Don Lancaster
Guest
Mon Aug 30, 2010 7:39 pm
.... at <
http://www.tinaja.com/glib/gjstab01.pdf >
It is on avoiding blowups when doing Gauss Jordan solution of linear nxn
equations.
Sourcecode is separately available at <
http://www.tinaja.com/glib/gjstab01.psl >
Additional GuruGrams at <
http://www.tinaja.com/gurgrm01.asp >
--
Many thanks,
Don Lancaster voice phone: (928)428-4073
Synergetics 3860 West First Street Box 809 Thatcher, AZ 85552
rss:
http://www.tinaja.com/whtnu.xml email: don_at_tinaja.com
Please visit my GURU's LAIR web site at
http://www.tinaja.com
Tim Williams
Guest
Mon Aug 30, 2010 7:57 pm
"Don Lancaster" <don_at_tinaja.com> wrote in message
news:8e2c75F5ppU1_at_mid.individual.net...
Quote:
Easy.
Open MATLAB, or if you prefer, GNU Octave,
block = [data];
y = rref(block);
For e.g. a 3x3 inhomogeneous system Ax = b, specify block = [A1, A2, A3,
b1; ...], a 4x3 matrix. Result (if nonsingular) is the identify matrix
with the rightmost column equal to the result (i.e., x1 + 0 + 0 = y1,
....).
Tim
--
Deep Friar: a very philosophical monk.
Website:
http://webpages.charter.net/dawill/tmoranwms
Don Lancaster
Guest
Tue Aug 31, 2010 2:17 am
On 8/30/2010 11:57 AM, Tim Williams wrote:
Quote:
"Don Lancaster"<don_at_tinaja.com> wrote in message
news:8e2c75F5ppU1_at_mid.individual.net...
... at<
http://www.tinaja.com/glib/gjstab01.pdf
It is on avoiding blowups when doing Gauss Jordan solution of linear nxn
equations.
Easy.
Open MATLAB, or if you prefer, GNU Octave,
block = [data];
y = rref(block);
For e.g. a 3x3 inhomogeneous system Ax = b, specify block = [A1, A2, A3,
b1; ...], a 4x3 matrix. Result (if nonsingular) is the identify matrix
with the rightmost column equal to the result (i.e., x1 + 0 + 0 = y1,
...).
Tim
And the results for a 65x65 set of linear equations are... ?
--
Many thanks,
Don Lancaster voice phone: (928)428-4073
Synergetics 3860 West First Street Box 809 Thatcher, AZ 85552
rss:
http://www.tinaja.com/whtnu.xml email: don_at_tinaja.com
Please visit my GURU's LAIR web site at
http://www.tinaja.com
Tim Williams
Guest
Tue Aug 31, 2010 9:29 am
"Don Lancaster" <don_at_tinaja.com> wrote in message
news:8e33gdF87tU1_at_mid.individual.net...
Quote:
And the results for a 65x65 set of linear equations are... ?
They are...whatever you get? It takes longer to enter than to process.
In that size, of course, you'd automate the process by loading a file,
which is another single function.
There's an awful lot in MATLAB, and it's all right there, a function away.
You might like it a lot better than plodding everything in PS. I bet you
could find the next 666 bit magic sinewave in a few minutes.
Tim
--
Deep Friar: a very philosophical monk.
Website:
http://webpages.charter.net/dawill/tmoranwms
Don Lancaster
Guest
Tue Aug 31, 2010 7:49 pm
On 8/31/2010 1:29 AM, Tim Williams wrote:
Quote:
There's an awful lot in MATLAB, and it's all right there, a function away.
You might like it a lot better than plodding everything in PS. I bet you
could find the next 666 bit magic sinewave in a few minutes.
Tim
A 666 bit magic sinewave is pretty much useless.
But it does have a "Let's piss off an Arizona politician" air about it.
Present magic sinewave studies and aps are at 44,000 bits.
There are A LOT of 44,000 bit words. Several even.
<
http://www.tinaja.com/demo28a.asp >
--
Many thanks,
Don Lancaster voice phone: (928)428-4073
Synergetics 3860 West First Street Box 809 Thatcher, AZ 85552
rss:
http://www.tinaja.com/whtnu.xml email: don_at_tinaja.com
Please visit my GURU's LAIR web site at
http://www.tinaja.com
