D-39
===
I plagerized this code << &#8220;Enter a: &#8220; {&#8220; :a: &#8220;
differently from the PROMPT command, and what function does the {2
0} V serve???
Actual string should read as follows:
This prompts for input. How does this function differ from the
PROMPT command, and what function does the {2 0} V} serve???
INPUT Command uses V as to Verify the input as you would type it
into the command line eg. not just a text string, but a valid object.
{ 2 0 } specify the tabulator fields where you input goes.
Experiment a bit, and if it doesn't open up, call again.

The HP48 Advanced users Manual (Printed version) has this and a lot
of other goodies that are valid for the 49, and are not easily found
elsewhere.
Take a look at this *HUGE* PDF: (HP 48G Series User's Guide)
http://ca-on.hpcalc.org/details.php?id=3937
HP48G Series AUR, or some similar user-written document
at www.hpcalc.org, or Appendix G of HP48G Series User's Guide
(also at hpcalc), or Donnelly's HP48 Handbook (out of print).
Read the manuals that came with prior HP calcs,
the longer ago the better. Or the Quick Ref guide, sometimes.
How to use INPUT and PROMPT
http://www.engr.uvic.ca/~aschoorl/faq/48faq-8.html#ss8.22
Other simple User-Interface Program Commands
.
I used the Statistics aplet to fit an exponential equation to a set of
population data. (Given below if you want it)
No problems so far - the result was 128.35*EXP(0.0585x) (rounded off).
The strange result came when I tried to use PREDX. Entering PREDX(150)
gave the result below:
(-2.67319970364E-2+(0,6.28318530718)*n1)/-.782404601086
What gives? Even if you eliminate the complex portion and use
-2.673E-2/-.782 you don't get the correct answer of x=2.66
Year (yrs after 1975)   0     5     10     15      20      25
Pop. (in '000s)        128   170   232    316     412     549
I fixed a bug in all versions of PowerPlot and my Table program. It has
previously plot garbage and possibly cause memory corruption. You can
download the new versions at my website:
PPL:            http://leviathan.orblivion.com/hp48/software/ppl137.zip
PPL + Table:    http://leviathan.orblivion.com/hp48/software/pplt138.zip
Table only:     http://leviathan.orblivion.com/hp48/software/Table196.zip
You can also download tar.gz files from my website.
If you don't or cannot use ArcSin(x)=blah blah blah Tan(x) I'd recommend
the
following infinite series:
ArcSin(x)=x + 1/2 * t^3/3 + 1/2 * 3/4 * t^5/5 + 1/2 * 3/4 * 5/6 * t^7/7 +

...
for -1<t<1
It was obtained knowing that
Intg(0,x) dx/[sqrt(1-x^2)] =ArcSin(x)
and applying Taylor Series in the denominator before finally integrating
term by term...
What's the best pocket PC to run EMU48CE as an HP49 emulator? Do any
work well? Any help would be appreciated.
www.savagetranscenden.com/hp48links.html
I'm not sure about others, but my the iPAQ 3950 runs it very well.
Additionally, it's much quicker than my actual 49G...
It does take a while to get used to stylus input, and the display is
too small, but overall it's not bad when your real calc's not around.
I have trouble even when reading the display in my HP 720 Jornada
I wonder what happens with a PPC.
The display should have twice the pixels to be accurate
and a bit larger, too...
PS: Are there any *GOOD* pictures and KML scripts for both
PPC and HPC LCD formats???
I have got a Toshiba e310 with an ARM porcessor (200MHz) and 32MB
It looks good with this machine ! Now you can find PPC with 400MHz
processor and 64Mb
it looks like you will be able to have bluetooth for your HP4*
http://www.3eti.com/
on the website, click on the news tab (sorry, no direct link) and look
for RS-232-buetooth adapter.
That's good news for us TI users, too, since someone is working on an
RS-232 interface.
http://en1.endiva.net/3eti/pub/News.asp
What answer are you getting?
If 5, then you are probably doing 15^12 (mod 11).
If -10, then that's congruent to 1 (mod 11) and is therefore correct.
If something else, please let us know what it is and how you're obtaining
it.
since several commands who I have reviewed refer CARRY, NO CARRY.
That it means this exactly.
Carry is a flag that can be set by various events. It can be easily
tested for
by GOC and GONC.
It is set if a register overflows (or underflows). Remember way back in
school
that you have to carry the two(or any other nmber)? this works the same.
Take a look at  examle 3.2 from the HP49 Assembly Language Examples
http://www.hpcalc.org/details.php?id=5007
Tan(x) integral from 0 to oo crashes my HP 40G (all memory is
cleared). How about HP 49G?
Well My HP49G gives me an Error Message Box !

For some ungodly reason, I can't figure out how to delete a folder using
the
file manager on my HP49g.
I enter [2nd][Files].
I hit the down arrow until I am highlighting the directory I want to
delete.
You have to go into the parent directory and then highlight the dir you want
to purge
if you now press the purge button (NXT 'A') the dir will be gone
==
To delete a directory you first have to go(with the right arrow key in
your FILE MANAGER) in the directory wich containt the directory you
want to delete and than use keys NEXT F1(PURGE).
Do not be afraid to play with the machine!
Highlight the parent dir (the one before the
one you want to delete)and press OK from the
soft menu to enter into it. Now highlight
the file or directory you want to eliminate
and hit the NEXT key until you see the soft
entry PURGE. Press it to nuke the sucker!
!Demeter!
The 49-built-in Filer has been essentially improved
in Filer1 or alternatively, Filer2 on the site below.
In Filer1 you need not page the menu anymore. PURGE
is contained in the 1-page menu, at the same place
as in the TOOLS menu. Additional functionalies are
on hardkeys. Just press the menu option KEYS to see
their functionalities. You'll learn them very fast.
One can make a HOME backup in a scanned port, with
date & time appended to the backup name. One can also
hide some files in the currently scanned directory.
One can ask for time and date without leaving the
filer. There is a complete VERSION and CAS-version
informator, one may search for a global name in the
scanned directory etc etc. Just read Filers.txt.
Filer1/2 is a very rich tool, both for newbees and
for advanced users. Several experts helped me to make
ftp://ftp.math.fu-berlin.de/pub/usr/raut/HP49/tools
Vertical Polish Notation (now you know what my acronym means ;)

UserRPL programs in ASCII transfer mode should work.
For more details please read the FAQ at www.hpcalc.org
It's a german Webpage: www.cynox.de

I am a new hp 49g user (running ROM version 1.19-6).
I have been learning to use the calculator with the help
of Gilberto Urroz's _Science and engineering mathematics
with the hp 49 g volume 1_.
I seem to be doing something wrong while attempting to
generate a table of values for a function. The function is
Following Urroz's directions I --
-- choose plot-type function in the Plot Setup dialogue;
-- enter the expression/function into the
set the independent variable to 'x' (lower case x);
-- press [NXT][OK] to accept the changes made;
-- after returning to normal calculator display I execute
TBLSET and choose a starting value and an increment;
-- press [OK];
-- after returning to normal calculator display I execute TABLE;
Here is the problem: the expected independent variable (x)
values are generated correctly but the corresponding
dependent variable (Y1) values are all -1 without
exception (which is only correct for x=0). Similar
unexpected results were experienced with other
functions: the generated table values for the dependent
variable (Y1=x^2) yields the constant 0.
What have I done wrong?
or Is this a bug in the software?

that could be ported to a linux-pda.
or maybe to an educational handheld (like the new casio):
http://www.fisher-price.com/us/pixter/
Or a Windows PDA ;-)
But that's assuming the new Casio supports such C programs. Casios
have been very bad platforms for programming in the past.
You can swap the both sides, but no subexpression moves across the =
You have to use stack syntax to subtract it from both sides and
then simplify or collect to get the terms together.
Yeah. I had sort of come to that conclusion when no key combination
I could think of worked.
There are a lot of problems that would be easier if one could do it.
For example, eliminating square roots from equations, particularly
when there is more than one root to be eliminated.
X
Take a look at:
http://www.hpcalc.org/details.php?id=4360
BUGS:
EXPMUL2SUMEXP
EXPPOW2POWEXP
have they names & internals swapped, sorry!
Not sure but, try the stack syntax for IFTE and see what happens...
Toby
I am user of linux and i have an Hp 49G, and never have found a
software for pc-hp transfer for linux?....there is one?..
http://lcdx00.wm.lc.ehu.es/~svet/hp49g.html
What OS do you use? For MacOS X, I havn't tried yet, but any
communication software supporting the Kermit protocol should work (try
ZTerm, or anything Google Search offers).
Under MacOS 9, I use ZTerm 1.0.1 (
http://homepage.mac.com/dalverson/zterm/ )
The communication module of AppleWorks 5.0 also works fine.
You may have guessed: I'm a HP48 user and I seldomly touch the filer
softkey menu.
The XSEND and XRECV commands use the XModem protocol, which is usually
faster, but loses file names.


I am wondering if it is possible to use the
alarm as an interrupt to another program that
is running all the time?
Well, the alarm will first wait for the running program to finish and
it will *then* come up. That means that the running program must check
from time to time when the next alarm comes up and HALT itself and the
alarm must be a program that has CONT at its end, so that the halted
program can resume.
Unfortunately no HP41 behavior :-(
I would certainly want to have - in addition to the HP 41CX control alarms
also a HP 71B style ON TIMER GOTO/GOSUB style behavior or
DO...UNTILTIMER...END & WHILETIMER...REPEAT...END
naturally one should pass the timer number as an argument
instead of the usual truth-flag...
This behavior is coded into the Keyeval of the System Outer Loop if
I am
remembering correctly. So the behaviour you expect should be possible by
means of SysRPL when you use your own Keyeval. I have, several years
ago, written an almost similar, but simpler program which updates the
clock display on every second. The program puts the calculator into
light sleep, when the time update interrupt arrives, the program is
wakened, the display is updated and calculator goes again into light
sleep. That way it consumes even less power than system time display
routine. Maybe the same approach may be used to solve your problem? I
have not published my clock program, because it uses fonts and some
and there are lot of time display programs on hpcalc.org. But if there
is interest towards it then I can send it to hpcalc.org.
| http://www.umr.edu/~faucett    |
I tried to trasfer a 100k file from my pc to my 48 using kermit.
At 9600 baud, by the time 50k had been transferred, kermit showed over
200 retries. I noticed  on the Kermit screen that everytime these
retries occurred, the packet lengths would get shorter (56), then they
would go back to the normal 93 untill the next retry.
I aborted the transfer and tried at 4800 baud. Even so, by the time
the whole file was transferred, there had been over 40 retries.
I'd use XModem: it's faster.
What I do is load the file on Emu48 and then transfer it to the real calc.
I've never had problems.
On the transmitting 48 I open the I/O window choose xmodem, put the
name of the variable in the proper window and press xsend.
What do I do on the receiving calculator?
Is there a way to do this directly from the stack?
Luis
I have this task automatized since... uf!.
The sending program:
it will send the object in level 1, unless it was a name; in this case, the
content is sent. (Run this prg BEFORE than the other (in the other calc)
for
faster transfer)
.82 DUP TYPE 6 ==
IF
THEN RCL
.82 'P' -40 CF XSEND
CLOSEIO -40 SF
e
e
The reciving program:
When the transferring is ended, you'lll get the object in level 1
.82 'Pqte' DUP DUP
XRECV CLOSEIO RCL
NEWOB SWAP PURGE
e
me and I'll try again
I would like to add MASD documentation to the Debug4x help files.
I can only find .PDF files on HPCalc.org
Anyone have something besides .PDF so I can copy and edit the text?
49g rom 19.6, calc set to real and approx.:
Reason being is I'm trying to write a program that take two numbers (+
or -) and evaluates the equation. The problem appears to be in how
RPN handles negative numbers raised to a power.
i.e. 'k*q^1.85'

Both expressions are not equivalent:
In the first one you must use ( ) for raise the negative number to the
power: '(-600)^1.85' and then you get the same (right) answer than in the
second one.
http://staging.porternovelli.com/sanfrancisco/preview/hpcalc1
HP will continue to enhance its presence in the calculator market
I am using my 49g for this----
After some reading I have found out how to correctly enter the equations
I need. With that comes a question:
X(t)=5cos(2t)
Y(t)=5sin(2t)
therefore: 5cos(2t)+i*5sin(2t)
Why is it that we enter the equations the way we do?
Each point on the parametric plot has the form (X,Y). This is not
just a coordinate, but a complex number as well. Since X and Y are
each controlled by separate equations, it can be expanded to
(Xequation, Yequation), and another way of writing that is Xequation +
i*Yequation, as you have above.
The so-called complex plane is nothing fierce - it's just the usual
XY plane with Y redefined as i and -i. That's how you generate
fracs - each complex number out of the equation's iterations is
plotted as a coordinate.
They shouldn't treat that as a complex equation. They may be doing it
that way because it makes things easier to calculate X and Y, by simply
doing a RE(f(t)) and IM(f(t)). The real problem with doing it this way
is maybe you *really do* want a parametric equation, and you input this:
x=cos(t)
y=sqrt(t)
t={-pi pi}
On a true parametric, when t=-1
x=cos(-1)
y=i
But if you treat it as complex always, when t=-1:
f(t)=cos(-1)+i*sqrt(-1) = cos(-1)+i^2 = cos(-1)-1
therefore:
x= RE(f(t)) = cos(-1)-1
y= IM(f(t)) = 0
So if they are automatically treating parametric as complex (which on
the HP48, that's how you enter the equation - in complex form), then (at
least by my book) that's not really parametric at all! In fact, it's
not even possible to have a discontinuity because of an imaginary
result. They should have probably called it complex or something,
instead of parametric. Very strange. I wish I had my HP48 user
manual to see how it tells you to setup a parametric plot.. but it
appears to want it in f(t)+i*g(t) form. Another weird thing, if you
don't give it a complex part at all, you should only be able to plot
points right on the real axis.. which it doesn't do. Either I'm doing
something wrong or it's doing something weird. If someone knows how the
'48 expects parametric input, could you post it?
On the parametric plots, I just consulted my HP48 user manual and it
appears
that it does indeed interpret everything as complex (that is the form that
you
have to enter the expressions in.. x(t)+i*y(t) ). I don't know why they
did
it this way (although I could guess that it was to make their job easier!).
Anyway, I am quite certain that the HP49 would behave in a similar way.
Most
likely they just made a nice, pretty front-end to the same routines the 48
used.
Just as a last note, I was testing those expressions from my previous post,
and the two plots look the same (comparing the built-in plotter vs
PowerPlot,
which does true parametric). But I realized that the imaginary part was
always 0 on the build-in plotter, so you wouldn't see any bad plot because
it'd be drawing over the x-axis. To see the effect I was king about,
set
y(t)=sqrt(t)+1. That flat line at y=1 should not be there on a true
parametric plot.
What was the problem? It's likely that your tangent line was caused by
a
undefined point. If the calculator is trying to connect points across an
asymptotical discontinuity, then you'll get this sort of effect. The
reason
is often because the discontinuity does not fall exactly on a calculated
point, and therefore the program has no way of knowing it crossed an
undefined point.. so it goes ahead and connects the two points, with the
discontinuity in between. This will be a problem for any discrete data
being displayed as a continuous function. For example, if you set your
increment to .2 and you are plotting '1/x', what happens when -.1 is one
point, and the next x value is -.1+.2 = .1 ? It doesn't see the x=0
because of the discrete step size, so it just connects (-.1,-10) to
(.1,10),
which is basically drawing an asymptote. However, if it happens to work
out
that a plotted point falls at x=-.2 then the next point is x=0.. this
causes
an error, which on moderately sophisticated plotter, should trap and
subsequently skip drawing at that point, and continue to the next.. which
would be x=.2. The result is a discontinuity. Sometimes it's just dumb
luck. Hopefully your teacher realizes that the machine is a tool, and
not
a replacement for a brain.. and that an apt student should see some
artifact
and realize that this is simply the limitation of the machine.
Unfortunately, it sounds like your teacher doesn't realize this, and just
views the machine as a magical box that returns an answer, but has no idea
how the machine works beneath the surface. It's like doing (2/3)*3 on a
calculator, and claiming the solution of 1.99999999999 to be incorrect. It
is just as important to understand the limitations of your tool, as it is
to understand its capabilities.

I have asked this question earlier but recieved no satisfying answers,
so, how is it possible to wire up an 48 G to print normal ascii code
to a, lets say Canon BJ-10SX? Or any other brand that accepts ASCII
print out, for that er.
I also have access to the HP 4L laserjet and yes I have tried using
the PCLPRINT - anybody seen that work? I would love to be able to
print my strings on a normal printer instead of using the thermal
printer...

The 48GX and PCLPRINT with an HP LaserJet Series II works with no problem,
using the printer's serial port. Just make sure that the baud rate etc are
the same on both ends, and if that doesn't work try throwing a null
modem
doohickey onto the serial end of your cable (in case lines 2 and 3 need to
be swapped, as the Fates decree).
If your printer only has a parallel port, a parallel-to-serial converter
should do the trick.
PS: Make sure that you have the G/GX version of PCLPRINT! The older
version
only works on the 48S/SX.
peoples have asked me for:
The Red Eye format of the HP IR printer:
To send a byte (d0-d7) to the printer you must send through the IR led:
0 0 0 1 1 1  h0 !h0 h1 !h1 h2 !h2 h3 !h3 d7 !d7 d6 !d6 d5 !d5 d4 !d4 d3 !d3
d2 !d2 d1 !d1 d0 !d0
where h0 = !(d0^d1^d3^d7)
h1 = !(d0^d2^d4^d6^d7)
h2 = !(d1^d2^d5^d6^d7)
h3 = !(d3^d4^d5^d6)
Each bit 0 or 1 must be set valid on the led for 0.425 ms

Recently, I dug my HP 48GX out of storage. I was unable to find the
manual, and I had only begun to 'get to know' the calculator before
needing to put things into storage. I've had a series of HP
calculators (Dad's 35, my own 25c, 41c, 32s & 48gx) and have a few
questions.
1. how do I convert a number from one unit to another? Say, Gallons
to liters?
2. Is there an equivalent of the caog of functions found on the
41c?
The manual: http://www.hpcalc.org/hp48/docs/misc/hp48gug.zip
With RS 6, select the units menu and then enter inn the desired submenu
Write the number, then press the unit key and you'll get the number with
unit in level 1.
For converting, press LS and then, the unit key.
No, but this helps:
http://www.hpcalc.org/hp48/docs/programming/hp48gaur.zip
http://jamlab.home.att.net
Lets say I have the following formula (which I do):
R = ( ( P * I ) / ( 1 - ( 1 + I)^-N )
how can i get my hp49g to automagically rearrange the formula so that
it would show the formula for solving anotehr variable ( P, I, or N)?
I would store that into
'EL.EQ'
then I would start the Numeric Solver above [ 7 ] in red.
I think you can figure out it yourself once you CHOOS'
the right EQuation....
Here it is. On stack level 2 I place 'X'
(actually any algebraic) and on level 1
a list with wonderful bad things inside
like { 1 TAN LOG } which is by itself a
simple innocent program if you hit EVAL. BUT,
if instead of EVAL you type TABVAL and
ask for the values of {1 TAN LOG} out of
'X' (I know it doesn't make sense but keep
on reading) then you DO get a list of two
lists only that the second list contains
binary garbage. I guess that the argument
checking FAILED and the machine proceeded
with normal execution. You can save this to
a string and download it to the PC to view it
BUT do not do an UNDO or EDIT operation on
either the list or the string (my machine
crashed on these attempts). Better yet, after
that just hit a warm reset to be sure.
I get an Insufficient Memory error, which is also not correct
functionality.
BTW, how many bugs are still hidden in the HP49G?*If* there has been any
argument checking at all!
I think that the only argument checking done,
is for an alg. object on stack level 2 and for
a list on stack level 1. Or is that not true?
I'm trying to solve 3 * 2x - 9y' * 3y^2 = 0 for y' to get y' =
2x/9y^2
on my 48GX. It's very simple to do on paper, but on the calculator
the best I've been able to do is either 2/9 * xy^-2 or 6x/27y^2.
Those are of course correct, but it's driving me crazy that I can't
find the right combination of keystrokes to get it into the form I want.
Is there a way to do it (other than opening the equation editor and just
6x/27y^2
2x/9y^2
Assuming that the vector can also contain negative numbers and that
you work with the HP49G. GS is the greek S (Sigma) character 133 on
the HP49G.
<<
AXL
IF
THEN
DUP 1
<<
IF
DUP 0 <=
THEN
DROP
END
OVER SWAP POS 0 PUT
END
AXL
This will only replace the first occurence of the smallest positive
element by 0. If the smallest positive element appears more than once
and you would like to replace all its occurences with 0, one possible
way would be to use another DOSUBS.
<<
AXL
IF
THEN
DUP 1
<<
IF
DUP 0 <=
THEN
DROP
END
<<
1
<<
IF
DUP minposit ==
THEN
DROP 0
END
END
AXL
Can anything else but integers appear in the vector?
Entry points are supported for reasons.
Programmers should use official entry points and only that.
Yes using non-supported entry points could save a few bytes, but what is
more important: life time of an application, or around 50 bytes saved of a
to of thousands ?
By stable unsupported
pointers are meant the ones in the closed invervals
more than once that these will not move in future
ROMs for the HP49.

Thus, as long as the HP49 lives, these can faithfully
be used. Not only 50 but 500 or more bytes are saved
in complex applications. Besides, the joy of SysRPL
programming is twice as big if using such pointers :-)
The modified extable has several different purposes.
1. It greatly improves the readability of decompiled code. A lot of
name.
2. Programmer who wish can easily use the unsupported-but-stable
pointers. However, we mark all those entries with a final
underscore (like RunInNewContext_), to make users aware they are
using unsupported pointers.
I personally use UBS pointers mainly when it is difficult to see how
to achieve the same functionality with supported pointers.
But I agree with Wolfgang, that the programs are being written for the
HP49G now, and the reasons to worry about compatibility with future
devices are very limited.
- It seems to be highly unlikely that another high-level calculator
will be released by HP *and* still use the same operating system as
the HP48/49. Also the screen/memory/hardware will be different, and
all features of current programs which make things nicely fitting
into the screen/memory... will be obsolete. Everything will have to
be rewritten anyway. Even most HP48 programs required significan
attention before they run on the 49G.
- Future devices will very likely use different programming languages.
- The only real worry is that when the HP49 OS source code really
becomes public domain (I still have to see it before I believe
it...), people might create new ROM releases and ignore JYA's
promise to keep the stable regions.

Naah, they have just release new models, which have a real keyboard
faster Saturn-CPU and the USB pointers have changed!!!

[to use unsupported but stable entry points]
Mika Heiskanen and even Bill Wickes used them extensively, however,
in some of their own published additional HP48 software [e.g. Jazz]
Of course, the stability is much easier to guarantee
for discontinued models which have no flash ROM :)
Read my lips -- no new taxes! [G.H.W.Bush :]
Can they be promised to move no more than a few nibbles, then? ;-)
The only serious problem is that WR is torn between the Devil
and the Deep Blue Sea, eternally condemned not to achieve
the absolute minimum possible number of program bytes,
thus destroying their perfect beauty.