B25 > Short of changing batteries before every store or archive > command, how does one know when they are no longer Well, the indicator indicates when the battery level drops bellow 3.3V. and as far as I know, works. > Is there a low batt test before each flash write? Yes, there is, and it will not write in flash if the battery are low > There is such a test in the HP48 even during *serial*port* > operations, by the way, which tend to cleanly abort, > without memory wipeouts, if batts are low. Well, I guess the difference is that the serial command drain current according to the Yorke Power supply specifications. However, erasing the flash can burn up to 80mA, 4 times over the Yorke power supply specification! and if you put such a load on the batteries, it's more or less equivalent to shorting them... Then, dropping from 3.3V to 3V (very low bat, shut down the calc) is just a matter of chance... This means that the calculator can go directly from working to dead without going through the start box, without receiving 20000F (well, in the french monopoly, you get 20 000...) and there is unfortunatly no way for the software to test it as the saturn does not have an ADC to monitor the batteries, but only a 3 level digital test (> 3.3V, <3.3V/>3V, <3V) ==== which reminds me of a question that i had... is it possible to mod the 49 for more memory? unfortunatelly, no. the only thing that could be done, is to put more flash, change the bank switcher to a 8 D cell chip and change the flash access function to allow 4 or even 8M of flash.. but this require hw and sw changes... it's doable by someone who has teh will, the time, and access to the ROM sources... Why not change the bank switcher for the upcoming 1.19-7 right now? It should handle FFF times a 128KB Bank. ==== I have a 1M card on my gx... i have every lib i use BZed and duplicated in ports (the working lib and a backup), daily ARCHIVEs, games, the first 30 or so pages of sasm.doc (up to the opcode dictionary), most of saturn.doc, the hptabs (for 48 and 49) and i STILL have 4 ports *empty*!! ==== Because the HP 49G ports are totally automatically bank switched I _seems_ to have more than, say, 128KB free, but actually this is scattered in many different banks, which none is capable of holding more than e.g. 32KB, While my lib needs 36KB. Yes, I would be pleased to have 8MB Flash Memory, of which 2MB belongs to the OS ROM. Note that if you want to have both 41CX and 42S emulators in your HP 49G, then all your Port 1 RAM is gone. PS: I ought to have an Xpander with xcas & ... ==== Well, personally, I would put the file in the 49, and use the 49 to disasemble it... but still, it's not the easiest thing to do... > Once I've made the conversion to ROM format, how do I go about > disassembling First you should learn the C programming language. There's no program in the public so far, that can handle the HP42S ROM image without any modification. I have heard from someone, that the HP42S ROM image was loaded into a HP48 > RAM card emulated on Emu48 and then used the Emu48 Debugger for > disassembling. IMHO not really nice, but when it works, why not? > The decoded HP42S ROM has a size of 64KB. You can check the ROM CRC's of the binary image or the .core format file with a program that can be find here http://privat.swol.de/ChristophGiesselink/Lewis/LEWISCRC.ZIP. you can convert a packed ROM file into an unpacked (the format Sad use one. So your problem is having a converter for converting the Dump into Binary format. Have you tried the dump2core program delivered with every Sad version to convert the DMP into the .core format? I put the file conversion tools I made in the last years onto my page. http://privat.swol.de/ChristophGiesselink/filetool.zip The included DOS program DMP2ROM converts a 00000:0123456789ABCDEF 00010:FEDCBA9876543210 > ... style file to a packed byte binary file. > Rom2emu oldfile newfile for blind conversation you can generate a Sad core file. I'm attempting to run a disassembler on the rom dump from a 42S - something I haven't done with these tools, or this cpu :) I've got a dump of the ROM that looks like: 00000:236B9308DCE30005 > 00010:B087FB0020386FBF > 00020:061371F120051557 > 00030:D280C322450B0621 > 00040:B8E042083250B061 > 00050:7F14507174145174 > 00060:151717F091553172 > etc. neither sad, nor sadhp seem to be able to accept this input and make a .core > that sad can use--- or I'm using the tools wrong. The documentation is a > bit lacking in some details- and missing a couple actual tools that are mentioned. any help? ==== > No, it's due to the slowness of the LIB menu itself. ... > Worst-case scenario: with lots of key assignments and libraries in > memory, turn on USER mode and press LIBRARY. Then try to do some > simple arithmetic on several multi-digit numbers. It's downright > painful. ... I hope you are sure of talking about the HP49. Perhaps it should also be noticed that the LIB key itself becomes painful slow if there are more than 12 libs attached. All this badness has been overcome by the library Libman. It sets a menu of selected libs only, dependent on the current directory. Thus, libs like extable, APPSman and others whose command menu is practically never set, need not show up in the LIB menu. They are nevertheless active (if present). And, if such a menu is needed, Libman has the command LBox which yields a choose box of *all* attached libs together with a menu which permits, e.g., the libs list with titles in alphabetic order Obviously, you may also omit in the LIB menu prefered libs whose command menu is set by an attached key. This all keeps the LIB menu small and solves the problems you were talking about, and some others too ==== > No, it's due to the slowness of the LIB menu itself. Here's an > explanation called TOOSLOW.DOC on Goodies Disk #9: Well, that was true on the 48, but if I remember well, the LIB menu has been completly recoded in ASM in the 49 and is much faster than the 48 one.... ==== > No, it's due to the slowness of the LIB menu itself. Here's an > explanation called TOOSLOW.DOC on Goodies Disk #9: > Well, that was true on the 48, but if I remember well, the LIB menu has been > completly recoded in ASM in the 49 and is much faster than the 48 one.... This is simply untrue. I know how the LIB menu is coded and this is not done optimally. Apart from that FPTR 2 15 is too slow, the computation of the final menu by appending to it the port entries adds to the slowness and is not done in ASM. Seriously speaking, your way of appending the port entries to the lib menu instead of having them on an extra key as on the HP48 together with the important menu 110, was a bad idea. On the other hand, it is true that the VARS command is essentially faster on the 49. Irs essentials are indeed programmed in ASM ==== > the VARS command is essentially faster on the 49. > Its essentials are indeed programmed in ASM It can be done almost as fast with a 40-byte SysRPL program found in the 48 Hack library, which in turn calls existing, supported ML basic functions; it's curious that another VARS in the same Hack library is also much longer ML code, when nearly identical SysRPL (just two different entry point names) can also be used: http://groups.google.com/groups?selm=64knj9%24123%241%40news.iastate.edu which I hadn't even thought of; the system screen display (SysDisplay) is called in 3-4 separate parts, in this sequence: ?DispStack draws the stack area, which will take longer for more and complex objects, but is generally faster on the 49 than on the 48. ?DispStatus draws the status area at the top of the screen. DispEditLine displays the command line, if any. ?DispMenu draws the menu, which always includes building the six separate label grobs, overlaying them onto the single menu grob, and may also be slowed further under circumstances mentioned by Joe. The above functions beginning with ? start with KEYINBUFFER?, and skip their drawing if additional keyboard input is waiting to be processed, so the faster you type, the less you wait! Not at all! This program (which, by the way, everybody wnd is much slower than the 49 VARS or the other VARS from the HACK-lib. This one has already 171 bytes. But not without reason Andre Schoorl put the latter one in version of the HACK-lib. ==== Ok, here's what is happening: I compiled Sad1.08. No problem. I copied my GX rom dump to the file named '.core' since it is the only one. Copied .symbols.gx to .symbols and .formats.gx to .formats. When I execute 'sad -AgX 0 7ffff' I get the following error: ERROR - Location : 02910-47F9A ERROR - Pass : 0 ERROR - Source : ERROR - Code : 92810A0631431911577101....on and on ERROR - Instruction: xCONFIG LibCfgD81 It turns out that I can succesfully disassemble up to 28DE after which wierd things start happening. If I try to dis from 28DF to some higher number, the output seems to be a lot of nibhex and PTRs. I also tried to dis some 49 stuff but I have the same problem except the addresses where failure occur are different. Still the same situation though. I did this some years back with the 48 but now I can't remember if I missed some sort of setup step or a file. Could someone using Sad for the 48 and 49 please provide some pointers? Mozgy? Mika? Paul A? =harns SysRPL should study) is not particularly fo least for long Varset. It also occurs in my 48-HIDE-lib a=== Has anyone tried communication between HP48 and lego mindstorms. I have ordred the lego kit but did not receive it yet. It seems that the amount of memmory is verry limited. Maybe HP can help. ==== for documents from Infoclearinghouse.com. Dr. Urroz and I set up the Infoclearinghouse site and I would be interested in feedback from people that have downloaded documents. ==== I don't mind the password. I just have one small problem with the PDF files. At home, I have no problems. In my office, The OS image of the PC I use is locked down. Therefore, I can't install the necessary Acrobat plugin to open the books. I purchased both full volumes, and I am VERY pleased with them. I wanted to print a nice set on one of our professional printers at work, but the plugin is stopping me. I haven't yet received the printed edition I ordered, but even after I do I want to be able to print another copy to carry around and abuse... ==== Of course, upon reading this again, I realize we are talking about somewhat different things. I am referring to the two comprehensive books (Science and Engineering Mathematics with the HP 49 G - Vol I & II) by Urroz at greatunpublished.com. http://paulf.free.fr While compiling a collection of User RPL trivia, the following questions arose and stumped me. Anybody out there know the answers to any of these, or know where a pointer is to the answer(s)? (1) The 49G's XSERV command accepts l (a *lower-case* L, not a 1) as an input command. What does l do? (2) What companies other than HP *ever* made RPN calculators? (3) Is the 49G's MAP command the same as DOLIST? They seem identical. (4) How is the 49G's DRAW3DMATRIX command and/or plotter mode used? (5) On the 49G, XLIB 2 392 has no name, and when its contents are recalled to the stack, it is displayed as an invisible object which disassembles into System RPL as xCMDAPPLY. Is this command really User RPL? If so, what does it do and how is it used? ==== > (2) What companies other than HP *ever* made RPN calculators? i got an MK61 and an MK52 (their names sound like a rifle). they are russian. http://www.rskey.org/mk61.htm http://www.rskey.org/mk52.htm ==== > Is the 49G's MAP command the same as DOLIST? They seem identical. Why do CASCMD and HELP say that MAP Executes a program on the elem. of a list when this is only an accidental consequence of automatic list processing, plus a liberal attitude to what arg type (any!) is acceptable on level 2? Let's compare results: 1 { 2 3 } 1 << + >> DOLIST ==> 6 1 { 2 3 } << + >> MAP ==> Error: Too Few Arguments { 2 3 } 1 << DUP >> DOLIST ==> { 2 2 3 3 } { 2 3 } << DUP >> MAP ==> { { 2 2 } { 3 3 } } { 2 3 } 1 << DROP >> DOLIST ==> Empty stack { 2 3 } << DROP >> MAP ==> { NOVAL NOVAL } DEG { 30. } << SIN >> DOLIST ==> { .5 } DEG { 30. } << SIN >> MAP ==> { -.988031624093 } By the way, does DOLIST force you to change modes and delete your current variable? I thought that MAP (Matrix APply) was really intended to operate on the elements of an *array* (a built-in version of Teach: APLY from the 48G): [ 2 3 ] << SQ >> MAP ==> [ 4 9 ] [ 2 3 ] 1 << SQ >> DOLIST ==> Error: Bad Argument Type That HELP was probably written late on the night of the 49G's coming out party in France -- what were they drinking? ;-) ==== DOLIST must be the most complicated RPL command. My list of complicated RPLs: -DOLIST -SEQ -STREAM -DOSUBS -(pile_up)MATCH > Asked: Is the 49G's MAP command the same as DOLIST? They seem identical. Why do CASCMD and HELP say that MAP > By the way, does DOLIST force you to change modes > and delete your current variable? > I thought that MAP (Matrix APply) was really intended > to operate on the elements of an *array* > (a built-in version of Teach: APLY from the 48G): [ 2 3 ] << SQ >> MAP ==> [ 4 9 ] > [ 2 3 ] 1 << SQ >> DOLIST ==> Error: Bad Argument Type Complicated? Apart from the last one the others are perfectly described in the HP48 docs (i.e., in ENGLISH, not in FRANGLAIS, hardly readable at least for a German If you do not know what is complicated then learn SysRPL. There you meet commands (like ParOuterLoop) which are really complicated. PS. I've the feeling that precision and care in language gradually declines. Consider only the perfect description of Dave Arnett in http://groups.google.com/groups?selm=4olc27%24104%40hpcvsnz.cv.hp.com This was more or less the standard a couple of years ago. Clearly, one cannot expect a perfect english text from a Frenchman or a German, but IMHO, these (and others) should concentrate as much as possible before going to write something in English. ==== I can't find ParOuterLoop in my sysRPL-reference. ==== > { 2 3 } 1 << DROP >> DOLIST ==> Empty stack > { 2 3 } << DROP >> MAP ==> { NOVAL NOVAL } That <> DOLIST yields no output seems to be not very intuitive. The {NOVAL NOVAL} looks slightly better, but the most natural output seems to be the empty list. At least, the DOL command from OT49 and the DOL command from OT48 do it this way. That avoids unpleasant treatment of special cases in programming. For a SysRPL-programmer, DOLIST is an inacceptable command. It is awfull slow and not well programmed. then a list of 2 nulllams and performs the sedate &COMP command before binding 5 nulllams. > That HELP was probably written late on the night of the 49G's > coming out party in France -- what were they drinking? ;-) Interesting question. It seems to me that also DOLIST was written late on the night of the 48G's coming out party in Corvallis (what were they drinking there? ;-) Seriously, I believe that the Corvallis team thaught on using list and other user commands like VARS in simple UsrRPL-programs only. Hence, they didn't pay too much attention on speed. But the 49 has a larger memory base and is more ambitious in general. Hence, DOLIST should have been rewritten. The same concerns painful slow rompointers like RCLKEYS, STOKEYS etc. Both the OT48 DOL and the OT49 DOL are much faster than DOLIST. The OT48-DOL uses CACHE. But this is nearly as powerful as the Virtual stack used for the OT49-DOL because one may call nulllams from the cached stack ad libitum. Thus, a programmer can define his nulllams above the three ones used by the DOL program itself. ==== Asked: > On the 49G, XLIB 2 392 has no name, and when its contents are > recalled to the stack, it is displayed as an invisible object > which disassembles into System RPL as xCMDAPPLY. > Is this command really User RPL? > If so, what does it do and how is it used? `R->C(A,B)` Decompiles to: :: xEVAL> :: ID A ID B SYMBOL xR->C ; # 2 xCMDAPPLY ; ; Somewhat analogous to this: 'MYFCN(X,Y)' Decompiles to: SYMBOL ID X ID Y SYMBOL ID MYFCN ; BINT2 xFCNAPPLY ; Notice how the first example produces # 2 instead of BINT2? Okay, WR, better rewrite that ROM code for JYA to save 2.5 bytes! . ==== > Okay, WR, better rewrite that ROM code for JYA to save 2.5 bytes! Theoretically, I could do this and I guarantee that the rewritten ROM would be 25% smaller than the present one, without loss in functionality and speed. That would save about 250 KB, that is to mean 2 flashbanks! Unfortunately, such a project would need about 5 years work, day and night ==== Why not let a desktop computer to do the search & replace?! > Okay, WR, better rewrite that ROM code for JYA to save 2.5 bytes! Theoretically, I could do this and I guarantee > that the rewritten ROM would be 25% smaller than > the present one, without loss in functionality > and speed. That would save about 250 KB, that is > to mean 2 flashbanks! Unfortunately, such a project > would need about 5 years work, day and night ==== http://www.geocities.com/oldcalculators http://www.geocities.com/oldcalculators/contex.html http://vcalc.net/cu.htm Also, all the early electronic calculators that I ever saw, prior to TI's Datamath (2500), the first I ever saw which had a key containing no other symbol than [=]: http://www.datamath.org and http://www.datamath.org/Album_Basic.htm All calcs of my earlier experience required number entry *before* the [+=] or [-=] key, which then immediately performed addition or subtraction, although most were a little bit hybrid, in that [*] or [/] were infix rather than postfix with the [+=] or [-=] key completing the operation and also then accumulating results in a postfix fashion: http://www.classiccmp.org/calcmuseum/EarlySharp.htm In fact, here it is asserted that All calcs were once RPN!: http://groups.google.com/groups?selm=59dcb4%2499t%40news.iastate.edu . ==== More calc history: http://www.vintagecalculators.com Big collections: http://www.calculators.de http://www.rskey.org Mostly HP: http://www.hpmuseum.org . ==== Who know RPN or RPL or what, better than the lines in his ( not her ) right hand? > (2) What companies other than HP *ever* made RPN calculators? While not 'true' RPN (4 level stack, etc.), National Semiconductor made > the Mathbox series in the late '70s that worked via postfix. Their > calculator operation later went by the Novus name. ==== >(2) What companies other than HP *ever* made RPN calculators? > Sinclair sold an RPN pocket calculator. Pi was printed on the case, as well as a factor for Deg to radians. This calculator was even sold in kit form for a while. I still have mine around somewhere if you need the details. Also, if you want to go back even further, the old Friden mechanical calculators were RPN of sorts. You put in one operand, you put in the other operand on another keyboard, and you press multiply. BTW, Friden had a mechanical calculator that could do square roots. My favorite was a Monroe calculator, where you had to manually shift the carriage with a knob at the bottom of the calculator for both the multiplies and the divides. Bill alternate E-dress wtstorey@ieee.org.no.spam.please (Use the obvious) ==== > Would you Al, like to take the blame instead? it to several sites, including www.hpcalc.org - Provided it did not > contain copywrited code. Strange behaviour, the anonymously sending of your progs. That's too easy... you should accept to be RESPONSABLE of your acts. Camille ==== I'm just saying what I'd do if I was Jordi. The problem is he is under the threat of the DMCA. While he may be morally in the right, a lawsuit will cost alot of time and money - I wouldn't want to be a maytr. If I, personally, developed the applet, I'd take responsibility for it - but only because 1. I'm in australia where it would be legal (provided that it didn't contain copywrited code) and 2. I have free legal help through my university's student guild. My point was that by anonymously sending the programs, at least they'd be availble to the public. If I put my name to the applet and announced it to the world, then I could be in for alot of trouble. Chances are the applet would be surpressed. What use would that be? Al www.alborowski.tk ==== > Strange behaviour, > the anonymously sending of your progs. That's too easy... you should > accept to be RESPONSABLE of your acts. If the software is unsupported (and documented as such), I don't see anything wrong with publishing it anonymously, although I have never needed to such a thing. -- ==== Could someone please explain how to use the plot diff eq command for say y'=y passing through y(0)=1 paul Young ==== > Could someone please explain how to use the plot diff eq command for say > y'=y passing through y(0)=1 > paul Young Press simoultaneously [BLUE-SHIFT] and [F4] to go to the PLOT-SETUP screen. Choose PLOT TYPE: Diff Eq. Enter Y in input field F:. H-var: should be 0, V-var: should be 1. Enter X in the input field Indep: Press simoultaneously [BLUE-SHIFT] and [F2] to go to the PLOT WINDOW - DIFF EQ screen. Enter in H-View: (for example) -1 to 3 and in V-View: (for example) -5 to 20. In field Init: enter 0. In field Final: enter 2.5. Until now we have a viewing window from -1 to 3 (x-axis) and from -5 to 20 (y-axis) and the differential equation will be plotted from x=0 to x=2.5. Now, in field Init-Soln: enter 1. Now we have the initial x=0 and the initial y=1. Press the menu keys [ERASE] , then [DRAW] ( [F5], then [F6] ). Hope it helped, Nick. ==== > Hope it helped. It did, and thank you very much for that! > P.S. Bye and thanks for the fish. ;-) Hehe... YouÇre welcome... it just needs an eye in (x,y) = (4,1) ==== > Duh, no, stupid. > You're learning. > How dumb are you? > All is becoming clear. You're a wannabe Merkin. > Paul I'm sorry Paul, but your standard for discussion is beyond my american culture. You are welcome of course to continue such level of discussion in a pub's across Britain. http://HolyJoe.org ==== www.hpcalc.org www.alborowski.tk ftp://ftp.math.fu-berlin.de/pub/usr/raut/hp48/tools/ ==== > I'm starting to disassemble code on my 48GX [...] > to disassemble > the DROP command, it shows the address 03244 for the 48G series. Using > ASM->, I enter in this starting address, but how do I get the ending address > for this object? How will I know how long objects are from this list? It's not possible to know the length of code blocks in ROM, because they are not enclosed in any know structure, they're just 'free-floating code'. Well, almost: What you find in 03244 is 03249, which is 03244+5, which in turn tells you that 03244 is an RPL 'entry point', and that the assembly code starts at 03249. So, what you should do to find the end, is just guess! You're only limited by the available memory, so there's no problems if you disassemble too much. In your example, you can try something like #03249 DUP 100 + ASM->, you should get the code and find the sequence A=DAT0.A D0=D0+5 PC=(A) or sometimes a GOTO/GOLONG/GOVLNG to this sequence. In case you're disassembling an assembly routine, as opposed to an entry point, it usually stops with RTN (or related). If the ending address is too low, simply disassemble again with a bigger value. ==== > I'm starting to disassemble code on my 48GX and am using the HP48/49 Entry > Database 1.0 from Carsten Dominik, Thomas Rast, and Eduardo Kalinowski. The > object on level 1 or the starting and ending address of the object to be > disassembled. If I'm using the Entry Database for examble to disassemble > the DROP command, it shows the address 03244 for the 48G series. Using > ASM->, I enter in this starting address, but how do I get the ending address > for this object? How will I know how long objects are from this list? ==== I am not sure if Gilbert Urroz is a US professor, although he is involved in teaching. He has published a two volume set which I think must be the most detailed documentation on using the 49G, with common engineering problems as examples. This does include differential equations. You can get the publications from one of two Internet retailers, both are US based: greatunpublished.com and booksurge.com I bought electronic copies of both volumes and although I have not gone through them page by page, I do recommend them. I think the consensus is that volume 1 is a good manual on how to use the 49G. Volume 2 contains lots of applications of the 49G in engineering problems. I think it is probably worth buying the printed copies - you can get a free electronic copy if you seller that you want to do this in advance, to confirm that they will. Both volumes are weighty, at about 400 pages each, and I have heard (but cannot confirm) that the binding is not as strong as it needs to be for regular use with such large books. Someone posted to this group recently, mentioning that they had had their copies rebound with a spiral binding. A good idea, I think. I hope this is of some help. Dr Urroz has also written other volumes for the 48 series HP calculators. I assume that these are very similar to the 49 books. ==== http://www.engineering.usu.edu/cee/faculty/gurro/index.html ==== Does the 49 have a rectangular to polar conversion command? Or do you convert to polar by going to mode and setting it to polar? Is there a shortcut to do this? ==== I believe the command SPHERE will do it for you. Also, you can get to the MODES menu by the following methods: LS PRG, NXT, MODES Or << 65 MENU >> Does the 49 have a rectangular to polar conversion command? Or do you > convert to polar by going to mode and setting it to polar? Is there a > shortcut to do this? ==== Probabily this is a trivial question, but I am trying to update the value of a local variable with a given name (not nullname). I will appreciate your comments. ==== If you defined the local variables with the structure {{ A B C D }} Then you can store values by 123 A! 456 B! and so on. The RPL A! B! etc names are created by RPLCOMP when you compile. -- - - - - - - - - - - - - - - - - > Probabily this is a trivial question, but I am trying to update the value of > a local variable with a given name (not nullname). I will appreciate your > comments. ==== Because you mentioned LAM, I'm assuming this is a SystemRPL program. To update the contents of a local variable you use the ' character. Example: :: CK2NOLASTWD { LAM x LAM y } BIND :: LAM x LAM y %* ( Program assumes the two values in the stack are reals ) ' LAM x STO (Tick mark is used to leave the LAM name on the stack, STO uses it ) ; ABND ClrDAsOK ; If this isn't want you wanted answered, please let me know. I think this is what you were asking how to do, but if I was wrong, please forgive me. Doug > Probabily this is a trivial question, but I am trying to update the value of > a local variable with a given name (not nullname). I will appreciate your > comments. ==== ==== >So, I was investigating the posibility of getting two HP 39G and of >using its Applet feature to ease the transition. Some questions: > Can I run Applets with the 48 somehow? Not easily. It's a huge hassle (the 49G is easier in this respect so you may want to use that.) Frankly, if the money is not too much of a problem, you'll find life much easier if you buy yourself a 39 as well. Then you can spend your time helping your children, rather than learning to be a hacker. > Can I use a stack similar to the one in the 48, in the 39? No, not really. The 48 operates using Reverse Polish Notation only while the 39 uses algebraic notation only. (On the 39, the results of calculations appear on a stack, but note that it is a stack for results not operands.) > Is the 39 UserRPL programmable? Oh yes. > About IR; can I transmit via IR between 39 & 48? Yes, but... they are compatible at the infra-red level but not at the programming level because, as I have just said, the 39 is algebraic whilst the 48 is not. So, although programs can be transferred, the reality is that they will need to be significantly re-written in order to work. ==== Because the 39G uses the same chip as the 48 it is also programmable using sRPL but all the entry points are different and the operating system superimposed on top (aplets etc) makes programming much more are many games out there which have the source code included so that you could have a look at it. If you're after educational aplets for the 39G then see The HP HOME view at http://www.hphomeview.com So, I was investigating the posibility of getting two HP 39G and of The HP 48-49 and the HP 38-39-40 are totally incompatible. Only the HP 40 CAS and the HP 49 CAS are mostly compatible. Aplets can only be found in the 38-39-40 series. I am a father of two teens who are studying algebra, trigonometry and geometry. I recently purchased an HP 48GX to get my chops together in order to help out. I know about the virtues of the 48, but it seems to me that, as a teaching tool, it is not that streamlined. I am refering to teaching my kids, for my own refresher it is just what I like. And I really cannot get my kids interested in the 48 now (I know they will latter). So, I was investigating the posibility of getting two HP 39G and of using its Applet feature to ease the transition. ==== Is it possible to implicitly differentiate a function with more than one variable? An example would be: Find dy/dx given that y^3+y^2-5*y-x^2=-4 and the result would be: dy/dx = (2x)/(3*y^2+2*y-5) Find dy/dx given that y^3+y^2-5*y-x^2=-4 and the result would be: dy/dx = (2x)/(3*y^2+2*y-5) Replace all instances of y by y(x) to indicate that y is, implicitly, a function of x. On the HP48, put x on the stack below the equation and press the derivative operator, the right-shift SIN button. You shuld see an expression with 'dery(x,1)' representing the derivative, for which you can solve. On the HP49, you also start by replacing all y by y(x). Then do x DERIV, and then do 'd1(y(x)' SOLVE. You sould get 'd1y(x)=2*x/(3*y(x)^2+2*y(x)-5)' Your dy/dx is represented by 'd1y(x)' here. ==== Below follows a list of performance tests I ran a while ago using UserRPL in order to rank the abilities of my 49G. Please feel free to comment or enhance the list ROM 1.19.6 Related calc settings for both RPN and ALG tests: RPN, approximate mode, radians, clock off, complex off, functions-> numbers as answers, ENG to 4 decimals Timing obtained with command TEVAL. Answers for algebraics in RPN obtained with ->NUM. RPN PROGRAMS TIME in sec RATIOS 0-100 with counter x : 0.41 1 0-100 with RPN x^2 : 0.692 | 1 0-100 with algebraic x^2 : 13.4 | 19.4 0-100 with RPN SIN x : 1.977 | | 1 0-100 with algebraic SIN x : 14.68 | | 7.4 0-1000 with RPN SIN x : 19.9 | | | 1 0-1000 with algebraic SIN x : 275.04 | | | 13.8 | | | | | | | | ALG NOTATION PROGS in RPN MODE | | | | | | | | | | | | 0-100 with counter x : 2.5 6.1 | | | 0-100 with RPN x^2 : - | | | 0-100 with algebraic x^2 : 12.551 18.1 | | 0-100 with RPN SIN x : - | | 0-100 with algebraic SIN x : 13.833 7 | 0-1000 with RPN SIN x : - | 0-1000 with algebraic SIN x : 183.84 9.2 MODES : REAL, EXACT, 4 ENG inverse of random 80x80 matrix : 1335. (22.25 min.) sort list of 5000 random numbers 0 to 1 : 1102. (18.37 min.) sort list of 1000 random numbers 0 to 1 : 95.3 (1.58 min.) variance of 1000 random numbers 0 to 1 : 2.9 !Demeter! ==== Multiple precision library with + - * / sqrt y^x exp ln sin cos tan asin acos atan sinh cosh tanh and some auxiliary functions for both real and complex input. Integers, longfloats and long complex may generally be mixed. Results should be accurate to within +/- 1 in last digit. Precision is set in global or local variable DIGITS. Constants like pi ln(2), ln(10) are automatically stored when recalculated to higher precision in a library data. The library is to a great extent based on modified algorithms of ZMLIB ( a fortran package for multiple precision). pi is calculated using Werner Huysegoms implementation of Chudnowsky. And thanks to Carsten Dominik and Peter Geelhoed for emacs. Things not working: automatic list processing ( but DOLIST/DOSUBS ok) until it shows up at hpcalc. Some timings below: DIGITS = 1000, pi precomputed Sqrt(pi) 11.7 sec /<< DUP FSQRT FSQ FSUB />> DIGITS=100, x=sqrt(pi), z = longcomplex((123456789 , 0.123456789 )) sqrt(x) .46 sec ln(x) 9.2 sec exp(x) 5.1 sec sqrt(z) 1.6 sec tan(z) 7.0 sec z-exp(ln(z)) /<< DUP FLN FEXP SWAP FSUB />> ~ (-3.E-91, 4.E-96) 21.7 sec x = 2^330 y = 2^3330 ZSqrt( x ) -> TRUE .31 sec ZSqrt( y ) -> TRUE 7.4 sec ==== wrt the 49. No answers yet so I will assume no one uses these tools or knows anything since it would be out of the question to assume I am simply being ignored. :-] Now I will try to get some help with Hptools. I am using them to make a library and I have noticed that, although the compilation of my rpl source goes fine and the object is good when loaded to my calc, it appears that none of rplcomp, sasm, makerom or lbcrc want to alert me to errors in my code. I can merrily compile, link and upload source with bogus tokens in it like ISIX, FOO, DooptyDo etc and the only alert would be to look at the interim rplcomp and sasm output files. Is this supposed to be how it works? I seem to remember (or dreamed) that some syntax checking was done. As it is I made a script to handle all of the build work in one swoop and then delete any interim files. That is dangerous I now see. Any wisdom to share? -- It aint what ya don't know but what ya know that aint so that can get you in trouble. - P. Thomas ==== I am getting very close to releasing Debug4x for the HP48 and HP49. Debug4x is the second release of the much older and buggier Debug2. Debug4x has been in Beta test by a number of people on this newsgroup. It will solve all your problems and complaints with HPTools, error messages etc. It is fully integrated with the Windows Environment and supports source level debugging. files and samples). ==== The only problem with the Debug4x is that the other beta testers are so quick to document a problem that while I'm finished, there's another , new, better version available. All I can say now that with this Xtreme rate of development we will not have to wait until Xmas and the best thing is: the new tool is 48G/49G aware and enormously helps releasing applications simultaneously for both calculators. Amazing! > Al > I am getting very close to releasing Debug4x for the HP48 and HP49. Debug4x is the second release of the much older and buggier > Debug2. > Debug4x has been in Beta test by a number of people on this newsgroup. It will solve all your problems and complaints with HPTools, > error messages etc. It is fully integrated with the Windows Environment and supports source level debugging. > place to download the file (about 6 MB now with good help > files and samples). ==== wrt the 49. No answers yet so I will assume no one uses these tools or > knows anything since it would be out of the question to assume I am > simply being ignored. :-] Now I will try to get some help with Hptools. HPTools and Debug2 are the best development tools available for the HP4x series. I am using them to make a library and I have noticed that, although > the compilation of my rpl source goes fine and the object is good when > loaded to my calc, it appears that none of rplcomp, sasm, makerom or > lbcrc want to alert me to errors in my code. I can merrily compile, > link and upload source with bogus tokens in it like ISIX, FOO, > DooptyDo etc and the only alert would be to look at the interim > rplcomp and sasm output files. Is this supposed to be how it works? It all depends ... If you have an error in your SysRPL or ML code (e.g syntax error vs calling a non-existent entry point) then yes, rplcomp/sasm will error provided that you asked them to choke on errors. Regarding unresolved entry-points, you have to use SLOAD at one stage when you're compiling. SLOAD will generate a file with the .LR extension. You have to check this file to see if there's any unresolved entry-points. Alternatively, you can simply use Debug2 which will take care of everything (compile, link and create the library) > I seem to remember (or dreamed) that some syntax checking was done. > As it is I made a script to handle all of the build work in one swoop > and then delete any interim files. That is dangerous I now see. Any wisdom to share? I've made a shell file that checks if there's any errors between RPLCOMP and SASM, it's available on my HP38 web page ==== HPTools and Debug2 are the best development tools available for the HP4x > series. > It all depends ... > If you have an error in your SysRPL or ML code (e.g syntax error vs calling > a non-existent entry point) then yes, rplcomp/sasm will error provided that > you asked them to choke on errors. > Regarding unresolved entry-points, you have to use SLOAD at one stage when > you're compiling. > SLOAD will generate a file with the .LR extension. You have to check this > file to see if there's any unresolved entry-points. I realize this. As a matter of fact if the source is not a library sload will do just that. However it will not even generate the .lr file as requested in the control file. It just skips by all together. > I've made a shell file that checks if there's any errors between RPLCOMP and > SASM, it's available on my HP38 web page ==== How can I view a vector on several lines? eg. when I Put the vector [15. 14.] (in rectangular mode) and change to cylindrical mode, the vector is only shown on one line. [20.5182845287 < .75... How can I view this on several lines? thanks ==== Installed the new filer2 and I now get the error message Erreur: Type Argument Incorrect. It also appears to have set the calculator so some prompts and messages are now in french. How do I change my HP49 back to english? Found it 0 ->LANGUAGE, obvious! Also tried it in EMU and get same error but in english :). Temporarily gone back to the old Filer2. > I finished my work on Filer1 and Filer2 with > the realization of something I was dreaming of > When hiding files inside the filer you'll see > what is going on (creation of a nullname etc). An advantage of hiding is to keep a directory > clean, but the main advantage is to make hidden > files more safe against unintended overwriting > and other misuse. Since hiding is now readily done in the filer, > I eliminated the hiding tools from OT49 in the > latest revision on my site. These are replaced > by some important system parameter togglers: > ~ANG, ~COOR. ~BASE, ~FONT, ~HEAD and ~LANG for > toggling the three builtin languages which makes > OT49 still more user-friendly. The language will > perhaps be more interesting in future because the > ROM 19-7 filer will display its basic messages > language-sensitive. Therefore, the new filers > have also a language toggler integrated. People who want to be independent on filers may > also use the very small revised library HIDE or > compatible with all well-programmed hiding tools. Have fun > ftp://ftp.math.fu-berlin.de/pub/usr/raut/HP49/tools ==== Installed the new filer2 and I now get the error message Erreur: Type Argument Incorrect. It also appears to have set the calculator so some prompts and messages are now in french. How do I change my HP49 back to english? lot > to call up a list of what I may have hidden away. My favourite is in my CST > hidden variable so you can work on them with the function keys just like > normal variable when I want to. The CrM is another one of your utilities > that adds EXIT to the end of the menu to return you to the previous menu. Stephen.N > I finished my work on Filer1 and Filer2 with > the realization of something I was dreaming of > When hiding files inside the filer you'll see > what is going on (creation of a nullname etc). An advantage of hiding is to keep a directory > clean, but the main advantage is to make hidden > files more safe against unintended overwriting > and other misuse. Since hiding is now readily done in the filer, > I eliminated the hiding tools from OT49 in the > latest revision on my site. These are replaced > by some important system parameter togglers: > ~ANG, ~COOR. ~BASE, ~FONT, ~HEAD and ~LANG for > toggling the three builtin languages which makes > OT49 still more user-friendly. The language will > perhaps be more interesting in future because the > ROM 19-7 filer will display its basic messages > language-sensitive. Therefore, the new filers > have also a language toggler integrated. People who want to be independent on filers may > also use the very small revised library HIDE or > compatible with all well-programmed hiding tools. Have fun > ftp://ftp.math.fu-berlin.de/pub/usr/raut/HP49/tools ==== I again slightly reprogrammed Filer1 and Filer2 by reading now most menu names from ROM. Therein the names are generally language-sensitive, e.g., the english words VIEW and TREE will automatically translated into the elegant french words VOIR and ARBO whenever you switch to French. In Filer1 this is done with key L (NXT), dispensable in Filer1). In Filer2 there is a separate language toggler on page 2 of the filers appl menu. The problem is that I do not know exactly whether the words VOIR, ARBO etc already exist in ROM 19-6 (as I mentioned already I have a slightly more modern flashbank 0 containing all the above). Thus, please be so kind to load Filer1NEW and Filer2NEW from by site below and test whether the menu names really respect the language. As soon as I know that it works I will overwrite Filer1 by Filer1NEW and corresponding for 2. Please try also the following: Suppose you have a backup of the entire HOME in a port which is classified as a directory. Drive with righshift RIGHTarrow into this directory inside the filer (if this is possible) and look whether you see at the end a directory with a nullname. If yes, not drive into it with righshift RIGHTarrow! Your calculator may crash, in the best case it warmstarts ftp://ftp.math.fu-berlin.de/pub/usr/raut/HP49/tools ==== Please note that On starting i get an errn # 513d. with filer2. I again slightly reprogrammed Filer1 and Filer2 by > reading now most menu names from ROM. Therein the > names are generally language-sensitive, e.g., the > english words VIEW and TREE will automatically > translated into the elegant french words VOIR and > ARBO whenever you switch to French. In Filer1 this > is done with key L (NXT), dispensable in Filer1). > In Filer2 there is a separate language toggler on > page 2 of the filers appl menu. The problem is that I do not know exactly whether > the words VOIR, ARBO etc already exist in ROM 19-6 > (as I mentioned already I have a slightly more > modern flashbank 0 containing all the above). Thus, please be so kind to load Filer1NEW and > Filer2NEW from by site below and test whether > the menu names really respect the language. As > soon as I know that it works I will overwrite > Filer1 by Filer1NEW and corresponding for 2. Please try also the following: Suppose you have > a backup of the entire HOME in a port which is > classified as a directory. Drive with righshift > RIGHTarrow into this directory inside the filer > (if this is possible) and look whether you see > at the end a directory with a nullname. If yes, > not drive into it with righshift RIGHTarrow! > Your calculator may crash, in the best case > it warmstarts > ftp://ftp.math.fu-berlin.de/pub/usr/raut/HP49/tools ==== I'm a beginner at using my HP49g. Is there a quick way to apply an > operation to all elements on the stack? ie: > 1 2 3 4 5 And then a + to ALL or - to ALL to return: 1 + 2 + 3 + 4 + 5 or > 1 - 2 - 3 - 4 - 5 respectively. > For addition, use: << WHILE DEPTH 1 > REPEAT + END >> For subraction use << WHILE DEPTH 1 > REPEAT - END >> ==== > I'm a beginner at using my HP49g. Is there a quick way to apply an > operation to all elements on the stack? ie: > 1 2 3 4 5 And then a + to ALL or - to ALL to return: 1 + 2 + 3 + 4 + 5 or > 1 - 2 - 3 - 4 - 5 respectively. For addition, use: > << WHILE DEPTH 1 > REPEAT + END >> Or, << DEPTH ->LIST 0 + gSLIST >> ==== > I'm a beginner at using my HP49g. Is there a quick way to apply an > operation to all elements on the stack? ie: > 1 2 3 4 5 And then a + to ALL or - to ALL to return: 1 + 2 + 3 + 4 + 5 or > 1 - 2 - 3 - 4 - 5 respectively. For addition, use: > << WHILE DEPTH 1 > REPEAT + END > For subraction use > << WHILE DEPTH 1 > REPEAT - END >> for addition: << 3. DEPTH START + NEXT >> for subtraction: << 0. 4. DEPTH START + NEXT - >> remark Virgil that your version returns 1 - 2 + 3 - 4 + 5 ==== This is another way This sum the elements << DEPTH ->LIST (SUM)LIST >> 1+2+3+4+5 This rest the elements << DEPTH 1 - ->LIST 1 NEG * (SUM)LIST + >> 1-2-3-4-5 SUM is the greek symbol for sum (Right Shift + [SIN] )and the function is in the MTH LIST menu. > I'm a beginner at using my HP49g. Is there a quick way to apply an > operation to all elements on the stack? ie: > 1 2 3 4 5 And then a + to ALL or - to ALL to return: 1 + 2 + 3 + 4 + 5 or > 1 - 2 - 3 - 4 - 5 respectively. For addition, use: > << WHILE DEPTH 1 > REPEAT + END > For subraction use > << WHILE DEPTH 1 > REPEAT - END >> ==== A) -98 SF @ Vector vertical display you don't need several lines with B) 4. FIX @ or 3. ENG or... with good vision you could have C) -72 SF @ Stack minifont How can I view a vector on several lines? > eg. when I Put the vector [15. 14.] (in rectangular mode) and change to cylindrical mode, the vector is only shown on one line. [20.5182845287 < .75... How can I view this on several lines? thanks == 1439 ==== > some of the recycled electrons going around have been used too many > times and can produced a low power effect that will eventually affect > the virility of the computer Bovine faeces, electrons are the viagra of processors :-) -- This message was written with 100% recycled electrons Pivo ==== My interest was possibly being able to get a good bulk deal for my deal for delivery outside the US for a large order (>100). >>Anyone know anything about an online store called Fry's Electronics >>Outpost.com? They're advertising a 39G for $49.99 . . . . >Unfortunately it's apparently only for buyers in the US (which I'm not). > Looks like a good price, new or used. You might want to keep an eye on > eBay. Every so often some clearing house puts several up at a time, > for around $30. - Paul > ==== Forced CG: MEM > It seems that every few weeks if the stack has built up to something huge > Garbage Collection pauses reappear. Clearing the stack then gets rid of any > noticable pause. It may not be the size of stack causing the re-appearance > of the pauses, it may be something else but since clearing the stack fixes > it I haven't looked further. I've successfully completed the upgrade to 19.6, and the apparent > garbage collection problem I've been having looks much improved. I won't > know for sure until I use it for a couple of weeks, as I don't have a > good way to benchmark the problem. Steve Smith > ==== Is there an HP49 emulator for Windows XP? ==== I use both the original hp emulator (calc.exe) with new ROM and the Emu49, with 49G KML and ROM They work fine with any 32-bit windows that I have tested. They can be found in www.hpcalc.org > Is there an HP49 emulator for Windows XP? ==== > When we make a new rom, only the Stable Entry points are checked... stable > 'area' are not, and we had to modify one of those a couple of rom ago in > order to add some patch, so these area are not to be trusted as we might > have to change them at one point or another. I agree that things like the > system part (where stack manip, prolog and similar are) is unlikely to > change, but you never know... This completely contradicts to what JYA said in this NG: The pointer interval areas 25EC - B3C7 and 25565 - 4000 are stable and will not change, at least not for the HP49 in its present shape. We could perhaps live with some limited modifification if it is clearly defined and announced in due time. But not without using pointers from these stable areas. That would make many high-quality tools like Emacs, Unitman, Keyman etc worthless and unusable. Please tell us if such modification will take place in ROM 19-7. If yes, at least I will immedately stop any further tool developement for the 49. Do you know how much time we invested in developing our Wolfgang PS. If you plan to emulate the HP49 on some science fiction PDA, various things may happen. I refer here only to the HP49 itself. ==== > This completely contradicts to what JYA said in this NG: The pointer > interval areas 25EC - B3C7 and 25565 - 4000 are stable and will not > change, at least not for the HP49 in its present shape. And I maintain it. I've always made sure when I modified that area that all the other entry points didn't move !! I do not know what Cyrille is referring to. ==== > This completely contradicts to what JYA said in this NG: The pointer > interval areas 25EC - B3C7 and 25565 - 4000 are stable and will not > change, at least not for the HP49 in its present shape. > And I maintain it. > I've always made sure when I modified that area that all the other entry > points didn't move !! > I do not know what Cyrille is referring to. that takes a load off my mind. It will make tools development much easer and more efficient with the new extable presently moderated by Thomas Rast. << ... I can remember at least 1 situation where I did [some change] > Well, a single change doen't matter much and will discovered sooner or later. Furthermore, if some change is indeed unavoidable but announced in due time, this also would be no serious problem. The only way to be sure is perhaps that nobody else but you should be allowed to make changes in the ROM. I know Cyrille. One cannot 100% sure whether he remembers well. He is sometimes an upsentminded professor (of which I'm a constant example :-). my site (under Download julia.jpg)? ==== > my site (under Download julia.jpg)? No I didn't.. but I will surely have a look :) ==== Maybe you could extend the amount of stable entry points? When we make a new rom, only the Stable Entry points are checked... stable > 'area' are not, and we had to modify one of those a couple of rom ago in > order to add some patch, so these area are not to be trusted as we might > have to change them at one point or another. I agree that things like the > system part (where stack manip, prolog and similar are) is unlikely to > change, but you never know... X > stable regions in the ROM in order to indicate that these entries can > safely be used. Is this not the case? Are you indicating we should > better not produce a version of extable which contains all the entries > we found in the stable regions? - Carsten ==== I'm looking for some HP-41 accessories, including the following: SUP-R-ROM (surveying) CO-GO41 ROM (surveying) HEPAX modules ZENROM module CCDROM module EPROM or RAM based ROM module simulators (MLDL, Protocoder, ROMBOX, etc.) or other special modules or accessories for the HP-41. I will pay good price, or will trade for other HP calculators or accessories. Please contact me at peter@hst-netzwerke.de ==== have you been looking for them at Ebay?... in Germany too many collectors are raising one another making the mentioned modules reaching horrible prices... during the last month, I have seen the HEPAX- and the ZENROM module beeing sold for more than EUR 400,- !!! (for one module!!!). I would also like to have the HEPAX and the CCDROM modules but the prices are way too high... unfortunately... Martin > I'm looking for some HP-41 accessories, including the following: SUP-R-ROM (surveying) > CO-GO41 ROM (surveying) > HEPAX modules > ZENROM module > CCDROM module > EPROM or RAM based ROM module simulators (MLDL, Protocoder, ROMBOX, etc.) > > or other special modules or accessories for the HP-41. I will pay good price, or will trade for other HP calculators or > accessories. Please contact me at peter@hst-netzwerke.de > ==== Martin Pedersen schrieb im Newsbeitrag > have you been looking for them at Ebay?... in Germany too many > collectors are raising one another making the mentioned modules reaching > horrible prices... during the last month, I have seen the HEPAX- and the > ZENROM module beeing sold for more than EUR 400,- !!! (for one module!!!). > but note that they weren't bought by a German collector. Instead, someone from USA (p.s.) who seems to have way too much money is fishing here;-) Raymond ==== no, I wasn't aware of that... thanks for the info... Martin Pedersen schrieb im Newsbeitrag >have you been looking for them at Ebay?... in Germany too many >>collectors are raising one another making the mentioned modules reaching >>horrible prices... during the last month, I have seen the HEPAX- and the >>ZENROM module beeing sold for more than EUR 400,- !!! (for one module!!!). > but note that they weren't bought by a German collector. > Instead, someone from USA (p.s.) who seems to have way too much money is fishing > here;-) > Raymond ==== on a TI-89 i could type: solve(3x=18,x) how would i do this on the 49? ==== > on a TI-89 i could type: solve(3x=18,x) how would i do this on the 49? Type SOLVE(3x=18,x) or type SOLVEVX(3X=18) , if VX=X The result is the same in both. Grettings from Cartagena - Chile (sorry my english... ;-) ==== > on a TI-89 i could type: solve(3x=18,x) how would i do this on the 49? Assuming that you are in RPN mode and exact mode: '3*X = 18' ENTER 'X' SOLVE ==== Well, I now tried to transmit from one HP49 to another using Xmodem. I opened the I/O input form, set the transmit mode to Xmodem, in the next field down I enter the file to transmit. I set the receiving calculator to Xmodem server by pressing right shift-right arrow, finally I press the F key for XSEND and nothing happens. HELP!!!!! ==== Use Left Shift IO then NXT. XXMIT, XRECV are on the E and F keys respectively. Bill > alternate E-dress wtstorey@ieee.org.no.spam.please Ok, so I enter the file to send and choose XXMIT, but what do I do on the receiving machine (PC or another HP48) If the receiving machine is an HP49 then I guess I just set it to Xmodem server mode, but how do I do the equivalent on the HP48 or the PC? Luis ==== >Ok, so I enter the file to send and choose XXMIT, but what do I do on >the receiving machine (PC or another HP48) If the receiving machine is >an HP49 then I guess I just set it to Xmodem server mode, but how do I >do the equivalent on the HP48 or the PC? >Luis I have only used XModem for PC to HP48. However, since the receiving HP48 would be the same, put the name of the receive file on the stack and type XRECV. I can't help you on the PC end for the receive. ==== Looked in the manual(s), both in the index and contents... Wanting to know how to program/draw graphics objects (with text), and am also want to know if they can be combined with choose boxes. MC For the 49G sorry, my mistake, was wondering why there where no repleys :) ==== > I'm not sure I'm following the question. I have the URLs of some of the > sites, > I don't recall if any of them had available source code, though they might. > Is > that what you mean? well, yes I mean any information would be useful. Sorry for the delay. For anyone who was interested here's a few URL's of various battlebots. http://www.gamerz.net/~c++robots/ http://www.necrobones.com/atrobots/index.htm http://www.cyty.com/robocom/ -- Jeffery http://woodfanclub.com No, it's not vanity, just desperation. ==== The upgrade process stops for some reason, on file 3. I formatted bank zero for some reason (was confused by the instructions), don't know if this is the problem??? The calculator was resetting when I started it, thought this might fix it... MC ==== In Nick's Complex number marathon, he refers to 256 Menu Upon excuting this command one gets number of menu options. I cannot find these in the standard HP documentation (advnced user guide). Where can I find n explanation of all the options. Are there any other hidden menu options not properly documented ? Aubrey. ==== http://www.hpcalc.org/details.php?id=2990 > In Nick's Complex number marathon, he refers to 256 Menu > Upon excuting this command one gets number of menu options. I cannot > find these in the standard HP documentation (advnced user guide). > Where can I find n explanation of all the options. > Are there any other hidden menu options not properly documented ? > Aubrey. ==== (not the > 48 one). > Serge try http://www.google.ca/search?q=rpncalc&ie=UTF-8&oe=UTF-8&hl=en&btnG=Google+ Se arch&meta= in particular http://www.gnu.org/directory/Mathematics/rpncalc.html ==== try http://www.google.ca/search?q=rpncalc&ie=UTF-8&oe=UTF-8&hl=en&btnG=Google+Se a rch&meta= > in particular > http://www.gnu.org/directory/Mathematics/rpncalc.html I'm interested in a hp28s emulator as well- is this gnu project an rpn > calculator or a real RPL programmable HP28S emulator? It's not programmable ==== > I tried changing the with of browser lines by hand (message 59). I > could make the line less wide all right, but I could not increase the > width. So it seems to me that the truncation of strings to about 18 > characters may be hard-coded into the engine. Probably something >A workaround will be to use the message handler and do the string convesion >yourself. Quite easy things to do. >The only place where the full browser is used is in the Flag manager, and >it's using graphics rather than text. > Will this be fixed in an upcoming ROM 1.19-7 ? TIA Andreas ==== > Will this be fixed in an upcoming ROM 1.19-7 ? If it is lodged in bigs.hpcalc.org maybe ... It's the only way I have now to trace bugs or enhancements. ==== Lets see what HP folks think about this... > >Message-id: <6MPX8.504338$cQ3.44713@sccrnsc01 >Message-id: <3D2EAE15.9D17EB82@pobox.com> > >Message-id: <3d2ce911.329838465@news.west.cox.net> >Are there any manufacturers, other than the HP's (32,41,48,49), that > >use RPN? > >TIA I remember that TI had some a while back. I don't know if they > >still do? ( of course... I've never been able to figure out why so many people > >like > RPN. I view it as something that has some decent computer coding > >applications > - I've used it for such - but.. for day to day calculations, I have > >always > found RPN to be a pain in the backside. That's part of why I am not > >an HP > >user > myself ) Dan :-) >RPN is faster because there are no ( ) keys. *Real* engineers use RPN >: -- >________________ > >Rob Munach, PE > >Excel Engineering > >Carrboro, NC He he. I've heard that one many times!!! LOL. > The amazing thing I find is... so many engineers actually believe that > >one. > ( about being faster I mean ) Dan :-) >IT IS FASTER Chuck, Chuck, Chuck??? NO. IT's NOT!! hehe ;-) Fewer keystrokes does not equate to faster. If it did... then I would have > lost most of those just for fun calculator races ( where you both see who can > punch in the equation the fastest and get the correct answer ). Fact is... NO > ONE using an RPN calculator has ever beaten me ( a lot have tried ). And... > I'm talking big equations as well as small. That means... at least in my > experience ( which I believe is actually fairly substantial ) RPN is not > faster. Sure, HP has published that in their manuals and stood by it. But of > course they have!!! They DO want to sell their calculators don't they? RPN often does take fewer keystrokes. But... really... who gives a crap > about that. There's no advantage to that ( it certainly does not give a speed > advantage ). I've also noticed that a lot of engineers using RPN are actually > slow with their calculators. My guess there is that they have to think more > about how to punch their numbers correctly into the calculator simply because > the order of data entry using RPN is less intuitive to them than the algebraic > math they generally read and write. If someone wants to use RPN... that's fine. But... to date... I've never > heard an argument that's even remotely convincing on why RPN is better. > Personally... I think lots of engineers have been essentially brainwashed > into actually believing it is better ( usually because they are erroneously > told that it is faster due to the fewer keystroke thing ). Dan :-) ==== > Lets see what HP folks think about this... > >RPN is faster because there are no ( ) keys. *Real* engineers use > RPN That's right ! X > Fewer keystrokes does not equate to faster. If it did... then I would > have > lost most of those just for fun calculator races ( where you both see > who can > punch in the equation the fastest and get the correct answer ). Fact > is... NO > ONE using an RPN calculator has ever beaten me ( a lot have tried ). X And I won all the class math championships where we were not allowed to use even pencil and paper system. No errors here. All 10. Does that make human brains (not counting Rainmans) better on number-crunching than pen&paper system or a calculator, a PC? Nope! It just shows that one individual can be better than others in one subject. Maybe I can beat everyone in a calc race? Maybe I'll loose in a math_with_brains -contest. Surely a speedy man with a fast mind and even faster fingers would beat the calc race using ANY calculator system. ugh Brother-Peter ==== The screen is too small to show the calculator (HP48/49) See them side by side (a real calc & a WinPPC) I use Jornada 720 and even that is not so good looking. > has any tried using emulators with a pocket PC? ==== Why not design your own screen? It's not hard and you can get exactly the size and layout that you want. > The screen is too small to show the calculator (HP48/49) > See them side by side (a real calc & a WinPPC) > I use Jornada 720 and even that is not so good looking. >has any tried using emulators with a pocket PC? > ==== Colin Croft schrieb im Newsbeitrag > I'm not joking, it's really not that hard. Christoph clearly put a lot > of thought into making it as easy as possible for people to create their > own versions. TNX for the flowers, but the KML script design was already made by Sebastien Carlier. > [..] > All you have to do then is to create a KML file for your image > which just contains a whole set of key locations along the lines of: > Button 11 > Type 5 > Size 64 42 > Offset 26 344 > OutIn 1 16 > End And sorry that I have to disappoint you a 2nd time. It's true that I implemented the transparent circle as button type 5 in Emu48 (Win32 version since v1.21), but in all WinCE versions I know type 5 works like type 0 and that isn't the transparent circle mode. :( [..] Christoph ==== and others will actually have to put some serious thought and Colin Croft schrieb im Newsbeitrag >I'm not joking, it's really not that hard. Christoph clearly put a lot >>of thought into making it as easy as possible for people to create their >>own versions. > TNX for the flowers, but the KML script design was already made by Sebastien > Carlier. >[..] > All you have to do then is to create a KML file for your image >>which just contains a whole set of key locations along the lines of: >>Button 11 >>Type 5 >>Size 64 42 >>Offset 26 344 >>OutIn 1 16 >>End > And sorry that I have to disappoint you a 2nd time. It's true that I > implemented the transparent circle as button type 5 in Emu48 (Win32 version > since v1.21), but in all WinCE versions I know type 5 works like type 0 and > that isn't the transparent circle mode. :( [..] > Christoph > ==== I told you so! I tried once quickly a year ago and yes it takes some effort, but I'll try 2nd time after my VOCATION! > and others will actually have to put some serious thought and X > And sorry that I have to disappoint you a 2nd time. It's true that I > implemented the transparent circle as button type 5 in Emu48 (Win32 version > since v1.21), but in all WinCE versions I know type 5 works like type 0 and > that isn't the transparent circle mode. :( ==== I have a 620 myself; there is a script for EMU48 for WinCE 2 that splits in half and displays horizontally. It is photorealistic -- very nice. > The screen is too small to show the calculator (HP48/49) > See them side by side (a real calc & a WinPPC) > I use Jornada 720 and even that is not so good looking. > has any tried using emulators with a pocket PC? > -- ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Kenton W. Irby Mostly Harmless! kwirby@earthlink.net The answer to the ultimate question of life, the universe, and everything is ... * + * FORTY-TWO! * + * (Douglas Adams) ==== I really can't understand what's happening to my HP 48G. Suddenly, the Cancel button (the ON button) stopped working. I can turn it on, but when I try to use this button, it simply doesn't work. If you have any idea of what can I do, please tell me. I'll describe exactly what happened and what I've already tried. What happened.... I used the calc. the last time in the morning. In the afternoon, when I tried to turn it on, it simply didn't. I took out the batteries, reseted, but nothing. After sometime though, it did turn on, but when I pressed the cancel button, it didn't work (for example: press number 1 and press the Cancel/ON button - it should disappear, but it doesn't). Every single thing works just fine, the serial communication, beep, calculations, etc. The only thing is that Cancel/ON button. The button itself is working, because I can turn on the calc. and when I press and hold it and, for instance, digit 556, it doesn't appear on the display, just when I release the Cancel/On button. Strange. I can't turn the calc. Off by pressing shift ON, because it doesn't work. No probs, I write OFF and press Enter and it turns Off. But when I press the ON button, the HP doesn't turn on, only if I hold it by aprox. 5 seconds or if I wait some hours (can it be a capacitor??? Witch one??). One of the most weird thing is that, I don't know why, after I did lots of things, it started working just fine, I could turn it on and off repeatedly, cancel and so on, just OK. But, on the other day it stopped working again and since then (Friday July, 12 2002) it came this way again. What I tried.... I opened the HP (very hard job, I should say), took out the batteries, reseted the calculator with the reset button (I can't use ON+A+F) and I even try to (after take out the batteries) short cut the capacitor that holds RAM while you change batteries. It asked me to try to recover memory, I pressed NO, and so on. So, it's too weird for me. Everything works but the Cancel/ON button, except to turn the HP on and just after some hours or holding it for more than 5 seconds. If you've ever seen it or have any idea, PLEASE tell me. Luciano Enzweiler ==== This is a rare fault in the old HP 49G series. You should press just under the LCD to use the [ON]. Since you have already opened the calculator, try to get pressure in the area between the LCD and the top row of keys. In the future you may also loose control of the shift keys and finally also the alpha key will be dead. > I really can't understand what's happening to my HP 48G. Suddenly, the > Cancel button (the ON button) stopped working. I can turn it on, but > when I try to use this button, it simply doesn't work. If you have any > idea of what can I do, please tell me. I'll describe exactly what > happened and what I've already tried. What happened.... > I used the calc. the last time in the morning. In the afternoon, when > I tried to turn it on, it simply didn't. I took out the batteries, > reseted, but nothing. After sometime though, it did turn on, but when > I pressed the cancel button, it didn't work (for example: press number > 1 and press the Cancel/ON button - it should disappear, but it > doesn't). Every single thing works just fine, the serial > communication, beep, calculations, etc. The only thing is that > Cancel/ON button. The button itself is working, because I can turn on > the calc. and when I press and hold it and, for instance, digit 556, > it doesn't appear on the display, just when I release the Cancel/On > button. > Strange. I can't turn the calc. Off by pressing shift ON, because it > doesn't work. No probs, I write OFF and press Enter and it turns > Off. But when I press the ON button, the HP doesn't turn on, only if I > hold it by aprox. 5 seconds or if I wait some hours (can it be a > capacitor??? Witch one??). > One of the most weird thing is that, I don't know why, after I did > lots of things, it started working just fine, I could turn it on and > off repeatedly, cancel and so on, just OK. But, on the other day it > stopped working again and since then (Friday July, 12 2002) it came > this way again. What I tried.... > I opened the HP (very hard job, I should say), took out the batteries, > reseted the calculator with the reset button (I can't use ON+A+F) > and I even try to (after take out the batteries) short cut the > capacitor that holds RAM while you change batteries. It asked me to > try to recover memory, I pressed NO, and so on. So, it's too weird for me. Everything works but the Cancel/ON button, > except to turn the HP on and just after some hours or holding it for > more than 5 seconds. If you've ever seen it or have any idea, PLEASE > tell me. > ==== 1439 ==== > some of the recycled electrons going around have been used too many > times and can produced a low power effect that will eventually affect > the virility of the computer Bovine faeces, electrons are the viagra of processors :-) -- This message was written with 100% recycled electrons Pivo ==== My interest was possibly being able to get a good bulk deal for my deal for delivery outside the US for a large order (>100). >>Anyone know anything about an online store called Fry's Electronics >>Outpost.com? They're advertising a 39G for $49.99 . . . . >Unfortunately it's apparently only for buyers in the US (which I'm not). > Looks like a good price, new or used. You might want to keep an eye on > eBay. Every so often some clearing house puts several up at a time, > for around $30. - Paul > ==== Forced CG: MEM > It seems that every few weeks if the stack has built up to something huge > Garbage Collection pauses reappear. Clearing the stack then gets rid of any > noticable pause. It may not be the size of stack causing the re-appearance > of the pauses, it may be something else but since clearing the stack fixes > it I haven't looked further. I've successfully completed the upgrade to 19.6, and the apparent > garbage collection problem I've been having looks much improved. I won't > know for sure until I use it for a couple of weeks, as I don't have a > good way to benchmark the problem. Steve Smith > ==== Is there an HP49 emulator for Windows XP? ==== I use both the original hp emulator (calc.exe) with new ROM and the Emu49, with 49G KML and ROM They work fine with any 32-bit windows that I have tested. They can be found in www.hpcalc.org > Is there an HP49 emulator for Windows XP? ==== > When we make a new rom, only the Stable Entry points are checked... stable > 'area' are not, and we had to modify one of those a couple of rom ago in > order to add some patch, so these area are not to be trusted as we might > have to change them at one point or another. I agree that things like the > system part (where stack manip, prolog and similar are) is unlikely to > change, but you never know... This completely contradicts to what JYA said in this NG: The pointer interval areas 25EC - B3C7 and 25565 - 4000 are stable and will not change, at least not for the HP49 in its present shape. We could perhaps live with some limited modifification if it is clearly defined and announced in due time. But not without using pointers from these stable areas. That would make many high-quality tools like Emacs, Unitman, Keyman etc worthless and unusable. Please tell us if such modification will take place in ROM 19-7. If yes, at least I will immedately stop any further tool developement for the 49. Do you know how much time we invested in developing our Wolfgang PS. If you plan to emulate the HP49 on some science fiction PDA, various things may happen. I refer here only to the HP49 itself. ==== > This completely contradicts to what JYA said in this NG: The pointer > interval areas 25EC - B3C7 and 25565 - 4000 are stable and will not > change, at least not for the HP49 in its present shape. And I maintain it. I've always made sure when I modified that area that all the other entry points didn't move !! I do not know what Cyrille is referring to. ==== > This completely contradicts to what JYA said in this NG: The pointer > interval areas 25EC - B3C7 and 25565 - 4000 are stable and will not > change, at least not for the HP49 in its present shape. > And I maintain it. > I've always made sure when I modified that area that all the other entry > points didn't move !! > I do not know what Cyrille is referring to. that takes a load off my mind. It will make tools development much easer and more efficient with the new extable presently moderated by Thomas Rast. << ... I can remember at least 1 situation where I did [some change] > Well, a single change doen't matter much and will discovered sooner or later. Furthermore, if some change is indeed unavoidable but announced in due time, this also would be no serious problem. The only way to be sure is perhaps that nobody else but you should be allowed to make changes in the ROM. I know Cyrille. One cannot 100% sure whether he remembers well. He is sometimes an upsentminded professor (of which I'm a constant example :-). my site (under Download julia.jpg)? ==== > my site (under Download julia.jpg)? No I didn't.. but I will surely have a look :) ==== Maybe you could extend the amount of stable entry points? When we make a new rom, only the Stable Entry points are checked... stable > 'area' are not, and we had to modify one of those a couple of rom ago in > order to add some patch, so these area are not to be trusted as we might > have to change them at one point or another. I agree that things like the > system part (where stack manip, prolog and similar are) is unlikely to > change, but you never know... X > stable regions in the ROM in order to indicate that these entries can > safely be used. Is this not the case? Are you indicating we should > better not produce a version of extable which contains all the entries > we found in the stable regions? - Carsten ==== I'm looking for some HP-41 accessories, including the following: SUP-R-ROM (surveying) CO-GO41 ROM (surveying) HEPAX modules ZENROM module CCDROM module EPROM or RAM based ROM module simulators (MLDL, Protocoder, ROMBOX, etc.) or other special modules or accessories for the HP-41. I will pay good price, or will trade for other HP calculators or accessories. Please contact me at peter@hst-netzwerke.de ==== have you been looking for them at Ebay?... in Germany too many collectors are raising one another making the mentioned modules reaching horrible prices... during the last month, I have seen the HEPAX- and the ZENROM module beeing sold for more than EUR 400,- !!! (for one module!!!). I would also like to have the HEPAX and the CCDROM modules but the prices are way too high... unfortunately... Martin > I'm looking for some HP-41 accessories, including the following: SUP-R-ROM (surveying) > CO-GO41 ROM (surveying) > HEPAX modules > ZENROM module > CCDROM module > EPROM or RAM based ROM module simulators (MLDL, Protocoder, ROMBOX, etc.) > > or other special modules or accessories for the HP-41. I will pay good price, or will trade for other HP calculators or > accessories. Please contact me at peter@hst-netzwerke.de > ==== Martin Pedersen schrieb im Newsbeitrag > have you been looking for them at Ebay?... in Germany too many > collectors are raising one another making the mentioned modules reaching > horrible prices... during the last month, I have seen the HEPAX- and the > ZENROM module beeing sold for more than EUR 400,- !!! (for one module!!!). > but note that they weren't bought by a German collector. Instead, someone from USA (p.s.) who seems to have way too much money is fishing here;-) Raymond ==== no, I wasn't aware of that... thanks for the info... Martin Pedersen schrieb im Newsbeitrag >have you been looking for them at Ebay?... in Germany too many >>collectors are raising one another making the mentioned modules reaching >>horrible prices... during the last month, I have seen the HEPAX- and the >>ZENROM module beeing sold for more than EUR 400,- !!! (for one module!!!). > but note that they weren't bought by a German collector. > Instead, someone from USA (p.s.) who seems to have way too much money is fishing > here;-) > Raymond ==== on a TI-89 i could type: solve(3x=18,x) how would i do this on the 49? ==== > on a TI-89 i could type: solve(3x=18,x) how would i do this on the 49? Type SOLVE(3x=18,x) or type SOLVEVX(3X=18) , if VX=X The result is the same in both. Grettings from Cartagena - Chile (sorry my english... ;-) ==== > on a TI-89 i could type: solve(3x=18,x) how would i do this on the 49? Assuming that you are in RPN mode and exact mode: '3*X = 18' ENTER 'X' SOLVE ==== Well, I now tried to transmit from one HP49 to another using Xmodem. I opened the I/O input form, set the transmit mode to Xmodem, in the next field down I enter the file to transmit. I set the receiving calculator to Xmodem server by pressing right shift-right arrow, finally I press the F key for XSEND and nothing happens. HELP!!!!! ==== Use Left Shift IO then NXT. XXMIT, XRECV are on the E and F keys respectively. Bill > alternate E-dress wtstorey@ieee.org.no.spam.please Ok, so I enter the file to send and choose XXMIT, but what do I do on the receiving machine (PC or another HP48) If the receiving machine is an HP49 then I guess I just set it to Xmodem server mode, but how do I do the equivalent on the HP48 or the PC? Luis ==== >Ok, so I enter the file to send and choose XXMIT, but what do I do on >the receiving machine (PC or another HP48) If the receiving machine is >an HP49 then I guess I just set it to Xmodem server mode, but how do I >do the equivalent on the HP48 or the PC? >Luis I have only used XModem for PC to HP48. However, since the receiving HP48 would be the same, put the name of the receive file on the stack and type XRECV. I can't help you on the PC end for the receive. ==== Looked in the manual(s), both in the index and contents... Wanting to know how to program/draw graphics objects (with text), and am also want to know if they can be combined with choose boxes. MC For the 49G sorry, my mistake, was wondering why there where no repleys :) ==== > I'm not sure I'm following the question. I have the URLs of some of the > sites, > I don't recall if any of them had available source code, though they might. > Is > that what you mean? well, yes I mean any information would be useful. Sorry for the delay. For anyone who was interested here's a few URL's of various battlebots. http://www.gamerz.net/~c++robots/ http://www.necrobones.com/atrobots/index.htm http://www.cyty.com/robocom/ -- Jeffery http://woodfanclub.com No, it's not vanity, just desperation. ==== The upgrade process stops for some reason, on file 3. I formatted bank zero for some reason (was confused by the instructions), don't know if this is the problem??? The calculator was resetting when I started it, thought this might fix it... MC ==== In Nick's Complex number marathon, he refers to 256 Menu Upon excuting this command one gets number of menu options. I cannot find these in the standard HP documentation (advnced user guide). Where can I find n explanation of all the options. Are there any other hidden menu options not properly documented ? Aubrey. ==== http://www.hpcalc.org/details.php?id=2990 > In Nick's Complex number marathon, he refers to 256 Menu > Upon excuting this command one gets number of menu options. I cannot > find these in the standard HP documentation (advnced user guide). > Where can I find n explanation of all the options. > Are there any other hidden menu options not properly documented ? > Aubrey. ==== (not the > 48 one). > Serge try http://www.google.ca/search?q=rpncalc&ie=UTF-8&oe=UTF-8&hl=en&btnG=Google+ Se arch&meta= in particular http://www.gnu.org/directory/Mathematics/rpncalc.html ==== try http://www.google.ca/search?q=rpncalc&ie=UTF-8&oe=UTF-8&hl=en&btnG=Google+Se a rch&meta= > in particular > http://www.gnu.org/directory/Mathematics/rpncalc.html I'm interested in a hp28s emulator as well- is this gnu project an rpn > calculator or a real RPL programmable HP28S emulator? It's not programmable ==== > I tried changing the with of browser lines by hand (message 59). I > could make the line less wide all right, but I could not increase the > width. So it seems to me that the truncation of strings to about 18 > characters may be hard-coded into the engine. Probably something >A workaround will be to use the message handler and do the string convesion >yourself. Quite easy things to do. >The only place where the full browser is used is in the Flag manager, and >it's using graphics rather than text. > Will this be fixed in an upcoming ROM 1.19-7 ? TIA Andreas ==== > Will this be fixed in an upcoming ROM 1.19-7 ? If it is lodged in bigs.hpcalc.org maybe ... It's the only way I have now to trace bugs or enhancements. ==== Lets see what HP folks think about this... > >Message-id: <6MPX8.504338$cQ3.44713@sccrnsc01 >Message-id: <3D2EAE15.9D17EB82@pobox.com> > >Message-id: <3d2ce911.329838465@news.west.cox.net> >Are there any manufacturers, other than the HP's (32,41,48,49), that > >use RPN? > >TIA I remember that TI had some a while back. I don't know if they > >still do? ( of course... I've never been able to figure out why so many people > >like > RPN. I view it as something that has some decent computer coding > >applications > - I've used it for such - but.. for day to day calculations, I have > >always > found RPN to be a pain in the backside. That's part of why I am not > >an HP > >user > myself ) Dan :-) >RPN is faster because there are no ( ) keys. *Real* engineers use RPN >: -- >________________ > >Rob Munach, PE > >Excel Engineering > >Carrboro, NC He he. I've heard that one many times!!! LOL. > The amazing thing I find is... so many engineers actually believe that > >one. > ( about being faster I mean ) Dan :-) >IT IS FASTER Chuck, Chuck, Chuck??? NO. IT's NOT!! hehe ;-) Fewer keystrokes does not equate to faster. If it did... then I would have > lost most of those just for fun calculator races ( where you both see who can > punch in the equation the fastest and get the correct answer ). Fact is... NO > ONE using an RPN calculator has ever beaten me ( a lot have tried ). And... > I'm talking big equations as well as small. That means... at least in my > experience ( which I believe is actually fairly substantial ) RPN is not > faster. Sure, HP has published that in their manuals and stood by it. But of > course they have!!! They DO want to sell their calculators don't they? RPN often does take fewer keystrokes. But... really... who gives a crap > about that. There's no advantage to that ( it certainly does not give a speed > advantage ). I've also noticed that a lot of engineers using RPN are actually > slow with their calculators. My guess there is that they have to think more > about how to punch their numbers correctly into the calculator simply because > the order of data entry using RPN is less intuitive to them than the algebraic > math they generally read and write. If someone wants to use RPN... that's fine. But... to date... I've never > heard an argument that's even remotely convincing on why RPN is better. > Personally... I think lots of engineers have been essentially brainwashed > into actually believing it is better ( usually because they are erroneously > told that it is faster due to the fewer keystroke thing ). Dan :-) ==== > Lets see what HP folks think about this... > >RPN is faster because there are no ( ) keys. *Real* engineers use > RPN That's right ! X > Fewer keystrokes does not equate to faster. If it did... then I would > have > lost most of those just for fun calculator races ( where you both see > who can > punch in the equation the fastest and get the correct answer ). Fact > is... NO > ONE using an RPN calculator has ever beaten me ( a lot have tried ). X And I won all the class math championships where we were not allowed to use even pencil and paper system. No errors here. All 10. Does that make human brains (not counting Rainmans) better on number-crunching than pen&paper system or a calculator, a PC? Nope! It just shows that one individual can be better than others in one subject. Maybe I can beat everyone in a calc race? Maybe I'll loose in a math_with_brains -contest. Surely a speedy man with a fast mind and even faster fingers would beat the calc race using ANY calculator system. ugh Brother-Peter ==== The screen is too small to show the calculator (HP48/49) See them side by side (a real calc & a WinPPC) I use Jornada 720 and even that is not so good looking. > has any tried using emulators with a pocket PC? ==== Why not design your own screen? It's not hard and you can get exactly the size and layout that you want. > The screen is too small to show the calculator (HP48/49) > See them side by side (a real calc & a WinPPC) > I use Jornada 720 and even that is not so good looking. >has any tried using emulators with a pocket PC? > ==== Colin Croft schrieb im Newsbeitrag > I'm not joking, it's really not that hard. Christoph clearly put a lot > of thought into making it as easy as possible for people to create their > own versions. TNX for the flowers, but the KML script design was already made by Sebastien Carlier. > [..] > All you have to do then is to create a KML file for your image > which just contains a whole set of key locations along the lines of: > Button 11 > Type 5 > Size 64 42 > Offset 26 344 > OutIn 1 16 > End And sorry that I have to disappoint you a 2nd time. It's true that I implemented the transparent circle as button type 5 in Emu48 (Win32 version since v1.21), but in all WinCE versions I know type 5 works like type 0 and that isn't the transparent circle mode. :( [..] Christoph ==== and others will actually have to put some serious thought and Colin Croft schrieb im Newsbeitrag >I'm not joking, it's really not that hard. Christoph clearly put a lot >>of thought into making it as easy as possible for people to create their >>own versions. > TNX for the flowers, but the KML script design was already made by Sebastien > Carlier. >[..] > All you have to do then is to create a KML file for your image >>which just contains a whole set of key locations along the lines of: >>Button 11 >>Type 5 >>Size 64 42 >>Offset 26 344 >>OutIn 1 16 >>End > And sorry that I have to disappoint you a 2nd time. It's true that I > implemented the transparent circle as button type 5 in Emu48 (Win32 version > since v1.21), but in all WinCE versions I know type 5 works like type 0 and > that isn't the transparent circle mode. :( [..] > Christoph > ==== I told you so! I tried once quickly a year ago and yes it takes some effort, but I'll try 2nd time after my VOCATION! > and others will actually have to put some serious thought and X > And sorry that I have to disappoint you a 2nd time. It's true that I > implemented the transparent circle as button type 5 in Emu48 (Win32 version > since v1.21), but in all WinCE versions I know type 5 works like type 0 and > that isn't the transparent circle mode. :( ==== I have a 620 myself; there is a script for EMU48 for WinCE 2 that splits in half and displays horizontally. It is photorealistic -- very nice. > The screen is too small to show the calculator (HP48/49) > See them side by side (a real calc & a WinPPC) > I use Jornada 720 and even that is not so good looking. > has any tried using emulators with a pocket PC? > -- ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Kenton W. Irby Mostly Harmless! kwirby@earthlink.net The answer to the ultimate question of life, the universe, and everything is ... * + * FORTY-TWO! * + * (Douglas Adams) ==== I really can't understand what's happening to my HP 48G. Suddenly, the Cancel button (the ON button) stopped working. I can turn it on, but when I try to use this button, it simply doesn't work. If you have any idea of what can I do, please tell me. I'll describe exactly what happened and what I've already tried. What happened.... I used the calc. the last time in the morning. In the afternoon, when I tried to turn it on, it simply didn't. I took out the batteries, reseted, but nothing. After sometime though, it did turn on, but when I pressed the cancel button, it didn't work (for example: press number 1 and press the Cancel/ON button - it should disappear, but it doesn't). Every single thing works just fine, the serial communication, beep, calculations, etc. The only thing is that Cancel/ON button. The button itself is working, because I can turn on the calc. and when I press and hold it and, for instance, digit 556, it doesn't appear on the display, just when I release the Cancel/On button. Strange. I can't turn the calc. Off by pressing shift ON, because it doesn't work. No probs, I write OFF and press Enter and it turns Off. But when I press the ON button, the HP doesn't turn on, only if I hold it by aprox. 5 seconds or if I wait some hours (can it be a capacitor??? Witch one??). One of the most weird thing is that, I don't know why, after I did lots of things, it started working just fine, I could turn it on and off repeatedly, cancel and so on, just OK. But, on the other day it stopped working again and since then (Friday July, 12 2002) it came this way again. What I tried.... I opened the HP (very hard job, I should say), took out the batteries, reseted the calculator with the reset button (I can't use ON+A+F) and I even try to (after take out the batteries) short cut the capacitor that holds RAM while you change batteries. It asked me to try to recover memory, I pressed NO, and so on. So, it's too weird for me. Everything works but the Cancel/ON button, except to turn the HP on and just after some hours or holding it for more than 5 seconds. If you've ever seen it or have any idea, PLEASE tell me. Luciano Enzweiler ==== This is a rare fault in the old HP 49G series. You should press just under the LCD to use the [ON]. Since you have already opened the calculator, try to get pressure in the area between the LCD and the top row of keys. In the future you may also loose control of the shift keys and finally also the alpha key will be dead. > I really can't understand what's happening to my HP 48G. Suddenly, the > Cancel button (the ON button) stopped working. I can turn it on, but > when I try to use this button, it simply doesn't work. If you have any > idea of what can I do, please tell me. I'll describe exactly what > happened and what I've already tried. What happened.... > I used the calc. the last time in the morning. In the afternoon, when > I tried to turn it on, it simply didn't. I took out the batteries, > reseted, but nothing. After sometime though, it did turn on, but when > I pressed the cancel button, it didn't work (for example: press number > 1 and press the Cancel/ON button - it should disappear, but it > doesn't). Every single thing works just fine, the serial > communication, beep, calculations, etc. The only thing is that > Cancel/ON button. The button itself is working, because I can turn on > the calc. and when I press and hold it and, for instance, digit 556, > it doesn't appear on the display, just when I release the Cancel/On > button. > Strange. I can't turn the calc. Off by pressing shift ON, because it > doesn't work. No probs, I write OFF and press Enter and it turns > Off. But when I press the ON button, the HP doesn't turn on, only if I > hold it by aprox. 5 seconds or if I wait some hours (can it be a > capacitor??? Witch one??). > One of the most weird thing is that, I don't know why, after I did > lots of things, it started working just fine, I could turn it on and > off repeatedly, cancel and so on, just OK. But, on the other day it > stopped working again and since then (Friday July, 12 2002) it came > this way again. What I tried.... > I opened the HP (very hard job, I should say), took out the batteries, > reseted the calculator with the reset button (I can't use ON+A+F) > and I even try to (after take out the batteries) short cut the > capacitor that holds RAM while you change batteries. It asked me to > try to recover memory, I pressed NO, and so on. So, it's too weird for me. Everything works but the Cancel/ON button, > except to turn the HP on and just after some hours or holding it for > more than 5 seconds. If you've ever seen it or have any idea, PLEASE > tell me. > ==== I have these these two problems, which nobody has yet help me solve (inluding h.p. costumer service) my calc is 49G: 1. I can`t perform any substitution using subst(). I get the error message: non algebraic expression. 2. I can`t multiply two 2X2 matrixes, after stoing them in different variables the same error message appears.. I changed all my flags to be like the ones in my friends calculator in which these operations work well, so I think it`s not a matter of flags. I even reinstalled the operation system, and the problem still exists. -- PS: Get rid of that annoying ALG mode and switch to RPN! When I do A*B I get this message: non algebric in expression. The same occours when trying to use subst(). Another example: When typing A*B, the error message is: invalid dimention -- Here is an example: When I do A*B I get this message: non algebric in expression. The same occours when trying to use subst(). Another example: When typing A*B, the error message is: invalid dimention Please help! -- Could someone give a brief description of the different message handlers codes and usage. I can't find anything on the web and the Donnelly book seems to be out of print. I have never used the message handlers doing an input form, maybe because mines are far too simple. Anyway, from Donelly's Book: The message handler is expected to return TRUE if the message was handled, along with any required results on the stack, or FALSE if the message was not handled. A message handler has the following stack diagram: There are many messaged, but the messages most likely to be of interest are documented as follows. Input Form Messages: These messages are processed by the main input form message handler. (#n) are the decimal message numbers Title Grob (#2) input form Menu (#15) Three Menu Keys (#16) Cancel Key Event (#28) OK Key Event (#29) Field Messages: These messages are processed by the individual field message handlers and are specific to the related field. Check Object Type (#45) Check Object Value (#46) -- That's it, another useful information is: there is also: gFldVal (recall the values for an individual field) GetFieldVals (recall the values for all the fields) the last two ones may noy apply to the 49, I don't know. And that's it. -- PS: Did you try AMAZON to get Donelly's book, I got mine there last year and is part of the fifth printing from January 2000. redirect me to ask for an used one (??). Luis. -- --------------------------------------- Luis Morales Boisset IT Manager Infonis International - Spain better take a used one than nothing;-) Donnelly's book is a very good starting point (IMHO). Raymond redirect Have someone been able to use the IfMain outside the Debug2 editor. It's a bit buggy and it seems to use some compressed strings so I don't know if is possible to use it manually (to make a program I mean). --------------------------------------- I'm not sure if it's useful for your needs, but did you see the document elaborated by Cyrille??? -- Yes, I read it but it talks about the Debug2 editor. And then the header for the IF2 stuff is a bit soft for the general entry IfMain2 or at least I need some explanations on creating programs with it and without Debug2. --------------------------------------- I was just curious if anyone knows of a debugger for Sys that functions similar to DBUG (the built-in userrpl debugger). I found some for asm on hpcalc but I can't find any for sys. BTW I do all programming on the calc so a calc based debugger is preferred, but anything that'll work with sys will do just fine. Or as an alternative, anybody have any good suggestions how they do their debugging? It would be very nice to be able to run the debugger and see exactly what is being fed to my program at various points... almost all of my troubles are caused by haveing not enough/ wrong / too many arguments on the stack. and its a pain to search through programs time and agian trying to isolate the problem. John Mart, II Solution 1: Insert xHALT in different and strategic parts of the code, then use the UserRPL debugger, but be careful with the SysRPL commands that works with the runstream (ITE,IT,',?SEMI, etc.) you have to use CONT before any of this commands, so Insert a xHALT after them. Solution 2: Use Jazz, it has a SysRPL debugger, but I'm not sure if the debugger works in the 49 version, I have never used Jazz. -- Try jazz, you have SDB which functions like DBUG but better. Now you can set breakpoints, browse loop environments, browse lams, skip next command, you also have a machine language debugger you can start directly from SDB (DB.) Read jazz docs for more info. Use Jazz Debugger -- ive only just noticed when reading throught the users guide, that there is no goto command, i remember my IT lecturer used to say using goto and labels was bad practice even though the language supported them, so why doesnt the 49 have these commands?? Don't you remeber E.Dijkstra's (probably one of the most important Harmful. Goto shoudn't be used at all!!! It's completely unnecesary, you can always solve any problem without using goto; just loops, conditionals structures..... http://www.acm.org/classics/oct95/ any others that you pile onto the same stack after them. As higher-level programs call lower-level programs, or perform branches and loops, the set of bookmarks to return to, in the execution runstream, forms another internal stack, known as the return stack. If you create any local variables in your programs (or even when you use such structures as START.. or FOR...) then you are also creating a set of nested environments, which once again acts just like another stack. In each of these three kinds of stacks, you have to finish with the latest things that were added before you can go back to using the previous things; you are not free to jump around at random, because there is only one single pointer which marks the current location (or level) in each of these stacks, so each has to be unwound in precisely the reverse order from which its items were created. Another way to think of it is like a bunch of rooms within rooms -- every time you go into an inner room, some new resources valid for that room alone are allocated to you, on either the return stack or the nested environments stack, and if you don't exit back out through exactly the same doors through which you came in, and in the same (actually reverse) order, returning each resource as you back yourself out, then you will be using the wrong resources for where you are, or perhaps you will never return those resources back to the operating system at all. It's clear enough when you explicitly call other programs or functions, or have programs nested within programs, that each of these creates a new return stack level, but also, there are even unseen inner programs within UserRPL branching and looping structures, required by the RPL mechanisms of either executing or skipping over exactly one object at a time, e.g. IF ... THEN [ ... ... ] ELSE [ ... ... ] END (square brackets indicate hidden nested program levels) And that's why you're not allowed to jump around in UserRPL using labels, goto, or even exit -- for every inner room you enter, you must return all the resources that you got for that room before you leave it, and the structures that the UserRPL compiler builds for you automatically guarantee that this will occur, including for all the inner invisible nested program levels that you don't even see. But in SysRPL and ML, lo and behold, you can free yourself from the chains and seat belts which force you to do things right, and protect you from crashing your calc, and you can go wherever you want; the responsibility is yours to understand and correctly handle every detail and subtlety of the operating system, and caveat emptor if you don't ;-) ----------------------------------------------------------- even more valuable, in the end, than only learning a single set of dogmatic rules -- but it's slightly harder to grade properly ;) And you can always compute any computable function using only a Turing Machine -- but who would want to? ----------------------------------------------------------- body acceptable but as the preferred method in this situation, which was more than I asked for. // Check data for various error situations if (data == 27) { errno = ERR_ESC; goto databad;} if (data == 0) { errno = ERR_NULL; goto databad} if (data == various_other_error_values) { errno = ERR_VARIOUS; goto databad} goto datagood: databad: do_error_processing(errno); datagood: do_processing(data); . . . important Considered unnecesary, you conditionals that using supported body then you can write in SysRPL/ML and do what you want, if you are mindful of the system's requirements to keep track and clean up, e.g. to do an ABND (environment) for every BIND, RDROP return stacks, LOOP for DO, ERRSET/ERRTRAP to catch errors, etc. (all of which is taken care of automatically for you in UserRPL). To illustrate exactly what has to be done to simply let the user jump to the end of his current program and exit it, see this: An EXIT command for UserRPL programs: http://groups.google.com/groups?ic=1&selm=74j0li%241qm%241%40news.iastate.ed u Other related links: A much simpler EXIT command for UserRPL programs: http://groups.google.com/groups?oi=djq&ic=1&selm=an_420382122 Re: UserRPL: missing BREAK command? http://groups.google.com/groups?ic=1&selm=8d2d9d%241tt%241%40nnrp1.deja.com Re: Cont, Kill, Exit, Break, and Structured UserRPL http://groups.google.com/groups?ic=1&selm=6vmo9d%24jis%241%40news.iastate.ed u ----- Of course! (see my other post for being on your side about this :) The HP48/49 ROMs themselves are just full of GOTO, GOLONG, and GOVLNG, so these must be useful, even to HP :) ----------------------------------------------------------- Is there a way to quote a ID that is already on the stack? Yes I know that you quote something to PUT it on the stack... Perhaps I should explain: command puts the ID XXXX on the stack. I now want to do: ' xPURGE BINT2 ::N This results in :: ID XXXX xPURGE ; Of course for this piece of code to be of any use the ID XXXX must already exist. AND if ID XXXX already exists then the code above will EXECUTE ID XXXX and then xPURGE will have To few arguments What I need is a command that quotes the object on stack level 1 lets call it 'PREV Now, with the ID XXXX on stack level 1, I can do: 'PREV 'xPURGE BINT2 ::N This will give me the secondary: :: ' ID XXXX xPURGE ; Any help is appreciated -John Mart, II :: ' ' SWAP ' xPURGE 3 ::N ; body My solution would be quoting a quoter: with ID XXXX on level one, :: ' ' SWAP ' xPURGE THREE ::N ; which returns :: ' ID XXXX xPURGE ; -- PS: For pure SysRPL, try using @ and PURGE or ?PURGE_HERE. 13:27:55 MET DST) time No, that's what alarms are for. yeah i did get left behind in what you guys were talking about, but the other post you put a link to explained what i needed to know thanks END DROP [H3O+] H DUP LOG NEG pH SWAP {V1 H C} PURGE In this way the loop goes fine and I get a pretty effect with DISP. How global vars are problematic and I knew that I could do better I changed to this: The problem is that now there is no loop! I debugged it and noticed that the program is not overwriting the local variable H every time it loops. What I am Rafael (don't 0 =/ global KEY END the value of the H, you created a new local H that will exist only in the second subprogram and when leaving this subprogram, the old H will keep its old value. Instead of making a new subprogram, just do 'H' STO. -- How global this: UNTIL KEY END that the What I am Replace: With: DUP 'H' STO 7 DISP James END DROP [H3O+] H DUP LOG NEG pH SWAP {V1 H C} PURGE In this way the loop goes fine and I get a pretty effect with DISP. How global vars are problematic and I knew that I could do better I changed to this: The problem is that now there is no loop! I debugged it and noticed that the program is not overwriting the local variable H every time it loops. What I am Rafael body in the string, and to leave them as null characters while compiling, would it not be possible for 00 to also be compiled as a null, which would then ease the situation for those who desire symmetric decompile/compile of UserRPL? (I hope). Too bad that it's not in the CHARS menu with the other string manipulation commands (but please ACO, don't hold up the next ROM to put it there) and not in the See also: of the REPL command in the User's Guide. On the other hand, it seems to me that if the original object actually does happen to contain 00 (admittedly unlikely), then EDITDECOMP$ would change it to 00, and 00 SREPL would change that to . I'll play with this a bit.... Sounds easy at least. Of course, I wouldn't object to having a method for forcing recompilation to the counted string form, although that wouldn't allow direct editing of the null string. 13:01:39 MET DST) null I've founded a previous thread started by James M. Prange on 2001-06-18 talking of this... the title is: (49G) Strange treatment of NUL There I've founded all the answers I need. Suggested reading ontopic with my message! :) Bye, Kickaha I don't know of any way to put in a NUL when you're editing (although work on the 49G). You can put a character string consisting of the null character on level 1 of the stack simply by executing 0 CHR. Once you have this string on the stack, you can use the + command to put it together with any other string, POS to find its first occurrence within a string, SUB to extract a substring, REPL to replace part of a string, in short, anything you can do with any other character, except that you can't edit any object containing a null character. Well, actually, you can (sort of) edit such an object on the 49G. For editing, the 49G changes each null character to 00 (and each character embedded within a string to , and each to ). For character strings, the CursorDown key returns the familiar Can't Edit Null Char. error, but you can force it by using LeftShift CursorDown. The problem with this is that the calculator changes the 00 to 00 instead of to the null character when you press ENTER. On the 48, perhaps the easiest method to edit objects containing the null character is to first, if it's not a character string, then use 64 character with some character that you're sure wouldn't otherwise be used and then edit the object. After editing, change the dummy character back to the null character, and, if the object was not originally a wordsize or display mode that you changed earlier. A program to change any null character to character 255d: %%HP: T(3)A(R)F(.); << WHILE DUP 000 POS DUP REPEAT 255 REPL END DROP On the 49G, the above method works if the object is already a character could change 00 to a seldom-used character, but what if the original object literally contained 00? This would be changed to 00, because the is changed to . Also, any embedded within a string that the object contains will be changed to . The following program leaves and the following character as is, unless the following character is 0, in which case the 00 is changed to an ASCII 255 character. %%HP: T(3)A(R)F(.); << SWAP WHILE DUP POS DUP REPEAT DUP2 1. SWAP OVER - SUB 4. ROLL SWAP + UNROT IF DUP2 1. + DUP SUB 0 == THEN ROT 255 + UNROT 3. + OVER SIZE SUB ELSE DUP2 DUP 1. + SUB 4. ROLL SWAP + UNROT 2. + OVER SIZE SUB END END DROP + Also note that in the 49G editor, if you want to insert a character within a string, you have to insert it as , and to insert a , double it up to . Of course, if you're doing your editing on a PC, you can simply use a literal NUL (as long as your editor provides a way to do that), or, assuming that you're using translate mode 2 or 3, you can use 000 for NUL. But note that the 49G changes NUL to 00, embedded to , and to when sending (this is before any Kermit ASCII mode translation). When it's receiving it has the same problem with changing 00 to 00 instead of to NUL. My understanding is that the null character marks the end of the command line. So what would happen if you did manage to insert a null character into the command line? I don't know; I suppose that everything preceding the NUL would be parsed, but what about anything following the first NUL? Is it simply discarded? Maybe a SysRPL guru could tell us. With the 49G you can print to a serial printer, although some of the 49G print commands change any NUL, embedded , or to 00, , or , respectively; some don't. Go figure. Examples of the use of NUL with an Epson LQ-850/950/1050 printer: For the graphics, I'll use the HP 82240A/B for an example, even though the 49G can't print to it, because of its simplicity. The Epson uses a similar method, but complicated by 11 graphics modes, up to 65536 columns per graphics command and either 8 or 24 dot positions in each column. To send column graphics to the 82240A/B: Print x columns of graphics, where xchar represents represent the character with decimal value x (x=1 through 166), and C1, C2,...,Cx are characters whose values determine where dots are printed in each column. The most significant bit represents the bottom dot, the least significant represents the top dot. To put it another way, assign each dot, starting at the top, a value of 2^R, where R (0 through 7) represents the row, and then sum the values. Thus, to print a blank column, the value would be 0, to print all eight dots in a column the value would be 255 decimal. JMP I have also encountered this strange behaviour. It was exactly the same problem as you described with the NULLLAMs. This type of problem is very annoying, When I come across a problem like this I get the feeling that I'm wasting my time with an inconsistant/buggy machine:( As a matter of fact my OS seems to be self destructive. Perhaps the errors I get while debugging my programs are corrupting the OS... I've been through MASD bugs that are very similar: A command compiles fine one day and the next the same exact source will not compile. ( remember my complaints about IREMAINDER and IDIV2 ) Anyway you are not the only one with these sorts of problems, -John Mart, II 13:14:04 MET DST) al dgeve: Same thought come to my mind, and that's why I reflashed my calc. But this move didn't solve problems. I feared about a permanent damage of my calc... (please, reassure my that every bad move I may make can be recovered with reflashing/backuping) Same thing happen to me: sometimes I take one once perfect running sysRPL program on the stack, edit it with Emacs and add some harmless commands, recompiling it and guess... can't compile... I think it was my poor experience in sysRPL programming, but if someone else get the same problems... My expectations with calculator are that identical procedures give identical results. Solution: I will start to program in sysRPL on emulator on my PC, and I will use MASD only in emergency situation. At least crashes will not severe my lovely calc... /-----------------------/ Marco Tinarelli - Bologna - Italy tittiXXYYZZ_tt@tin.it (togliere XXYYZZ dall'indirizzo per rispondere) 13:20:10 MET DST) The problem is caused by bad programming, not an error in the calc. hine:( You should learn to program then, instead of blaming the machine. SysRPL errors can very well corrupt the RAM and/or stored variables. This can make commands crash the machine. ON+C will often correct this, but sometimes an ON+A+F is needed. This is not possible. You must have done something wrong. No, every one in a learning process will have problems with the tools once in a while. A hammer can be fine even though it bends nails. 13:26:50 MET DST) Programs can't damage the calc physically. That's not without changing it. I think so too. It could of course also be a bug in Emacs (unlikely, just by running the editor, whihc is in fact the built-in one). Exactly. If users get different results from time to time, it can only be because they have changed something. Or use a back-up program. I back everything up in the Flash, when I'm about to execute a new potentially damaging SysRPL program. There the data will be safe - afterwarsd you can just compare checksums to see if anything unknown changed in tha stored variables. Unstable RAM can usually not be detected, but if a crash should happen later, you'll always have a recent backup. I use MASD all the time (I've never coded a single nibble for the calc on the PC) and it's currently very stable. A least as stable as the PC development kits. 14:01:41 MET DST) So please, can you explain why NULLLAM example I talk about (taken from Kalinowsky's manual) worked before and now don't? I didn't change anything of my programs. (after performing ON+C, ON+A+F, hardware reset by putting a paper clip end in the hole in the back of the calc and reloading again extable). I assure you, they worked before (I used my program with this structure A LOT). I use Back Manage 49 1.0 since my first attemp in sysRPL (I've read a lot about potential damage caused by bad programming in this newsgroup). My behaviour is the same as yours. It'll work whenever it's not assigned to anything else (when no other environments are running). Have you used it in a POL or something? You should quote it for safety though. This is easy. They did not work CORRECTLY before! Since the usage was improper, they could not. It is possible to exit a sysrpl program and leave with too many or not enough environments properly exited without immediately crashing the calculator. Also you can attempt to take more objects from the stack than are available. This can give the appearance of everything working great, then things may happen later that we don't attribute to our program, but to the operation just executed (why can't I add these numbers!!!) When testing a new sysrpl program, instead of starting with an empty stack, leave the numbers 1-4 on stack levels 1-4. When you exit your program, the stack should stay the same. If you only have numbers 1-3, you have a problem... Dennis But, where can i find it? Is it on a RAM card, or a Lib. file? More info please 119H2+18Ca(CN)2+6NaAlF4+12FeSO4+6MgSiO3+6KI+2H3PO4+12PbCrO4+24BrCl+6CF2Cl2+2 4SO2 ClearFlag Lastcmd saves your last result in comand to be recovered if it is lost. Greetins Bernardo Aguilera Quick Tour, Bare Necessities, Memory Management, Helpful Programs, More Trig Functions, Symbolic Differentiation and Definite Integration, Examples, Practice Worksheet and Quiz. Part I - Advanced Placement Calculus Materials. Parts III-IV - TICAP (Technology Intensive Calculus for Advanced Placement). Part V - Additional Calculus Classroom Activities. $15 including Priority Mail. HP 48 Mugs 2 of the rare and wonderful mugs from EduCalc for the HP 48SX and GX respectively. Both as new. Best Offer... Hewlett-Packard Thermal Paper 82045A for Portable Printer Box of 6 rolls. $10 plus shipping. Entity Dennis Straley spoke thus: This is CalcWare and includes serial cable, instructions and a disk. It is free but whan you receive it I will ask you to pay for shipping with PayPal. Respond to Cogent@GNTech.net. speech... Dennis OK, I think I got it right this time: This is CalcWare and includes serial cable, instructions and a disk. It is free but whan you receive it I will ask you to pay for shipping with PayPal. Respond to Cogent@GNTech.net. It is PayPal. Hell all, I am trying to multiple 2 matrixes 2X2 and get thg message non algebraic expression. do you know what should I do? -- 052-584131 I think your problem might be a flag , that's why i recommend push modes , then next and push RESET All. Greetings Bernardo Aguilera What is your calculator ? 48,49...? It it's a 49 what Rom version you are using. A 2x2 matrix is like [[ a11 a12][a21 a22]]. In hp48 you cannot put letters or algebraics in a matrix. Is it possible to make, or is there already, a HP49 emulator for Palm? Does this one fit your needs? http://www.hpcalc.org/hp48/pc/emulators/emu48ce120.exe Emu48 for Windows CE 1.23 (details) 3285KB Emu48 1.23 ported to Windows CE. Emulates an HP48, HP49, or HP39/40 ROMs included) and includes several KML scripts and bitmaps for palmtop (240x320 solution) computers. Works on MIPS-based systems (Cassiopeia E-105) and ARM-based systems (iPaq) only. By Sebastien Carlier (H), Juan Pablo Lopez Anadon, Leopoldo Bueno Castillo (H) and others. I just searched fort PALM in hpcalc... Ricardo -- http://xie121.infovia.xtec.es/~rblasco Un222os, hermanos linuxeros I asked the same question a while ago, and, according to JYA, the hardware of the Palm (memory and speed limitations) just wouldn't make such a project feasible... AL I posted a message outlining a problem with quoting an ID that is already in the stack. Assume ID xxxx is on the stack level 1. Here is the solution: :: ' :: 'R ' xPURGE ; SWAP BINT2 ::N ; This code will output a program that deletes the ID on level 1. I guess that I should have read my manual a little more before posting. -John Mart, II the Saturn processor in assembly language. I have already looked at the Introduction to Saturn Assembly Language by Gilbert and Eric on HPCALC.org, however, that has yet to be completed. I was wondering if anyone one knew where I could get a very good tutorial. If it makes a I've heard of a very good one written in french, it was called Voyage au centre of HP48GX or something like that. I remember it became available for free a while ago, so search google about it! Anyway I think Introduction to Saturn Assembly Language by Gilbert and Eric is pretty good, even though it's incomplete. -- 17:11:21 MET DST) How about upgrading the rom? am i dreaming again? do this: INV(TAN(X)) [EXPLN] i got the wrong answer. to verify (easy to be full around), i do a 'x=2' to my result in the stack, then a [SUBST] [EVAL].(-.4576575554361,6.05329943752E-13) i apply the same in MAPLE: the answers are different, it happens the same with TAN(X) [EXPLN], but works ok with SIN(X), COS(X). CAN SOMEBODY WAKE ME UP FROM MY DREAMS? rcobo@eng.morgan.edu schreef: TRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRIIIIIIIIIIIIIIIIIIIIII IIIIIIIIIIIIIINNNNNNNNNNNNGGGGG the answers are the same: -.4576575543. When the second term is 9 orders of magnitude smaller I usually become suspicious -- This message was entirely written with recycled electrons Pivo 13:34:03 MET DST) [EVAL].(-.4576575554361,6.05329943752E-13) [...] Aren't they nearly identical? The differences in the imaginary parts (should be zero) emerge from the different algorithms used. Do you think MAPLE uses the same algorithm to calculate COT as the HP49G uses to calculate EXP? ok :-) (playing around in the middle of the night) says... BRING ON THE ASM =) for asm examples like this, and find nothing. I am looking forward to studying this. I have briefly looked over it and I can tell it's going to be fun;) However I have a question about the note you made concerning the HEX2DEC conversion. Currently I use EMACS for programming. I have never used JAZZ and am unfamiliar with the DB command? that you referred to. I assume that you meant that this command? was to be used with jazz? Perhaps I should try JAZZ, if it is of any aid to programming in asm. Anyway thank you for both of your solutions to my question, and particularly for your in depth explanation of how to complete this task in asm. -John Mart, II John Mart, II schreef: Indeed! Keep on doing that! DB is an ASM debugger from Jazz and is really usefull for learning ASM, however jazz is ported from the 48 and is still not finished so you have to live with a few drawbacks but if you use Emacs to program and Jazz to debug(ASM&SysRPL) you can do anything :-) HEX2DEC was just a name I gave to part of the program the source an be compiled with MASD and run on any 49(and 48 I guess) I told you to debug this, because I didn't want to explain how it worked (lengthy) BTW if you want a real ASM source take a look at the source of LibEx although this is for advanced ASM'ers http://ca-on.hpcalc.org/hp49/programming/libs/libex30.zip -- This message was entirely written with recycled electrons Pivo Although you may have meant it as a bit of a joke (BTW if someone knows why we are here, where are we going and why, I encourage him to give me some hints on that ;)), I think I may shed some light on the subject (as far as my knowledge is concerned): You may want to peruse these two sites; I've found the information there invaluable: http://www.lds.org/library/display/0,4945,105-1-11-1,FF.html http://www.lds.org/library/display/0,4945,11-1-13-1,00.html AL I'm still looking for a good ANOVA proggy for the HP49. I#m using the one from S.Guth in the StatPro Pack which is great for fixed factors 1 and 2 factorial ANOVA's but in our statistics course we need to calculate 1,2 and 3 factorial ANOVA's with random or fixed factors, and either crossed or nested or both factors (in the Honeck, Kibler & Sugar Notation). Maybe you guys can help me out... Thanx Sven