Collatz-Intro - The General loop

Workshop
Recreational
Mathematics

Gottfried Helms Univ. Kassel

mailto: helms at uni-kassel

www: math-homepage

 

 

 

A "general" loop is a distinguishing from special forms of loops, where no a-priori restriction is made to the exponents. So it may be written as

 

 

a' = T(a;A,B,C,D,...G,H)  with N exponents (=steps) and a'=a

 

 

Since the exponents are arbitrarily mixed, ascending and descending steps may follow in arbitrary sequence, and the only things, that are known a priori is the length N and that the sum of all exponents 1  must roughly agree with the sum of all exponents >1, since the former are ascending steps and the latter descending. This can be estimated from the following form:

 

L = number of 1's

Ascending factor u ~ (3/2)^L

M = number of exponents > 1  = N - L

S = sum of all exponents ; S-L = sum of descending exponents

descending factor d ~ 3^M / 2^S

 

ascending factor * descending factor = 1 // required to satisfy a ~ a * u * d

 

 u * d ~  1

 

3^L/2^L ~ 3^(N-L)/2^(S-L)

2^S  ~ 3^N

 

S ~ N*log(3)/log(2) ~ N * 1.59

 

 

In contrast to a general loop is the so called "primitive" loop, which is defined that after N-1 ascending steps one single descending step follows:

 

a' = T(a;1,1,1,1...,1,A) with N-1 1's and a'=a

 

Then A must be roughly

 

2^A ~ (2/3)^(N-1) or

 

A ~ (N-1) (log(3)/log(2)  -1)
A ~ (N-1)*0.59...

and also be integer, by definition.

 

This type of loop is easier accessible, and remarkable results were reached, especially the existence of such a loop of *any* length could be disproven with one proof. This is a qualitative difference to the problem of the general loop, where only disproofs are reached for a specified length or a interval of lengthes of the form "from length 1 up to length x". (currently ~ 250000, see [Rosendaal])

 

 

 

Conditions of a general loop, some disproving examples

The most simple case: is there another 1-step-loop?

 

A 1-step-loop means, whether after only one transformation the same number occurs as output, which was given as input. One known exemplar is 1; in the original Collatz-transformation the loop is 1 -> 4 -> 2 -> 1, and this loop is usually called the "trivial loop".

 

In my notation this is only one step, namely

 

 1 = T(1;2)

 

But also:

 1 = T(1;2,2) = T(1;2,2,2...)

 

which rhs-transformations indicates, that this 1-step-loop (as any) can concatenated infinitely many times.

 

Now I show, how to disprove, the possibility of another loop of the same length, by first finding out that this is one solution for a = T(a;A) and then showing, that there are no others.

This example may be found too simple, but I use it to introduce my general line of arguing, which is  only a schematic extension from this most simple case.

 

 

Assume general 1-step-transformation

a' = T(a;A) 

 

where a' = a   or

 

 a'
 -- = 1
 a

which makes it a loop, where se seek, whether there exists any exponent of A, which allows a to be an odd integer>1 or controversely, whether exist any a, which allows the exponent A be integer.

 

The transformation-formula expanded says:

 

     3*a + 1
a' = -------
       2^A

 

 

Now, if a'/a = 1 we have:

 

     3*a + 1    1
1  = ------- * ---
       2^A      a

 

 

or

 

      3*a + 1
2^A = -------
         a   

 

which is

 

           1
2^A = 3 + ---
           a

 

 

Now the qestion was: is there any A, that allows an integer solution for a, or conversely: is there any a which allows an integer exponent A?

 

We see, that if a = 1 then the rhs is 4, so indeed we have a solution for A namely A=2

 

           1
2^A = 3 + ---  --> A=1 
           1

 

This result means, we have a loop with

 

 a = T(a;A) with a=1 and A=2  --->  1 = T(1;2)

 

 

Now we look, whether there could be another solution:

this is not possible, since if we increase a, then the rhs would decrease, but cannot decrease below 3. But between 3 and 4 are no more powers of 2.

 

We see, that the range of the rhs for any choice of a is

 

   3<=rhs<=4

 

But since the lhs must be an integer power of 2 , this can be expanded to

 

   3<=2^A = rhs<=4

 

The function powerceil2(x) denotes the smallest power of 2 equal or greater than x so

 

   powerceil2(3)<= 2^A = rhs <=4

 

 

This will be a standard notation for the critical-loop-criteria from now.

 

 From this we know, there can only be one solution, since for the critical inequality

 

   powerceil2(3) = 4  <= 2^A = rhs <=4

 

there can only be one solution for A, namely 2. Conversely, from that we can uniquely determine a

 

2^2 = 3+ 1/a  ---> a = 1/(4-3) = 1

 

 

 

The 2-step-loop

Now we extend this to the question, whether a 2-step-loop exists:

 

We write both transformations with their two members a,b

a -> b -> a -> b -> a ....

 

a' = T(a;A) where b=a' and b'=T(b;B), where a = b'

 

Then we state the trivial product

 

                   3a +1    3b +1

a * b  = a' * b' = -----  * -----

                    2^A      2^B

 

 

and reorder like in the previous example:

 

           3a + 1    3b + 1

 2^(A+B) = -----  * ------ 

             b          a  

 

Rotating the denominator, and cancelling, this can be expressed as

 

                 1           1
 2^(A+B) = (3 + --)  * (3 + -- )
                 a           b  

 

and exibits a pretty simple exact extension of the one-step-condition before.

 

For the smallest possible a,b (a=b=1) the rhs calculates to 4^2, and even if a and b grow to infinity, the rhs will always be greater or equal 3^2.

 

So we have again the form of the critical inequality, only with the exponent N=2 and two parentheses:

 

                                 1        1

powerceil2(3^2) <= 2^(A+B) =(3 + --)(3 + --)  <= 4^2

                                 a        b

 

 

Now we can insert on the lhs:

 

                                         1        1

powerceil2(3^2) = 16    <= 2^(A+B) =(3 + --)(3 + --)  <= 16

                                         a        b

 

that means, we can have only the solution, where A+B=4 . Since the parentheses need their maximum 4 to form their product 16, a and b both must take their minimum, which is 1.

 

So we have a solution for the 2-step-loop with

 

a = T(a;A,B) -->  a=1, A=B=2 and  1 = T(1;2,2)

 

and this is the only solution.

 

 

 

More steps of a general loop

For more steps in the loop, the formula is simply extensible. The critical inequality for the general case with N=number of steps, S = sum of exponents we have:

 

[Eq 2.1] "critical inequality" for loops:

 

powerceil2(3^N) <= 2^S = (3+1/a)(3+1/b)(3+1/c)...(3+1/h) <= 4^N

 

 

This formula has the nice properties:

one can either simply put candidates for a,b,c into the product, and see, whether a power of 2 occurs -

or conversely one could search this way...(but that would result in too many combinatorical possibilities).

 

 

Unfortunately weaken the bounds, that this inequality states, with growing N: there are multiple possibilities for the sum of exponents S to fit between powerceil2(3^N) and 4^N, so it is *generally* not yet sufficient.

 

But still there are some lengthes of loops, which can be denied by this formula.

The reason for this is, that there is another requirement on all members a,b,c... of a projected loop: they must all be different (otherwise the loop would already be shorter).

Since they all

·         should be greater than 1,

·         should be odd, and

·         cannot be divisible by 3,

the theoretically lowest values can be: a=5,b=7,c=11,...

and this allows to disprove several loops up to the length of 100 to 200 as it is shown in the following table.

 

 

Remarks to the approximation-approach

The following table lists results for the tests for general loops for lengthes up to 200 other than the trivial loop.

N:=      looplength

ug:=     powerceil2(3)/3^N

prod:=   (3+1/a)(3+1/b).../3^N

ratio:=  ug/prod must be <=1 to make this loop possible by satisfying

                 the critical equation

 

 

 

    n    ug               prod          ug/prod    ug > prod:
                                                   loop cannot exist

------------------------------------------------------------

    1    1.333333         1.066667     1.250000     -true-

    2    1.777778         1.117460     1.590909     -true-

    3    1.185185         1.151323     1.029412     -true-

    4    1.580247         1.180844     1.338235     -true-

    5    1.053498         1.203998     0.875000           

    6    1.404664         1.225120     1.146552     -true-

    7    1.872885         1.242876     1.506897     -true-

    8    1.248590         1.259447     0.991379           

    9    1.664787         1.273924     1.306818     -true-

   10    1.109858         1.287622     0.861944           

   11    1.479811         1.299885     1.138416     -true-

   12    1.973081         1.311596     1.504336     -true-

   13    1.315387         1.322259     0.994803           

   14    1.753850         1.332509     1.316201     -true-

   15    1.169233         1.341960     0.871288           

   16    1.558977         1.351089     1.153868     -true-

   17    1.039318         1.359586     0.764437           

   18    1.385758         1.367826     1.013110     -true-

   19    1.847677         1.375554     1.343224     -true-

   20    1.231785         1.383070     0.890616           

   21    1.642379         1.390163     1.181429     -true-

   22    1.094920         1.397079     0.783720           

   23    1.459893         1.403638     1.040078     -true-

   24    1.946524         1.410048     1.380467     -true-

   25    1.297683         1.416152     0.916344           

   26    1.730243         1.422127     1.216659     -true-

   27    1.153496         1.427838     0.807861           

   28    1.537994         1.433438     1.072941     -true-

   29    1.025329         1.438807     0.712625           

   30    1.367106         1.444077     0.946699           

   31    1.822808         1.449144     1.257852     -true-

   32    1.215205         1.454124     0.835696           

   33    1.620274         1.458923     1.110596     -true-

   34    1.080182         1.463644     0.738009           

   35    1.440243         1.468204     0.980956           

   36    1.920324         1.472694     1.303954     -true-

   37    1.280216         1.477038     0.866746           

   38    1.706955         1.481319     1.152321     -true-

   39    1.137970         1.485469     0.766068           

   40    1.517293         1.489561     1.018618     -true-

   41    1.011529         1.493533     0.677273            

   42    1.348705         1.497453     0.900666           

   43    1.798274         1.501263     1.197840     -true-

   44    1.198849         1.505026     0.796564           

   45    1.598465         1.508688     1.059507     -true-

   46    1.065644         1.512306     0.704648           

   47    1.420858         1.515831     0.937346           

   48    1.894477         1.519316     1.246928     -true-

   49    1.262985         1.522714     0.829430           

   50    1.683980         1.526076     1.103471     -true-

   51    1.122653         1.529358     0.734068           

   52    1.496871         1.532605     0.976684           

   53    1.995828         1.535778     1.299555     -true-

   54    1.330552         1.538919     0.864602           

   55    1.774069         1.541990     1.150506     -true-

   56    1.182713         1.545032     0.765494           

   57    1.576951         1.548009     1.018696     -true-

   58    1.051300         1.550957     0.677840           

   59    1.401734         1.553845     0.902106           

   60    1.868978         1.556707     1.200598     -true-

   61    1.245986         1.559512     0.798959           

   62    1.661314         1.562292     1.063383     -true-

   63    1.107543         1.565018     0.707687           

   64    1.476724         1.567721     0.941956           

   65    1.968965         1.570374     1.253819     -true-

   66    1.312643         1.573004     0.834482           

   67    1.750191         1.575587     1.110818     -true-

   68    1.166794         1.578149     0.739343           

   69    1.555725         1.580666     0.984221           

   70    1.037150         1.583163     0.655113           

   71    1.382867         1.585618     0.872131           

   72    1.843823         1.588053     1.161058     -true-

   73    1.229215         1.590449     0.772873            

   74    1.638954         1.592826     1.028960     -true-

   75    1.092636         1.595165     0.684967           

   76    1.456848         1.597487     0.911962           

   77    1.942463         1.599772     1.214212     -true-

   78    1.294976         1.602041     0.808328           

   79    1.726634         1.604276     1.076270     -true-

   80    1.151089         1.606495     0.716522           

   81    1.534786         1.608680     0.954065           

   82    1.023191         1.610851     0.635186            

   83    1.364254         1.612991     0.845792           

   84    1.819006         1.615116     1.126238     -true-

   85    1.212670         1.617211     0.749853           

   86    1.616894         1.619292     0.998519           

   87    1.077929         1.621344     0.664837           

   88    1.437239         1.623384     0.885335           

   89    1.916319         1.625395     1.178986     -true-

   90    1.277546         1.627395     0.785025           

   91    1.703394         1.629367     1.045433     -true-

   92    1.135596         1.631328     0.696118           

   93    1.514128         1.633263     0.927057           

   94    1.009419         1.635187     0.617311           

   95    1.345892         1.637086     0.822126           

   96    1.794522         1.638974     1.094906     -true-

   97    1.196348         1.640839     0.729108           

   98    1.595131         1.642693     0.971046           

   99    1.063421         1.644524     0.646643           

  100    1.417894         1.646345     0.861237           

  101    1.890526         1.648145     1.147063     -true-

  102    1.260350         1.649934     0.763879           

  103    1.680467         1.651703     1.017415     -true-

  104    1.120311         1.653462     0.677555           

  105    1.493749         1.655200     0.902458            

  106    1.991665         1.656930     1.202021     -true-

  107    1.327777         1.658640     0.800521           

  108    1.770369         1.660341     1.066268     -true-

  109    1.180246         1.662023     0.710126           

  110    1.573661         1.663697     0.945882           

  111    1.049107         1.665352     0.629961           

  112    1.398810         1.667000     0.839118           

  113    1.865080         1.668629     1.117732     -true-

  114    1.243387         1.670251     0.744431            

  115    1.657849         1.671855     0.991622           

  116    1.105233         1.673452     0.660451           

  117    1.473643         1.675032     0.879770           

  118    1.964858         1.676605     1.171927     -true-

  119    1.309905         1.678162     0.780560           

  120    1.746540         1.679711     1.039786     -true-

  121    1.164360         1.681245     0.692558           

  122    1.552480         1.682772     0.922573           

  123    1.034987         1.684284     0.614497           

  124    1.379982         1.685789     0.818597           

  125    1.839977         1.687280     1.090499     -true-

  126    1.226651         1.688764     0.726360           

  127    1.635535         1.690234     0.967638           

  128    1.090356         1.691697     0.644534           

  129    1.453809         1.693147     0.858643           

  130    1.938412         1.694590     1.143882     -true-

  131    1.292274         1.696020     0.761945           

  132    1.723032         1.697444     1.015075     -true-

  133    1.148688         1.698855     0.676154           

  134    1.531584         1.700260     0.900794           

  135    1.021056         1.701653     0.600038           

  136    1.361408         1.703040     0.799399           

  137    1.815211         1.704414     1.065006     -true-

  138    1.210141         1.705783     0.709434           

  139    1.613521         1.707140     0.945160           

  140    1.075681         1.708492     0.629608           

  141    1.434241         1.709832     0.838820           

  142    1.912321         1.711167     1.117554     -true-

  143    1.274881         1.712490     0.744460           

  144    1.699841         1.713808     0.991850           

  145    1.133227         1.715116     0.660729           

  146    1.510970         1.716418     0.880304            

  147    1.007313         1.717709     0.586428           

  148    1.343084         1.718996     0.781319           

  149    1.790779         1.720272     1.040986     -true-

  150    1.193853         1.721544     0.693478           

  151    1.591804         1.722805     0.923960           

  152    1.061202         1.724062     0.615525           

  153    1.414937         1.725308     0.820107           

  154    1.886582         1.726550     1.092689     -true-

  155    1.257721         1.727783     0.727940           

  156    1.676962         1.729011     0.969897           

  157    1.117975         1.730229     0.646143           

  158    1.490633         1.731443     0.860919           

  159    1.987510         1.732648     1.147094     -true-

  160    1.325007         1.733849     0.764200           

  161    1.766676         1.735041     1.018233     -true-

  162    1.177784         1.736228     0.678358           

  163    1.570379         1.737407     0.903863           

  164    1.046919         1.738582     0.602168           

  165    1.395892         1.739748     0.802353           

  166    1.861189         1.740910     1.069090     -true-

  167    1.240793         1.742063     0.712255           

  168    1.654391         1.743213     0.949047           

  169    1.102927         1.744355     0.632284           

  170    1.470569         1.745493     0.842495           

  171    1.960759         1.746623     1.122600     -true-

  172    1.307173         1.747749     0.747918           

  173    1.742897         1.748867     0.996586           

  174    1.161931         1.749982     0.663968           

  175    1.549242         1.751088     0.884731           

  176    1.032828         1.752192     0.589449           

  177    1.377104         1.753288     0.785441           

  178    1.836138         1.754380     1.046602     -true-

  179    1.224092         1.755465     0.697304           

  180    1.632123         1.756547     0.929166           

  181    1.088082         1.757621     0.619065           

  182    1.450776         1.758692     0.824918           

  183    1.934368         1.759756     1.099225     -true-

  184    1.289579         1.760817     0.732375           

  185    1.719438         1.761870     0.975916           

  186    1.146292         1.762921     0.650223           

  187    1.528390         1.763965     0.866451           

  188    1.018926         1.765005     0.577294           

  189    1.358569         1.766039     0.769274           

  190    1.811425         1.767070     1.025100     -true-

  191    1.207616         1.768095     0.683004           

  192    1.610155         1.769116     0.910147           

  193    1.073437         1.770131     0.606417           

  194    1.431249         1.771143     0.808093           

  195    1.908332         1.772149     1.076846     -true-

  196    1.272221         1.773152     0.717492           

  197    1.696295         1.774149     0.956118           

  198    1.130864         1.775143     0.637055           

  199    1.507818         1.776130     0.848934           

  200    1.005212         1.777116     0.565642           

 

Again it is to say, that already loops smaller than N=250000 (collatz-transformations) are disproven, since all numbers < 2^40 are excluded.

If we would start with such a number as smallest entry for a, we could get such a table:

 

 

    n    ug               prod          ug/prod    ug > prod:
                                                   loop cannot exist

------------------------------------------------------------

    1    1.333333         1.000000     1.333333     -true-

    2    1.777778         1.000000     1.777778     -true-

    3    1.185185         1.000000     1.185185     -true-

    4    1.580247         1.000000     1.580247     -true-

    5    1.053498         1.000000     1.053498     -true-

    6    1.404664         1.000000     1.404664     -true-

    7    1.872885         1.000000     1.872885     -true-

    8    1.248590         1.000000     1.248590     -true-

    9    1.664787         1.000000     1.664787     -true-

   10    1.109858         1.000000     1.109858     -true-

   11    1.479811         1.000000     1.479811     -true-

   12    1.973081         1.000000     1.973081     -true-

   13    1.315387         1.000000     1.315387     -true-

   14    1.753850         1.000000     1.753850     -true-

   15    1.169233         1.000000     1.169233     -true-

   16    1.558977         1.000000     1.558977     -true-

   17    1.039318         1.000000     1.039318     -true-

   18    1.385758         1.000000     1.385758     -true-

   19    1.847677         1.000000     1.847677     -true-

   20    1.231785         1.000000     1.231785     -true-

...

10000    1.296833     1.000e+0000     1.296833     -true-

10001    1.729111     1.000e+0000     1.729111     -true-

10002    1.152741     1.000e+0000     1.152741     -true-

10003    1.536987     1.000e+0000     1.536987     -true-

10004    1.024658     1.000e+0000     1.024658     -true-

10005    1.366211     1.000e+0000     1.366211     -true-

10006    1.821615     1.000e+0000     1.821615     -true-

10007    1.214410     1.000e+0000     1.214410     -true-

10008    1.619213     1.000e+0000     1.619213     -true-

10009    1.079475     1.000e+0000     1.079475     -true-

10010    1.439300     1.000e+0000     1.439300     -true-

10011    1.919067     1.000e+0000     1.919067     -true-

10012    1.279378     1.000e+0000     1.279378     -true-

10013    1.705838     1.000e+0000     1.705838     -true-

10014    1.137225     1.000e+0000     1.137225     -true-

10015    1.516300     1.000e+0000     1.516300     -true-

10016    1.010867     1.000e+0000     1.010867     -true-

10017    1.347822     1.000e+0000     1.347822     -true-

10018    1.797096     1.000e+0000     1.797096     -true-

10019    1.198064     1.000e+0000     1.198064     -true-

10020    1.597419     1.000e+0000     1.597419     -true-

 

 

This was computed with the multi-precision interpreter Aribas with 1000-digits float-precision; but checks up to lengthes of N>250000 seemed to be too time consuming for this program.

 

The above critical inequality shows, that we have N factors of the form

(3 + 1/a)

where a is assumed the minimal number. The maximum value for a member x is not known; if we set it to an astronomical optimistic value of a+N*3/2 (considering N in the area of 2^40) then we have

powerceil2(3^N)        1              1
--------------- <=(3+ ---- )(...)(3 + ----------)  / 3^N
     3^N               2^40           2^40+N*3/2

 

I don't have a function for bounds of the lhs depending on N for such high N at hand, but from some empirical calculations I have

 

             N

       N   powerceil2(3)/3^N

       1   1 + 3.3333333e-0001

       3   1 + 1.8518519e-0001

       5   1 + 5.3497942e-0002

      17   1 + 3.9318248e-0002

      29   1 + 2.5329408e-0002

      41   1 + 1.1528852e-0002

      94   1 + 9.4188494e-0003

     147   1 + 7.3132484e-0003

     200   1 + 5.2120395e-0003

     253   1 + 3.1152137e-0003

     306   1 + 1.0227618e-0003

     971   1 + 9.7906399e-0004

    1636   1 + 9.3536809e-0004

    2301   1 + 8.9167411e-0004

    2966   1 + 8.4798202e-0004

    3631   1 + 8.0429185e-0004

    4296   1 + 7.6060358e-0004

    4961   1 + 7.1691722e-0004

    5626   1 + 6.7323277e-0004

    6291   1 + 6.2955022e-0004

    6956   1 + 5.8586958e-0004

    7621   1 + 5.4219085e-0004

    8286   1 + 4.9851402e-0004

    8951   1 + 4.5483910e-0004

    9616   1 + 4.1116609e-0004

   10281   1 + 3.6749498e-0004

   10946   1 + 3.2382578e-0004

   11611   1 + 2.8015849e-0004

   12276   1 + 2.3649310e-0004

   12941   1 + 1.9282962e-0004

   13606   1 + 1.4916804e-0004

   14271   1 + 1.0550838e-0004

   14936   1 + 6.1850612e-0005

   15601   1 + 1.8194754e-0005

   47468   1 + 1.0929714e-0005

   79335   1 + 3.6647274e-0006

  190537   1 + 6.4507505e-0008

 

 

and I don't think, that an approximation to that of the rhs side, where "a" was assumed to be minimum 2^40 (and the minimal member of the loop) can be reached when n=250000, but with a much larger N. That means:

 

the given critical formula seems to give much better restrictions than the currently used formulas and suggest, that with the exclusion of numbers a, where 1..a..2^40 the lengthes for excluded general loops are far higher than 250000.

 

 

 

 

 

 


                                                                                                                                last update: 15.8.2004