Concerning FAMP program
creigh at o2online.de
creigh at o2online.de
Fri Aug 27 12:04:07 CEST 2004
It is my opinion that "FAMP" (Floretion Algebra Multiplication Program)
should be extended, improved, and even regularly updated.
I feel that FAMP holds a cornucopia of beautiful sequences
(including many unknown ones).
Evidence has been accumulated that the program
get's interesting for large numbers (for examples, see below)
but my limited programming skills as well as the time I, a part time
English teacher and physics student, have available to put into the
project mean that much time will pass before this will ever be
investigated. Specifically, the Java class MultiplyFlorets (included in
Jantje.zip at http://www.crowdog.de -> The Floretions) should probably
make use of the BigDecimal.class (perhaps I've misspelled this)
available as part of the standard Java library, and not the way I did it,
below:
public double[] MultiplyFlorets() {
double z[] = new double[16]; // To hold result of multiplication
// 'i, 'j, 'k, i', j', k', 'ii', 'jj', 'kk', 'ij', 'ik', 'ji', 'jk',
'ki', 'kj', 1 x
// 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
15 x
// detemines 'i only
z[0] = ( x[0]*y[15] + x[1]*y[2] - x[2]*y[1] - x[3]*y[6] -
x[4]*y[9] - x[5]*y[10] -x[6]*y[3] - x[7]*y[14] + x[8]*y[12]
- x[9]*y[4] - x[10]*y[5] - x[11]*y[13] -x[12]*y[8] + x[13]*y[11]
+ x[14]*y[7] + x[15]*y[0] );
// 'i note: #+ = 6, #- = 10
I would be very happy to find a small group of interested people to
assist me with this. The claim given at
http://mathforum.org/discuss/sci.math/t/622432
provides one example of why the program should include larger numbers.
Here is another:
From the options menu, choose "Load Floret's Star".
Change B = .5(- 'i + 'j - 'kk' - 'ki' - 'kj' - 1)
(represented by the row to the left of the button "B") to
B = .5(- 'i + 'j - 'kk' - 'ki' - 'kj' - 3).
Now, enter x= 'i + 'j + i' + j' +'ii' + 'jj' + 'ji' + 'ij' + 1 by marking
"1's" in
the appropriate boxes next to any button which does not already have
entries filled in. Pressing button B and then the button for x gives Bx
and is displayed at the bottom of the screen. Finally, from the pull down
menu, click "vesseq" to give the sequence
a(n) = ves( (Ex)^n ) (= -17, 27, -117, 1971 ...). The sequences "tesseq",
"lesseq" and "jesseq"
are related to vesseq by vesseq = jesseq + lesseq + tesseq (termwise
addition). "lesseq" and "tesseq" must be multiplied by 2 to give an
integer sequence. I then went to http://www.btinternet.com/~se16/js/factor.
htm
to factorize the elements of these sequences and thought "jetzt
verstehe ich die Welt nicht mehr". The elements quickly
become to big to see exactly what's happening. It would also be great if
the program could do calculations with fractions without turning them
into a decimal first.
Sincerely,
Creighton
More information about the SeqFan
mailing list