B13
====

Am Besten erkl.8art anhand eines Beispiels:
folgendes Gleichungssystem ist zu l.9asen:
2x + 5y + 7z = 33
2x + 2y + 2z = 12
3x + 4y + 2z = 17
Man erh.8alt die erweiterte Matrix:
|    2    5    7    33  |
|    2    2    2    12  |
|    3    4    2    17  |
Diese Matrix gibst Du in den Matrixwriter ein (LeftShift und MTRW)
Danach f.9frst Du den Befehr RREF aus und bekommst.
|    1    0    0    1 |     X = 1
|    0    1    0    2 |     Y = 2
|    0    0    1    3 |     Z = 3
HTH
Bernd
====
de ton nombre avant de faire la somme ou diff.8erence.
pour une soustraction
avfl2o$s02$1@rex.ip-plus.net...
====
card) and initially was delighted.  Soon, however, I got into a big
trouble-vortex with the fonts.
CLLCD text 1 DISP 7 FREEZE 0 WAIT, etc,
CLLCD.  I tried all kinds of flags, downloaded Ytrik and mkf flags, to no
avail.  On the HP49 the txt is FONT6.
Even more disappointing, when applying Menu 30.01 (SOLVE ROOT SOLVR) the
menu labels still show the old clunky HP CAPITALIZED letters.  There must 

be
a way around?
====
card) and initially was delighted.  Soon, however, I got into a big
trouble-vortex with the fonts.
CLLCD text 1 DISP 7 FREEZE 0 WAIT, etc,
CLLCD.  I tried all kinds of flags, downloaded Ytrik and mkf flags, to no
avail.  On the HP49 the txt is FONT6.
Even more disappointing, when applying Menu 30.01 (SOLVE ROOT SOLVR) the
menu labels still show the old clunky HP CAPITALIZED letters.  There must 

be
a way around?
====
Try DISP2 (decribed in the documentation)
be
No, for these things MK is using the original ROM
====
a
be
====
card) and initially was delighted.  Soon, however, I got into a big
trouble-vortex with the fonts.
CLLCD text 1 DISP 7 FREEZE 0 WAIT, etc,
CLLCD.  I tried all kinds of flags, downloaded Ytrik and mkf flags, to no
avail.  On the HP49 the txt is FONT6.
Even more disappointing, when applying Menu 30.01 (SOLVE ROOT SOLVR) the
menu labels still show the old clunky HP CAPITALIZED letters.  There must 

be
a way around it?
Walt.
====
card) and initially was delighted.  Soon, however, I got into a big
trouble-vortex with the fonts.
CLLCD text 1 DISP 7 FREEZE 0 WAIT, etc,
CLLCD.  I tried all kinds of flags, downloaded Ytrik and mkf flags, to no
avail.  On the HP49 the txt is FONT6.
Even more disappointing, when applying Menu 30.01 (SOLVE ROOT SOLVR) the
menu labels still show the old clunky HP CAPITALIZED letters.  There must 

be
a way around it but is it worth bothering?
====
with:
MK can't change the internal commands from HP48, so you have to
manually replace them with the MK commands built on the MK library,
this way you have to change DISP to DISP2, that comes:
CLLCD text 1 DISP2 7 FREEZE 0 WAIT
What you have to do is replace in your programs the original command
with the MK one, this way all will become with coherent fonts and will
run with a faster speed. This is why there are different versions of
the programs to MK and HP (for example Organizer).
About the menus, in fact MK can't change their appearence in all
places, specially in those ones (SOLVER), and MK don't replace the
library menus too (I've compiled a version of FastLibMenu from Java to
be usable with MK to put the fonts coherent, and much faster;).
Paulo Pinheiro
p_pinhas@clix.pt
====
card) and initially was delighted.  Soon, however, I got into a big
trouble-vortex with the fonts.
CLLCD text 1 DISP 7 FREEZE 0 WAIT, etc,
CLLCD.  I tried all kinds of flags, downloaded Ytrik and mkf flags, to no
avail.  On the HP49 the txt is FONT6.
Even more disappointing, when applying Menu 30.01 (SOLVE ROOT SOLVR) the
menu labels still show the old clunky HP CAPITALIZED letters.  There must 

be
a way around?
====
My english is not perfect, I wan't to say i have an HP 28 for sale....
;-)
opriqaxfwekts43u@shawnews...
====
I need help to complete the new version of Navigator : that program is
able to download, read online or offline HTML pages, and you can
interact with the forms (write inside text areas and text input forms,
push radio buttons, check checkboxes, select options, etc.) but I need
to know how to send a form using the POST method.
I read the RFC of the HTTP protocol version 1.1, but I still don't
understand how you can send full texts :
you must send something like :
POST http://www.domain.com/abs_path HTTP/1.1
Host: www.domain.com
Data1: value1
Data2: value2
...
but when the value is a text, how to explain to the server that the
: of the text is not the sign which separates a name of data from
its value ?
Could somebody write me an example of form sent with the http protocol
?
My adresse : yoanndesir@yahoo.com
Yoann D.8esir.
====
I guess graphing something like y=sin(x) would be pretty hard, especially
since it is not truly a function like you mentioned.
I was just looking at my friend's TI, and it has a DrawInv command that
flips a graph along the y=x axis, which is basically the 'inverse 
function'.
I cannot really find practical use for this, but is it possible to 'flip' a 

graph
along the y=x axis on the HP49G?  How about actually being able to access
the function to find the zeros, intersections, etc?  On a machine like this
there should be a way just to flip a graph like that... ;)
-- 
Al
====
Lets visit www.HP-Network.com
Eric Vialle
dd553de1.0212130523.7576115b@posting.google.com...
====
Does anyone have a survey levelling program that I can manually enter in my
HP.
Any  websites or actual written progs appreciated.
JB
====
The procedure is the same on the '48 & '49:
On the HP48GX, TEVAL is included in an Erable or Alg48 utility lib - I 
don't
remember which.
====
I had problems logging on to this newsgroup because flash.net merged with
prodigy and prodigy with sbcyahoo.
Until they finally figured out how to upload all me previous (futile)
attempts that were stored in the Outbox  were also uploaded.  That's why my
message is 8-fold.
SO Sorry!
====
A pre-version of Usinagaz is available on Hpcalc.org, it now has an
assembly and a sysrpl interface. It also now supports PAP & Dynamic DNS
setting.
I reimplemented tftp in assembly as a separated package, so it is now very
fast (getting files was sooo slooow when in RPL)
This is still a pre-version since both interfaces have not been thoroughly
tested, but as for User interface, it does work, and Navigator 0.4 needs
it, so...
-- 
(By Dr. Greg Wettstein, Roger Maris Cancer Center)
====
I can confirm that Usinagaz pre-0.3 is really faster that the previous 
versions !
Conclusion : if you own an external phone modem, you MUST try it on your 
HP49 !!
====
Since you have not put in a valid reply-to address I'll answer here and 
you can get back to me.  I repair these calculators and am willing to 
try to fix yours if you like.  Let me know.
-- 
Dave Mabry             dmabry@mich.com
Dossin Museum Underwater Research Team
NACD #2093
====
IĒve been using the hp49 for about 3 months now and 
IĒve learned a lot, but
I have 2 questions at this time:
1. In a book about the hp48 I remember something about a multiple level
solver or something like that. In my right-shift + num.slv gives a 
menu
oportunity which name is MSLV. WhatĒs the use of that 

function?
2. I often get very big results or very small results in rpn mode, like
14328438284321.54 or 0.00000000000038, but how can I make that into
scientific notation without having to go into calculator modes + 
number
format + fix + scientific/engineering + all the way back again, 

when I
want to calculate in standard notation...
There must be an easier way, right??????
There has to be.......
ItĒs not always easy and I think itĒs very 

annoying to count small digits in
the display and get a result like 1,433E?
I really appreciate any help!
Martin J.
====
This is the Multiple Equation solver. It can solve pretty much any number 

of
equation with the same or less number of unknown
I
in
you can use the STD or FIX or ENG command (FIX and ENG take the number of
displayed digit as arguement)
====
Brķ©bisson <cyrille_de- 
Don't forget the SCI command for scientific notation. :)
-- 
Al
====
If you look in CAT, it's called a multi-variate solver and there's a
brief Help screen (at least with my version).  The MSOLVR command is
the multi-eq solver - it expects to find a list of equations in the
variable EQ and will write the variables into a white softkey menu. 
Enter the known values, pressing the associated key afterwards.  To
solve for an unknown, press Leftshift and then the key for the
unknown.
 
This may only apply to later ROM upgrades, but if you press and hold
the Leftshift key and then the Mode key, you get the HP48-style
softkey menu.  FMT gives you the display modes.  What I do: since I go
with the naames F1, F2, etc.  You could also assign them to custom
keys.  (I work with HMS time a lot, so I have 4 Fix and STD assigned
to keys.)
BTW, if you haven't upgraded to v1.19-6, it's well worth it.
Bill
====
Den Befehl kannst Du eintippen oder vom command catalog ausw.8ahlen.
Mit A vor dem x meinst Du bestimmt einen symbolischen Koeffizienten
von x, wie z.B. A*x+3*y=2. Im prinzip kannst Du exakt so vorgehen, wie
f.9fr numerische Koeffizienten.
OK, lass uns mal ein Beispiel probieren.
Mit hilfe des Matrix Writer bilde den Vektor:
und lege ihn in stack ebene 1. Dann, bilde den
Vector der Unbekanten [X Y Z] und lege ihn auch
in den Stack.
Dann benutz den Befehl LINSOLVE, um das Gleichungssystem
zu l.9asen. (LINSOLVE is im Menu S.SLV)
Ergebnis: [X=29/(5*A+3) Y=(11*A-5)/(5*A+3) Z=(4*A-15)/(5*A+3)]
War ich verst.8andlich? Wenn nicht, sag mir bitte, was ich besser
erkl.8aren soll, OK?
Gruss,
Nick.
====
 I always remember it as CdB ;-)
Gjermund
====
How about that!  What a great program OT49 is - it does good things
even when I'm not using it.  :-)
Bill
====
Though I agree in general, I doubt that me, you or anybody else will
recognize erroneous results if the differential equation exceeds some
degree of complexity.
Greetings,
Nick.
====
I got the permission from Philippe Pamart (HPCooL)
to use his minifont tools in Fontman. The library
has been rewritten this night. It doesn't use slow
lib 256 commands anymore. The command FPAR sets new
font parameters instanteneously (due to fast string
prosessing). I don't know whether JKH was the first
who applied this idea in his font manager, at any
rate he got credits in Fontman.txt.
The dialog box FPAR accepts *any* string and *any*
number as font name and ID, resp. The latter will
automatically be recalulated modulo 256, the number
of possible fonts. I would be happy if somebody tests
this small library and reports here in the NG.
Have fun, Wolfgang
ftp://ftp.math.fu-berlin.de/pub/usr/raut/HP49/fonttools/
PS. Since a font uses over 2 KB you may compress them
(e.g. with the powerful BZ-tools from 0T49) and store
in port 2. Fontman has a font browser looking also for
compressed fonts (the builtin font browser doesn't).
====
Suggestion: recalculate the font ID modulo 245 instead of 256, in light of
the following...
-Joe-
----- flashback to before the dark times, before the Empire -----
2: Italic
3: Underline
4: Inverse
To change the font it will be:
11+ID_FONT
So you won't have a conflict when calculating the number of carriage 
returns
in the text (easier to program).
As you can see, the font ID number can't be more than 244 otherwise a carry
will occur and may conflict with the style coding.
There is enough memory allocated in the system to handle 256 different
fonts, but it's in fact not possible to use more than 244. In fact, any
the memory for fonts..
Hope this help
====
Wenn ich Dich richtig verstanden habe, m.9achtest Du den 
Durchsto¤punkt
der Gerade[x y z]=[1+2*t 1+t -1-t] mit der Ebene 2*x-z+1=0 finden,
oder?
Eine der zahlreichen Methoden w.8are:
1) Aus der Gleichung der Ebene bilde den Vektor, dessen Spitze
   auf der selben Ebene liegt.
   Dann benuthen wir den Befehl ZEROS und finden: 2*X+1. Also ist
   der gesuchte Vektor [X Y 2*X+1]. F.9fr jede m.9agliche Kombination
   von Werten f.9fr X und Y, wird die Spitze des Vektors auf der
2) Der Durchsto¤punkt befindet sich genau da, wo die Vektoren der
   Ebene und der Gerade alle Koordinaten gleich haben. D.h.:
   wir mussen die beiden Vektoren [X Y 2*X+1] und [1+2*t 1+t -1-t]
   einander gleich setzen. Da der HP49G gegen algebraische Objekten
   mit Vektoren ist, benutzen wir Listen um.
   Mit der L.9asung von Schritt 1 auf Stack-Ebene 1, gib ein 'X' und
   Benutze den Matrix Writer, um den Vektor der Gerade [1+2*t 1+t
-1-t]
   einzugeben. Dann benutze den Befehl AXL, um den Vektor zu einer
   Liste umzuwandeln. Dr.9fck =, um eine Liste mit drei Gleichungen zu
   bekommen.
3) Jetzt m.9fssen wir nach X,Y,t l.9asen. Gib den Vektor [X Y t] ein 
und
   dr.9fck SOLVE. Du kriegst: {[X=-3/5 Y=1/5 t=-4/5)}. Jetzt hast Du die
Benutze
   den Befehl HEAD, um den Vektor der L.9asungen aus der Liste heraus-
   zu nehmen. Dr.9fck ENTER um eine Kopie des Vektors zu machen. Die Z-
   Koordinate kann man aus der Z-Koordinate der Gerade mit hilfe von t
   berechnen. Die Z-Koordinate der Gerade war -t-t. Gib diesen
Ausdruck ein.
   Dann druck SWAP und 3 GET, um t=-4/5 aus dem Vektor zu bekommen.
   Druck SUBST und dann EXPAND. Das Ergebnis, -1/5 ist die Z
koordinate
   des Durchsto¤punktes.
4) Durch R.9fcksubstitution der X,Y,Z Koordinaten des 
Durchsto¤punktes
kannst
   Du auch nachschauen, ob die L.9asung stimmt.
Hoffentlich hat es ein wenig geholfen,
Nick.
====
I've got a HP48G and I'm looking for RPL sources of programs (I don't
have any cable to transfer libraries to my caculator and I don't have
time to make one right now) that compute results about graphs.
Especially an implementation of the Warshall algorithm for calculating
the transitive enclosure of a graph using its matrix.
Does anyone know a website where I can find that kind of stuff. I've
already searched on hpcalc.org and other greate hp48 source repositories
but I didn't find anything.
rozwel
====
There is a way, but I cn't remember where I read about it. Sorry
====
Yes, it is possible with Emu48. Therefore you must generate a HP48G state
file with the MKE48 command line tool. An example how to do this is inside
the MKE48.TXT file.
Christoph
====
Is the HP 49G as crash prone as the TI-89?
====
You're very welcome - but I suspect it's *Joe* Horn who posts much more 
(and more useful) items here.  Someday I'll have to answer for giving my 
Jim (Younger than Uncle Henry, older than Joe)
====
I think this car argument is flawed...
I've seen plenty of 30+ years old cars built in the 60s, I'm yet to see any
car of that age built in the 90s !!!
====
Is it Virtual private network? On HP48/49? :-)
 Demo
====
LOL
====
Veli-Pekka Nousiainen: a great one in this forum. And a great 49borg
====
Not that hard. Just:
1) solve for independend variable,
2  decide which of the branches of the inverse relation
   you want to plot, by giving some integer value to the
   variable n1, which is in the solutions,
3) Exchange the roles of the dependend and the independend
   variables,
4) and plot.
OK, I put all the above actions in a program, which then we baptize
DRAWINV.
<<
  IF                        @If current equation isn't of the form
    HEAD POS NOT
  THEN                      @then convert it to the form
    RCEQ PPAR 3 GET =       @y=f(x)
    STEQ
  END
  PPAR 7 GET                @Get current independend variable
  ZEROS                     @Solve for current independend variable
  STEQ                      @Store solutions as equations to plot
  PPAR 7 GET DEPND          @Set independend variable as dependend
variable
  PPAR 3 GET INDEP          @Set dependend variable as independend
variable
  0 'n1' STO                @Here you can store any integer value in
n1,
                            @the variable that is created when trig.
functions
                            @are solved.
  DRAX DRAW PICTURE         @Plot and switch to plotting environment
This is of course not perfect but as you can imagine it can be made
much better. When the plot is finished you can do tracing, find roots
graphically and so on.
The HP49G has no dedicated built-in command for drawing the inverse,
but, he, he, again we find the unexpected but nonetheless useful
commands like ISOM and MKISOM, with which we can not only graphically
mirror on y=x, but do anything else, like plotting graphs rotated, or
mirrored on (0,0) or any combination of symmetry operations like that.
Of course the prise to pay is coding more, but you know, we can't get
something for nothing ;-)
Greetings,
Mirrored Nick.
====
I have an mint condition, UNUSED, out of the box HP49G and separate
Connectivity Kit  and all documentation along with the original packing  
FOR
SALE.   This may be a last chance thing as it is no longer available at HP.
Will ship only to US.
I purchased this because it was the only RPN calculator left besides the 
12C
but after reading the manual, I realized it would take too much of my time
to become proficient at.
You can respond here with information that will allow me to contact you
directly.
Barry Shaklan, M.D.
====
I am using it for ~4-5 years...But I could not emulate HP48G with 32kb
of RAM. It emulates HP48GX with 128kb of RAM.
  Demo
And PEOPLE HAS ANYONE TRIED THAT NewOS? DOES IT WORK?
FEEEEEDDDDBBBBAAACCCKKK.
====
If you load an SX rom, it will limit you to 32kb.  This is what I did
(well, I used x48).  I wasn't prepared to put in it my real 48SX
because the backup would be long--it will erase port 1, I read. But I
couldn't get it to work.  Running the Code in 'loader48' with the
'system' Code on the stack simply warm-starts the calculator.  The
same thing happened on x48 with a GX rom too, though, so I may be
doing something else wrong.
Karl.
====
**** Post for FREE via your newsreader at post.usenet.com ****
I don't know what happened. I tried it & it worked. Do you have this
version:
http://www.twincar.sk/download/NewOS_0742.zip      ?
And what revision of rom do you have in 48GX ? I have L in emulator and my
brother was able to run it 48G with R revision. Of course he used 
loader48G.
  Demo
I
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
 *** Usenet.com - The #1 Usenet Newsgroup Service on The Planet! ***
                      http://www.usenet.com
Unlimited Download - 19 Seperate Servers - 90,000 groups - Uncensored
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
====
Maybe I've been sleeping through the discussions but just what is NewOS?
WigglePig
====
NewOS is a multitasking operating system currently capable of running
at HP49/48GX/48G maybe also SX. Current version is available @
http://www.twincar.sk/download/NewOS_0742.zip     there no web site
yet.
Make sure to read readme.txt before running!
   Enjoy 
      Demo
====
with games, programs, etc.
http://www.geocities.com/gaak_2001/prgm/index.html
Soon at www.hpcalc.org
 GaaK
====
I think that you first have to setup the printer (using the Mac) to
the correct parameters (baud etc.). Then connect it to the HP and just
XMIT the strings you want to have printed.
You don't have to emulate protocols.
Greetings,
Nick.
====
You make it sound so easy Nick :-)
The StyleWriter was probably the first all graphic printer in that it 
doesn't have a built in character set so sending it normal ascii won't 
result in anything recognisable. He needs to know how to generate a bit 
pattern and send that in order for it to be printed out properly. This 
is what he means by emulate protocols.
There's a website - http://www.lowendmac.com (or similar, I'm going from 
memory here) that specialises in providing information about the older 
Macintosh machines. Maybe there is something there about the 
StyleWriter.
-- 
Bruce Horrocks
Hampshire
England
bh@granby.demon.co.uk
====
If any content of text is important, then this kind of content, Joe.
I'm full of ears :-)
Hmmm, OK, though accepting the opposite is also allowed.
OK, any real existing ruler will fail to give the exact point where pi
is. (And also any other number, but that's rather a technology
problem.)
As said, the opposite can also be assumed, but let's get the above for
granted.
You just did, by saying to realize that pi REALLY IS an exact point
that lies exactly SOMEWHERE on every slide rule.
I don't see how this relates to pi. The rules for finding the number
of stones are not known to us. The rule for finding pi is known to us.
Here you get extraordinary philosophical. That's a problem that is
100000 times harder to comprehend ;-)
Anyway, you talked about another definition of precision and accuracy.
Where is it? I don't see any here.
Greetings,
Nick.
P.S. Not every number has to be located the way you described.
====
Send me the schematic of what you did, and i'll take a look.
Don't use this address, use:
Steve Sousa
====
Well, with the exception of the CAS commands, the commands of the
machines are identical, aren't they? A big part of the commands of the
HP49G can be used on the HP48 series, and so the manuals of the 49 are
useful also for 48 users. Am I wrong here?
Greetings,
Nick.
====
Me toooo... where and when will the future HP-Party be in Paris?
Lilian.
very
thoroughly
====
It was talked about on the HPParty forum on www.hp-sources.com,
discussed about february holidays, but no more news since november...
-- 
====
Bonsoir,
Voici un programme pour mettre un polyn.99me du second degr.8e ax^2+bx+c 

sous
forme canonique c'est-.88-dire a[(x+b/(2a)]^2-delta/(4a^2)] avec
delta=b^2-4ac.
%%HP: T(1)A(R)F(.);
.82 0 ō a b c a1
  .82 a 'a1' STO
1 'a' STO
b a1 / 'b' STO
c a1 / 'c' STO
a X 2 ^ * b X * + c +
y b 2 / a / EVAL X + =
DUP ROT ROT SOLVEVX
SUBST DISTRIB EVAL DISTRIB
SWAP DUP2 SUBST a1
SWAP *
  é
é
Avec les donn.8ees 1 2 10 3 / +/-  j'obtiens
y^2 - 16/3
y=1+X
(1+X)^2-16/3
mais avec
1 2 3 +/- j'obtiens
y^2 +-4
y=1+X
(1+X)^2+-4
Comment .8eviter l'.8ecriture +-4 ?
Merci d'avance pour vos r.8eponses.
--
@+
Jean-Marc Rebert
jm.rebert@calixo.net
====
Petite faute de frappe, lire  13/3 au lieu de 16/3
avfl2o$s02$1@rex.ip-plus.net...
====
[this is about neither True type fonts nor Debug4x,
 but it does relate to what was written before]
WG asserts:
That's how eBay turned the legitimate auction format into a kind
of sham which it keeps trying to persuade people is a real auction;
so you've decided that your last bid is the end of the discussion,
the item is now sold, and no follow-up bids are allowed?
But you keep inserting it into your sig every time, right?
A far too narrow consideration; better to find out what *else*
might have a much better overall benefit to an entire society,
especially if it would, in the end, lead to reductions
in the presence of material of basically destructive-only nature,
and its eventual supplantation by something of more positive value.
Again a tunnel-vision approach, as above, especially when only
two possibilities are compared (more or fewer private firearms),
not anything else which might affect the entire society
in such manner as to reduce crime without increasing firearms at all,
or to even reduce crime to such a degree that the incentive
to stock firearms would significantly decline.
The infamous red-herring of this topic...
So, if only Iraqi citizens had more private handguns, then perhaps
they would be able to overthrow SH's WMD-equipped military?
Or is it, at the opposite extreme, that the military primarily
consists of citizens grabbing their muskets whenever civil defense
is needed?  (This is the explicit rationale stated in the U.S.
Constitution for RKBA, but it doesn't fit present day life on Earth).
And if you think that your own government might be ready to turn arms
against you, remember these few things:  You freely elected them
to their offices, you authorized and paid them to maintain armies
(for your benefit), many of us make our livings from their industries,
and those people who joined them are your fellow citizens, so any
such event would really be a civil war, of one part of a nation
against another, which was fostered by our own behavior.
Reflect on how such things can come about,
and find a more plausible means of prevention than the notion
of your personal guns against tanks, gunships, bombers and WMD,
or even of one socially isolated group waging war against another
in the streets with their small arms, if it came to that.
Your emphasis, like that of governments during arms races
(which are continuing into the WMD arena today), is primarily
on the *hardware*, and is not directed towards the *people*
whose collective and individual feelings are the only fundamental
cause of whatever is either good or bad in society,
and of the actions which will follow from those feelings.
Narrowing the vision to think that human discontent can be corralled
and kept away from oneself through possession and use of personal
deadly arms, is to me a recipe for disastrous final consequences,
some of which are very near at hand in the world today,
where massively destructive capability can be concentrated
in even one person's hands, and often falls
into very mentally disturbed hands,
whether a rampaging student or a rampaging dictator.
Want protection against personal assault?
There are plenty of less deadly things to carry or keep at home;
even the more civilized police forces use them when possible.
The cause of all violence starts in a human mind,
and its most effective prevention for entire societies lies again
in its relief within human minds; explaining how this is
achievable is the hardest part of my task, because our
entire educational orientation is so materialistic
that it leaves us far less attentive to the experience
within our own consciousness, and between ourselves
and other people.
If fatigue frazzles you nerves and lowers your emotional
boiling point, only a reversal of fatigue can cure you,
and only some time spent in a different state of consciousness
(known to you as sleep) is effective, as well as absurdly simple.
You have to balance your activity with rest
to remain a happy, productive and peaceful human being,
rather than degenerating into a state of lower level experience
of life, which could include poorer health, poorer disposition,
imbalance of judgment, lack of satisfaction,
anger, misery and violence.
You have to personally alternate waking with sleeping,
have alternating periods of different brain functioning,
to live life in a state of greater creativity, accomplishment,
and fulfillment.
Your metabolic rate is generally higher while awake,
lower while asleep, but in fact it is possible to be completely
awake and alert, yet have a lower metabolic rate than in deep sleep,
and it is found that one experiences finer well being, health,
judgment, creativity, satisfaction, achievement, sense of true
freedom, and peace of heart when this is added to one's daily routine.
So we add it to the daily routine, and everyone feels finer;
feeling finer, they act more positively.  This gets around,
it affects the whole community, people act more supportively,
and the feeling by each person of a need to keep a gun at hand
becomes more remote.  So I'd rather emphasize the right for
everyone to learn how to generate this finer experience of
life and spread it around the community, and become a more
harmonious society, than to dwell on the empty and inferior right
to possess weapons of devastation to turn against one another.
Oh look, we've been discussing this already for about 3.5 years:
http://groups.google.com/groups?threadm=19990629005603.29980.00005033%40ng63

.aol.com
Well, you have spoken and I have answered,
so please, let's not hear anything more about it, okay?  :-)
http://www.mum.edu
.
====
More or less agree, but for the sake of completness:
Some additional arms free and also nearly crime free zones:
http://www.uni-duisburg.de
http://www.unilu.ch
http://www.ethz.ch/
http://www.imos-consulting.com
And some additional nearly crime free but in no sense arms free zones.
(A gun in almost every house, and nonetheless very very low
criminality.)
http://www.hergiswil.ch
http://www stans.ch
And what a mystery it is, to know that in the above nearly crime free
zones there are no gurus, bodhisattvas, saints, dervishes, prophets,
holy men, and all the other drugs that humans took for thousends of
years in order to have some good wise guy in heavens to look at,
when we want somebody to tell us the truth, because we are too
convenient to find it out ourselves and for ourselves. And maybe
because we are too scared to face the fact that the truth could be
that there is no truth at all.
Oops, I was mistaken. In fact at the arms free zone
www.imos-consulting.com you will find a guru, our boss, a technology
guru, who of course as John says, seems (has?) to be the leader in a
world of violence, since he is an atheist and science-technology-freak
(like me). He is one of the most pacifistic, good hearted, and
peaceful persons that I have ever known, but if the
we-know-all-truth-of-the-universe-gurus declare him (and me) to
follow a way that will lead to destruction, then this has to be the
truth.
Greetings from somebody who lives since three years in a house full of
guns, and still hasn't experienced any violence, or even bad behavior
by the bad neighbours that have guns. ;-)
The snake in the grass of your paradise. Ssssssssss ;-)
Nick.
====
I can't use my 49g because it is too powerful. 89 is the same, and it 
has been a LOOOONG time since I have really used my 48gx.
Sooo, I just don't remember, can the 48gx do all the symbolic without 
erable and alg48? I didn't think so, but I just don't remember! Is there 
anything else on the 48gx that might disqualify it?
I really don't want to have to use an algebraic calc, I am just too used 
to rpn for regular stuff, and I really don't want to have to buy a ti.
Matt
====
I just finished DiffyQs last emester and the only thing I used a
calculator for was Euler's Method and runga-Kutta (sp?) method  which
are decimal approximation for DEs.  For that I used a TI 83+ and a
program my prof gave to us. 
You will just have to do some integration and differentiation on some
problems that hopefully your book and prof will rig so that the stuff
you are integrating/differentiating is simple first semester calculus
stuff. 
Bye
   http://www.newsfeed.com       The #1 Newsgroup Service in the World!
====
There are roughly 17 *other* TI emulators on ticalc.org
http://www.ticalc.org/programming/emulators/software.html
as well...surely there must be a bug or two in there somewhere....but, yes,
Virtual TI is an excellent piece of work.
    Isn't TI's RPN an environment and not run natively...and, you know, 

I
think it is somewhat remarkable that the ancient Saturn-based HP holds up 

as
well as it does in comparison to the much more modern 68000-based TI's....I
would think that these Mac-based TI's would just blow the doors off all the
competiton, but that's not the case--the HP's still acquit themselves very
well....
the
think
calc.
The number of users doesn't have much to do with whether or not they
understand what they're doing.....
Happy holidays,
Roger
====
If RPN is only an a non native environment of the TIs then the
comparison is not very fair. If someone ever makes a native RPN
environment for the TI, then I bet the TI will be much faster.
If the software of the TIs would be optimized in a comparable degree
to the software of the HP49, then the TIs would really blow the doors
off all competition. If software optimization is approx. the same,
then it is the hardware that decides about performance, even if this
group doesn't want to hear this. As the time passes by the software of
the TIs will be optimized, and I guess that the TIs still have much
much optimization potential, while the HP49G represents more or less
the endth degree of any tricks that you can find in order to increase
its performance. So the future, as it looks today, doesn't belong to
the archaic HP49G and its paleontological processor.
In addition we have the following fact: If we compare performances of,
say adding two numbers on both machines, we are not going to find
anything special. For me as a user, it is totally uninteresting if
this operation takes 1 millisecond or 100 microseconds. I just can't
tell the difference. If we compare, we should compare lengthy
operations or programs. Guess what happens when the software of the
TIs is as highly optimized as the software of the HP49G. It is short
sighted to believe that this optimization will never happen and that
HP is the one and only company that can do that.
This is a constructed and very cheap excuse - and also short sighted.
Of course everybody that works with Maple and Mathematica also doesn't
know what he/she is doing? And only HP-users are smart guys because
they can work with RPN and because they can find the bugs of the 49?
The bugs of the TI92 were also found but there was not so much noise
because the bugs can be counted on the fingers of one hand, quite
contrary to the HP49G. And don't forget that if you keep examining
each and every result that the HP49G or any other CAS returns, then
the whole concept of using the CAS becomes questionable. If we do all
by hand again, just to check the correctness of the results, then what
did we buy the machine for? Imagine what would happen if you had to
manually check the correctness of a variation of parameters for energy
optimization of a wave function of a molecule. How long will it take
to do that? One life? Do all people that use mopac, or gaussian have
to repeat by hand the hundreds or even thousands of numeric
integrations that the programs do? Do designers and graphics editors
need to repeat, say rotation or blur of a graphic, just to check that
they got the correct result? And do I, normal user of my refrigerator,
need to repeat thermodynamics calculations everyday just to check that
the machine really cools down to the the correct temperature?
Indeed, quantity doesn't have to do with quality, but if so, then
neither in positive *nor in negative* sense. We would better stop
implying that the big number of TI users is automatically a dumpness
indicator of TI users and that the few of us are the smarties just
because we are few.
All of the above is without any intention to harm. It's only that my
sight presents me a rather sad picture when I think about the future
of HP advanced calculators. Perhaps I should change my glasses..;-)
Greetings,
Nick.
====
Imagine what will happen if there were no people checking the CAS
results -- your fridge would certainly fail to function after 3 days
of normal work (or would keep the good and pretty constant 25C degrees
in the camera ;), designer's computers would refuse to do any blur
filters, etc., etc.
Those, who do check the results are called testers -- and most people
in this group cannot be called just bare users of HP calcs.
-- 
Computers are not intelligent. They only think they are.
====
Both the saturn and the 68000 were developed in the late seventies,
both were first deployed round 1979. In this terms they have both the
same paleontological  age..;)
I personally think, that the TiOS software *is* optimized, but not to
an extend found in the HP49G. Where do you nowadays find an OS (even
in  embedded hardware) which only uses 70 to 150K of RAM (seems to
depend on the actual hardware, my Voyage 200 has only ca 109000 bytes
free, whereas an TI89 should have 188K free), when started? But
compare it to an HP49G, which uses only 3K ram after warmboot. 
After all, the HP is optimized in such an extrem way, because its the
only chance to compete with the better hardware of the Ti68ks. 
IĒm
sure, Ti will never optimize TiOS so much, because the corresponding
development cost are just to high. If Ti feel the necessarity of
making an faster calc, the will mainly use an faster coldfire or
dragonball cpu core. Maybe they will port their TiOS or AMS to a fast
Risc-CPU, which should be much easier then porting SysRPL because of
the much more conventional OS design of TiOS.
-- 
Mathias Habel
mathias.habel_no-spam_@t-online.de
Remove _no-spam_ before replying
====
That's what I did with my Atari 8 bit games and computers...now I feel
like an idiot!
====
Are you referring to the file rpn2022.zip? I could not find any help
file associated with it... or is it self explanatory if you're
familiar with HP's RPN.
====
You can get the (pretty extensive) documentation here:
http://www.paxm.org/symbulator/download/rpn.html
--
Bhuvanesh
====
Speaking of optomizations...
I hate to give Hp people fodder, but some of the code that is used in
the AMS is absolute junk.  I've been coding and reverse engineering
parts of it for a couple years.
Here's an example of the code the compiler ti used has been know to
generate:
 swap d0
 swap d0
 swap d0
here is another:
 swap d0
 swap d0
For those of you who don't know what the swap instruction does I'll
tell ya.  It exchanges the upper half of the register with the lower
half.  So having two swaps in a row is redundant.  Having three is
stupidity.
====
I have a 38G and a comms package for Mac. When I do an image transfer the
program works fine, when I try a file trasnfer it hangs. Doe anyone have 
any
sugestions?
====
AFAIK, you can't with built-in commands
====
If your current directory is not HOME itself,
then PATH RCL recalls the current directory (this errors in HOME),
and BYTES SWAP DROP then gives you its size, in all HP48/49.
The following similar SysRPL program returns identical results,
plus a result for HOME which in the HP49G
appears only slightly inflated, due to some extra data
(besides the hidden directory) which is stored only in HOME:
:: CONTEXT@ OSIZE UNCOERCE %2 %/ ;
In the HP48, however, this may produce a highly inflated,
meaningless value (492150.5 in my 48GX!)
The following 65-byte SysRPL program instead sums the sizes
of every individual object in the current directory;
it works equally well in HP48 and HP49,
and works fine in HOME, where it predicts
the exact size of an archived home directory, for example;
its 49G checksum is 7F4Fh, 48 checksum is #3855h:
:: CK0NOLASTWD THIRTEEN :: CONTEXT@ LASTRAM-WORD NOT?SEMI BEGIN
   DUP RAM-WORDNAME OSIZE SEVEN OVER#< IT #2+ OVER OSIZE #+
   ROT#+SWAP PREVRAM-WORD NOT_UNTIL ; UNCOERCE %2 %/ ;
All of the above methods include all objects, even if null-named;
therefore they include any hidden variables,
and the entire hidden directory in HOME.
The last program above also avoids the 'name' BYTES bug
(when 'name' contains a ROMPTR, then the library command's size
is counted, instead of the 5.5 bytes of the ROMPTR object itself).
Here also is a 40-byte VARS2 which lists all variable names,
including those with null names
(checksum #7EC4h on 49G, #3BDEh on 48)
:: CK0NOLASTWD CONTEXT@ LASTRAM-WORD NOTcase NULL{} ZEROSWAP BEGIN
   DUP RAM-WORDNAME UNROT SWAP#1+SWAP PREVRAM-WORD NOT_UNTIL {}N ;
.
====
I guess I missed something.
Is it the case that HP49 and HP48 are being discontinued?
Is it the case that HP is not going to continue the calculator line?
What is the true story here????
====
It's item #HP90136, on the Books- Sci. & Graphing Calc. page
This is where I got my copy a couple of months ago.
Monte
====
was trying to type in 90136, HP, without luck.  Once I got to the right
sub-page (the Sci, & Graphing Calculator one), it was easy.
Also ordered The HP 48 Handbook - 2nd Ed. #AR04 (wt.: 1,00), on the
promise of many examples.  No idea if it is the Donnelly authored book
or not, but at least it is available.
I appreciate the help.
-- 
Richard...
DH/DSS: 0x886DFF31 
====
I usually do 2 3 / ENTER
--Andrew Huey
====
You can also do '2/3' ENTER
You can do algebraic entry provided you enclose the 
algebraic in tick marks as shown, or you can enter things 
using RPN entry mode, in each case followed by ENTER.
You can also key in several entries in either format before 
pressing the enter key and the HP will parse them correctly.
====
hi,
the ez way is to press EQW and enter what ever you want, then hit enter 
my two pezs
====
====
http://www.faqs.org/faqs/usenet/creating-newsgroups/part1
http://www.faqs.org/faqs/usenet/creating-newsgroups
.
====
my question is the next: 
i'm tryging do an Input (USER-RPL), it helped with temporal menu (TMENU)
and i want can it to cancel or to acept the entry with one key of said menu
as choice the ENTER o the CANCEL key of the HP48GX.
does it posible with something LIBEVAL O SYSEVAL?
is it necesary SYSRPL lenguage ?
thanks on the hand...
====
On the HP49G, how do you graph and display the inverse function of an 
equation?
ie. replacing X with Y, or f^-1(x)
Is there a command such as finverse?
-- 
Al
====
Well, it works perfectly for me. What makes you think it doesn't?
You can check if it works, by doing what you described above, that is
should return the same number like that you used (300).
[red-shift], then [0].
Greetings,
Nick.
====
changing
Keytime
enter.
somewhere
I see my problem now, I was putting in a minus sign and a greater-than sign
Alex.
====
::
  FPTR 2 A5
  DUPLEN$
  #5+
  FPTR 2 A7
  88B20
  SWAP&$
  SWAP&$
  FPTR 2 A4
;
the program that return the object is:
::
  FPTR 2 A5
  BINT11
  LAST$
  FPTR 2 A4
;
supposing that lib 256 is dettach mode.
it is correct?
suggestions to optimize memory?
 GaaK
====
I don't know what your program is doing,
since I don't use my 49G;-)
At least on the HP-48 there's a standard format
for 'Library Data'. It's not only the prologue.
To construct a LD object, you pass a list of objects,
where the first list element is a BINT containing the Library ID,
to a suitable function:
Internally,
the function calculates the list length (or size),
then, it replaces the list prolog by the =DOEXT0 prolog,
and the BINT *prolog* by the object length .
The corresponding LD extractor does the job backwards.
This way the extractor can see if the Library Data object
fed to it even belongs to its library.
As a side effect, the HP-48 routine consumes less memory,
and doesn't need to convert a HEX stream from the original object
and back, because it all happens 'in place' .
Raymond
====
Actually some of this impression is false,
as may surface below.
It's been reported that some do, but that these
failed to meet specs that HPACO had set
(leaving open the question of who is responsible if so).
However, the legends on the face plates of the ancient and revered
HP55 used to rub off pretty readily, as any owners might recall,
although those were pretty pricey, and otherwise well-made.
Hmm.. the key legends on the 6 month old HP brand Chinese made PC
keyboard on which I'm typing also seem to be starting to rub off,
so I guess I have to become a touch-typist if I keep it ;-)
I did once keep a Timex calculator watch whose face plate legend
completely flaked off, but I managed to always remember
what the keys formerly said, like gambler Big Jule
in Guys and Dolls:
Nathan Detroit:
These dice ain't got no spots on 'em; they're blank.
Big Jule:
Oh, I had the spots removed for luck,
 but I remember where the spots formerly were.
Sound clip from the 1955 film (Frank Sinatra as Nathan Detroit):
http://www.fortunecity.com/skyscraper/emacs/1438/guysndolls/dice.wav
I've had Sharp wizards like this (gold-plated on PCB, however)
which have never worn out, and both Sharp and Casio calculators
(cheapies) with very easy-to-press keys that also defy ever wearing
out or ever failing.  I don't pound on or pour things over
my keyboards to give them more abuse, however, but the one
series of HP keyboards that all failed on me were the old
HP30E/C Spice series -- 34C, 33E, and even a Woodstock HP22.
HP acknowledged and backed off those calamities by physically
redesigning and replacing them with metal-disk keyboards,
and by keeping that kind of keyboard through the
Voyager series (10C), but all of these older models
were designed for and bought by a more selective clientele,
who no doubt expected top quality and long life,
and were willing to pay some premium both for the
better operational design and the physical quality.
HP was already heat-staking keyboards together
in the Woodstock series (e.g. HP22),
and only *pressed* the parts onto the PCB
(i.e. used no solder at all) in the original Spice series --
but the manual forgot to mention that when these started failing,
the user should twist the case ends back and forth a few times,
to restore the electrical contact between parts :)
See http://www.hpmuseum.org/tech30.htm for details.
Since these assemblies are no longer ever repaired
(just completely replaced), there's no longer any real need
to spend extra time (and money) on screws;
all that matters is that the parts stay fastened,
which they seem to do very well in these HP cases.
Same as HP48; doesn't give me any trouble,
and nothing can pop out when the battery door is closed.
About as cheap as TI?  Then they should be about the same,
although some say that HP is still better.
But are there not changing expectations in today's
consumer society that to some degree prompt all this?
Some years ago, when Western Electric manufactured most of the
telephones in the USA, I heard them criticized for making them
*too*well* -- they were practically indestructable,
but people now expect everything either to soon become obsolete
or to be anyway to some extent throw away and buy another, and
they seem to shop more on price, when it comes to that trade-off;
it's what drives the consumption-oriented economy, in which
cheap stuff prevails, and thoughts of longer time scales
and the global world quality of life ahead are less important.
.
====
question Re: Normal Probability Plot. The small HP49
UserRPL program below will do the Normal Probability 
Plot by transforming the Median Rank so it can be 
plotted on ordinary rectangular axes. The transformation
equation can be found in statistics reference material.
Input is a LIST of the data in stack Level 1. Output is 
a scatter plot. The program uses the Stat Matrix of the 
HP49 so be sure to save any previous data that may be 
needed later on.
Random cheers,
Santos Lucero
=========================================================
@ NPPLOT49 -  Normal Probability Plot for HP49.
@ Language:   HP49 UserRPL.
@ Written by: Santos Lucero       rev 1   12/11/01.
@ Input:      A LIST of data in level 1 of stack.
@             Ex: {7.9 3.7 12.5 10.9 13.8 6.6 8.8}
@ Output:     Normal Probability Plot with input
@             data in horizontal axis.
@ Affects:    (BigSigma)DAT, the Statistical Matrix.
@             Previous data in Stat Matrix is wiped
@             out! Save previous data if needed!!
@ Algorithm:  1. Sort data in ascending order
@             2. n = # of data points; i = 1, 2, ...n 
@             3. Find Benard's Median Rank
@                   p[i] = (i - .3)/(n + .4)
@             4. Transform median rank to plot on
@                ordinary rectangular graph
@                  y[i] = (4.91)(p[i]^.14 - (1-p[i])^.14)
@             5. Scatter plot y[i] vs. SortedData[i]
@ Note:       V-axis units are standard dev units!
@             If hardcopy is needed(???) don't use 
@             fancy-schmancy paper! Use regular
@             rectangular graph paper to plot
@             2nd vs. 1st column of Stat Matrix.
@
<< SORT DUP SIZE DUP
    'X' DUP 1 4 ROLL
    1 SEQ .3 - 
    SWAP .4 + / DUP
    .14 ^ SWAP 1 - NEG
    .14 ^ - 4.91 * AXL
    SWAP AXL SWAP 2
    SCATRPLOT PICTURE 
====
http://www.cse.iitk.ac.in/news/primality.pdf
====
Oops!  Here it is:
http://mathworld.wolfram.com/news/2002-08-07_primetest/
====
hi NK
yea, but maybe some people happen to use it (and i use it sometimes).
and if you got a bussines or a very curious girlfriend, then you may
want to throw dirt some day. but i seriously doubt anybody need this
unless they are in trouble.
====
Hey, check out http://www.concise.co.jp/eng0731/top_eng.html
They don't make that particular slide rule anymore, but they're
still in business making rules!  :-)
        Mike (a recent and happy Concise customer!)
====
I'm very happy with mine.
It's interesting that we're discussing slide rules in comp.sys.hp48
today, and RPN calculators on sliderule@yahoogroups.com.  :-)
-- 
Wayne Brown                | When your tail's in a crack, you improvise
fwbrown@bellsouth.net      |  if you're good enough.  Otherwise you give
                           |  your pelt to the trapper.
e^(i*pi) = -1  -- Euler  |           -- John Myers Myers, 
Silverlock
====
No, not arbitrary precision, only infinite precision (exact numbers like 
pi^2).
--
Bhuvanesh
====
When I came across my old serial# ID93... 49G
(the units which have the incredible serial port hardware bug)
in a packing case, where it has been sitting for a year or two,
and found a PC which can still communicate with it,
I decided to have a go at upgrading it from 1.17 to 1.19-6,
and since I already have 1.19-6 on my emulator,
why not, I thought, just type a ROMUPLOAD command
on that emulator, and let it upload its rom to my real calc?
Apparently this was not so great an idea after all,
since immediately after the expected end-of-download message
System may not be installed properly on the calculator,
when I select Download Pack as instructed, I do not get
System Present, as assumed by the instructions;
instead I get Error - Full Check Bad CRC,
while the emulator still waits for something else to happen.
Testing Full ROM [#7] on the calculator now gives,
sure enough, Object with bad CRC, Bank 1
After wasting more time erasing banks again and repeating,
with identical results, I finally thought to try the Full ROM
test on the emulator -- bingo! the emulator's ROM also has that
same bad CRC in its own bank 1, even though it has been
operating all this while.
It is possible, I reflect, that I may have applied the last
ROM updates to the emulator using program upd49rom.exe
(which updates the emulator's rom.e49 file from a .flash file),
so perhaps I should go back and reinstall the emulator's original
ROM, for comparison, without relying on that program at all.
So now I'm setting aside my current rom.e49 (or rom.49g) file,
I'm copying the original 1.19-6 emulator ROM file (possibly from
epita.fr?) in its place, and I'm restarting the emulator,
and guess what -- already an Object with bad CRC, Bank 1,
although this is the *original* emulator ROM file for 1.19-6,
untouched by anyone, directly downloaded from the source.
I also did another 1.19-6 download from www.hpcalc.org
and got identical md5 checksums on its emulator rom file
which was dated 2001/10/03:
d5fddb6f16b443b6749a36996b1053f8  rom.49g [new]
d5fddb6f16b443b6749a36996b1053f8  rom.e49 [old]
The flash files also match:
d881753cebe3318f86e5b5615e301ebf  hp49119-6.flash [new]
d881753cebe3318f86e5b5615e301ebf  hp491196.fla [my old one]
But the emulator ROM file apparently comes with a built-in
Object with bad CRC, Bank 1,
and this in turn appears to infect any calculator
which tries to download ROM from the emulator.
Is there any way to fix the bad ROM in the emulator?
I thought perhaps to try to re-upload to Emu48 a good ROM from my
calculator (meanwhile re-updated via FDP.EXE from the flash file),
but I can't seem to get the emulator to go into that mode
which is required to get to the no system menu.
Any ideas?
Ciao.
.
====
Hallo John,
short answer to your problem:
Please _remove_ the
  Patch  BEEP.49G
line in your KML script. The ROM is patched for beep emulation.
But I must confess that one Emu48 version, but I don't know which one (any
final or beta in the last time) made the PATCH changes permanent on my
systems. In such case I update the ROM on the PC with the upd49rom.exe
tool to fix this.
Christoph
====
This indeed corrects the Bad CRC in Bank 1, thanks!
I wonder whether there exists a second patch
which could correct the CRC to match the first patch
(perhaps by changing something other than the CRC itself,
so that it could remain constant for all ROM versions)?
.
====
Changing another unused byte(s) to get the correct CRC is quite impossible,
therefore we use CRC's to protect the data. The only solution is to modify
the CRC itself. The calculation methods, or better the position of the 
CRC's
are different in the calculators emulated by Emu48, so there's no general
solution. When you know the position of the CRC in ROM you can patch the 
CRC
by the same way like the beep patch (by adding one or more lines in the
patch file). And here you are right, that only works for one ROM revision. 

I
use this method in the HP28C skin for Emu28 for example.
In a Lewis chip family calculator emulator (not published) I use the
technique that after patching (and only then) all ROM CRC's are
automatically updated. This is possible because all emulated calculators
(with only one very minor exception) use the same calculation scheme and 
the
same positions of the CRC's in ROM.
Christoph
====
I have been having difficulty in finding a way to transferr files between
linux and my 49. The ckermit rpm i dowloaded requires glibc 2.3 which I
don't have time to download over dialup. Can anyone help?
====
you can use sx / rx, which is in package lrzsz, like this :
but on 49, a silly delay happens. I know a software exists which avoids
it, but I can't remember its name. Yoann, can you remember ?
++
-- 
====
Could anyone let me know whether official pdf copies of the HP48G
Series User Guide and HP48G Series Advanced User Reference from HP
are available for download anywhere? HP Support does not provide them
and Google searches have been unsuccessful. (Note: The HP49G guides
are available.)
I do have Eric Rechlin's HpCalc.org scanned pdf copy of the HP48G
Series User Guide (which is much appreciated).  However, the HP48G
Advanced User Reference does not appear to be available for download
in either official or unofficial form anywhere.
Rodney
====
Where are the HP 49G guides available? Not at the link given in the manual
that comes with it. Although I haven't really looked either... ;)
Alex
====
It would be an HP-12C. -- about $50-70 new.
Gene
--
* These statements and opinions are mine alone and do not reflect my
employer's views. *
====
I agree. the 12C is still available, and the last of the golden era. Get 
one
soon, you never know!
--
Thierry Morissette
thm47@msn.com
of
====
I own an HP Infrared Printer ver A and one ver B for the HP-41/28/48
etc. My intention is to get a voltage adaptor for it so I need not use
expensive batteries all the time.
Is there a difference between the thermal printer ver A respectively
ver B when it comes to the volatges?
I live in Sweden where normal output voltage is 230 V.
I have a manual for the A version stating 9-12 V DC with 500-1500 mA.
But where is the positive and where is the negative to be inserted?
So, anybody have a hint?
/Matti
====
One of the previous link URLs actually ends with a period,
which news readers may not recognize as part of the URL,
hence it may appear to be a broken link, unless re-written:
Re: AC adapter for HP 95LX
http://groups.google.com/groups?selm=65300034%40hpcvra.cv.hp.com%2e
.
====
When looking at the latest HP calculators, I noticed that RPN is going
to be abandonned? (I was looking at HP's website and noticed that the
HP49 and the HP32 are discontinued?)
I sincerely hope not... If I would be obliged to use an algebraic
calc. again it would be like driving a car with 5 gears reverse and
one forward.
I wonder if any other brand offers RPN calcs?
I use my HP48G and GX every day, but for the ordinary  +-*/ operations
it is sometimes a bit too complicated. (Not to me, but other people in
our company)
Did the HP32 have multiple stacklines? I'm just truying to find a
solution for the future to secure the usage of RPN calcs.
The only problem with RPN is, that once tried and mastered, there is
no turn back... RPN is just too easy to use.
Many people who use RPN now in our company have real problems using an
algebraic calc again.
Greetings,
Frank
====
Frank,
I sincerely hope RPN won't be abandoned! It is the only entry mode that
really makes sense! As for the cheapest calc it is probably still an HP of
some flavor...
Rod
====
Well Kiel, I have a 49G with a screen that I'm pretty sure is a good one
because no matter how hard I try I can't get the rainbow effect, but I did
recently find these pictures..
http://holyjoe.net/rainbow.htm
Only the first two are of the rainbow effect, but I think it should give 
you
an idea of wheather or not you have a bad screen.
Alex.
====
Bonsoir,
Voici un programme pour mettre un polyn.99me du second degr.8e ax^2+bx+c 

sous
forme canonique c'est-.88-dire a[(x+b/(2a)]^2-delta/(4a^2)] avec
delta=b^2-4ac.
%%HP: T(1)A(R)F(.);
.82 0 ō a b c a1
  .82 a 'a1' STO
1 'a' STO
b a1 / 'b' STO
c a1 / 'c' STO
a X 2 ^ * b X * + c +
y b 2 / a / EVAL X + =
DUP ROT ROT SOLVEVX
SUBST DISTRIB EVAL DISTRIB
SWAP DUP2 SUBST a1
SWAP *
  é
é
Avec les donn.8ees 1 2 10 3 / +/-  j'obtiens
y^2 - 16/3
y=1+X
(1+X)^2-16/3
mais avec
1 2 3 +/- j'obtiens
y^2 +-4
y=1+X
(1+X)^2+-4
Comment .8eviter l'.8ecriture +-4 ?
Merci d'avance pour vos r.8eponses.
--
@+
Jean-Marc Rebert
jm.rebert@calixo.net
====
Petite faute de frappe, lire  13/3 au lieu de 16/3
avfl2o$s02$1@rex.ip-plus.net...
====
[this is about neither True type fonts nor Debug4x,
 but it does relate to what was written before]
WG asserts:
That's how eBay turned the legitimate auction format into a kind
of sham which it keeps trying to persuade people is a real auction;
so you've decided that your last bid is the end of the discussion,
the item is now sold, and no follow-up bids are allowed?
But you keep inserting it into your sig every time, right?
A far too narrow consideration; better to find out what *else*
might have a much better overall benefit to an entire society,
especially if it would, in the end, lead to reductions
in the presence of material of basically destructive-only nature,
and its eventual supplantation by something of more positive value.
Again a tunnel-vision approach, as above, especially when only
two possibilities are compared (more or fewer private firearms),
not anything else which might affect the entire society
in such manner as to reduce crime without increasing firearms at all,
or to even reduce crime to such a degree that the incentive
to stock firearms would significantly decline.
The infamous red-herring of this topic...
So, if only Iraqi citizens had more private handguns, then perhaps
they would be able to overthrow SH's WMD-equipped military?
Or is it, at the opposite extreme, that the military primarily
consists of citizens grabbing their muskets whenever civil defense
is needed?  (This is the explicit rationale stated in the U.S.
Constitution for RKBA, but it doesn't fit present day life on Earth).
And if you think that your own government might be ready to turn arms
against you, remember these few things:  You freely elected them
to their offices, you authorized and paid them to maintain armies
(for your benefit), many of us make our livings from their industries,
and those people who joined them are your fellow citizens, so any
such event would really be a civil war, of one part of a nation
against another, which was fostered by our own behavior.
Reflect on how such things can come about,
and find a more plausible means of prevention than the notion
of your personal guns against tanks, gunships, bombers and WMD,
or even of one socially isolated group waging war against another
in the streets with their small arms, if it came to that.
Your emphasis, like that of governments during arms races
(which are continuing into the WMD arena today), is primarily
on the *hardware*, and is not directed towards the *people*
whose collective and individual feelings are the only fundamental
cause of whatever is either good or bad in society,
and of the actions which will follow from those feelings.
Narrowing the vision to think that human discontent can be corralled
and kept away from oneself through possession and use of personal
deadly arms, is to me a recipe for disastrous final consequences,
some of which are very near at hand in the world today,
where massively destructive capability can be concentrated
in even one person's hands, and often falls
into very mentally disturbed hands,
whether a rampaging student or a rampaging dictator.
Want protection against personal assault?
There are plenty of less deadly things to carry or keep at home;
even the more civilized police forces use them when possible.
The cause of all violence starts in a human mind,
and its most effective prevention for entire societies lies again
in its relief within human minds; explaining how this is
achievable is the hardest part of my task, because our
entire educational orientation is so materialistic
that it leaves us far less attentive to the experience
within our own consciousness, and between ourselves
and other people.
If fatigue frazzles you nerves and lowers your emotional
boiling point, only a reversal of fatigue can cure you,
and only some time spent in a different state of consciousness
(known to you as sleep) is effective, as well as absurdly simple.
You have to balance your activity with rest
to remain a happy, productive and peaceful human being,
rather than degenerating into a state of lower level experience
of life, which could include poorer health, poorer disposition,
imbalance of judgment, lack of satisfaction,
anger, misery and violence.
You have to personally alternate waking with sleeping,
have alternating periods of different brain functioning,
to live life in a state of greater creativity, accomplishment,
and fulfillment.
Your metabolic rate is generally higher while awake,
lower while asleep, but in fact it is possible to be completely
awake and alert, yet have a lower metabolic rate than in deep sleep,
and it is found that one experiences finer well being, health,
judgment, creativity, satisfaction, achievement, sense of true
freedom, and peace of heart when this is added to one's daily routine.
So we add it to the daily routine, and everyone feels finer;
feeling finer, they act more positively.  This gets around,
it affects the whole community, people act more supportively,
and the feeling by each person of a need to keep a gun at hand
becomes more remote.  So I'd rather emphasize the right for
everyone to learn how to generate this finer experience of
life and spread it around the community, and become a more
harmonious society, than to dwell on the empty and inferior right
to possess weapons of devastation to turn against one another.
Oh look, we've been discussing this already for about 3.5 years:
http://groups.google.com/groups?threadm=19990629005603.29980.00005033%40ng63

.aol.com
Well, you have spoken and I have answered,
so please, let's not hear anything more about it, okay?  :-)
http://www.mum.edu
.
====
More or less agree, but for the sake of completness:
Some additional arms free and also nearly crime free zones:
http://www.uni-duisburg.de
http://www.unilu.ch
http://www.ethz.ch/
http://www.imos-consulting.com
And some additional nearly crime free but in no sense arms free zones.
(A gun in almost every house, and nonetheless very very low
criminality.)
http://www.hergiswil.ch
http://www stans.ch
And what a mystery it is, to know that in the above nearly crime free
zones there are no gurus, bodhisattvas, saints, dervishes, prophets,
holy men, and all the other drugs that humans took for thousends of
years in order to have some good wise guy in heavens to look at,
when we want somebody to tell us the truth, because we are too
convenient to find it out ourselves and for ourselves. And maybe
because we are too scared to face the fact that the truth could be
that there is no truth at all.
Oops, I was mistaken. In fact at the arms free zone
www.imos-consulting.com you will find a guru, our boss, a technology
guru, who of course as John says, seems (has?) to be the leader in a
world of violence, since he is an atheist and science-technology-freak
(like me). He is one of the most pacifistic, good hearted, and
peaceful persons that I have ever known, but if the
we-know-all-truth-of-the-universe-gurus declare him (and me) to
follow a way that will lead to destruction, then this has to be the
truth.
Greetings from somebody who lives since three years in a house full of
guns, and still hasn't experienced any violence, or even bad behavior
by the bad neighbours that have guns. ;-)
The snake in the grass of your paradise. Ssssssssss ;-)
Nick.
====
I can't use my 49g because it is too powerful. 89 is the same, and it 
has been a LOOOONG time since I have really used my 48gx.
Sooo, I just don't remember, can the 48gx do all the symbolic without 
erable and alg48? I didn't think so, but I just don't remember! Is there 
anything else on the 48gx that might disqualify it?
I really don't want to have to use an algebraic calc, I am just too used 
to rpn for regular stuff, and I really don't want to have to buy a ti.
Matt
====
I just finished DiffyQs last emester and the only thing I used a
calculator for was Euler's Method and runga-Kutta (sp?) method  which
are decimal approximation for DEs.  For that I used a TI 83+ and a
program my prof gave to us. 
You will just have to do some integration and differentiation on some
problems that hopefully your book and prof will rig so that the stuff
you are integrating/differentiating is simple first semester calculus
stuff. 
Bye
   http://www.newsfeed.com       The #1 Newsgroup Service in the World!
====
There are roughly 17 *other* TI emulators on ticalc.org
http://www.ticalc.org/programming/emulators/software.html
as well...surely there must be a bug or two in there somewhere....but, yes,
Virtual TI is an excellent piece of work.
    Isn't TI's RPN an environment and not run natively...and, you know, 

I
think it is somewhat remarkable that the ancient Saturn-based HP holds up 

as
well as it does in comparison to the much more modern 68000-based TI's....I
would think that these Mac-based TI's would just blow the doors off all the
competiton, but that's not the case--the HP's still acquit themselves very
well....
the
think
calc.
The number of users doesn't have much to do with whether or not they
understand what they're doing.....
Happy holidays,
Roger
====
If RPN is only an a non native environment of the TIs then the
comparison is not very fair. If someone ever makes a native RPN
environment for the TI, then I bet the TI will be much faster.
If the software of the TIs would be optimized in a comparable degree
to the software of the HP49, then the TIs would really blow the doors
off all competition. If software optimization is approx. the same,
then it is the hardware that decides about performance, even if this
group doesn't want to hear this. As the time passes by the software of
the TIs will be optimized, and I guess that the TIs still have much
much optimization potential, while the HP49G represents more or less
the endth degree of any tricks that you can find in order to increase
its performance. So the future, as it looks today, doesn't belong to
the archaic HP49G and its paleontological processor.
In addition we have the following fact: If we compare performances of,
say adding two numbers on both machines, we are not going to find
anything special. For me as a user, it is totally uninteresting if
this operation takes 1 millisecond or 100 microseconds. I just can't
tell the difference. If we compare, we should compare lengthy
operations or programs. Guess what happens when the software of the
TIs is as highly optimized as the software of the HP49G. It is short
sighted to believe that this optimization will never happen and that
HP is the one and only company that can do that.
This is a constructed and very cheap excuse - and also short sighted.
Of course everybody that works with Maple and Mathematica also doesn't
know what he/she is doing? And only HP-users are smart guys because
they can work with RPN and because they can find the bugs of the 49?
The bugs of the TI92 were also found but there was not so much noise
because the bugs can be counted on the fingers of one hand, quite
contrary to the HP49G. And don't forget that if you keep examining
each and every result that the HP49G or any other CAS returns, then
the whole concept of using the CAS becomes questionable. If we do all
by hand again, just to check the correctness of the results, then what
did we buy the machine for? Imagine what would happen if you had to
manually check the correctness of a variation of parameters for energy
optimization of a wave function of a molecule. How long will it take
to do that? One life? Do all people that use mopac, or gaussian have
to repeat by hand the hundreds or even thousands of numeric
integrations that the programs do? Do designers and graphics editors
need to repeat, say rotation or blur of a graphic, just to check that
they got the correct result? And do I, normal user of my refrigerator,
need to repeat thermodynamics calculations everyday just to check that
the machine really cools down to the the correct temperature?
Indeed, quantity doesn't have to do with quality, but if so, then
neither in positive *nor in negative* sense. We would better stop
implying that the big number of TI users is automatically a dumpness
indicator of TI users and that the few of us are the smarties just
because we are few.
All of the above is without any intention to harm. It's only that my
sight presents me a rather sad picture when I think about the future
of HP advanced calculators. Perhaps I should change my glasses..;-)
Greetings,
Nick.
====
Imagine what will happen if there were no people checking the CAS
results -- your fridge would certainly fail to function after 3 days
of normal work (or would keep the good and pretty constant 25C degrees
in the camera ;), designer's computers would refuse to do any blur
filters, etc., etc.
Those, who do check the results are called testers -- and most people
in this group cannot be called just bare users of HP calcs.
-- 
Computers are not intelligent. They only think they are.
====
Both the saturn and the 68000 were developed in the late seventies,
both were first deployed round 1979. In this terms they have both the
same paleontological  age..;)
I personally think, that the TiOS software *is* optimized, but not to
an extend found in the HP49G. Where do you nowadays find an OS (even
in  embedded hardware) which only uses 70 to 150K of RAM (seems to
depend on the actual hardware, my Voyage 200 has only ca 109000 bytes
free, whereas an TI89 should have 188K free), when started? But
compare it to an HP49G, which uses only 3K ram after warmboot. 
After all, the HP is optimized in such an extrem way, because its the
only chance to compete with the better hardware of the Ti68ks. 
IĒm
sure, Ti will never optimize TiOS so much, because the corresponding
development cost are just to high. If Ti feel the necessarity of
making an faster calc, the will mainly use an faster coldfire or
dragonball cpu core. Maybe they will port their TiOS or AMS to a fast
Risc-CPU, which should be much easier then porting SysRPL because of
the much more conventional OS design of TiOS.
-- 
Mathias Habel
mathias.habel_no-spam_@t-online.de
Remove _no-spam_ before replying
====
That's what I did with my Atari 8 bit games and computers...now I feel
like an idiot!
====
Are you referring to the file rpn2022.zip? I could not find any help
file associated with it... or is it self explanatory if you're
familiar with HP's RPN.
====
You can get the (pretty extensive) documentation here:
http://www.paxm.org/symbulator/download/rpn.html
--
Bhuvanesh
====
Speaking of optomizations...
I hate to give Hp people fodder, but some of the code that is used in
the AMS is absolute junk.  I've been coding and reverse engineering
parts of it for a couple years.
Here's an example of the code the compiler ti used has been know to
generate:
 swap d0
 swap d0
 swap d0
here is another:
 swap d0
 swap d0
For those of you who don't know what the swap instruction does I'll
tell ya.  It exchanges the upper half of the register with the lower
half.  So having two swaps in a row is redundant.  Having three is
stupidity.
====
I have a 38G and a comms package for Mac. When I do an image transfer the
program works fine, when I try a file trasnfer it hangs. Doe anyone have 
any
sugestions?
====
AFAIK, you can't with built-in commands
====
If your current directory is not HOME itself,
then PATH RCL recalls the current directory (this errors in HOME),
and BYTES SWAP DROP then gives you its size, in all HP48/49.
The following similar SysRPL program returns identical results,
plus a result for HOME which in the HP49G
appears only slightly inflated, due to some extra data
(besides the hidden directory) which is stored only in HOME:
:: CONTEXT@ OSIZE UNCOERCE %2 %/ ;
In the HP48, however, this may produce a highly inflated,
meaningless value (492150.5 in my 48GX!)
The following 65-byte SysRPL program instead sums the sizes
of every individual object in the current directory;
it works equally well in HP48 and HP49,
and works fine in HOME, where it predicts
the exact size of an archived home directory, for example;
its 49G checksum is 7F4Fh, 48 checksum is #3855h:
:: CK0NOLASTWD THIRTEEN :: CONTEXT@ LASTRAM-WORD NOT?SEMI BEGIN
   DUP RAM-WORDNAME OSIZE SEVEN OVER#< IT #2+ OVER OSIZE #+
   ROT#+SWAP PREVRAM-WORD NOT_UNTIL ; UNCOERCE %2 %/ ;
All of the above methods include all objects, even if null-named;
therefore they include any hidden variables,
and the entire hidden directory in HOME.
The last program above also avoids the 'name' BYTES bug
(when 'name' contains a ROMPTR, then the library command's size
is counted, instead of the 5.5 bytes of the ROMPTR object itself).
Here also is a 40-byte VARS2 which lists all variable names,
including those with null names
(checksum #7EC4h on 49G, #3BDEh on 48)
:: CK0NOLASTWD CONTEXT@ LASTRAM-WORD NOTcase NULL{} ZEROSWAP BEGIN
   DUP RAM-WORDNAME UNROT SWAP#1+SWAP PREVRAM-WORD NOT_UNTIL {}N ;
.
====
I guess I missed something.
Is it the case that HP49 and HP48 are being discontinued?
Is it the case that HP is not going to continue the calculator line?
What is the true story here????
====
It's item #HP90136, on the Books- Sci. & Graphing Calc. page
This is where I got my copy a couple of months ago.
Monte
====
was trying to type in 90136, HP, without luck.  Once I got to the right
sub-page (the Sci, & Graphing Calculator one), it was easy.
Also ordered The HP 48 Handbook - 2nd Ed. #AR04 (wt.: 1,00), on the
promise of many examples.  No idea if it is the Donnelly authored book
or not, but at least it is available.
I appreciate the help.
-- 
Richard...
DH/DSS: 0x886DFF31 
====
I usually do 2 3 / ENTER
--Andrew Huey
====
You can also do '2/3' ENTER
You can do algebraic entry provided you enclose the 
algebraic in tick marks as shown, or you can enter things 
using RPN entry mode, in each case followed by ENTER.
You can also key in several entries in either format before 
pressing the enter key and the HP will parse them correctly.
====
hi,
the ez way is to press EQW and enter what ever you want, then hit enter 
my two pezs
====
====
http://www.faqs.org/faqs/usenet/creating-newsgroups/part1
http://www.faqs.org/faqs/usenet/creating-newsgroups
.
====
my question is the next: 
i'm tryging do an Input (USER-RPL), it helped with temporal menu (TMENU)
and i want can it to cancel or to acept the entry with one key of said menu
as choice the ENTER o the CANCEL key of the HP48GX.
does it posible with something LIBEVAL O SYSEVAL?
is it necesary SYSRPL lenguage ?
thanks on the hand...
====
On the HP49G, how do you graph and display the inverse function of an 
equation?
ie. replacing X with Y, or f^-1(x)
Is there a command such as finverse?
-- 
Al
====
Well, it works perfectly for me. What makes you think it doesn't?
You can check if it works, by doing what you described above, that is
should return the same number like that you used (300).
[red-shift], then [0].
Greetings,
Nick.
====
changing
Keytime
enter.
somewhere
I see my problem now, I was putting in a minus sign and a greater-than sign
Alex.
====
::
  FPTR 2 A5
  DUPLEN$
  #5+
  FPTR 2 A7
  88B20
  SWAP&$
  SWAP&$
  FPTR 2 A4
;
the program that return the object is:
::
  FPTR 2 A5
  BINT11
  LAST$
  FPTR 2 A4
;
supposing that lib 256 is dettach mode.
it is correct?
suggestions to optimize memory?
 GaaK
====
I don't know what your program is doing,
since I don't use my 49G;-)
At least on the HP-48 there's a standard format
for 'Library Data'. It's not only the prologue.
To construct a LD object, you pass a list of objects,
where the first list element is a BINT containing the Library ID,
to a suitable function:
Internally,
the function calculates the list length (or size),
then, it replaces the list prolog by the =DOEXT0 prolog,
and the BINT *prolog* by the object length .
The corresponding LD extractor does the job backwards.
This way the extractor can see if the Library Data object
fed to it even belongs to its library.
As a side effect, the HP-48 routine consumes less memory,
and doesn't need to convert a HEX stream from the original object
and back, because it all happens 'in place' .
Raymond
====
Actually some of this impression is false,
as may surface below.
It's been reported that some do, but that these
failed to meet specs that HPACO had set
(leaving open the question of who is responsible if so).
However, the legends on the face plates of the ancient and revered
HP55 used to rub off pretty readily, as any owners might recall,
although those were pretty pricey, and otherwise well-made.
Hmm.. the key legends on the 6 month old HP brand Chinese made PC
keyboard on which I'm typing also seem to be starting to rub off,
so I guess I have to become a touch-typist if I keep it ;-)
I did once keep a Timex calculator watch whose face plate legend
completely flaked off, but I managed to always remember
what the keys formerly said, like gambler Big Jule
in Guys and Dolls:
Nathan Detroit:
These dice ain't got no spots on 'em; they're blank.
Big Jule:
Oh, I had the spots removed for luck,
 but I remember where the spots formerly were.
Sound clip from the 1955 film (Frank Sinatra as Nathan Detroit):
http://www.fortunecity.com/skyscraper/emacs/1438/guysndolls/dice.wav
I've had Sharp wizards like this (gold-plated on PCB, however)
which have never worn out, and both Sharp and Casio calculators
(cheapies) with very easy-to-press keys that also defy ever wearing
out or ever failing.  I don't pound on or pour things over
my keyboards to give them more abuse, however, but the one
series of HP keyboards that all failed on me were the old
HP30E/C Spice series -- 34C, 33E, and even a Woodstock HP22.
HP acknowledged and backed off those calamities by physically
redesigning and replacing them with metal-disk keyboards,
and by keeping that kind of keyboard through the
Voyager series (10C), but all of these older models
were designed for and bought by a more selective clientele,
who no doubt expected top quality and long life,
and were willing to pay some premium both for the
better operational design and the physical quality.
HP was already heat-staking keyboards together
in the Woodstock series (e.g. HP22),
and only *pressed* the parts onto the PCB
(i.e. used no solder at all) in the original Spice series --
but the manual forgot to mention that when these started failing,
the user should twist the case ends back and forth a few times,
to restore the electrical contact between parts :)
See http://www.hpmuseum.org/tech30.htm for details.
Since these assemblies are no longer ever repaired
(just completely replaced), there's no longer any real need
to spend extra time (and money) on screws;
all that matters is that the parts stay fastened,
which they seem to do very well in these HP cases.
Same as HP48; doesn't give me any trouble,
and nothing can pop out when the battery door is closed.
About as cheap as TI?  Then they should be about the same,
although some say that HP is still better.
But are there not changing expectations in today's
consumer society that to some degree prompt all this?
Some years ago, when Western Electric manufactured most of the
telephones in the USA, I heard them criticized for making them
*too*well* -- they were practically indestructable,
but people now expect everything either to soon become obsolete
or to be anyway to some extent throw away and buy another, and
they seem to shop more on price, when it comes to that trade-off;
it's what drives the consumption-oriented economy, in which
cheap stuff prevails, and thoughts of longer time scales
and the global world quality of life ahead are less important.
.
====
question Re: Normal Probability Plot. The small HP49
UserRPL program below will do the Normal Probability 
Plot by transforming the Median Rank so it can be 
plotted on ordinary rectangular axes. The transformation
equation can be found in statistics reference material.
Input is a LIST of the data in stack Level 1. Output is 
a scatter plot. The program uses the Stat Matrix of the 
HP49 so be sure to save any previous data that may be 
needed later on.
Random cheers,
Santos Lucero
=========================================================
@ NPPLOT49 -  Normal Probability Plot for HP49.
@ Language:   HP49 UserRPL.
@ Written by: Santos Lucero       rev 1   12/11/01.
@ Input:      A LIST of data in level 1 of stack.
@             Ex: {7.9 3.7 12.5 10.9 13.8 6.6 8.8}
@ Output:     Normal Probability Plot with input
@             data in horizontal axis.
@ Affects:    (BigSigma)DAT, the Statistical Matrix.
@             Previous data in Stat Matrix is wiped
@             out! Save previous data if needed!!
@ Algorithm:  1. Sort data in ascending order
@             2. n = # of data points; i = 1, 2, ...n 
@             3. Find Benard's Median Rank
@                   p[i] = (i - .3)/(n + .4)
@             4. Transform median rank to plot on
@                ordinary rectangular graph
@                  y[i] = (4.91)(p[i]^.14 - (1-p[i])^.14)
@             5. Scatter plot y[i] vs. SortedData[i]
@ Note:       V-axis units are standard dev units!
@             If hardcopy is needed(???) don't use 
@             fancy-schmancy paper! Use regular
@             rectangular graph paper to plot
@             2nd vs. 1st column of Stat Matrix.
@
<< SORT DUP SIZE DUP
    'X' DUP 1 4 ROLL
    1 SEQ .3 - 
    SWAP .4 + / DUP
    .14 ^ SWAP 1 - NEG
    .14 ^ - 4.91 * AXL
    SWAP AXL SWAP 2
    SCATRPLOT PICTURE 
====
http://www.cse.iitk.ac.in/news/primality.pdf
====
Oops!  Here it is:
http://mathworld.wolfram.com/news/2002-08-07_primetest/
====
hi NK
yea, but maybe some people happen to use it (and i use it sometimes).
and if you got a bussines or a very curious girlfriend, then you may
want to throw dirt some day. but i seriously doubt anybody need this
unless they are in trouble.
====
Hey, check out http://www.concise.co.jp/eng0731/top_eng.html
They don't make that particular slide rule anymore, but they're
still in business making rules!  :-)
        Mike (a recent and happy Concise customer!)
====
I'm very happy with mine.
It's interesting that we're discussing slide rules in comp.sys.hp48
today, and RPN calculators on sliderule@yahoogroups.com.  :-)
-- 
Wayne Brown                | When your tail's in a crack, you improvise
fwbrown@bellsouth.net      |  if you're good enough.  Otherwise you give
                           |  your pelt to the trapper.
e^(i*pi) = -1  -- Euler  |           -- John Myers Myers, 
Silverlock
====
No, not arbitrary precision, only infinite precision (exact numbers like 
pi^2).
--
Bhuvanesh
====
When I came across my old serial# ID93... 49G
(the units which have the incredible serial port hardware bug)
in a packing case, where it has been sitting for a year or two,
and found a PC which can still communicate with it,
I decided to have a go at upgrading it from 1.17 to 1.19-6,
and since I already have 1.19-6 on my emulator,
why not, I thought, just type a ROMUPLOAD command
on that emulator, and let it upload its rom to my real calc?
Apparently this was not so great an idea after all,
since immediately after the expected end-of-download message
System may not be installed properly on the calculator,
when I select Download Pack as instructed, I do not get
System Present, as assumed by the instructions;
instead I get Error - Full Check Bad CRC,
while the emulator still waits for something else to happen.
Testing Full ROM [#7] on the calculator now gives,
sure enough, Object with bad CRC, Bank 1
After wasting more time erasing banks again and repeating,
with identical results, I finally thought to try the Full ROM
test on the emulator -- bingo! the emulator's ROM also has that
same bad CRC in its own bank 1, even though it has been
operating all this while.
It is possible, I reflect, that I may have applied the last
ROM updates to the emulator using program upd49rom.exe
(which updates the emulator's rom.e49 file from a .flash file),
so perhaps I should go back and reinstall the emulator's original
ROM, for comparison, without relying on that program at all.
So now I'm setting aside my current rom.e49 (or rom.49g) file,
I'm copying the original 1.19-6 emulator ROM file (possibly from
epita.fr?) in its place, and I'm restarting the emulator,
and guess what -- already an Object with bad CRC, Bank 1,
although this is the *original* emulator ROM file for 1.19-6,
untouched by anyone, directly downloaded from the source.
I also did another 1.19-6 download from www.hpcalc.org
and got identical md5 checksums on its emulator rom file
which was dated 2001/10/03:
d5fddb6f16b443b6749a36996b1053f8  rom.49g [new]
d5fddb6f16b443b6749a36996b1053f8  rom.e49 [old]
The flash files also match:
d881753cebe3318f86e5b5615e301ebf  hp49119-6.flash [new]
d881753cebe3318f86e5b5615e301ebf  hp491196.fla [my old one]
But the emulator ROM file apparently comes with a built-in
Object with bad CRC, Bank 1,
and this in turn appears to infect any calculator
which tries to download ROM from the emulator.
Is there any way to fix the bad ROM in the emulator?
I thought perhaps to try to re-upload to Emu48 a good ROM from my
calculator (meanwhile re-updated via FDP.EXE from the flash file),
but I can't seem to get the emulator to go into that mode
which is required to get to the no system menu.
Any ideas?
Ciao.
.
====
Hallo John,
short answer to your problem:
Please _remove_ the
  Patch  BEEP.49G
line in your KML script. The ROM is patched for beep emulation.
But I must confess that one Emu48 version, but I don't know which one (any
final or beta in the last time) made the PATCH changes permanent on my
systems. In such case I update the ROM on the PC with the upd49rom.exe
tool to fix this.
Christoph
====
This indeed corrects the Bad CRC in Bank 1, thanks!
I wonder whether there exists a second patch
which could correct the CRC to match the first patch
(perhaps by changing something other than the CRC itself,
so that it could remain constant for all ROM versions)?
.
====
Changing another unused byte(s) to get the correct CRC is quite impossible,
therefore we use CRC's to protect the data. The only solution is to modify
the CRC itself. The calculation methods, or better the position of the 
CRC's
are different in the calculators emulated by Emu48, so there's no general
solution. When you know the position of the CRC in ROM you can patch the 
CRC
by the same way like the beep patch (by adding one or more lines in the
patch file). And here you are right, that only works for one ROM revision. 

I
use this method in the HP28C skin for Emu28 for example.
In a Lewis chip family calculator emulator (not published) I use the
technique that after patching (and only then) all ROM CRC's are
automatically updated. This is possible because all emulated calculators
(with only one very minor exception) use the same calculation scheme and 
the
same positions of the CRC's in ROM.
Christoph
====
I have been having difficulty in finding a way to transferr files between
linux and my 49. The ckermit rpm i dowloaded requires glibc 2.3 which I
don't have time to download over dialup. Can anyone help?
====
you can use sx / rx, which is in package lrzsz, like this :
but on 49, a silly delay happens. I know a software exists which avoids
it, but I can't remember its name. Yoann, can you remember ?
++
-- 
====
Could anyone let me know whether official pdf copies of the HP48G
Series User Guide and HP48G Series Advanced User Reference from HP
are available for download anywhere? HP Support does not provide them
and Google searches have been unsuccessful. (Note: The HP49G guides
are available.)
I do have Eric Rechlin's HpCalc.org scanned pdf copy of the HP48G
Series User Guide (which is much appreciated).  However, the HP48G
Advanced User Reference does not appear to be available for download
in either official or unofficial form anywhere.
Rodney
====
Where are the HP 49G guides available? Not at the link given in the manual
that comes with it. Although I haven't really looked either... ;)
Alex
====
It would be an HP-12C. -- about $50-70 new.
Gene
--
* These statements and opinions are mine alone and do not reflect my
employer's views. *
====
I agree. the 12C is still available, and the last of the golden era. Get 
one
soon, you never know!
--
Thierry Morissette
thm47@msn.com
of
====
I own an HP Infrared Printer ver A and one ver B for the HP-41/28/48
etc. My intention is to get a voltage adaptor for it so I need not use
expensive batteries all the time.
Is there a difference between the thermal printer ver A respectively
ver B when it comes to the volatges?
I live in Sweden where normal output voltage is 230 V.
I have a manual for the A version stating 9-12 V DC with 500-1500 mA.
But where is the positive and where is the negative to be inserted?
So, anybody have a hint?
/Matti
====
One of the previous link URLs actually ends with a period,
which news readers may not recognize as part of the URL,
hence it may appear to be a broken link, unless re-written:
Re: AC adapter for HP 95LX
http://groups.google.com/groups?selm=65300034%40hpcvra.cv.hp.com%2e
.
====
When looking at the latest HP calculators, I noticed that RPN is going
to be abandonned? (I was looking at HP's website and noticed that the
HP49 and the HP32 are discontinued?)
I sincerely hope not... If I would be obliged to use an algebraic
calc. again it would be like driving a car with 5 gears reverse and
one forward.
I wonder if any other brand offers RPN calcs?
I use my HP48G and GX every day, but for the ordinary  +-*/ operations
it is sometimes a bit too complicated. (Not to me, but other people in
our company)
Did the HP32 have multiple stacklines? I'm just truying to find a
solution for the future to secure the usage of RPN calcs.
The only problem with RPN is, that once tried and mastered, there is
no turn back... RPN is just too easy to use.
Many people who use RPN now in our company have real problems using an
algebraic calc again.
Greetings,
Frank
====
Frank,
I sincerely hope RPN won't be abandoned! It is the only entry mode that
really makes sense! As for the cheapest calc it is probably still an HP of
some flavor...
Rod
====
Well Kiel, I have a 49G with a screen that I'm pretty sure is a good one
because no matter how hard I try I can't get the rainbow effect, but I did
recently find these pictures..
http://holyjoe.net/rainbow.htm
Only the first two are of the rainbow effect, but I think it should give 
you
an idea of wheather or not you have a bad screen.
Alex.
====
1) About keytime:
The 49 stores a value that represents how long it takes before evaluating 

a
key.  This is to prevent accidental double presses of keys.  The regular
value is too high for most people.  To change it, put a number like 300 
into
I press 300 + (rightshift) + 0 + enter keytime + enter and get the 
message
Invalid syntax. I canĒt seem to find the right 
keystroke-combination...
What am I doing wrong?
2) Whenever you guys are working with variables isnĒt there 
a 
faster way to
purge them, instead having to go into the files menu or instead of having 

to
press rightshift + eqw + var + corresponding function key  + 

tool
+ F5...... It must be possibly to make it faster or anyone???
Martin J.
====
**** Post for FREE via your newsreader at post.usenet.com ****
a
into
message
Make sure you have RPN mode on than:
300  (SPC) (right-shift) 0 (ALPHA) (ALPHA) K E Y T I M E  (enter)
I think waht you did wrong was that you did not press SPC after 300.
to
to
tool
Well, maybe it is not the best way to do that, but I:
clear the stack
1. press VAR
2. left-shift
3. push one of the soft buttons with VARiable name
You WILL get Too few arguments ERROR !
Now you have var ID on stack
4. press TOOL
5. press PURGE soft button
        Demo
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
 *** Usenet.com - The #1 Usenet Newsgroup Service on The Planet! ***
                      http://www.usenet.com
Unlimited Download - 19 Seperate Servers - 90,000 groups - Uncensored
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
====
message
actually the first character in the command name.  So try
300 [enter] [rightshift]  0  [alpha] [alpha] keytime [enter]
to
I use something one of the guru's posted sometim back.
- Use the VAR key to get the normal menu of variables,
- then {} to start a list, then select variables from the function keys 

to
fill list with variable names you want to purge.
- Select TOOL menu then PURGE
{ X Y Z } PURGE
You can feed a list of names to many commands.  I find  creating a list of
variables directly from the variable menu very useful.
====
**** Post for FREE via your newsreader at post.usenet.com ****
Ooops: It should have been:
Make sure you have RPN mode on than:
300  (SPC)  (ALPHA) (ALPHA) (right-shift) 0 K E Y T I M E  (enter)
Sorry,
    Demo
evaluating
type
keystroke-combination...
having
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
 *** Usenet.com - The #1 Usenet Newsgroup Service on The Planet! ***
                      http://www.usenet.com
Unlimited Download - 19 Seperate Servers - 90,000 groups - Uncensored
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
====
(rightshift) + 0 + enter KEYTIME + enter will give you a 
space between the right arrow and KEYTIME, which parses as 
two commands.
Put 300 on the stack
then key in
(alpha)(rightshift)0KEYTIME
and press ENTER.
====
OOPS!  Forgot to mention that before using MSOLVR, you have to run
MINIT.  It checks the validity of EQ and creates the control variable
Mpar.  The reason I missed the first time around is that I did a quick
test of a multiple-eq program and it ran without MINIT.  Of course,
I'd forgotten that I'd used the same program recently and there was a
valid Mpar in the dir.
To make up for this blunder, I'll reveal an earth-shaking secret.  You
can use MSOLVR with a single equation, the advantage being that the
labels turn dark when you put in a value, unlike the single solver (30
MENU).  Just put two copies of the equation on the stack, make a list,
and store it in EQ.  Then do MINIT MSOLVR.
Bill
====
IĒll do the same as you, because I also think that 
IĒll get to switch a lot
between scientific and standard notation.
I knew it! I downloaded a program equation library and this program 
also
makes use of the MSOLVR (I think) and itĒs a very cool 
function - you just
(without knowing it) answered the question I had in my mind, that I wanted
to try to figure out myself, but luckily you revealed the secret... 
Thanx,
again.
Martin J.
====
of
But if I have a number like 0.00000000004 in standard notation and I 
donĒt
want to count all the small numbers/decimals and enter SCI + ENG, 

then
the calculator deletes the number (0.00000000004) and then I should
problem... Here the calc does the right thing without dropping the 
number
on the stack.
So what should I do???
Martin J.
====
Read The manual.  The SCI command takes an argument, the number of
decimals you want to have displayed.  Use
 4 SCI 
or something like this.
- Carsten
====
I'd like some kind of utility that you could personalise that would
automaticaly set the display type depending on the size of the number.  Say
<10^5 use 4 FIX else 4 ENG etc.
Also like to have option for 1000's seperater in STD mode.
Stephen N
lot
Thanx,
====
Is there any way to easily convert a 48 XLIB's rom adresses to the 49's rom
adresses? i.e. a conversion program...
====
If you don't find one, you can always do it the hard way, using the 48/49
cross-reference tables here:
http://holyjoe.org/HP49.htm
-Joe-
====
There should be such a program on www.hpcalc.org
Raymond
====
**** Post for FREE via your newsreader at post.usenet.com ****
I'm looking for source code for BZ uncompress tool, because I want to
compress NewOS kernel and uncompress it directly into Port 1.
     Demo
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
 *** Usenet.com - The #1 Usenet Newsgroup Service on The Planet! ***
                      http://www.usenet.com
Unlimited Download - 19 Seperate Servers - 90,000 groups - Uncensored
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
====
You can see the source (HP48 source) here:
http://www.hpcalc.org/viewzip.php?id=2355&file=bz.s
Aaron
(reply to: awallace AT utdallas DOT edu)
====
maybe someone is able to set the FAQ and www.hpcalc.org links as
repeated postings or alike to c.s.hp48, since there seem to be many
newbies and people who don't know about those helpful links,
and some who always first ask before RTFM;-)
Raymond
====
That message seems to be obsolate for ROM 19-6. It
is definitely obsolate for the font browser ChFnt
of Fontman which looks for fonts in any port. There
is absolutely no need for recalculation modulo 245.
FPAR by giving it the number 255 and store it in 
2:F255.F, say. Now run the font browser ChFnt. It
contains the choose item 2:F255.F. Just press on
this item. That loads font F255 with the right ID.
This also works if 2:F255 is a self-decompressing
font, easily made with OT49, for instance.
If somebody uses dozens of fonts, these should be
This is no problem for Fontman, but a problem for
the builtin font browser which, IMHO, is useful
only for newbees.
- Wolfgang 
====
What about styles embedded in strings?  For example:
and
fails if the font's ID is between 245 and 255.
Therefore, font ID's above 244 should never be used, unless the font will
NEVER be used for embedded string styles, which would be an unnecessary
limitation for an HP49G font.
-Joe-
====
Talked the store manager into selling me 3 of the HP38s for a discount 
of the $16.  Still has 3 or 4 left so in a month I might have another 
one or two for $5.
I tore one apart as I was curious about the HP keyboard.  Pretty simple 
once you have it apart.  Even the circuits for the main board look 
fairly straight forward.  Know I have parts now worth more than I paid 
for at least one of the calculators and probably all three.
The first (now parts) and 2nd calculators have (or had) dismal displays 
unless you have the calculator at just the right angle w/ the right 
light source.
The menu system (aplets) is interesting and I can see the advantage to 
it in certain instances.  I don't know that it is any easier to learn 
than an HP48 though.  Maybe once I play with it someone more it will 
seem 'better'.
The different registers to hold variables of different types is nice and 
reminds me of the HP11 except the 38G has a better access to the variables.
The 38G processing seems quicker than the 48G, and there are several 
built in functions that are handy for one of my classes this semester.
Maybe I can hack the rom eventually and have rpn ...
John
====
only if you're a juggler for a living........
====
especially memorable!
Jim (still laughing)
====
I'm taking this one to work tomorrow...
Toby
any
====
This is bogus.  Cars last *ages* nowadays, they're much better than
they were 20 years ago, because people understand how to stop things
rusting and computer design has made them stiff and reduced water
traps. Engines &c just go on for ever.  Sure, cars probably don't last
as long as chassis-based cars can do *if* they're properly maintained,
which means `taken to bits and repainted every 10 years', but you
probably don't want to get involved in that kind of maintenance, and
you almost certainly can't afford the fuel they burn (oh well, in the
US you probably can: I bet my 73 year old car gets more MPG than the
average SUV, - it's probably safer too, and definitely more fun to
drive).
It's really not true that nothing is built to last now.  Some things
are, when they need to be.  Mobile phones are a good example - some of
them are crap obviously, but good ones need to be tough enough to be
dropped regularly, sat on, rained on &c &c, and they are well-made as
a result - I have a fairly old motorola and a new nokia both of which
are just obviously well-made and will end up being thrown away because
they're technologically obsolete (or, more likely, because it will
cost more to buy new batteries than it will to buy a new phone with
the way the phone market works in the UK) long before they wear out.
CDs, while not as nice as records, last for ever (I have 19 year old
CDs which play as well now as when they were new).
HP calculators clearly *aren't* made the way they used to be, but that's
because there's not enough sales in the calculator market to support
well-made products at a price people will pay (and also because HP
are a dying company, of course).
--tim
====
Have you *driven* a car made in the 1920s?  I have, and they are *not
the same thing* as cars made now.  No synchromesh, unreliable brakes,
32 greasing points which you are meant to grease every 500 miles, 60
miles an hour if you are lucky (and brave).  Heating? Weatherproofing?
Petrol gauge?
Actually: what an amazing bit of design, it worked well in the 20s and
still works well now.  And modern batteries almost never die, too
(when was the last time you had to jump-start a car?)
--tim
====
Gee, don't be so anxious!  Just wait a while and it will come to pass!
Geoff
 
-- 
====
There is probably enough expertise on this newsgroup to design a 
calculator that is better than is currently available.
After destructing the HP38G yesterday, hardware is fairly straight 
forward especially if you do most of the features in the firmware/software.
The starting point is to figure out the requirements and if it is 
designed should it be a home built kit or just have a parts list available.
====
Ouch! The above sounded rather like, I'll show you easyness ;-)
What? No built-in character set?
That means the Mac has to built the bit patterns before sending data
to the printer?
Anyway, even then we could just send the bit patterns as strings
without having to re-code serial transmission. Only the data itself
has to be converted to something that the printer can understand.
very much.
Greetings,
Nick.
====
Hey
Is it (of course it is, more correctly how) possible to make a programme 
for the HP49G with eg. 2 variables that when on execution the programme 
asks you for the values to fill into this variables?
Eg.
Make a programme called funny that calculates:
x^2-log(y^2)
But when you type funny you don't have to write:
funny,1,2 or similar. The programme simply makes a dialog box asking for 
the values (one at a time) for x and y, respectivly.
How to do that?
Anders
====
These are just a few quick questions I was wondering as I look at NewOS:
 Are all of the compiled apps that come with NewOS reverse compatable
with older versions the NewOS kernel/loader?
Maybe some type of versioning system would be useful here.
 Also, will there be some way of perfoming inter-thread communication
and managing threads from within other NewsOS programs?
 In what way does NewOS currently handle threads? Would there be
significantly greater overhead in using a scheduler that would allow
an arbitrary number of these at once?
 Finally, maybe there's a way of enforcing a 'user mode' of operation so
that unprivilaged apps could be limited somehow. This is obviosly
difficult without direct support of this feature by the Saturn but
perhapse the use of some API that encapsulates certain sensitive
operations and manages permissions could be encouraged by developers
of newOS programs.
====
**** Post for FREE via your newsreader at post.usenet.com ****
Applications are reverse compatible, because I am using static GOLONG
functions @ the beginning of port 1. Those jump to non-static
functions(those change version by version)
Current out:        kernel: 0.8.2.3
                          shell: 0.21
Did you read the readme.txt file ? :-)
Currently every register of process have static place in memory a every
process has it's own save & restore function in interrupt handler.
In future I plan to use ONLY 1 restore & 1 save function in order to make
kernel smaller & I should probably dynamicaly allocate momory for 
processes,
so there will be no limitation.
Small note: use word developer instead of  developers
Currently shell suspends, GUI suspends & for future it will be supported 
for
all programs.
    Demo
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
 *** Usenet.com - The #1 Usenet Newsgroup Service on The Planet! ***
                      http://www.usenet.com
Unlimited Download - 19 Seperate Servers - 90,000 groups - Uncensored
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
====
[OT]
Yes, if you define really is as can be observed (that's positivism 

in a
nutshell).
No, if you define really is as regardless of observation (that's 
realism
in a nutshell).
Example: Let's call the Exact Moon-to-Earth Distance EMED.  Note well: 

I
do NOT mean the *mean* distance, since that changes over the millennia.  I
reather mean the *exact* distance *right now*, which also changes over 
time,
but always is an exact distance at any given moment, even before there were
any people on Earth to observe, measure, or even think about it.
Furthermore, there always exists in space a line segment that starts at the
exact center point of the Earth (Zero EMED), intersects the exact center
point of the Moon (One EMED) and continues outwards from the Earth to the
*exact* point PI EMED.  It's a precise *point*, tracing a strange path
through space as it orbits the Earth which is orbiting the Sun which is
orbiting Sagittarius A* which is orbiting God-knows-what... but it's an
exact, precise, real point, even though I highly doubt whether *anybody*
ever gave that point any thought before.
The positivists would say, no, there is no such point, because it cannot be
observed.
I would say, yes, there is such a point, because reality is perfectly
content to exist with or without our awareness of it (see
http://holyjoe.org/teasdal2.htm for a tragic application of that thought).
Ergo, by my claim that pi is real, I do not confuse reality with the
observed.  In less precise terms, pi is a real number, not an imaginary 
one.
:-)
Correspondence between two aspects of reality, whether observed or not.
True, but real numbers do, no?
-Joe-
====
First of all, thank you so much for this. We should talk more often
about such matters here.
And please don't get me wrong. None of my answers below are meant in
the way you are wrong, I am right. Just think of me with my face
turned to a question mark, OK? Something like :-?
If you omit the really from really is, or if you rather say is,
under some certain premises, then it will fit better. Really has
always been too much of a saying.
This is a much more complicated problem than the existence of a
number, as many other factors play a role here. So let's stay at
numbers, we mix up too many things otherwise.
Only if we stay at maths and accept that any real number represents
some distance. But you bring physics in the game (Do you?). If you do,
then the above paragraph about the distance PI EMED is true only if
space itself isn't quantized (discrete) in elementary segments of
which no int. multiple results in PI EMED.
It is dangerous to project physical reality and existence to
anything that can't be measured. Do you say here that electrons have a
real path and a real location and a real velocity, but we can
only not measure them?
If you can't prove or much more disprove something, let it be and find
something different. If I extrapolate the above sentence of yours in
an admitedly exagerated way, then I find out: Anything exists, the
existance of which I can't disprove.
Again that hard thing, reality. (If John hears this then we'll both
run for cover ;-)) I don't know even what reality is. I don't have any
means to define the one and only absolute reality. (John apparently
does ;-)) So I find it very hard to talk about that thingie. But
awareness always plays a role. Do you thing that we would find the
same reality if our sences were those of an alien? If we could
smell colors and hear photons? I doubt that very much.
for searching reality.
No, it seems to me that you sometimes confuse reality with what you
wish to exist. It all fits so wonderful in a simple picture, if the
wished to be distance PI EMED, can be assumed to exist a priori. But
the wish for an easy picture is not a proof.
Yes. But though not imaginary, it is in transe ;-)
Still I see no definition. Only interpretations.
No. Depends on what you accept as a valid way to construct the
corresponding distance. Locate PI on a straight line. You can't, if
you are allowed to use only ruler and.. the other thing with the two
legs, what's its name, compass?, Well I think you know which
instrument I mean.
Anyway, we started with precision and accuracy. There are strict
definitions of those quantities. The philosophical part doesn't offer
definitions but rather what each an every brain conceives and
thinks/wishes true, because it seems to make sence.
Greetings,
Nick.
====
They sell a few different models ranging from about US$8 to US$25.
Check the link above for details.  I don't recall what shipping
was, but it was reasonable - roughly $5, I think.  Shipping was
pretty fast, too, taking 2 or 3 weeks from Japan to east coast
The rules are nicely made and the people at the company are very
friendly & helpful.  A nice place to deal with.
        Mike Markowski
====
[OT.  Yes, Nick, I'm in the same boat as you: thinking out loud and always
eager to learn, because learning is massively fun. -jkh-]
Sorry.  The very thought that things perdure even when I'm not thinking
about them (proven repeatedly by ordinary experience every day) is
sufficient to convince me that there is something (called reality) 
outside
of my mind which exists independently of my mind and which is not effected
by my thoughts about it.
Ok, here's a PROOF:
Let's suppose that the point PI EMED does NOT exist in reality, but only
exists as a concept in my mind.  Ah, but point and location are
identical.  So, if a point does not really exist, then that location does
not exist, and is a mere fiction, like the Land of Nod.  One thing is
certain about locations that don't exist: you can't go there.  Therefore, 

if
PI EMED does not really exist, then you would be unable to go there.  Since
that is patently absurd (what's gonna happen if you try? gonna run into a
brick wall that stops you?), the original supposition must be false (by
process of reductio ad absurdum) and we have thus PROVED that the point PI
EMED *does* exist in external reality, Q.E.D.
Suppose I say, A and B are in EXACTLY the same place (or PRECISELY the 
same
location).  I'm not interpreting anything, or measuring anything.  I'm
saying that their locations CORRESPOND.  In mathematical terms, they
coextend, and if they are points, they are identical points.  This is true
whether measured or not, whether known or not.  Example: Is the warped
spiral galaxy ESO 510-13 http://antwrp.gsfc.nasa.gov/apod/ap010803.html a
single galaxy that's warped (as everybody claims) or *two* galaxies seen
edge-on, one in front of the other, and slightly askew from it (as I
believe)?  It *really is* one or the other (or some third as-yet-unimagined
situation!) and really WAS that before mankind came along and might very
well continue being what it really is long after we're gone.  Scientists
take great delight in cooking up hypotheses like these, and then VERIFYING
them, which means seeing whether they CORRESPOND to REALITY or not.
AAuuggghhh!!!  Yes, WE can't LOCATE it... but it's REALLY THERE!  If it 
were
not, then what the heck is the point of math (no pun intended)?  Proof: If
it were NOT really there, then ANY talk about locating it with accuracy 

is
pointless (pun intended) since that kind of accuracy is determined by the
error, the distance between where I think pi is and where the REAL pi is.
How can that distance exist if there is no REAL pi?
Similar situation to PI EMED: If there were no real pi on a number line,
then there would have to be a discontinuity there and you wouldn't be able
to go through that point, but you can, so there isn't, ergo pi exists.  I
suppose a black hole might exist at PI EMED, but if it did, it'd be even
EASIER to go there, although the final approach would suck.  ;-)
Ah, but you accept those strict definitions ONLY because THEY make 
sense.
Otherwise you're caught in an infinite regression: These strict 
definitions
are based solidly on other solid definitions, which are based on something
else, which, um, well, at some point it becomes *obvious*.  In math, we
call such a thing an axiom, defined as a self-evident truth that requires
no proof.  But some people claim that even self-evident things need 
proof.
So I did (see above).  What say you?
-Joe-
Ever rule has an exception is its own sole exception.
====
Is there any command to simplify exp(n*ln(x)) to x^n?  I have some large 
equations, and they'd be easier to read if this was simplified.
A small example:
I want 5^n - 1 instead of (exp((n+1)*ln(5))-5)/5
It gets more complicated with larger equations, any ideas?
-- 
Al
====
You were trying to transfer a text file which was not
a calculator program, using Kermit in ascii mode, right?
Please look here for explanation and solution:
http://groups.google.com/groups?selm=3DA22AD5.805A0176%40miu.edu
.
====
Steve VanDevender is back!
Yes, still ISO-8859-1,
with a few altered characters (e.g. Angle at 128, Euro at 160,
and some control characters used for graphics).
Characters 32-126 (decimal) are the same as 7-bit ascii,
so imported ascii strings are completely readable in the calc,
and when exporting using internal HP Kermit with translation
(mode 3), characters 128-255 are also readable in plain ascii.
.
====
For what it's worth, both Mathematica and Derive give 6561 (my personal
preference).
Both QuickBASIC and UBASIC give 729.
What about other programming languages?
-Joe-
====
G'day!
This has already been discussed...
http://groups.google.com/groups?hl=en&lr=&ie=UTF-8&oe=UTF-8&q=2%5E1%5E3+grou

p%3Acomp
.sys.hp48
We chose mathematical correctness (in this case, established convention) 
over
backward compatibility, which seemed a minor problem since the only problem 

would
arise from sending/copying a program containing an algebraic object 
containing ^, as
text between a 48 and a 49.
The 49 market is primarily high-school+ rather than engineers (and it 
includes a
CAS), which reinforces this choice.
In fact, I'd be interested in why the 48 (28?) team chose the 'wrong' case, 

if
anyone of them is reading this, just curious, please...
Gerald.
====
A few lying around on my computer:
GiNaC: 6561
Python: 6561
Yabasic: 729
Gnumeric spreasheet: 6561
Bash shell: same as Maple - doesn't parse the second portion
-Mike
====
Exactly what I guessed in my post from yesterday.
This was a necessity for maintaining compatibilty
with an elaborated CAS.
This problem is easily overcome by setting flag -53
on the *sending* calculator (either 48 or 49) before
starting to transfer a UserRPL program in ASCII mode.
Rather strange that this seems nowhere to have been
mentioned. However, I doubt as to whether this is the
only serious problem in backward compatibilty :-)
Wolfgang
====
Exactly what I guessed in my post from yesterday.
This was a necessity for maintaining compatibilty
with an elaborated CAS.
This problem is easily overcome by setting flag -53
on the *sending* calculator (either 48 or 49) before
starting to transfer a UserRPL program in ASCII mode.
Hence, include an algebraic object containing ^ at
ASCII-text. Strange that this seems nowhere to have
been mentioned. However, I doubt as to whether this is
the only serious problem in backward compatibilty :-)
Wolfgang
====
Just another guess (only for fun) before somebody
from the 28/48 design team will answer :-)
They clearly realized that the tradition convention
in reading a^b^c as a^(b^c) is in conflict with RPN.
What to do in this dilemma? Since they had no plans
to join a CAS to the 48 in the future, they decided
that RPN should have priority above all tradition.
Interesting only that neither the ordinary nor the
advanced users manual make any notice about this.
   
- Wolfgang
====
I'm missing why it's in conflict with RPN.
a^(b^c) is done easily in RPN as a b c ^ ^.
Or are you referring to the *internal RPL parsing* of 'a^b^c'?
-Joe-
====
 
Thats how the 49 compiles the term 'a^b^c' while
the 48 compiles it differently.
What else could I refer to?. The normal user
communicats with the machine on the base of
what he sees in the display :-) 
- Wolfgang
====
 
Thats how the 49 compiles the term 'a^b^c' which
is in accordance with the tradition of reading
the term 'a^b^c'. The 48 compiles it as a b ^ c ^
(the wrong case according the Gerald Sqelart).
The term 'a^b^c' is as meaningless as is the
term 'a+b+c' as long as there are no definite
and tacedly accepted rules of bracketing. Such
rules exist indeed, but are, unfortunately,
different in the two examples.
Since the right-bracketing convention in 'a^b^c'
was ignored by the Corvallis people, they should
have at least commented on this in either of the
manuals. On the other hand, from the 49-designers
we get comments only if we explicitely state some
problem (and it they are in a good mood :-)
What else could I refer to?. The normal user
communicats with the machine on the base of
what he sees in the display :-) 
- Wolfgang
====
Some additional remarks on bracketing
-------------------------------------
It might seem at the first glance that
bracketing convention for + is irrelevant
since + is assiciative.
But this is not the case at all. On the set
of all 48/49 objects on which + is defined,
it is definitely not: (a + {}) + b is
different from a + ({} + c). Clearly, if
some operation is not associative, bracketing
rules become important. As a matter of fact,
on both calcs, left-bracketing is tacidly
implemented both for + and *.
- Wolfgang
PS. On the 49, left- or tight-bracketing is
not shown in pretty-print even if flag -53
is set. Isn't that a bit dangerous?
====
Some additional remarks on bracketing
-------------------------------------
It might seem at the first glance that
bracketing convention for + is irrelevant
since + is associative.
But this is not the case at all. On the set
of all 48/49 objects on which + is defined,
it is definitely not: (a + {}) + b is
different from a + ({} + b). Clearly, if
some operation is not associative, bracketing
rules become important. As a matter of fact,
on both calcs, left-bracketing is tacidly
implemented both for + and *.
- Wolfgang
PS. On the 49, left- or tight-bracketing is
not shown in pretty-print even if flag -53
is set. Isn't that a bit dangerous?
====
Try with ... http://ca-on.hpcalc.org
Next step ... type Mika HEISKANEN in the window search.
Miguel Angel CAPORALINI HERK
****************************************************************************

***
====
Seen:
If you lie to the CAS (storing 'R' in 'VX')
and then evaluate expressions where 'X',
a variable of summation, already has a defined value,
but a different name in 'VX' causes SAVECASFLAGS
to look for a different variable than the one needing to be purged,
then varied results (or sometimes error messages)
should not be surprising.
Of course, it would be nicer if the CAS never had
any conflicts with existing variables, just as
the 48S/48G never had any, so long as flag -3 was clear.
But one questionable basic design decision killed that nice hope,
given which it would next have been nice for commands
which don't actually use the name in 'VX' at all
(most CAS commands, actually) to recognize instead
the name(s) of the actual variables which would need
to remain undefined for each particular command to work,
and try to prompt you to purge those variables instead.
Neither of these nice things happened, however, so it remains
up to the user to gain a much deeper insight into the inner workings
of the CAS, and to defend himself against these anomalies
by purging his own variables, or by setting 'VX' himself,
to cause the uninformed SAVECASFLAGS to check for and
purge his variables for him.
Uh-oh, the above expression actually has *two* different variables
('X' and 'Y') which *each* must be purged, but 'VX' does not
accept lists of variables, so in fact it is truly impossible
for the user to get the CAS to do the work using 'VX' alone!
So the broad decision never to leave symbolic names unevaluated
has proved to be a very fateful decision, alas -- would you like
to be able to go back and re-think that one?
The reliance by SAVECASFLAGS upon whatever name is in 'VX'
is also unfortunate, as most CAS commands use some other variable,
which is supplied directly instead, as a stack argument.
But listen, if even the blurry Hubble space telescope could be
fixed by a retrofit, maybe something can yet be devised
to fix these things -- why not take it as a mini-challenge,
whose solution would be very useful indeed!
.
====
---snip---
- Just a naive thought
Why don't  we forget about 'VX' ?
 In the event of formal variables, wouldn't it be easier just to do the
calculation, then give warning, stating that values' of those questionable
variabels  have been disregarded?
( f.i. when you have a formal variable, scan the expression, temporarily
rename all occurences to something undefined ( an reserved generic name 
like
cas#001.. ) , do the calculation, and rename back again.)
Gjermund Skailand
====
The last extable I have is dated November, 1999.
I have a pretty good entries (Suprom49.a) table for the 49 but it does not 

have many of the 49 user ROMPTR calls.  For example
xFONTn and xINFORM are missing.
Does anyone have a better list that includes the above commands and other 

user command entries.  (I can get there one at a time from
Carsten's terrific database but I am hoping for a simple list).
-- - - - - - - - - - - - - - - - -
Bill Graves   RKBA!
bgraves@ix.netcom.com
====
After the program
   << ENTER A NUMBER
you should have typed
   TEST    (this appears as 'TEST')
   STO
because STO wants two arguments: any object in level 2 and a name (no 
string
delimiters) in level 1.
I advice you to get the doc Programming in User RPL, by Eduardo
Kalinowski, from http://www.hpcalc.org/ .
Paolo Cavallo
====
yes, the officials are VERY aware of the HP's superiority. anyone
caught using just a TI on the test is almost laughed right out of the
competition becuase literally everyone else knows they have no chance.
Most people use a ti-83+ or 86 for graphing or conversions etc but a
large portion of the test is straight number crunching where the
32sii's speed is completly unparalled (because the test is practially
written for it).  next best thing is the hp 49g which i can beat the
average 32sii user at the competition although the good people would
just trounce me if i used it reguarly.  at the state level
competitions EVERYONE uses a 32sii.
        The non-HP-using people simply cower in fear when the Hp-using
people show up.
====
I've purchased an HP49G - and I'm having all sorts of probs with the
simplist of programming - I'm sure I'm missing the (very) obvious ...
(Touch of history - still own and love my trusty HP41CX (was right in to
programming with it) - and in the past I've been a pretty serious (80x86)
Assembly Language programmer)
I prefer to work with RPN - and I'm trying to create the simplist of
programs - that prompts for a value - and then accepts the value.
I've tried entering the following from the user guide ...
<< ENTER A NUMBER
But all I'm getting is ...
STO ERROR - TOO FEW ARGUMENTS
Can anyone spot the obvious?
 - HP48g is in RPN mode
arrow is much longer in book) - I'm getting it by typing:
ALPHA ALPHA OBJ RIGHT-SHIFT ZERO
- Have also typed   without a space in between - didn't look like there
was one in the example
All help really appreciated.
Many thanks
====
In RPN mode, you have to put the argument(s) before the command or operator. 

In your case, the STO command expect a number and a name to store it in 
before the STO command is executed. Swap TEST and STO in your program, 
then I think it should work.
/Nikodemus
-- 
Get my public pgp key at http://www.algonet.se/~nickek/pubkey.asc
====
Download from http://ca-on.hpcalc.org :
Nosy
If you have a HP-49 with ROM 1.19-6 (B), you can download:
Extable (of J.Y.Avenard)
Miguel Angel CAPORALINI HERK
****************************************************************************

***
====
You should remind us that your questions refer to HP48GX,
to avoid getting answers that apply only to 49G :)
So you are using Hack and Jazz, too :)
In general, you may find it most convenient to use
DIS and then ED; within ED, every time you put the
cursor on the word PTR or ROMPTR (as well as on any symbol),
you can press Right-shift then [+/-] to explore that entry.
ED is pretty powerful; perusing its manual may be worthwhile.
ROMPTR 0AB 06D (in 48G) happens to be REVLIST,
so you can use either { REVLIST } DIS ED
or { REVLIST } XRCL DIS ED  [but there's no need for XRCL]
The latest versions of Jazz may not even require DIS before ED,
but mine does (mine is on an old ROM card, not updateable).
Here are some little helper programs which I use,
to make exploring ROM even easier:
http://groups.google.com/groups?selm=6h4571%24ufg%241%40news.iastate.edu
HP48GX, you mean?   Yes, it was wonderful, yet:
Full many a gem of purest ray serene
 The dark unfathom'd caves of ocean bear;
 Full many a flower is born to blush unseen,
 And waste its sweetness on the desert air.
Elegy in a Country Churchyard, Stanza 14
Thomas Gray (1716.9a1771)
.
====
  Colin, thanks for your link...   Actually I had bumped into your
homepage before when checking out the calculator for reviews... but
didn't know about that library.  Well I would actually like to have
just readily available binary/etc operations (add/ ands/ xor,
substraction, complement, etc.) for binafry/hex/octal.
   I am truly debating returning this and getting an HP48G+ instead. 
Are there any advantages of the 39G over the HP48Gxx series? (doubt it
but who knows)  The price is one.... and the learning applets are
cool but I wouldn't die for them.  OH the ability to put graphs side
by side is another I guess...
   Maybe I should keep it and use my default Windows XP calculator in
scientific mode for those other operations I say I miss.
- Raist
====
I think the thing you need to remember is that the 39G is aimed at high 
school students.  As such it is IMHO *far* easier to use than is the 
48/49 - it seems to me that this was one of the prime design priorities. 
  The aplets were also a major factor in our school's decision to use it 
(well the 38G originally) as was the way it had an IR link so that you 
could quickly and easily transfer a teaching aplet to your whole class 
in < 5 minutes.
However no-one would even try to claim that it has the same level of 
power as the 48 and 49G. If that's the level of power that you need then 
you'll need to swap to one of those.
====
i've got an 48GX+2 MB RAM and a nokia 3320 cell phone(IIRC it's TDMA). the
phone has tools for sending contact list, agenda, and uploading the SMS
messages to a computer or another 3320; via IR or via cable to access
internet. can I use it in to sync. my contact list to my HP48?
====
The 49G is picky about keystroke order.  If you press ON, *then* press A,
*then* press F, and then release them one by one in reverse order (F, then
A, then ON), you'll get the TTRM prompt every time.  Some other orders work
too, but I only guarantee this one.
Hope this helps!
-Joe-
====
KP says:
Have you figured out a method which wouldn't be readily hacked?
Well, in asking for the serial number, be sure to instruct
your clients to get their serial number from the SERIAL command,
rather than from the case, and decide what to do if they
have no serial number at all (some calcs and all emulators).
So, where's all the commercial software for the 49G, anyway?
.
====
I don't think such a method exists. On the HP49, you'll most probably
need ML code; but even the most obfuscated code may be single-stepped
in DB or Emu48.
Greetings
Thomas
-- 
If you cannot convince them, confuse them.
    -- Harry S. Truman
====
marketing strategy ???
--
Carlos
====
I'm looking for an exact solution (if possible) to the problem that involves 

differential equation as follows:
cos(t) + x * cos(t) + y(x) * sin(t) = 1   and   y'(x) = tan(t)   and   y(0) 

= 0
where t is a parameter to be eliminated from the first two equations.
Any ideas how to find either y(x) or in the parametric form of y(s) and x(s) 

?
Jack
====
The solution is (and is valid for all t):
x(t) = t*sin(t) + cos(t) -1
y(t) = -t*cos(t) + sin(t)
First, you can check that this is the solution:
cos(t)+x*cos(t) + y*sin(t) = cos(t) + t*sin(t)*cos(t) + cos^2(t) -
cos(t) - t*cos(t)*sin(t) + sin^2(t) = cos^2(t)+sin^2(t) = 1
y'(x) = (dy/dt)/(dx/dt) = [t*sin(t)- cos(t) +
cos(t)]/[t*cos(t)+sin(t)-sin(t)] = tan(t)
x(t=0) = 0 and y(t=0) = 0, which means y=0 when x=0 as required
I obtained the solution by first writing the second equation as:
y'(t)/x'(t) = tan(t) = sin(t)/cos(t)
or
cos(t)*y'(t) = sin(t)*x'(t)
This can be rewritten as:
d/dt(y*cos(t))+ y*sin(t) = d/dt(x*sin(t)) - x*cos(t)
or
d/dt(x*sin(t) - y*cos(t)) = y*sin(t)+ x*cos(t)
Using the first equation, the right hand side may be written as a
function of t:
d/dt(x*sin(t) - y*cos(t)) = 1 - cos(t)
This is integrated to produce:
x*sin(t) - y*cos(t) + K = t - sin(t)
where K is a constant of integration.  The initial condition is:
y(x=0) = 0.
For convenience, take t=0 at this point as well (the t coordinate can
be translated so this will hold).  Then K = 0.
Then:
Eq.(1): x*cos(t) + y*sin(t) = 1 - cos(t)
Eq. (2):  x*sin(t) - y*cos(t)= t - sin(t)
Multiply Eq.(1) by cos(t) and Eq. (2) by sin(t) and add, to get:
x(t) = t*sin(t) - sin^2(t) - cos^2(t)+cos(t) = t*sin(t) + cos(t) -1
Multiply Eq.(1) by sin(t) and Eq.(2) by -cos(t) and add to get:
y(t) = sin(t) - t*cos(t)
These final 2 equations are the solution.
Geoff Garner
====
This will work. The two solutions I was talking about are actually hidden
in your equations.
When you look on your equations:
x =  t * sin(t) + cos(t) - 1
On the other hand:
y = -t * cos(t) + sin(t)
reasonable t within -pi/2 to pi/2).
system you will get two curves that start at (0,0) and one is 
describing
system behaviour for positive t and the second curve is describing system
behaviour for negative t as I was expecting. Actually of course those are
two branches of the same parametric curve.
Jack
====
This is all getting to sound somewhat familiar:
Lord of the Flies
http://www.ricochet-jeunes.org/eng/biblio/books/lordflies.html
http://www.ricochet-jeunes.org/eng/biblio/films/fliescinema.html
http://www.dvdverdict.com/reviews/lordotf.shtml
-[]-
.
====
Fractions are really general algebraic expressions, but
the HP48 supports only numeric array elements (real or complex).
While creating or editing elements, the matrix editor
actually allows you to input any RPL input stream
which leaves numeric values on the stack,
so when creating or editing your matrix,
you may enter the element 1/2 in any of these equivalent ways:
  1 2 /
 '1/2' EVAL
   .5
By the way, you may know that you don't have to press ENTER
after every single array element; you can enter a whole bunch
at once, separated by spaces (just don't enter past the end
of the very first row when creating a brand new matrix).
If you would like to *display* an already existing array
(or result) as a set of fractions, however,
then proceed as follows:
Put these programs in your HOME directory [48G/GX only!]
(enter #E803Ch exactly as shown and spell LIBEVAL correctly :)
Then you can display matrices as fractions, e.g.:
.
====
I recently saw one of my friends studyinng for his history exam, on
his TI-83+ silver edition! He downloaded a program were he could type
questions on his computer and turn them into virtual study cards.
Questions would be on one side and there was a flip command where the
answers were on the other side. There was also a key for changing the
cards. I was wondering if there was any such program for the 49G and
how to make these cards.
Thanxs in Advance,
CID
====
Isn't that supposed to be (+ a b) ?
-- 
john R. Latala
jrlatala@golden.net
====
You're right, of course.
Too long ago that I even made a syntax error;-)
Raymond
====
[OT]
Sounds like you're in a bad mood.  Maybe you need a nap, John.  ;-)
It seems to me that you contradict yourself here.  First you say that 
most
of the human race are blithering idiots.  Then you say that those same
people ought to look inward to who they really are.  But who they really
are, according to you, is blithering idiots.  So, looking inward 
would
either do them no good (since idiots are incapable of deep learning), or
make them even more obnoxious (since nobody like to be told that they're an
idiot, especially idiots).
My opinion is the opposite of yours.  I believe that most of the human race
is quite intelligent, with infinite untapped potential for even greater
intelligence.  Yes, blithering idiots exist, but I believe them to be a
small minority.  Unfortunately, there are enough of them to be a bother,
such as in Washington DC where they abound.
-Joe-
And that's all I have to say about that. -- Forrest Gump, IQ 70; 
idiot
means IQ < 25 and mental age < 3.
====
Some absolutely vital units of measurement:
1 football field = 2.963368 femtoparsecs (fpc) + 576 angstroms (.81)
1 Yottabarn (1_Yb) = 1_cm^2
1 microfortnight = 1.2096 seconds
1 millifortnight = 20 minutes + 9.6 seconds
1 attoparsec per microfortnight is approximately one inch per second
A nanocentury is approximately pi seconds
-Joe-
And that's all I have to say about that. -- Forrest Gump
====
As Raymond says, no part of the screen display is normally
re-drawn by WAIT [for a keypress], so if you had just stored
a new menu (via a MENU or TMENU command) to have some new labels
displayed for the top row keys, the menu would not yet
be re-drawn by 0 WAIT, whereas -1. WAIT will cause the
current (new) menu to be displayed before the program pauses.
Any negative numeric argument also has the same effect.
It takes a detectable amount of time to redraw a menu,
so you might want to use -1. WAIT only when necessary,
and 0. WAIT otherwise (when you haven't changed the menu).
.