Fw: why the mantissas used below are all roots of powers of 10
- To: mathgroup at smc.vnet.net
- Subject: [mg131887] Fw: why the mantissas used below are all roots of powers of 10
- From: Marvin Burns <marvin at marvinrayburns.com>
- Date: Thu, 24 Oct 2013 23:48:10 -0400 (EDT)
- Delivered-to: l-mathgroup@mail-archive0.wolfram.com
- Delivered-to: l-mathgroup@wolfram.com
- Delivered-to: mathgroup-outx@smc.vnet.net
- Delivered-to: mathgroup-newsendx@smc.vnet.net
- References: <1382136837.63494.YahooMailNeo@web2801.biz.mail.ne1.yahoo.com>
- Reply-to: Marvin Burns <marvin at marvinrayburns.com>
We get nearly identical answers without the sine function
Try
Table[{x, (10 MantissaExponent[N[10^(-100 - 1/x)], 10][[1]])^x}, {x,1,10}] // TableForm
replacing the 1/x with 2/x,3/x and 4/x, etc.
I think that can be explained! Anyone care to do it?
----- Forwarded Message -----
From: Marvin Burns <marvin at marvinrayburns.com>
To: "mathgroup at smc.vnet.net" <mathgroup at smc.vnet.net>
Sent: Friday, October 18, 2013 6:53 PM
Subject: [mg131887] why the mantissas used below are all roots of powers of 10
I sent this in yesterday but only part of the message was posted.
From Marvin Ray Burns
to mathgroup
We will use the Mathematica code
Table[{x, (10 MantissaExponent[N[Sin[10^(-100 - 1/x)], 10]][[1]])^ x}, {x, 1, 10}] // TableForm
I was playing around with small sines and posted this on another blog and didn't get much response. However, I would like to know why the mantissas used below are all roots of powers of 10. .
There seems to be patterns for sin(10^-k) for rational k;
Here we have the "floats."
n sin(10^(-n-1/2))
1 0.03161750640
2 0.003162272390
3 0.0003162277607
4 0.00003162277660
5 0.000003162277660
6 0.0000003162277660
7 0.00000003162277660
Basically the mantissa of 316227766 is being "floated out," because for small x sin(x)~=x.
Here we use the mantissa:
Noticing that 3.162277660^2~=10 we have below a more subtle and beautiful pattern for sin(10^-k), using sufficiently large integral value for k. Here we use 100 but 9 is usually sufficient.
x (10 Mantissa[sin(10^(-100 - 1/x))])^x
1 1.* 10^1
2 1.* 10^1
3 1.* 10^2
4 1.* 10^3
5 1.* 10^4
6 1.* 10^5
etc.
x (10 Mantissa[sin(10^(-100 - 2/x))])^x
1 1.* 10^0
2 1.* 10^2
3 1.* 10^1
4 1.* 10^2
5 1.* 10^3
6 1.* 10^4
etc.
x (10 Mantissa[sin(10^(-100 - 3/x))])^x
1 1.* 10^1
2 1.* 10^1
3 1.* 10^3
4 1.* 10^1
5 1.* 10^2
6 1.* 10^3
etc.
x (10 Mantissa[sin(10^(-100 - 4/x))])^x
1 1.* 10^1
2 1.* 10^0
3 1.* 10^2
4 1.* 10^4
5 1.* 10^1
6 1.* 10^2
7 1.* 10^3
etc.
x (10 Mantissa[sin(10^(-100 - 5/x))])^x
1 1.* 10^1
2 1.* 10^1
3 1.* 10^1
4 1.* 10^3
5 1.* 10^4
6 1.* 10^1
7 1.* 10^2
8 1.* 10^3
etc.
The Mathematica code for this is
Table[{x, (10 MantissaExponent[N[Sin[10^(-100 - 1/x)], 10]][[1]])^ x}, {x, 1, 10}] // TableForm
Change 1/x to 2/x,3/x, etc .
Can anyone figure out the pattern here?
Replacing 1/x with (3/2)/x and ^x to ^(2x) we find
x (10 Mantissa[sin(10^(-100 - (3/2)/x))])^(2x)
1 1.* 10^1
2 1.* 10^1
3 1.* 10^3
4 1.* 10^5
5 1.* 10^7
6 1.* 10^9
etc.
Replacing 1/x with (5/2)/x and ^x to ^(2x) we find
x (10 Mantissa[sin(10^(-100 - (5/2)/x))])^(2x)
1 1.* 10^1
2 1.* 10^3
3 1.* 10^1
4 1.* 10^3
5 1.* 10^5
6 1.* 10^7
etc.
Replacing 1/x with (5/3)/x and ^x to ^(3x) we find
x (10 Mantissa[sin(10^(-100 - (5/3)/x))])^(3x)
1 1.* 10^1
2 1.* 10^1
3 1.* 10^4
4 1.* 10^7
5 1.* 10^10
6 1.* 10^12
etc.