Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

entering complex numbers in polar form (hp49g+)

24 views
Skip to first unread message

chell...@gmx.net

unread,
Nov 28, 2006, 11:52:28 AM11/28/06
to
Hello folks,

I've set my calc to Deg and cylin [ R<)Z ]. (Algebraic mode

If I enter a complex number like
(2, <)2 ) everything just works fine..

But is there a way to get something like
(2*2, <) 2) working? If I try typing this in my calc retruns an error
to me: "Invalid Syntax".

Is there a way to make it possible entering something like I mentioned
above?

(it is also not possible of entering a complex number with variables
e.g.: (m, <)2) does not work too)

Thank you very much

Veli-Pekka Nousiainen

unread,
Nov 28, 2006, 2:02:24 PM11/28/06
to

<chell...@gmx.net> wrote in message
news:1164732748....@14g2000cws.googlegroups.com...

You simply can't - using numeric complex expressions
BUT
using algebraic notation you can:
'x+y*i'
you may also set the flag
-27
if you want to see:
(X,Y)


John H Meyers

unread,
Nov 28, 2006, 4:05:31 PM11/28/06
to
On Tue, 28 Nov 2006 10:52:28 -0600:

> (2*2, <) 2)

'(2*2, \<)2)' \->NUM

> (m, <)2)

'(m, <)2)'

"You can get anything you want;
don't believe any 'no you cannot'"

[r->] [OFF]

Veli-Pekka Nousiainen

unread,
Nov 29, 2006, 8:56:12 PM11/29/06
to
"John H Meyers" <jhme...@nomail.invalid> wrote in message
news:op.tjqvj...@isdept23.ia.mum.edu...

> On Tue, 28 Nov 2006 10:52:28 -0600:
X
>> (m, <)2)
>
> '(m, <)2)'

With angle sign I get
'(m*COS(2),m*SIN(2))'
BUT
when I use EQW (must start with [m] [ , ] () comes then automagically) I get
'(m,2)' @ no angle
checking back the (X,Y) flag to show X+Y*i
reveales
'm+2*i' @ no angle

> "You can get anything you want;
> don't believe any 'no you cannot'"

What went wrong?


John H Meyers

unread,
Nov 29, 2006, 10:41:18 PM11/29/06
to
On Wed, 29 Nov 2006 19:56:12 -0600, for VPN:

The "angle symbol" is recognized (and parsed)
by the UserRPL text compiler (command line),
not by the 49/50 Equation Writer
(although the 48 Equation Writer understood it).

[r->] [OFF]

Veli-Pekka Nousiainen

unread,
Nov 30, 2006, 12:48:20 AM11/30/06
to

"John H Meyers" <jhme...@nomail.invalid> wrote in message
news:op.tjs8i...@isdept23.ia.mum.edu...

Repeat: command line ended up with COS and SIN parts


John H Meyers

unread,
Nov 30, 2006, 2:23:41 AM11/30/06
to
On Wed, 29 Nov 2006 23:48:20 -0600, Veli-Pekka Nousiainen wrote:

> command line ended up with COS and SIN parts

for input like '(m,\<)2)' [with "angle" symbol]
producing algebraic result '(m*COS(2),m*SIN(2))'

Isn't that obviously correct?

What result are you expecting, if not that?

For others, who don't yet have more than a decade's experience
(or time to reflect, which would undoubtedly result
in a sudden flash of insight anyway):

Note that the "angle symbol" is only a compiler syntax word,
which must be interpreted and changed into some other action;
it will not exist within any result object.

When the components are both pure [real] numbers,
then the compiler can produce a "complex number" object type
(internally performing the equivalent of CYLIN \->V2
to produce the final result from individual components,
which can again be decompiled using a reverse procedure),
but when any component is symbolic, the compiler can only
express the result as a symbolic complex-valued result,
with each component expressed symbolically.

And that's what it did, in the only form possible;
there is no other way it could have turned out.

This reminds me of famed musical teacher Nadia Boulanger,
who often required her students to follow through
the consequences of the rules of classical composition,
until they realized for themselves that certain works,
once begun in a certain way, had to inevitably proceed
to the perfect compositions that their composers
had already completed -- there was simply no other way
that they could have unfolded, just like that above.

Too bad that so few take the time to follow through
and finish any composition, or to answer their own questions,
which almost everyone certainly could,
if there weren't a newsgroup to turn to
for skipping the steps of personal discovery
(I thought that was a "TI thing" -- an "answer machine,"
rather than a machine to help develop thinking :)

http://en.wikipedia.org/wiki/Nadia_Boulanger
http://www.encyclopedia.com/doc/1E1-BoulangrN.html
http://www.ambache.co.uk/wBoulanger.htm
http://members.aol.com/aaocompose/boulanger.html
http://www.americansymphony.org/dialogues_extensions/97_98season/6th_concert/leon.cfm
http://www.nadiaboulanger.org/

[r->] [OFF]

Veli-Pekka Nousiainen

unread,
Nov 30, 2006, 4:22:35 AM11/30/06
to

"John H Meyers" <jhme...@nomail.invalid> wrote in message
news:op.tjqvj...@isdept23.ia.mum.edu...

What does THAT mean then?!


David Combs

unread,
Dec 11, 2006, 5:57:10 AM12/11/06
to
Silly, even ignorant question, but in the manuals
I notice no mention of "e to complex exponent" -- seen,
apparantely, throughout mathematics.

Like within the integral for fourier transforms --
neat, concise.

Does the 50g (et al) "know" this notation -- and if
not, when (presumably) will it?

Thanks,

David


John H Meyers

unread,
Dec 11, 2006, 12:20:07 PM12/11/06
to
On Mon, 11 Dec 2006 04:57:10 -0600, David Combs wrote:

> ... in the manuals


> I notice no mention of "e to complex exponent" -- seen,

> apparently, throughout mathematics.


>
> Like within the integral for fourier transforms --
> neat, concise.
>
> Does the 50g (et al) "know" this notation -- and if
> not, when (presumably) will it?

HP symbolic calcs know the 'Y^X' notation
(and the ^ command) generated by the [Y^X] key),
which handles all exponents, real or complex.

Sorry to hear this isn't in the manuals yet :)

[r->] [OFF]

John H Meyers

unread,
Dec 11, 2006, 1:18:42 PM12/11/06
to
... and of course 'EXP(x)' handles "e to a power"
for both real and complex exponent
(unless the CAS interferes,
which can be avoided by a flag setting)

If you mean that you'd like the *decompiler*
to display in some other notation, well,
you could always write a new decompiler :)

[r->] [OFF]

Veli-Pekka Nousiainen

unread,
Dec 12, 2006, 8:09:01 AM12/12/06
to
"John H Meyers" <jhme...@nomail.invalid> wrote in message
news:op.tkeqh...@isdept23.ia.mum.edu...

TI has that build into it's decompiler
eg. TI calcs understand exponential complex number format
parallel to standard (x,y) pair
that's why he asked (my educated guess)


0 new messages