick, that takes seconds to find. 
> > > > > > It appears all that matters is the two powers of 2 are (4) apart and even and the larger power of 2 is the divisor. 
> > > > > Continuing this as a new question about pi--- 
> > > > > 
> > > > > pi ~ 4 - 1/n - 1/(2*n) - 1/(3*n) - 1/(4*n) - 1/(5*n) ... 
> > > > > 
> > > > > What is the largest value of n possible where the convergence is always >pi? 
> > > > > n =? 
> > > > I should have stated, n as an integer. 
> > > My next one -- 
> > > (((2^952)*113)+8104404765941412889702286114479258869520636215834051707467153534189991882023910364929120556800695084661119334983817134490962909515315408427569855783444429124224573862605853199320109614444831382137591181375499492619928811818429775149930083922377757409905197865478907212143145193701698)/2^956 
> > > pi correct to 290 decimal places.. 
> > In case anyone is wondering how I came up with the right integer to add 
> > to the left side of the equation 
> > (2^956) * (1/(pi -3) - ((2^252)*113) = k . if the remainder of k is > 0.5 then round up and <0.5 just int(k) . 
> > This k gets added to the final equation -- 
> > (((2^952)*113)+k)/(2^956) = n 1/n +3 ~pi. 
> > The actual correct decimal expansion is 289 decimal places. 
> > The length of the divisor ( 2^956) = 288 
> > I am not sure if any of these powers separated by 4 will produce 
> > much better than 3 decimal digits greater in length than (2x).

There is a mistake in the numerator, not in keeping with the 4 apart of the powers of 2 so 2^252 should be
2^952. I should have proofread it.

(2^956) * (1/(pi -3) - ((2^952)*113) = k . if the remainder of k is > 0.5 then round up and <0.5 just int(k) . 
 This k gets added to the final equation -- 
 (((2^952)*113)+k)/(2^956) = n 1/n +3 ~pi. 
The actual correct decimal expansion is 289 decimal places. 
The length of the divisor ( 2^956) = 288

[toc] | [prev] | [next] | [standalone]


#514462

FromMostowski Collapse <bursejan@gmail.com>
Date2021-03-16 07:23 -0700
Message-ID<f5998004-bde6-4fbc-bb02-d4f352e84db4n@googlegroups.com>
In reply to#514460
Ok, you got 113 from:

?- X is 1/(pi-3)*2^4.
X = 113.00021289489683

Oki Doki

[toc] | [prev] | [next] | [standalone]


#514464

Fromdjoyce099 <hlauk.h.bogart@gmail.com>
Date2021-03-16 07:43 -0700
Message-ID<d9c97191-a70d-4fbd-9011-2367e6072f79n@googlegroups.com>
In reply to#514462
On Tuesday, March 16, 2021 at 10:23:56 AM UTC-4, Mostowski Collapse wrote:
> Ok, you got 113 from: 
> 
> ?- X is 1/(pi-3)*2^4. 
> X = 113.00021289489683 
> 
> Oki Doki

Thanks, I never looked at it that way.

[toc] | [prev] | [next] | [standalone]


#514156

FromSergio <invalid@invalid.com>
Date2021-03-13 14:09 -0600
Message-ID<s2j65a$1run$1@gioia.aioe.org>
In reply to#514139
On 3/13/2021 10:43 AM, djoyce099 wrote:
> On Saturday, March 13, 2021 at 11:32:56 AM UTC-5, djoyce099 wrote:
>> On Thursday, March 11, 2021 at 10:39:59 AM UTC-5, djoyce099 wrote: 
>>> On Thursday, March 11, 2021 at 1:44:06 AM UTC-5, djoyce099 wrote: 
>>>> On Thursday, March 11, 2021 at 1:25:46 AM UTC-5, djoyce099 wrote: 
>>>>> On Thursday, March 11, 2021 at 12:35:08 AM UTC-5, Michael Moroney wrote: 
>>>>>> On 3/7/2021 3:35 PM, djoyce099 wrote: 
>>>>>>> The first is a well-known approximation in one simple fraction. 
>>>>>>>
>>>>>>> 113/(2^4) = 7.0625 
>>>>>>> 1/7.0625 +3 ~ pi 
>>>>>>> Or 355/113 
>>>>>>>
>>>>>>> A better one below 
>>>>>>> --------------------------------------------------------------------------------- 
>>>>>>> (((2^12)*113)+1)/(2^16) =n 
>>>>>>> 1/n +3 ~pi 
>>>>>>> Or 462849/(2^16) 
>>>>>>> --------------------------------------------------------------------------------- 
>>>>>>> What is the next one? 
>>>>>> The answers so far all seem to involve 113 somehow. That is unusual. 
>>>>>>
>>>>>> I will reword the question slightly: What is the smallest n, where k/n 
>>>>>> is an approximation of pi, where k and n are both integers, and k/n is a 
>>>>>> better approximation of pi than 355/113? 
>>>>>>
>>>>>> ( |pi - k/n| < |pi - 355/113| ) 
>>>>>>
>>>>>> Just like 113 is the smallest n where k/n [k=355] is a better 
>>>>>> approximation of pi than 22/7, another common approximation of pi. 
>>>>> 113 is used because to start these equations where n = 7.0625 then 1/7.0625 = 0.1415929203... +3 = 355/113. 
>>>>> Here is the next one-- 
>>>>> (((2^328)*113)+116413629295091038246264489256036237717551434878188590657475340766437085830461471938008682720975)/2^332 correct to 103 decimal digits of pi 
>>>> Should have added above at the end of the equation = n then 1/n +3 ~ pi too 103 correct decimal digits. 
>>> So basically the trick is to find the right integer to add to the smaller power of 2 on the left side of the equation. 
>>> It is just that, a trick, that takes seconds to find. 
>>> It appears all that matters is the two powers of 2 are (4) apart and even and the larger power of 2 is the divisor.
>> Continuing this as a new question about pi--- 
>>
>> pi ~ 4 - 1/n - 1/(2*n) - 1/(3*n) - 1/(4*n) - 1/(5*n) ... 
>>
>> What is the largest value of n possible where the convergence is always >pi? 
>> n =?
> I should have stated, n as an integer.
> 

pi ~ 4 - 1/n - 1/(2*n) - 1/(3*n) - 1/(4*n) - 1/(5*n) ...

pull the 1/n out;

pi ~ 4 - 1/n * ( 1 + 1/2 + 1/3 + 1/4 + 1/5 ...+1/k+...  )

inside the brackets is roughly about log(k), does not converge,

so in counting down from 4 to Pi by subtracting 1/n of an increasing
series that keeps going up...

unless you truncated the series



[toc] | [prev] | [next] | [standalone]


#514172

Fromdjoyce099 <hlauk.h.bogart@gmail.com>
Date2021-03-13 14:46 -0800
Message-ID<ea49a0a9-f594-425d-aa2a-202c71130ed8n@googlegroups.com>
In reply to#514156
On Saturday, March 13, 2021 at 3:09:24 PM UTC-5, Sergio wrote:
> On 3/13/2021 10:43 AM, djoyce099 wrote: 
> > On Saturday, March 13, 2021 at 11:32:56 AM UTC-5, djoyce099 wrote: 
> >> On Thursday, March 11, 2021 at 10:39:59 AM UTC-5, djoyce099 wrote: 
> >>> On Thursday, March 11, 2021 at 1:44:06 AM UTC-5, djoyce099 wrote: 
> >>>> On Thursday, March 11, 2021 at 1:25:46 AM UTC-5, djoyce099 wrote: 
> >>>>> On Thursday, March 11, 2021 at 12:35:08 AM UTC-5, Michael Moroney wrote: 
> >>>>>> On 3/7/2021 3:35 PM, djoyce099 wrote: 
> >>>>>>> The first is a well-known approximation in one simple fraction. 
> >>>>>>> 
> >>>>>>> 113/(2^4) = 7.0625 
> >>>>>>> 1/7.0625 +3 ~ pi 
> >>>>>>> Or 355/113 
> >>>>>>> 
> >>>>>>> A better one below 
> >>>>>>> --------------------------------------------------------------------------------- 
> >>>>>>> (((2^12)*113)+1)/(2^16) =n 
> >>>>>>> 1/n +3 ~pi 
> >>>>>>> Or 462849/(2^16) 
> >>>>>>> --------------------------------------------------------------------------------- 
> >>>>>>> What is the next one? 
> >>>>>> The answers so far all seem to involve 113 somehow. That is unusual. 
> >>>>>> 
> >>>>>> I will reword the question slightly: What is the smallest n, where k/n 
> >>>>>> is an approximation of pi, where k and n are both integers, and k/n is a 
> >>>>>> better approximation of pi than 355/113? 
> >>>>>> 
> >>>>>> ( |pi - k/n| < |pi - 355/113| ) 
> >>>>>> 
> >>>>>> Just like 113 is the smallest n where k/n [k=355] is a better 
> >>>>>> approximation of pi than 22/7, another common approximation of pi. 
> >>>>> 113 is used because to start these equations where n = 7.0625 then 1/7.0625 = 0.1415929203... +3 = 355/113. 
> >>>>> Here is the next one-- 
> >>>>> (((2^328)*113)+116413629295091038246264489256036237717551434878188590657475340766437085830461471938008682720975)/2^332 correct to 103 decimal digits of pi 
> >>>> Should have added above at the end of the equation = n then 1/n +3 ~ pi too 103 correct decimal digits. 
> >>> So basically the trick is to find the right integer to add to the smaller power of 2 on the left side of the equation. 
> >>> It is just that, a trick, that takes seconds to find. 
> >>> It appears all that matters is the two powers of 2 are (4) apart and even and the larger power of 2 is the divisor. 
> >> Continuing this as a new question about pi--- 
> >> 
> >> pi ~ 4 - 1/n - 1/(2*n) - 1/(3*n) - 1/(4*n) - 1/(5*n) ... 
> >> 
> >> What is the largest value of n possible where the convergence is always >pi? 
> >> n =? 
> > I should have stated, n as an integer. 
> > 
> 
> pi ~ 4 - 1/n - 1/(2*n) - 1/(3*n) - 1/(4*n) - 1/(5*n) ...
> pull the 1/n out; 
> 
> pi ~ 4 - 1/n * ( 1 + 1/2 + 1/3 + 1/4 + 1/5 ...+1/k+... ) 
> 
> inside the brackets is roughly about log(k), does not converge, 
> 
> so in counting down from 4 to Pi by subtracting 1/n of an increasing 
> series that keeps going up... 
> 
> unless you truncated the series

No. the series continues. Note   (...).
From my original, how big is integer n?
Is there a limit to the size of n? 
Also, note that what  I am saying is if say n=10 then --

pi~4- {1/10 - 1/20- 1/60 - 1/80 -...}
Not sure as to what size n can go and still have the value >pi?
Whereas n+1 would be <pi.
 

 

[toc] | [prev] | [next] | [standalone]


#514194

FromSergio <invalid@invalid.com>
Date2021-03-13 20:22 -0600
Message-ID<s2js0j$1rat$1@gioia.aioe.org>
In reply to#514172
On 3/13/2021 4:46 PM, djoyce099 wrote:
> On Saturday, March 13, 2021 at 3:09:24 PM UTC-5, Sergio wrote:
>> On 3/13/2021 10:43 AM, djoyce099 wrote: 
>>> On Saturday, March 13, 2021 at 11:32:56 AM UTC-5, djoyce099 wrote: 
>>>> On Thursday, March 11, 2021 at 10:39:59 AM UTC-5, djoyce099 wrote: 
>>>>> On Thursday, March 11, 2021 at 1:44:06 AM UTC-5, djoyce099 wrote: 
>>>>>> On Thursday, March 11, 2021 at 1:25:46 AM UTC-5, djoyce099 wrote: 
>>>>>>> On Thursday, March 11, 2021 at 12:35:08 AM UTC-5, Michael Moroney wrote: 
>>>>>>>> On 3/7/2021 3:35 PM, djoyce099 wrote: 
>>>>>>>>> The first is a well-known approximation in one simple fraction. 
>>>>>>>>>
>>>>>>>>> 113/(2^4) = 7.0625 
>>>>>>>>> 1/7.0625 +3 ~ pi 
>>>>>>>>> Or 355/113 
>>>>>>>>>
>>>>>>>>> A better one below 
>>>>>>>>> --------------------------------------------------------------------------------- 
>>>>>>>>> (((2^12)*113)+1)/(2^16) =n 
>>>>>>>>> 1/n +3 ~pi 
>>>>>>>>> Or 462849/(2^16) 
>>>>>>>>> --------------------------------------------------------------------------------- 
>>>>>>>>> What is the next one? 
>>>>>>>> The answers so far all seem to involve 113 somehow. That is unusual. 
>>>>>>>>
>>>>>>>> I will reword the question slightly: What is the smallest n, where k/n 
>>>>>>>> is an approximation of pi, where k and n are both integers, and k/n is a 
>>>>>>>> better approximation of pi than 355/113? 
>>>>>>>>
>>>>>>>> ( |pi - k/n| < |pi - 355/113| ) 
>>>>>>>>
>>>>>>>> Just like 113 is the smallest n where k/n [k=355] is a better 
>>>>>>>> approximation of pi than 22/7, another common approximation of pi. 
>>>>>>> 113 is used because to start these equations where n = 7.0625 then 1/7.0625 = 0.1415929203... +3 = 355/113. 
>>>>>>> Here is the next one-- 
>>>>>>> (((2^328)*113)+116413629295091038246264489256036237717551434878188590657475340766437085830461471938008682720975)/2^332 correct to 103 decimal digits of pi 
>>>>>> Should have added above at the end of the equation = n then 1/n +3 ~ pi too 103 correct decimal digits. 
>>>>> So basically the trick is to find the right integer to add to the smaller power of 2 on the left side of the equation. 
>>>>> It is just that, a trick, that takes seconds to find. 
>>>>> It appears all that matters is the two powers of 2 are (4) apart and even and the larger power of 2 is the divisor. 
>>>> Continuing this as a new question about pi--- 
>>>>
>>>> pi ~ 4 - 1/n - 1/(2*n) - 1/(3*n) - 1/(4*n) - 1/(5*n) ... 
>>>>
>>>> What is the largest value of n possible where the convergence is always >pi? 
>>>> n =? 
>>> I should have stated, n as an integer. 
>>>
>>
>> pi ~ 4 - 1/n - 1/(2*n) - 1/(3*n) - 1/(4*n) - 1/(5*n) ...
>> pull the 1/n out; 
>>
>> pi ~ 4 - 1/n * ( 1 + 1/2 + 1/3 + 1/4 + 1/5 ...+1/k+... ) 
>>
>> inside the brackets is roughly about log(k), does not converge, 
>>
>> so in counting down from 4 to Pi by subtracting 1/n of an increasing 
>> series that keeps going up... 
>>
>> unless you truncated the series
> 
> No. the series continues. Note   (...).
> From my original, how big is integer n?
> Is there a limit to the size of n? 
> Also, note that what  I am saying is if say n=10 then --
> 
> pi~4- {1/10 - 1/20- 1/60 - 1/80 -...}
> Not sure as to what size n can go and still have the value >pi?
> Whereas n+1 would be <pi.
>  
> 

the series keeps going up, there is no upper limit

https://www.quora.com/What-is-the-limit-of-the-series-1-1+1-2+1-3+1-4+-1-n-if-n-10-5?share=1

 1 + 1/2 + 1/3 + 1/4 + 1/5 ...+1/k+...

=> The first k = 100,000 terms of the harmonic series is equal to about
12.09.


when k = 10^8    series sum is  18.9979 or about 19


4 - pi = 0.8584074...

so sum of 1/n*(1 + 1/2 + 1/3 + 1/4 + 1/5 ...)

    1/n * 19  =>  n = 1/19 or .0526

but the sum will still increase but slowly, so n will move but slowly
slowly => go to differentials to calculate rates

[toc] | [prev] | [next] | [standalone]


#514177

FromJames Waldby <j-waldby@no.no>
Date2021-03-13 23:46 +0000
Message-ID<s2jitg$3d6$1@dont-email.me>
In reply to#514138
djoyce099 <hlauk.h.bogart@gmail.com> wrote:
...
> Continuing this as a new question about pi---
> 
> pi ~ 4 - 1/n - 1/(2*n) - 1/(3*n) - 1/(4*n) - 1/(5*n) ...
> 
> What is the largest value of n  possible where the convergence is always >pi? 
> n =?


Your question appears to be not well-posed.  First, suppose you
mean pi < 4 - 1/n - 1/(2*n) - 1/(3*n) - 1/(4*n) - 1/(5*n).
Let s = 1 + 1/2 + 1/3 + 1/4 + 1/5 = 137/60, and let d = 4-pi.
Solving d > s/n we get n > s/d ~ 2.66, so any integer n > 2
will work.  Now n can be arbitrarily large; it has no largest value.

Second, suppose you mean for the sequence to continue indefinitely.
The sum, if defined, would be 1/n times the sum of the harmonic 
sequence.  But the harmonic series [the sequence of partial sums]
does not converge.  It grows arbitrarily large.  For any fixed 
value of n, there is a value of m such that 4 - (1 + 1/2 +
1/3 + ... + 1/m)/n is less than pi.  There is no n that is large
enough to do what you want.

-- 
jiw

[toc] | [prev] | [next] | [standalone]


#514179

Fromdjoyce099 <hlauk.h.bogart@gmail.com>
Date2021-03-13 16:20 -0800
Message-ID<e5d10524-727e-4651-b47a-8fed093b8041n@googlegroups.com>
In reply to#514177
On Saturday, March 13, 2021 at 6:47:04 PM UTC-5, James Waldby wrote:
> djoyce099 <hlauk.h...@gmail.com> wrote: 
> ...
> > Continuing this as a new question about pi--- 
> > 
> > pi ~ 4 - 1/n - 1/(2*n) - 1/(3*n) - 1/(4*n) - 1/(5*n) ... 
> > 
> > What is the largest value of n possible where the convergence is always >pi? 
> > n =?
> Your question appears to be not well-posed. First, suppose you 
> mean pi < 4 - 1/n - 1/(2*n) - 1/(3*n) - 1/(4*n) - 1/(5*n). 
> Let s = 1 + 1/2 + 1/3 + 1/4 + 1/5 = 137/60, and let d = 4-pi. 
> Solving d > s/n we get n > s/d ~ 2.66, so any integer n > 2 
> will work. Now n can be arbitrarily large; it has no largest value. 
> 
> Second, suppose you mean for the sequence to continue indefinitely. 
> The sum, if defined, would be 1/n times the sum of the harmonic 
> sequence. But the harmonic series [the sequence of partial sums] 
> does not converge. It grows arbitrarily large. For any fixed 
> value of n, there is a value of m such that 4 - (1 + 1/2 + 
> 1/3 + ... + 1/m)/n is less than pi. There is no n that is large 
> enough to do what you want. 
> 
> -- 
> jiw

So no matter what the size of integer (n) with enough iterations the value will eventually <pi?

Thanks

[toc] | [prev] | [next] | [standalone]


#514180

Fromdjoyce099 <hlauk.h.bogart@gmail.com>
Date2021-03-13 16:22 -0800
Message-ID<c9c6890e-e701-445f-953f-030cfd3e2c5cn@googlegroups.com>
In reply to#514179
On Saturday, March 13, 2021 at 7:20:48 PM UTC-5, djoyce099 wrote:
> On Saturday, March 13, 2021 at 6:47:04 PM UTC-5, James Waldby wrote: 
> > djoyce099 <hlauk.h...@gmail.com> wrote: 
> > ... 
> > > Continuing this as a new question about pi--- 
> > > 
> > > pi ~ 4 - 1/n - 1/(2*n) - 1/(3*n) - 1/(4*n) - 1/(5*n) ... 
> > > 
> > > What is the largest value of n possible where the convergence is always >pi? 
> > > n =? 
> > Your question appears to be not well-posed. First, suppose you 
> > mean pi < 4 - 1/n - 1/(2*n) - 1/(3*n) - 1/(4*n) - 1/(5*n). 
> > Let s = 1 + 1/2 + 1/3 + 1/4 + 1/5 = 137/60, and let d = 4-pi. 
> > Solving d > s/n we get n > s/d ~ 2.66, so any integer n > 2 
> > will work. Now n can be arbitrarily large; it has no largest value. 
> > 
> > Second, suppose you mean for the sequence to continue indefinitely. 
> > The sum, if defined, would be 1/n times the sum of the harmonic 
> > sequence. But the harmonic series [the sequence of partial sums] 
> > does not converge. It grows arbitrarily large. For any fixed 
> > value of n, there is a value of m such that 4 - (1 + 1/2 + 
> > 1/3 + ... + 1/m)/n is less than pi. There is no n that is large 
> > enough to do what you want. 
> > 
> > -- 
> > jiw
> So no matter what the size of integer (n) with enough iterations the value will eventually <pi? 
> 
> Thanks
But >0

[toc] | [prev] | [next] | [standalone]


#514203

FromJames Waldby <j-waldby@no.no>
Date2021-03-14 04:29 +0000
Message-ID<s2k3f8$bmt$2@dont-email.me>
In reply to#514180
djoyce099 <hlauk.h.bogart@gmail.com> wrote:
> On Saturday, March 13, 2021 at 7:20:48 PM UTC-5, djoyce099 wrote:
>> On Saturday, March 13, 2021 at 6:47:04 PM UTC-5, James Waldby wrote: 
>> > djoyce099 <hlauk.h...@gmail.com> wrote: 
>> > ... 
>> > > Continuing this as a new question about pi--- 
>> > > 
>> > > pi ~ 4 - 1/n - 1/(2*n) - 1/(3*n) - 1/(4*n) - 1/(5*n) ... 
>> > > 
>> > > What is the largest value of n possible where the convergence is always >pi? 
>> > > n =? 
>> > Your question appears to be not well-posed. First, suppose you 
>> > mean pi < 4 - 1/n - 1/(2*n) - 1/(3*n) - 1/(4*n) - 1/(5*n). 
>> > Let s = 1 + 1/2 + 1/3 + 1/4 + 1/5 = 137/60, and let d = 4-pi. 
>> > Solving d > s/n we get n > s/d ~ 2.66, so any integer n > 2 
>> > will work. Now n can be arbitrarily large; it has no largest value. 
>> > 
>> > Second, suppose you mean for the sequence to continue indefinitely. 
>> > The sum, if defined, would be 1/n times the sum of the harmonic 
>> > sequence. But the harmonic series [the sequence of partial sums] 
>> > does not converge. It grows arbitrarily large. For any fixed 
>> > value of n, there is a value of m such that 4 - (1 + 1/2 + 
>> > 1/3 + ... + 1/m)/n is less than pi. There is no n that is large 
>> > enough to do what you want. 

>> So no matter what the size of integer (n) with enough iterations the value will eventually <pi? 

For any positive n,  the finite sum
4 - 1/n - 1/(2*n) - 1/(3*n) - 1/(4*n) - ... - 1/(k*n)
can be made arbitrarily small by making k big enough. 

When we don't limit the number of terms,
  4 - 1/n - 1/(2*n) - 1/(3*n) - 1/(4*n) - ... 
is an undefined value -- the series is non-convergent.

> But >0

No, there is no lower limit on how negative you can make 4 - 1/n -
1/(2*n) - 1/(3*n) - 1/(4*n) - ... - 1/(k*n) by making k big enough.
As another poster intimated, with gamma = Euler???Mascheroni constant ~
0.5772, we have - 1/n - 1/(2*n) - 1/(3*n) - 1/(4*n) - ... - 1/(k*n)
approaching (-1/n) * (log(k) + gamma) as k increases.  See eg
https://www.math.drexel.edu/~tolya/123_harmonic.pdf

[toc] | [prev] | [next] | [standalone]


#514206

FromSergio <invalid@invalid.com>
Date2021-03-13 23:05 -0600
Message-ID<s2k5jk$pek$2@gioia.aioe.org>
In reply to#514203
On 3/13/2021 10:29 PM, James Waldby wrote:
> djoyce099 <hlauk.h.bogart@gmail.com> wrote:
>> On Saturday, March 13, 2021 at 7:20:48 PM UTC-5, djoyce099 wrote:
>>> On Saturday, March 13, 2021 at 6:47:04 PM UTC-5, James Waldby wrote: 
>>>> djoyce099 <hlauk.h...@gmail.com> wrote: 
>>>> ... 
>>>>> Continuing this as a new question about pi--- 
>>>>>
>>>>> pi ~ 4 - 1/n - 1/(2*n) - 1/(3*n) - 1/(4*n) - 1/(5*n) ... 
>>>>>
>>>>> What is the largest value of n possible where the convergence is always >pi? 
>>>>> n =? 
>>>> Your question appears to be not well-posed. First, suppose you 
>>>> mean pi < 4 - 1/n - 1/(2*n) - 1/(3*n) - 1/(4*n) - 1/(5*n). 
>>>> Let s = 1 + 1/2 + 1/3 + 1/4 + 1/5 = 137/60, and let d = 4-pi. 
>>>> Solving d > s/n we get n > s/d ~ 2.66, so any integer n > 2 
>>>> will work. Now n can be arbitrarily large; it has no largest value. 
>>>>
>>>> Second, suppose you mean for the sequence to continue indefinitely. 
>>>> The sum, if defined, would be 1/n times the sum of the harmonic 
>>>> sequence. But the harmonic series [the sequence of partial sums] 
>>>> does not converge. It grows arbitrarily large. For any fixed 
>>>> value of n, there is a value of m such that 4 - (1 + 1/2 + 
>>>> 1/3 + ... + 1/m)/n is less than pi. There is no n that is large 
>>>> enough to do what you want. 
> 
>>> So no matter what the size of integer (n) with enough iterations the value will eventually <pi? 
> 
> For any positive n,  the finite sum
> 4 - 1/n - 1/(2*n) - 1/(3*n) - 1/(4*n) - ... - 1/(k*n)
> can be made arbitrarily small by making k big enough. 
> 
> When we don't limit the number of terms,
>   4 - 1/n - 1/(2*n) - 1/(3*n) - 1/(4*n) - ... 
> is an undefined value -- the series is non-convergent.

agree,

> 
>> But >0
> 
> No, there is no lower limit on how negative you can make 4 - 1/n -
> 1/(2*n) - 1/(3*n) - 1/(4*n) - ... - 1/(k*n) by making k big enough.
> As another poster intimated, with gamma = Euler???Mascheroni constant ~
> 0.5772, we have - 1/n - 1/(2*n) - 1/(3*n) - 1/(4*n) - ... - 1/(k*n)
> approaching (-1/n) * (log(k) + gamma) as k increases.  See eg
> https://www.math.drexel.edu/~tolya/123_harmonic.pdf
> 

[toc] | [prev] | [next] | [standalone]


#514243

Fromdjoyce099 <hlauk.h.bogart@gmail.com>
Date2021-03-14 09:48 -0700
Message-ID<4438b0aa-7aaf-4a61-98ea-d9fa2925e6f7n@googlegroups.com>
In reply to#514203
On Saturday, March 13, 2021 at 11:29:36 PM UTC-5, James Waldby wrote:
> djoyce099 <hlauk.h...@gmail.com> wrote: 
> > On Saturday, March 13, 2021 at 7:20:48 PM UTC-5, djoyce099 wrote: 
> >> On Saturday, March 13, 2021 at 6:47:04 PM UTC-5, James Waldby wrote: 
> >> > djoyce099 <hlauk.h...@gmail.com> wrote: 
> >> > ... 
> >> > > Continuing this as a new question about pi--- 
> >> > > 
> >> > > pi ~ 4 - 1/n - 1/(2*n) - 1/(3*n) - 1/(4*n) - 1/(5*n) ... 
> >> > > 
> >> > > What is the largest value of n possible where the convergence is always >pi? 
> >> > > n =? 
> >> > Your question appears to be not well-posed. First, suppose you 
> >> > mean pi < 4 - 1/n - 1/(2*n) - 1/(3*n) - 1/(4*n) - 1/(5*n). 
> >> > Let s = 1 + 1/2 + 1/3 + 1/4 + 1/5 = 137/60, and let d = 4-pi. 
> >> > Solving d > s/n we get n > s/d ~ 2.66, so any integer n > 2 
> >> > will work. Now n can be arbitrarily large; it has no largest value. 
> >> > 
> >> > Second, suppose you mean for the sequence to continue indefinitely. 
> >> > The sum, if defined, would be 1/n times the sum of the harmonic 
> >> > sequence. But the harmonic series [the sequence of partial sums] 
> >> > does not converge. It grows arbitrarily large. For any fixed 
> >> > value of n, there is a value of m such that 4 - (1 + 1/2 + 
> >> > 1/3 + ... + 1/m)/n is less than pi. There is no n that is large 
> >> > enough to do what you want.
> >> So no matter what the size of integer (n) with enough iterations the value will eventually <pi?
> For any positive n, the finite sum 
> 4 - 1/n - 1/(2*n) - 1/(3*n) - 1/(4*n) - ... - 1/(k*n) 
> can be made arbitrarily small by making k big enough. 
> 
> When we don't limit the number of terms, 
> 4 - 1/n - 1/(2*n) - 1/(3*n) - 1/(4*n) - ... 
> is an undefined value -- the series is non-convergent. 
> 
> > But >0 
> 
> No, there is no lower limit on how negative you can make 4 - 1/n - 
> 1/(2*n) - 1/(3*n) - 1/(4*n) - ... - 1/(k*n) by making k big enough. 
> As another poster intimated, with gamma = Euler???Mascheroni constant ~ 
> 0.5772, we have - 1/n - 1/(2*n) - 1/(3*n) - 1/(4*n) - ... - 1/(k*n) 
> approaching (-1/n) * (log(k) + gamma) as k increases. See eg 
> https://www.math.drexel.edu/~tolya/123_harmonic.pdf

Thanks.

[toc] | [prev] | [next] | [standalone]


Page 3 of 4 — ← Prev page 1 2 [3] 4  Next page →

Back to top | Article view | sci.math


csiph-web