HP 50g help
netsez
the HP-41C, and now the HP-50g.
I just bought it Friday and I have some nagging issues with it. This
is my first CAS calculator. First thing I did was to put the
calculator in DEG and numeric. Things were going perfectly until I
tried to enter lim(x+1,x=1) The calculator responded with
Error: Undefined Name
This error is not mentioned anywhere in the 800 page user guide! I
tried reentering it many times, tried making some changes, but same
error every time. Finally I thought to switch back to RAD and turn
numeric off and it worked!
The calculator never prompted to switch modes (as it did other times)
VERY FRUSTRATING!
Now the question. Why do I need to be in RAD mode to do limits? Why
do I need to turn numeric AND approx off?
I'd like my answers in decimals not fractions.
Thanks
John H Meyers
> I tried to enter lim(x+1,x=1)
Still in ALGebraic mode? (as opposed to RPN mode)
> The calculator responded with Error: Undefined Name
Go to MODE > CAS and un-check both "numeric" and "approx"
> This error is not mentioned anywhere in the 800 page user guide!
It should have been expanded to 1600 pages, then :)
> Why do I need to turn numeric AND approx off?
Because God wrote it in the Great Book
(it must be in that 800 pages, or try the other books on the CD :)
"Numeric" (system flag -3) tries to evaluate all variables,
often stumbling if any are undefined.
The CAS often demands integers (which you can enter
only in Exact mode, which is the opposite of Approximate mode),
and is very stubborn about what it will or won't eat,
like a fussy cat, so either feed it the food it wants,
or else get ignored :)
[r->] [OFF]
netsez
> On Mon, 02 Jul 2007 23:20:16 -0500, netsez wrote:
> > I tried to enter lim(x+1,x=1)
>
> Still in ALGebraic mode? (as opposed to RPN mode)
Yes. As much as I love RPN there are some operations
that require holding down multiple keys at once when RPN mode. I just
don't have the dexterity for that.
> Go to MODE > CAS and un-check both "numeric" and "approx"
As you read later I figured that out.
> "Numeric" (system flag -3) tries to evaluate all variables,
> often stumbling if any are undefined.
>
> The CAS often demands integers (which you can enter
> only in Exact mode, which is the opposite of Approximate mode),
> and is very stubborn about what it will or won't eat,
> like a fussy cat, so either feed it the food it wants,
> or else get ignored :)
So you are telling me that I now have to see silly things like
radical-sign(37) and 59/9 instead of decimals?
Also why does it force me to use RAD mode?
Giancarlo
> So you are telling me that I now have to see silly things like
> radical-sign(37) and 59/9 instead of decimals?
I don't think so: sometimes a "silly thing" is just the "most
rigorous" representation of a number the 50G can provide :-)
If you want to see the decimal values of "radical-sign(37) and 59/9",
simply set the 50G to "Approx" mode, by entering the "MODE" "CAS"
display and checking the relevant field, or, more straightforwardly,
holding down [RIGH ARROW] while pressing [ENTER] (this will toggle
between Exact and Approx mode).
If you don't want that setting to be permanent and wanna see the
decimal values anyway, just press [RIGHT SHIFT] [->NUM].
> Also why does it force me to use RAD mode?
I'm afraid I can't give you a detailed answer....
But the Advanced Uaser's Reference Guide says that, when using the
"lim" CAS function, the following settings are to be in place:
"Exact mode must be set (flag -105 clear).
Numeric mode must not be set (flag -3 clear).
Radians mode must be set (flag -17 set)."
Please keep asking if you need more info - Many will be glad to help
around here :-)
Best regards.
Giancarlo
netsez
wrote:
> Hi netsez.
>
> > So you are telling me that I now have to see silly things like
> > radical-sign(37) and 59/9 instead of decimals?
>
> I don't think so: sometimes a "silly thing" is just the "most
> rigorous" representation of a number the 50G can provide :-)
Yes true, just as pi is not 3.141592
Still my mind can visualize decimals easier than 59/9
> If you want to see the decimal values of "radical-sign(37) and 59/9",
> simply set the 50G to "Approx" mode, by entering the "MODE" "CAS"
> display and checking the relevant field, or, more straightforwardly,
> holding down [RIGH ARROW] while pressing [ENTER] (this will toggle
> between Exact and Approx mode).
I know how to switch modes, but I have limited dexterity so key
presses take effort. I was hoping I could stick to ONE mode and be
able to do all my calculations and use the CAS without ever having to
switch modes. I see now that this is impossible :{
Paul Schlyter
netsez <nets...@yahoo.com> wrote:
>Also why does it force me to use RAD mode?
Probably because it's a little more complex - and perhaps also confusing
to the user - to take the integral and the derivative of trig functions
properly in DEG or GRAD mode.... :-)
"But I'm not using trig functions!", I hear you complain. True, but
the CAS is able to handle trig functions too - at least in RAD mode...
--
----------------------------------------------------------------
Paul Schlyter, Grev Turegatan 40, SE-114 38 Stockholm, SWEDEN
e-mail: pausch at stockholm dot bostream dot se
WWW: http://stjarnhimlen.se/
John H Meyers
> So you are telling me that I now have to see silly things
> like radical-sign(37) and 59/9 instead of decimals?
In general, symbolic math can be rigorously done only in "exact" mode.
> Why does it force me to use RAD mode?
Calculation of limits in general may require taking derivatives,
and calculation of derivatives in general may involve functions
whose properties are valid only in this mode; it appears that
even if some specific CAS function, with some particular argument(s),
could be derived without depending upon this, it would require
more of limited ROM (or leave room for fewer CAS functions)
to predict the required environment in advance for all
individual cases, so what seems to happen is that a common
environment is set up at the outset for most CAS commands.
If you want to suppress the mode-change dialog,
and also automatically restore modes after every operation,
to put that whole issue out of mind,
you could store the following simple program,
which acts automatically after every keyboard operation
(in RPN mode only, which is a small incentive to use it :)
\<< PATH POP PUSH EVAL { -62 -63 -120 } SF
DEPTH { DROP } IFT -55 DUP SF CF \>> '\GbENTER' STO
Where \Gb represents the Greek Beta character
(Alpha Right-shift "B")
You should first set your preferred modes and do one PUSH
(turning off USER mode while you do this),
so that there will be something to POP the very first time;
type any digit and invoke the program once manually
to initiate its subsequent automatic operation
(turn off USER mode to suspend automatic invocation).
The above was taken from a longer discussion
of varied uses and programs for '\GbENTER':
http://groups.google.com/group/comp.sys.hp48/msg/a5716f4bf1cca011
(and the entire thread).
Much less is available in ALG mode,
but the following can be manually invoked to restore modes,
*and* to display the numeric value of the last result
(no arguments are required):
\<< PATH POP PUSH EVAL -120 SF 1 ANS \->NUM \>> 'SHO' STO
If you assign that to any main keyboard key
which you otherwise infrequently use
(perhaps any of APPS, TOOL, HIST, EVAL, ['], ...)
then it would involve minimal key pressing.
[r->] [OFF]
netsez
> Calculation of limits in general may require taking derivatives,
> and calculation of derivatives in general may involve functions
> whose properties are valid only in this mode; it appears that
> even if some specific CAS function, with some particular argument(s),
> could be derived without depending upon this, it would require
> more of limited ROM (or leave room for fewer CAS functions)
> to predict the required environment in advance for all
> individual cases, so what seems to happen is that a common
> environment is set up at the outset for most CAS commands.
Yes, I can see now how forcing an environment where CAS -always- works
makes sense.
> If you want to suppress the mode-change dialog,
> and also automatically restore modes after every operation,
> to put that whole issue out of mind,
> you could store the following simple program,
Wow, very impressive stuff! Thanks for the info!