34
   

The worlds first riddle!

 
 
Tryagain
 
  1  
Reply Sat 22 Mar, 2008 05:58 pm
The only reason I post this is because it looks impossible! I strongly suggest you don't drive yourself mad by attempting it; I will post the answer tomorrow and you will see why! Twisted Evil

Integral calculus and trigonometry required.

What is the mean distance between two random points in a unit square Question
0 Replies
 
Izzie
 
  1  
Reply Sat 22 Mar, 2008 06:10 pm
Look forward to seeing you tomorrow a? Laughing
0 Replies
 
markr
 
  1  
Reply Sat 22 Mar, 2008 06:34 pm
Tryagain wrote:
interestingly enough each of the denominators of the fractions is prime


Not true: 1807 = 13 * 139
0 Replies
 
Tryagain
 
  1  
Reply Sat 22 Mar, 2008 06:47 pm
Interestingly enough I was quoting Rap! Although, I must confess I agreed with the way he said it! :wink:
0 Replies
 
markr
 
  1  
Reply Sat 22 Mar, 2008 07:29 pm
Sorry. If they were all prime, we could generate arbitrarily large primes at will.

Can you hold off on posting the answer to the random points in a square question? I'm taking a shot at it. Thanks.
0 Replies
 
raprap
 
  1  
Reply Sat 22 Mar, 2008 08:04 pm
mea culpa, on the primes. It is not the first time, I've previously done the same with Mersennes.

Rap
0 Replies
 
Izzie
 
  1  
Reply Sat 22 Mar, 2008 08:16 pm
((rap))
0 Replies
 
markr
 
  1  
Reply Sat 22 Mar, 2008 11:12 pm
I give up on the random points in a square problem. Assuming I set it up properly, the integrals I get are too much for me. I wrote a simulation and get something a bit greater than .521.
0 Replies
 
raprap
 
  1  
Reply Sun 23 Mar, 2008 05:22 am
Like markr I wrote a simple routine. After > 50,000 iterations my routine gets somewhat greater than 0.521.

Rap
0 Replies
 
Tryagain
 
  1  
Reply Sun 23 Mar, 2008 07:55 am
Modern-day Easter is derived from two ancient traditions: one Judeo-Christian and the other Pagan. Both Christians and Pagans have celebrated death and resurrection themes following the Spring Equinox for millennia. Most religious historians believe that many elements of the Christian observance of Easter were derived from earlier Pagan celebrations.

The equinox occurs each year on March 20, 21 or 22. Both Neopagans and Christians continue to celebrate religious rituals linked to the equinox in the present day. Wiccans and other Neopagans usually hold their celebrations on the day or eve of the equinox. Western Christians wait until the Sunday on or after the next full moon. The Eastern Orthodox churches follow a different calculation; their celebration is often many weeks after the date selected by the Western churches.

Therefore to you all on this Easter Sunday (except you wiccans; because that wicker chair you sold me collapsed) a very happy time too you.



Mark:

I wrote a simulation and get something a bit greater than .521. Cool Cool Cool


Rap:

I wrote a simple routine. After > 50,000 iterations my routine gets somewhat greater than 0.521. Cool Cool Cool



Dagnabit! The hardest question in the world; and you both come up with the correct answer. And Rap got a sympathy ((hug))! Happy Easter dudes… Razz

The answer is [2^1/2 + 2 + 5*ln(1+2^1/2)] / 15 =~ 0.521405433

And this is why…

First, let I(x,y) stand for "the integral from x to y of ...", and sqr stand for "the square root of...".

To find the answer we must solve I(0,1) I(0,1) I(0,1) I(0,1) sqr((x-a)^2+(y-b)^2) db da dy dx.

Even the first integration of this is very messy, requiring a big table of integrals, and I think at best a small book of integrals would only take you through the second one. So we need to be more creative to make the problem more manageable. First we introduce a change of variables:

• u=x-a
• s=x
• v=y-b
• t=y

Or:

• x=s
• y=t
• a=s-u
• b=t-v

(1): Now we have I(0,1) I(t,t-1) I(0,1) I(s,s-1) sqr(u^2 + v^2) * |J| du ds dv dt

Where |J| = the Jacobian =
| db/du db/ds db/dv du/dt | | 0 0 -1 1 |
| | | |
| da/du da/ds da/dv da/dt | | -1 1 0 0 |
| | = | | = -1
| dy/du dy/ds dy/dv dy/dt | | 0 0 0 1 |
| | | |
| dx/du dx/ds dx/dv dx/dt | | 0 1 0 0 |

Now let's solve the following inner two integrals of equation (1):
(2): I(0,1) I(s,s-1) sqr(u^2 + v^2) du ds.

The area of integration is bounded by the quadrilateral with vertices (0,0), (1,1), (0,1), and (-1,0), with s being the horizontal axis and u being the vertical.

Changing the order of integral we get:
I(-1,0) I(0,u+1) sqr(u^2 + v^2) ds du + I(0,1) I(u,1) sqr(u^2 + v^2) ds du =
I(1,0) sqr(u^2 + v^2) * (1+u) du + I(0,1) sqr(u^2 + v^2) * (1-u) du =
-I(1,0) sqr(p^2 + v^2) * (1-p) dp + I(0,1) sqr(u^2 + v^2) * (1-u) du (letting p=-u in the first integral) =
I(0,1) sqr(p^2 + v^2) * (1-p) dp + I(0,1) sqr(u^2 + v^2) * (1-u) du (Notice the two terms are equal)=
2* I(0,1) sqr(u^2 + v^2) * (1-u) du.

Now put this into equation (1), and moving the constant terms to the left:
-2* I(0,1) I(t,t-1) I(0,1) sqr(u^2 + v^2) * (1-u) du dv dt =
-2* I(0,1) I(0,1) I(t,t-1) sqr(u^2 + v^2) * (1-u) dv dt du (changing the order of integration)=
-2* I(0,1) (1-u) I(0,1) I(t,t-1) sqr(u^2 + v^2) dv dt du =
Note that the inner two integral are of the same form as equation (2), thus:

4* I(0,1) I(0,1) sqr(u^2 + v^2) * (1-u) * (1-v) dv du =
4* I(0,1) I(0,1) sqr(u^2 + v^2) * (1-u) * (1-v) du dv (changing the order of integration)

Next we make another substitution:
• u=r*sin(w)
• v=r*cos(w)

This time the Jacobian shall equal:

| du/dr du/dw | | sin(w) r*cos(w) |
| | = | | = -r
| dv/dr dv/dw | | cos(w) -r*sin(w) |
Thus the integral as a function of w and r is:
-8 * I(0,pi/4) I(0,1/cos(w)) -r * sqr(r*sin2(w) + r*cos2(w)) * (1-r*cos(w)) * (1-r*sin(w)) dr dw.

Note that we halved the area of integration, going from 0 to pi/4 as opposed to pi/2, and multiplying by 8 instead of the 4. This is permissible because of the symetry of the two regions.

= 8 * I(0,pi/4) I(0,1/cos(w)) r^2 * (1-r*cos(w)) * (1-r*sin(w)) dr dw.
= 8 * I(0,pi/4) I(0,1/cos(w)) r^2 - r^3*(cos(w)+sin(w)) + r^4*(cos(w)+sin(w)) dr dw.
= 8 * I(0,pi/4) r^3/3 - r^4*(cos(w)+sin(w))/4 + r^5*(cos(w)+sin(w))/5 ( from 0 to 1/cos(w) ) dw.
= 8 * I(0,pi/4) (3*cos^3(w))-^1 - (cos(w)+sin(w))/(4*cos^4) + (cos(w)+sin(w))/(5*cos5(w)) dw.
= 8 * I(0,pi/4) (3*cos^3(w))-^1 - (4*cos^3(w))-^1 - sin(w)*(4*cos^4(w))-^1 + sin(w)*(5*cos4(w))-^1 dw.
= 8 * I(0,pi/4) (12*cos^3(w))-^1 - sin(w)*(20*cos^4(w))-^1 dw.
(3) = 2/15 * I(0,pi/4) 5/cos^3(w) - 3*sin(w)/cos^4(w) dw.

Next we must dust off our table of integrals to help us with the integral of cos-^3(w).

I dx/cos^n(x) = (1/(n-1)) * sin(x)/cos^n-1(x) + ((n-2)/(n-1)) * I dx/cos^n-2(x).

In our case n=3 so, I dx/cos^3(x) = (1/2) * sin(x)/cos^2(x) + (1/2) * I sec(x) dx.

Now lets look up the integral of sec(x):

I sec(x) dx = (1/2) * ln((1+sin(x))/(1-sin(x))).

Now we are ready to integrate cos-^3(w):

I(0,pi/4) cos-^3(w) dw = sin(w)/(2*cos^2(w)) (from 0 to pi/4) + (1/2)*I(0,pi/4) sec(x) dx.
= (1/2)*(2-^1/2/2-^1 - 0) + (1/4)*ln((2+sqr(2))/(2-sqr(2))).
= (1/2)*sqr(2) + (1/4)*ln(((2+sqr(2))/2+sqr(2))/((2-sqr(2))*(2+sqr(2))).
= 1/sqr(2) + (1/4)*ln((4+4*sqr(2)+2)/(4-2)).
= 1/sqr(2) + (1/4)*ln(3+2*sqr(2)).
= 1/sqr(2) + (1/2)*ln(3+2*sqr(2))^1/2.
= 1/sqr(2) + (1/2)*ln(1+sqr(2)).

Now lets integrate sin(w)/cos^4(w).

let m=cos(w), thus dm=-sin(w) dw.
I(0,pi/4) sin(w)/cos^4(w).
= I(1,1/sqr(2)) -1*m^-4 dm.
= 1/(3*m^3) from 0 to 1/sqr(2).
= (1/3) * (2*sqr(2) - 1).

Not it is time to plus these integrals into equation (3):
(2/15) * [ 5*(1/sqr(2) + (1/2)*ln(1+sqr(2))) + 3*((1/3) * (2*sqr(2) - 1)) ]
= (2/15) * [ 5*sqr(2)/2 + (5/2)*ln(1+sqr(2)) - 4*sqr(2)/2 + 1 ]
= (2/15) * [ sqr(2)/2 + (5/2)*ln(1+sqr(2)) + 1 ]
= (sqr(2) + 2 + 5*ln(1+sqr(2))) / 15.

Easy freeking peasy! Laughing Laughing Laughing
0 Replies
 
Tryagain
 
  1  
Reply Sun 23 Mar, 2008 06:20 pm
Ok! This is a good one…

You have a circular piece of paper with radius 1 and wish to form a cone shaped cup.

You may cut out any wedge you like from the paper, call the angle of the wedge x.

The point of the wedge must be at the center of the circle. After cutting out the wedge you then attach the two straight edges remaining to form a cone.

Assuming the paper cone can hold water, what should x be to maximize the water holding capacity of the cone Question
0 Replies
 
raprap
 
  1  
Reply Sun 23 Mar, 2008 08:16 pm
Wedge cutting

[size=7]Note: Initially all angles are in radians
pi is pi
Volume of cone is V=(1/3)*h*Ab where h is height and Ab is the area of the base
Both are going to be functions of the wedge angle x
Crcumference (Cb)and radius (Rb) of the base is determined from
Cb=(2pi-x)*1=2pi*Rb
Rb=(2*pi-x)/(2*pi)
Ab=pi*Rb^2=pi*[(2*pi-x)/(2*pi)}^2
Now to determine height (h) of the cone
h^2+Rb^2=1
h^2=1-Rb^2=1-[(2*pi-x)/(2*pi)]^2=x(4*pi-x)/(2*pi)^2
h= sqrt[x(4*pi-x)]/(2*pi)
now we got the formula for the volume of the cone in terms of x (the wedge angle)
V=(1/3)*h*Ab=(1/3)* pi*[(2*pi-x)/(2*pi)}^2* sqrt[x(4*pi-x)]/(2*pi)
V=(1/12*pi^2)*(2*pi-x)^2*sqrt[x(4*pi-x0]
Now differentiate V with respect to x and set equal to 0
dV/dx=0
going through a half page of algebra
-2x(4*pi-x)+(2*pi-x)^2=0
5*x^2-12*pi*x+4*pi^2=0
using the good old quadratic equation
x=[12*pi+/-sqrt(144*pi^2-80*pi^2)]/10=(12*pi+/-8*pi)/20
x=20Pi/10=2*pi or x=4*pi/20=0.2*pi are 0's on dV/dx
x=2*pi=360 degrees which is a complete removal and the cone holds nothing, so this is a minimum
x-.2*pi=72 degrees which results in a maximum

So the maximum wedge is 72 degrees or 1/5th of the circle
[/size]

Rap
0 Replies
 
markr
 
  1  
Reply Sun 23 Mar, 2008 11:48 pm
CONE
[size=7]Using the same method as rap, but with degrees instead of radians, I get 360*(1-sqrt(2/3)) degrees, which is approximately 66.06123 degrees. If you pour the contents of my cup into rap's, his cup runneth over (it's too small) by approximately 0.00094 cubic units. :wink: [/size]
0 Replies
 
Tryagain
 
  1  
Reply Mon 24 Mar, 2008 10:03 am
Rap:

Note: Initially all angles are in radians
pi is pi
Volume of cone is V=(1/3)*h*Ab where h is height and Ab is the area of the base
Both are going to be functions of the wedge angle x
Crcumference (Cb)and radius (Rb) of the base is determined from
Cb=(2pi-x)*1=2pi*Rb
Rb=(2*pi-x)/(2*pi)
Ab=pi*Rb^2=pi*[(2*pi-x)/(2*pi)}^2
Now to determine height (h) of the cone
h^2+Rb^2=1
h^2=1-Rb^2=1-[(2*pi-x)/(2*pi)]^2=x(4*pi-x)/(2*pi)^2
h= sqrt[x(4*pi-x)]/(2*pi)
now we got the formula for the volume of the cone in terms of x (the wedge angle)
V=(1/3)*h*Ab=(1/3)* pi*[(2*pi-x)/(2*pi)}^2* sqrt[x(4*pi-x)]/(2*pi)
V=(1/12*pi^2)*(2*pi-x)^2*sqrt[x(4*pi-x0]
Now differentiate V with respect to x and set equal to 0
dV/dx=0
going through a half page of algebra
-2x(4*pi-x)+(2*pi-x)^2=0
5*x^2-12*pi*x+4*pi^2=0
using the good old quadratic equation
x=[12*pi+/-sqrt(144*pi^2-80*pi^2)]/10=(12*pi+/-8*pi)/20
x=20Pi/10=2*pi or x=4*pi/20=0.2*pi are 0's on dV/dx
x=2*pi=360 degrees which is a complete removal and the cone holds nothing, so this is a minimum
x-.2*pi=72 degrees which results in a maximum

So the maximum wedge is 72 degrees or 1/5th of the circle

Sorry Rap, I have gotta go with Mark on this one; but I did enjoy your way of thinking!


Mark:

CONE
Using the same method as rap, but with degrees instead of radians, I get 360*(1-sqrt(2/3)) degrees, which is approximately 66.06123 degrees. Cool Cool

If you pour the contents of my cup into rap's, his cup runneth over (it's too small) by approximately 0.00094 cubic units Cool



X = 360*(1-(2/3)^1/2) = ~ 66 degrees.

The radius of the base of the cone is (360-x)/360.
The height of the cone is (1-((360-x)/360)^2)^1/2.
The area of the cone is 1/3*pi*radius^2*height.
Let u=((360-x)/360)^2.
The area becomes: pi/3 * (1-u)^1/2 * u.
Take the deriviate and set equal to 0: u=2/3.
If u=2/3 then x=360*(1-(2/3)^1/2).
Thus the radius = (2/3)^1/2.
Thus the height = 3-^1/2.
The area of the cone is (2*pi)/(9*3^1/2) = ~ .403.
0 Replies
 
raprap
 
  1  
Reply Mon 24 Mar, 2008 10:31 am
Try

I think my logic was correct, the error was in the execution. Ah well--the condemned was hung by arithmetic.

Rap
0 Replies
 
Tryagain
 
  1  
Reply Mon 24 Mar, 2008 06:16 pm
Rap wrote, "I think my logic was correct, the error was in the execution. Ah well--the condemned was hung by arithmetic."

I agree; I blame Mark, if he had gone with your thinking, I would have surrendered! Laughing


Suppose you have a coin in which the probability of flipping a heads is p, where p>=0.5

What is the expected number of flips it will take for the number of heads to equal the number of tails, assuming the first flip is a tails Question
0 Replies
 
markr
 
  1  
Reply Mon 24 Mar, 2008 07:27 pm
FLIPS
[size=7]1 + SUM[i=0 to infinity, C(i) * (1-p)^i * p^(i+1) * (2i+1)]
Where C(i) is the ith Catalan Number. For purposes of this formula, define C(0) = 1 and 0^0 = 1.

This includes the first flip (the 1 before the sum).

I don't know how to put this in closed form, but after a little experimentation with Excel, I came up with this little gem that seems to work:
1 + [1 / (2p - 1)] (which includes the initial flip)
[/size]
0 Replies
 
Izzie
 
  1  
Reply Mon 24 Mar, 2008 08:26 pm
raprap wrote:
Try
I think my logic was correct, the error was in the execution. Ah well--the condemned was hung by arithmetic.
Rap


aw heck, I'm gonna hug (Rap) again. (I know, get off the thread girl and hug somewhere else Rolling Eyes )
0 Replies
 
Tryagain
 
  1  
Reply Tue 25 Mar, 2008 06:37 pm
Mark:

FLIPS

1 + SUM[i=0 to infinity, C(i) * (1-p)^i * p^(i+1) * (2i+1)]
Where C(i) is the ith Catalan Number. For purposes of this formula, define C(0) = 1 and 0^0 = 1.

This includes the first flip (the 1 before the sum).

I don't know how to put this in closed form, but after a little experimentation with Excel, I came up with this little gem that seems to work:

1 + [1 / (2p - 1)] (which includes the initial flip) Cool Cool Razz


I am delighted we are in accord.

Think of the problem as a random walk where the probability is p of moving to the left and 1-p of moving to the right. The expected change in the position on the number line per move is p*(+1) + (1-p)*(-1) = 2p-1. The first flip resulted in moving one space to the right. The expected number of moves to move one space to the left is the inverse of the expected movement, which equals 1/(2p-1).

Is there anything that guy cannot do with Excel! Laughing



Lzzie wrote, "aw heck, I'm gonna hug (Rap) again."

What have you done with him; he's gone missing? Shocked



Given an ordinary deck of cards, what is the probability of drawing each kind of hand in poker Question

No discards, just the first five cards. Remember, aces can be high or low.
0 Replies
 
Tryagain
 
  1  
Reply Wed 26 Mar, 2008 09:15 am
Well here I sit, gettin' ideas
Ain't nothing but a fool would live like this
Out all night and runnin' wild
Woman sittin' home with a month old child

CHORUS:
Dang me, dang me
They oughta take a rope and hang me
High from the highest tree
Woman would you weep for me.

Just sittin' around drinkin' with the rest of the guys
Six rounds bought, and I bought five
Spent the groceries and half the rent
Like fourteen dollars and twenty seven cents.

(CHORUS)

They say roses are red and violets are purple
Sugar is sweet and so is maple surple
Well I'm the seventh out of seven sons
My pappy was a pistol
I'm a son of a gun.

Dang me, dang me
They oughta take a rope and hang me
High from the highest tree
Woman would you weep for me.




ECSHE Question

RICOONTTAS Question
0 Replies
 
 

Related Topics

Alternative Einstein's riddle answer - Discussion by cedor
Urgent !!! Puzzle / Riddle...Plz helpp - Question by zuzusheryl
Bottle - Question by Megha
"The World's Hardest Riddle" - Discussion by maxlovesmarie
Hard Riddle - Question by retsgned
Riddle Time - Question by Teddy Isaiah
riddle me this (easy) - Question by gree012
Riddle - Question by georgio7
Trick Question I think! - Question by sophocles
Answer my riddle - Question by DanDMan52
 
Copyright © 2025 MadLab, LLC :: Terms of Service :: Privacy Policy :: Page generated in 0.1 seconds on 01/12/2025 at 09:33:16