A140 ==== I have a couple of 48sx's and I'd like to know how to get the rom version of every machine. I know is not like the 48g or 49 because I've type VERSION and it don't work. Joan. ==== I have a couple of 48sx's and I'd like to know how to get the rom > version of every machine. I know is not like the 48g or 49 because I've type VERSION and it > don't work. Any binary transfer from any 48 series starts with a header, HPHP48-x, where x is the ROM version, so you could transfer something in binary mode and look at the file on the other device. Or you could run (be careful to type in the argument exactly, including as a hex binary integer, if you want to retain memory): # 30794h SYSEVAL to get the binary transfer header as a string in level 1. This also works on any 48 series. Or, on the 48S series only: First, be prepared for a warmstart. You could press ON and D together to get into the interactive self-test mode, then press backspace to start the memory browser. You should see something starting with 705D9: at the top of the display. Press EVAL, and you should see something like: Copyright HP 1989 briefly displayed. Press EVAL again if you want another look. Press ON and C together to do a warmstart and get out of self-test mode. -- James ==== I've found one of them is version E and the other is J. Curious at least. ==== > > I've found one of them is version E and the other is J. > > Curious at least. The answer to this question could be found in the HP48 FAQ at www.hpcalc.org, also lost of other usefull information about the 48 series, which is helpfull if you own a 48S and lost all manuals and documentation. By the way, where is coma mode ( [ON][SPC] on the 48) on the 49? Kaspaas ==== The Canadian Site is http://store.axidata.ca/kore/search/search.html?searchDomain=0&searchExpr=00 048-90136 and I will try it if the calcpro site doesn't work, but the Canadian site will be more expensive for an American. :-) I found http://store.axidata.ca through http://groups.google.com by clicking on the web section for google. I entered 00048-90136 as the search term. Good luck! > Try: http://www.calcpro.com I ordered one this week, but the sales rep. wasn't so sure they actually had > it in stock. He mentioned something about a shipment coming in (?) and I > was in a hurry so I just placed the order and hoped for the best. You might > try your luck. BTW, what's the Canadian source you found if I may ask??? Best, > Eduardo ==== >the sales rep. wasn't so sure they actually had >it in stock. He mentioned something about a shipment coming in (?) and I >was in a hurry so I just placed the order and hoped for the best. Sales Rep ??.... Calc Pro is just ONE guy... and he is blissfully unconcerned (or else extremely busy... like... too busy to give good service) Keep hoping for the best.... you may get it eventually.... sometime in 2005 ==== >Recently, it seems that prices of HP-48SX models have gone up a notch (to >around $50 from a low of $20). Maybe this is the beginning of the same for >that line? Gene >-- >* These statements and opinions are mine alone and do not reflect my >employer's views. * > I'm a bit surprised that the prices for used 28 series, 48 series, and >> 49G calculators haven't gone up nearly as much. >> James FWIW the HP-12Cs seem to have dropped a little over the last month or two. Were going regularly for $40-$50 now $30-$40. -- Bruce Horrocks Hampshire England bh@granby.demon.co.uk ==== Finally getting around to convert some USER RPL programs for my HP49. I have two books, The Gray Book {Donnelly} and Programming in System RPL by Eduardo de Mattos Kalinowski. Both are good, the later is really good. And yet I am not sure how to exactly convert. 2: real 1: real IF == THEN DROP 21. ELSE 21 + END I am thinking is should be: example: LAM x1 LAM x2 EQUAL IT :: DROP %21 ; ITE :: %21 %+ ; I am using ASM2 on the HP49G with ROM 1.19-6. Is this close? Also, I noticed that if I place ASM2 source as described by Mr. Kalinowski. shows syntax errors in debuger4x, why are the file syntaxes different? Can debuger4x be set to look at ASM2 sources? Sincerely, Kevin Waite ==== The answer is: ITE takes a flag from level one. The flag is dropped. If TRUE, it evaluates the next object in runstream and skips the second one. If FALSE. skips the next object and evaluates the second one. In both cases, the program execution is resumed after the second object. So, in your example: EQUAL ITE :: DROP %21 ; :: %21 + ; Assuming that there are three reals on the stack: two of them are compared and the other (on level three) is either dropped and replaced by 21 if real1=real2 or added to 21 if real1=!real2. Kevin Finally getting around to convert some USER RPL > programs for my HP49. I have two books, The Gray Book {Donnelly} and > Programming in System RPL by Eduardo de Mattos Kalinowski. Both are good, the later is really good. And yet I am not sure how to exactly convert. 2: real > 1: real > IF == THEN DROP 21. ELSE 21 + END I am thinking is should be: example: LAM x1 > LAM x2 > EQUAL > IT :: DROP %21 ; > ITE :: %21 %+ ; I am using ASM2 on the HP49G with ROM 1.19-6. Is this close? Also, I noticed that if I place ASM2 source as > described by Mr. Kalinowski. shows syntax > errors in debuger4x, why are the file syntaxes > different? Can debuger4x be set to look at > ASM2 sources? > Sincerely, > Kevin Waite > ==== >You can't open an HP calculators without doing any visible damage. I suppose HP believes they last forever.....forever may be too strong a word BUT... I have calculators here... dating back to the seventies.... and working perfectly. I have always been told.... pocket calculator failure is VERY exceptional.... and even more unlikely if there is no AC adapter. Battery leakage or physical damage is about the only thing that takes them out. ==== Would both programs FEM48 v5.3 and VigaG v4.1b fit in the memory of a HP-48G+ (128kb)? Would the memory be sufficient? Arnold ==== 128KB? Yes, I think so. FEM max 52KB VigaG 24KB Sum max = 76KB + some data You will have more than 32KB left Would both programs FEM48 v5.3 and VigaG v4.1b fit in the memory of a > HP-48G+ (128kb)? Would the memory be sufficient? > Arnold > ==== The size of FEM48 depends on the modules you install, and if you use standard or compressed modules (as can be read in section 13.2 of the manual). The sizes quoted are for the compressed modules. Minimum size is 22586.5 bytes (FEM48 module), which gives you only basic features like FEA (only data at nodes), structure plots and file management. To do beam analysis you also need the QUERY module, at 12896 bytes. This yields a total of 35482.5 bytes. Note that other modules might come in handy too (load cases, wizards, pretty print output, moving loads)... Maximum size of FEM48, with all modules installed is either 58582 bytes (all modules compressed) or 73164 bytes (all modules standard = fastest). Note that the memory requirements (section 11.1 of manual) of FEM48 are: bytes needed = 79 x (nodes)^2 (good guesstimate) Caspar Arnold FRANCK schreef in bericht Would both programs FEM48 v5.3 and VigaG v4.1b fit in the memory of a > HP-48G+ (128kb)? Would the memory be sufficient? > Arnold > ==== What solvents can be used for removing grease of calculator keyboards without doing damage? Arnold ==== Soapwater moistured non-rubbing clean cloth, (I use an old frotee sock) What solvents can be used for removing grease of calculator keyboards > without doing damage? > Arnold > ==== > Soapwater moistured non-rubbing clean cloth, > (I use an old frotee sock) A psychological tactic? (scare the grease away) Greetings Steve Sousa ==== > Soapwater moistured non-rubbing clean cloth, > (I use an old frotee sock) A psychological tactic? (scare the grease away) AND if that doesn't work, you could use XO Cognac instead of soap The combination with and old sock with certainly get the grease away like greased lightning. PS: *don't use too much moisture* OR if you insist (Steve?) then lay on your back and hold the calculatrice above your face while cleaning so that water doesn't penetrade the innards of the HP. ==== > A directory CASDIR, was placed under the HOME directory but when trying to > remove CASDIR it will not remove per the instructions in the user manual. like > on my other HP 49Gs. I tried reloading 1:19-5 but the CASDIR was not > removed. thx > CASDIR is a reserved variable in the HP49G from ROM 1.19-5. You can remove it, but it will be re-created almost instantly (the Main Stack display when trying to access VX will recreate the directory) You can re-order your HOME directory so CASDIR will be the last object and you won't have to worry about it. ==== > CASDIR is a reserved variable in the HP49G from ROM 1.19-5. You can > remove it, but it will be re-created almost instantly (the Main > Stack display when trying to access VX will recreate the directory) > You can re-order your HOME directory so CASDIR will be the last > object and you won't have to worry about it. With my Filer2 (for instance) you may even hide the CASDIR. Hence, it disappears from the VAR menu. It remains active. For instance, if 'z' is stored in its variable VX, 'z' is displayed in the Header instead of 'X'. One has still direct access to all variables in CASDIR via Filer2, not using the HP49 CAS. - Wolfgang ==== > not using the HP49 CAS. I thought the whole point of using an HP49G was for the CAS. -- ==== > not using the HP49 CAS. I thought the whole point of using an HP49G was for the CAS. It's a nice add-on, but I use it primarily for numerics. The equation solver is nice, and Sune Bredahl's simultaneous equation solver is even nicer. It is sometimes useful to see symbolic results for integrals and derivatives, but it's not part of my everyday-life. The 3d graphing helps me in modelling. ==== > I also have EQNLIB (library 999) but I don't think it's quite as nice as the > 48GX eq'n library and was wondering if ECLIB is any better but of course if > it ain't run, it ain't better. So now that you have it working, is it any better? ==== I have to admit I like EQLIB better, but it is an empowering feeling to have both of them at hand :-) Also, the BZIP library I installed by your Best, Eduardo > I also have EQNLIB (library 999) but I don't think it's quite as nice as > the > 48GX eq'n library and was wondering if ECLIB is any better but of course > if > it ain't run, it ain't better. So now that you have it working, is it any better? ==== Today I noticed I had to slow PC keyboard input or it would drop characters. Is there a way around this? Also, is there, or will there be newer rom version (than 1.18) available? ==== I'm using 1.30 and there could be something newer. Check the web page! http://privat.swol.de/ChristophGiesselink/emu48.htm Look's like 1.30 is the latest. Try it out and see if it fixes your problem Rod > Today I noticed I had to slow PC keyboard input or it would drop characters. Is there a way around this? Also, is there, or will there be newer rom version (than 1.18) available? ==== > I'm using 1.30 and there could be something newer. I use 1.32 from some time ago... ==== > I'm using 1.30 and there could be something newer. > > I use 1.32 from some time ago... > Sorry, I meant the HP 49 ROM image was 1.18, the emulator is 1.32. FWIW I am using it under XP. I don't recall this happening with an older version of EMU 48 and using it as a HP 48. I know a few of you have TI-89 as well as HP48 and 49 Well I just got my TI-89 a few days ago.... adding to my stock of HP's and other calcs.... I guess now I am a confirmed calculator freak, bona fide and getting worse... anyway, on page 8 and 9 of the TI-89 manual is the following example: Graph the function (abs(X^2 - 3) - 10) / 2 and find the value of the minimum in the fourth quadrant The calculator gives ( 1.7320508075707 - 4.9999999999969 ) but the picture in the book on page 9... clearly shows the screen shot as.... ( 1.7320508075682 - 4.9999.......... (it scrolls off the screen) All my HP's only go to 12 digits.... and they give ( 1.73205080756 - 4.99999999998 ) Can somebody with a TI-89 do the problem and tell me what they get ????? or... those with a long-float library... or Mathcad... tell me what the last digits should be ???? SM ==== > I know a few of you have TI-89 as well as HP48 and 49 I have an emulator... it just might survive longer than the real thing. > Well I just got my TI-89 a few days ago.... adding to my stock of HP's and > other calcs.... I guess now I am a confirmed calculator freak, bona fide and > getting worse... Welcome to the club! :-) > Can somebody with a TI-89 do the problem and tell me what they get ????? > > or... those with a long-float library... or Mathcad... tell me what the last > digits should be ???? You can easily get the exact result on the TI-68k using fMin(expression,x): fMin((abs(x^2-3)-10)/2,x)|x>=0 -- ==== > anyway, on page 8 and 9 of the TI-89 manual is the following example: Graph the function (abs(X^2 - 3) - 10) / 2 and find the value of the minimum in the fourth quadrant The calculator gives ( 1.7320508075707 - 4.9999999999969 ) but the picture in the book on page 9... clearly shows the screen shot as.... ( 1.7320508075682 - 4.9999.......... (it scrolls off the screen) With a HW2 TI-89 running AMS 2.08, I get: ( 1.7320508075704 - 4.9999999999974 ) It's not an error in the manual but a different ROM revision. Tom Lake ==== > You can easily get the exact result on the TI-68k using fMin(expression,x): fMin((abs(x^2-3)-10)/2,x)|x>=0 How come if you solve sign(x^2 - 3)*x [the derivative of the above function] for x, the Ti89 gives me x=0 while the HP49 gives me x=0, x=sqrt(3), and x=-sqrt(3)? Can the Ti89 not solve very well with 'sign' in the equation? ==== I want to order some Klotz memory cards. I noticed that you have to send the credit card numbers by fax. I dont have a fax machine and I have never used fax. Can my computer fax? Someone please explain to me how to do this. ==== You need a modem with fax support, and... If you have Windows XP you have fax. http://www.microsoft.com/WindowsXP/expertzone/columns/crawford/02october21.a sp You can also try one of the many fax software offered here: http://www.electrasoft.com/ Johannes >I want to order some Klotz memory cards. I noticed that you have to send the credit card numbers by fax. I >dont have a fax machine and I have never used fax. Can my computer >fax? Someone please explain to me how to do this. > ==== I am using a serial port on my PC to connect to an HP 48GX (Rev R ROM) using HP's Serial Interface Kit software (on Windows 98). I know the connection is good, because I can put the 48 in server mode, and I can execute remote commands using the UI on the PC. MKRAM.KER (from hpcc.org) to the 48, as instructed. The first time I tried it, most of the file was transferred, then a glitch occured (I assume) and so I reset the 48 using ON-A-F. [I have a 128k card in slot 1, merged with the 48's memory.] Now, whenever I try to transfer the file, the 48 complains Error: Bad Argument Value after about 2 seconds, and the UI on the PC says Error sendin files. I know it must be simple, but after restarting the PC, re-installing the 48 utilities on the PC and resetting the calculator, I am stuck. Also, whenever I go to Right-Shift I/O to invoke the comms on the 48, the default is always KERMIT. Can I set Xmodem to be the default? TIA, Mark. ==== Please, put your HP-48GX in RPN MODE. Now, configure for Binary and delete extension (.KER) and try again. Miguel Angel CAPORALINI HERK (M.A.C.H.) **************************************************************************** **** > I am using a serial port on my PC to connect to an HP 48GX (Rev R ROM) using > HP's Serial Interface Kit software (on Windows 98). I know the connection > is good, because I can put the 48 in server mode, and I can execute remote > commands using the UI on the PC. > > MKRAM.KER (from hpcc.org) to the 48, as instructed. The first time I tried > it, most of the file was transferred, then a glitch occured (I assume) and > so I reset the 48 using ON-A-F. [I have a 128k card in slot 1, merged with > the 48's memory.] > > Now, whenever I try to transfer the file, the 48 complains Error: Bad > Argument Value after about 2 seconds, and the UI on the PC says Error > sendin files. I know it must be simple, but after restarting the PC, > re-installing the 48 utilities on the PC and resetting the calculator, I am > stuck. > > Also, whenever I go to Right-Shift I/O to invoke the comms on the 48, the > default is always KERMIT. Can I set Xmodem to be the default? > > TIA, > > Mark. ==== See the setup.tx file of mk230.zip (www.hpcalc.org), and then try again. **************************************************************************** **** 1-Unprotect the card if necessary and merge it with MERGE1 (in the Left shift- Library menu). 2-Empty the whole port 0. For instructions, read chapter 28 of the User's Guide. If there is no important data in your calculator, it is easier to initialize it by pressing simultaneously the keys ON, A and F, and answering NO. 3-Transfer the program named RECV.KER from the PC to the calculator, with XMODEM or KERMIT. 4-Then transfer the file MKRAM.KER. XMODEM is recommended, because of the huge size of this file. In order to use XMODEM, with an HP48 REV R, just use the built-in command XRECV. With an HP48 REV M or P, first send the file XRECV.PRG from the PC to the calculator, then use it as if it was the standard XRECV command. 5-On the calculator, put the name of this file on the stack. and execute the program RECV.KER. **************************************************************************** **** Miguel Angel CAPORALINI HERK (M.A.C.H.) **************************************************************************** **** > I am using a serial port on my PC to connect to an HP 48GX (Rev R ROM) using > HP's Serial Interface Kit software (on Windows 98). I know the connection > is good, because I can put the 48 in server mode, and I can execute remote > commands using the UI on the PC. > > MKRAM.KER (from hpcc.org) to the 48, as instructed. The first time I tried > it, most of the file was transferred, then a glitch occured (I assume) and > so I reset the 48 using ON-A-F. [I have a 128k card in slot 1, merged with > the 48's memory.] > > Now, whenever I try to transfer the file, the 48 complains Error: Bad > Argument Value after about 2 seconds, and the UI on the PC says Error > sendin files. I know it must be simple, but after restarting the PC, > re-installing the 48 utilities on the PC and resetting the calculator, I am > stuck. > > Also, whenever I go to Right-Shift I/O to invoke the comms on the 48, the > default is always KERMIT. Can I set Xmodem to be the default? > > TIA, > > Mark. ==== don't feel too bored while I'm away... Enjoy ! >>>>>> My HP40G Gripes >>>>>> As your flamethrowers might get a little rusty nowadays (while HP tries very effectively to keep them in good working order), I believe a recent experience might fuel (!) some use for them. So read on, and start the blast ! I'm not a new user of calculators, and purchased the HP40G at the time when HP finally lowered its price in France (it had done so in the US many many years before - but we European users are just good old cash cows, you know). Recently, as I was happily calculating at home on a numerical problem, I said to myself : Why not use a portable tool and do this while commuting ? Why not give the good old HP40G another try ?. I recalled I had put it back in the box in despair a couple of years ago, but the idea looked interesting at the time. Not for long ! So off we go. The 40 is a durable, powerful, low price offer with a wonderful CAS and I like this. Now, I start by a little program to retrain my skills on the machine. No need to reread the manual (er, User Guide), this is targeted at the educational market you know. So let's switch it on and press [Program]. My simple goal so far is to display the date correctly formatted (DD/MM/YYYY) plus the day on screen. Hum, no string manipulation. No free position display of data on screen (only DISP ;). So let's display on consecutive lines. Well well, how was this 'difference of dates' function called by the way ? No catalog. No way to search for a function. Nothing in possibly related function catalogs. By the way, those menus never recall what was the last family used, so let's scroll up and down every time... OK, I'm done, this program can't be written. The input form capability can't display a screen for input of multiple values at once (while it is possible, the system does this all the time - see the [Plot Setup] for example). Now I'm in a bad mood, but let's try a small calculation for fun. Let's look at the history in search of an example I have done in the past. Easy to use, but the history displays only two past calculations at a time (one line for expression, one for result, this 2 times makes for the 4 available lines). Not very good, since there is a wasted line both at the top AND the bottom of the screen? Yes yes, TI does no such mistakes ! On the competition, there are 8 lines of which 7 keep previous calculations. I like the fact that you can remove flexibly past entries or results on the 40G, but this is a meager benefit. Why not the ability to switch to a small font, or remove those useless decorating lines ? OK, I just found a simple integral, so simple in fact that I can figure it in my head in under 5 seconds. But let's run it : integral of (x+1)/(x-1) from -1 to 0. Typing... Syntax Error ! A nice popup, but it obscures the screen and slows down the process : you must take an extra step to acknowledge the error (never too good to insist in saying that the user is dumb, right ?). Well, what I need is some help, and I recall there is a function to explain all operations. What the bad expression. Strange that the 40G doesn't take the 'standard' syntax used on every single graphing calculator I know : integral(function, var, start, stop). Well, why make it easy ? Now, I face an egg and chicken problem. I want to know the syntax, but I am still editing the equation. Note that as it was incorrect, it was not pushed on the history stack. Fair enough, but the help function is a command-line function itself, hence a conflict. Eventually, I find a solution : put it within quotes and prepend a DISP 1; to enter it in the history stack. A real p... in the a.. my friend. So now I have the syntax, I see that the parameters are expected in the 'wrong' order. Morale : Syntax should be an interactive tool not interfering with current entries, maybe even a catalog (!!). Note also that before it accepts to give you directions, you'll have to remove all opening parentheses that the typing aid could have inserted for you. Lose what you gain. Would it be hard to recognize any line and provide help on the first word encountered ?? I now just have to recall my DISP line and (painfully - did I tell you about the keys ?) edit it to calculate. But wait ! What's going on ? Yes, w-a-i-t really, the machine loops endlessly on that simple integral. After one minute, I kill the calculation. No machine on the market should be so excruciatingly slow ! It's just amazing for such a simple function. I go into the CAS and type my integral, and yes the result comes out nicely in a few seconds. Note that I had to press OK because step-by-step display of calculations was on, I think I recall this was mandatory to see some results (like TABVAR, which doesn't work otherwise). Just painful, but not too bad. But what about those complex numbers in the result (they cancel just right, but are presented to the amazed user - the 40 also asked if I wanted complex mode ON) ? Argh. Now is the time to do what I was aiming at : graph a function defined by a 'complex' calculation. My function takes one real number and spits out another. It's time for [Program] mode again. I must enter a program name, only the 40 hasn't the brains to put me in alpha mode on name entry. Just another pitfall, my friend. I like mixed-case names, capitalizing words. So my program will be called LimMoy. How do I type this ? Alpha lock is not available on the keyboard (you must look up the soft keys), and then switching between upper and lower case is a pain. Finally I get it right. You all know that all variables are global on the 40, and you only get 26 of these (ok, 27. Or 28. 29 ? Right. Changes everything). , . I don't recall if the A..Z memories can be used for complex numbers, and at this time I don't really care any more. Time for quick results, my patience is growing thin. Another probleem arises immediately : how do I type the FOR loop I need ? Of course, HELPWITH being a command line statement, I must go back to the Home screen. Too painful, so I look at old programs I back' from this place. HP has a name for this, I think it is something like 'contextless loop', so you can't go back : you always must select a function anew. So type [Program] or [Home] again and again. Now I have the syntax : how weird ! HP chose to use the comma to separate arguments in a function BUT the semicolon ; for arguments to commands. Why, Oh Mamma why ? Don't tell me the machine can't decide between these based on the context. This is just a p... in the a.. again. I believe a program must absolutely be easily readable to be of use, as you're bound to reuse it and edit it (if you haven't run away in disgust, that is). So I always use end of lines to separate long statements. What does this mean my friend ? When I say END of line, I mean this is the END of the line. The line is FINISHED. So why oooooh why do I still have to type the statement separators ? TI did it right again ! What a pity. Also, note that the separator MUST NOT be typed between the last statement of a loop and the closing END (the statement this time). Otherwise : Syntax Error ! OK HP, I know you hate me, but I hate you too now !!!! Do I have to say that verbose programming style is the norm on this machine (take a look at INPUT) ? Also, where is the [STO>] key on the keyboard ? Oh yes it is a soft key, but to me this function is so fundamental that it should be available at any time. We only have 6 soft keys, so reserving one space for the STO key is a waste. On the other hand, if HP programmers have forgotten to reserve a space for the STO key on *ONE* menu, you're just stuck (I haven't seen this happen, but why create a potential bug waiting to happen ?). At long last, my program is correct (albeit pitiful). To run it, you have two possibilities : either use the [Program] menu, but you can't see the result, or use the command line, but you can't get a typing aid for the program name. HP, you did it again ! (***Correction : as I reread this, I recall a solution on the command line : just use the [Vars] menu. Still 5 keypresses at minimum, but this is much less than retyping the name !) The results are correct, now I want to plug this program into the grapher. But this is not possible : RUN is not allowed in the grapher, neither is (X). No Graph(Something) capability from the command line either. So finally I'm stuck !! Yes they finally got me !! Too bad this toy is a little too expensive for the trash can. To be honest, I'm not convinced other graphing calculators (apart the TI92/89) are capable of defining user functions (the wonderful HP71B had them in good ol'times...). For the record, I finally did this and a lot more on the TI92 with great success, the TI92 has even a script capability which allows you to create worksheets mixing comments and calculations with the capability to run these on a split screen, what a nice feature (teachers must love this !). Well, goodbye HP40G. Tomorrow you're back in the box. But this time, I won't forget. Just to say farewell, I do some tests on the CAS. Nice feature, very badly integrated with the normal operation of the calculator. I know it is possible to transfer expressions back and forth between CAS and command line, but without the manual I dare you to find it. I didn't. Also, remember that the expressions will have to be carefully rewritten by hand, since the sommand line uses S1/S2/... variables for unknowns while the CAS accepts the normal X/Y/... The CAS settings are accessible if you search a little, but it is very unfriendly. For instance, the settings displayed are the reverse of what's in effect - logical if you consider that selecting that line will put the item in effect (You see, I just can't explain that screen !!!). Why didn't they use a configuration screen like the [Modes] screen ? Was this designed by another person without synchronization with the main team or what ? What strikes me too is the bad use of the available keys. The 6 Aplet management keys are of minor importance and could have been put as Shift- or Alpha- functions of the soft keys (these key combinations are unusable now as the soft keys have no secondary function labelled). Also, why put the alpha characters BELOW their corresponding key ? I typed the wrong key an uncountalbe number of times. There is no justification for this, space is available above the keys. And please please please give me alpha lock !!! Well, this is the end of the road. As for CAS, the overall slowness and strange opinions on a valid result prevent its use. As for the use of screen real estate, it is a joke (I would say a case in point). As for the typing facilities and ease of accessing the functions, it is a pain. Farewell. ==== > don't feel too bored while I'm away... Enjoy ! > >>>>>>> > My HP40G Gripes >>>>>>> > As your flamethrowers might get a little rusty nowadays (while HP > tries very effectively to keep them in good working order), I believe > a recent experience might fuel (!) some use for them. > So read on, and start the blast ! > It is unfortunate that you've spent more time writing this rant than reading the manual. You would have learned on how you can very easily display more than one item on a screen. The HP40G has an additional feature DISPXY that hasn't been documented due to its late introduction. To see how it works: HELPWITH DISPXY like with any other commands You would have got how to actually calculate an integral (which is the same way as any other HP calculators). The standard? HP calculators were the first to actually calculate an integral in a symbolic calculator. I guess you could say that HP way of writing it IS the standard. calculator or to spend a little bit of time reading the manual If I had to throw away any piece of machine I bought because I couldn't use it at its best within 5 minutes I wouldn't own anything. Have a look at the manual, I believe the HP40G is one of the easiest machine to use. It does work differently than some other machines but you will quickly see that the logic behind each action is very clear and very sharp. Sure the TI83 let you do things differently, but these two calculators are not in the same range. It's like comparing a DOS machine to a MacOS one and say: MacOS sucks it doesn't work like my good old DOS machine ==== > Strange that the 40G doesn't take the 'standard' > syntax used on every single graphing calculator I know : I for one am extremely happy that HP doesn't do things like everyone else. If they followed everyone else, we wouldn't have RPN calculators. > integral(function, var, start, stop). Hmmm...when I write the integral on paper, I say to myself in my head The integral from (start) to (stop) of (function) d(var). In your example, I would read it the integral from -1 to 0 of (x+1)/(x-1) dx. HP's way seems more logical, even if it isn't like every other [TI] calculator you've ever seen. > HP chose to use the comma to separate > arguments in a function BUT the semicolon ; for arguments to > commands. Isn't this the way C++ works? Semi-colons for loop commands and commas for functions? Aw, no one uses that old programming language anyway. You're comparing the HP-40G to a TI-92. This seems a little like comparing a Ford Mustang to a Chevrolet Cavalier, and coming to the conclusion that all the cars Chevy makes are slow and cheap. You can't expect HP's mid-range calculator to compete with TI's top of the line calculator. I own an HP-49G and haven't had any of the problems that you have described. The equation writer (which you have to buy separately on the TI) takes care of most of the formatting issues. The HP-49G evaluated the integral you mentioned in 2.8494 seconds. Granted, you could have still done it faster in your head (it took a couple seconds to type it in) but I would not consider this painfully slow. Of course, I got a decimal approximation in .1694 seconds. This is quite a bit faster than most people can do in their heads (unless you happen to have memorized the 2*Ln(2) ). You want to see slow, try using the numerical solver on the TI. --CS ==== > HP chose to use the comma to separate > arguments in a function BUT the semicolon ; for arguments to > commands. > Isn't this the way C++ works? Semi-colons for loop commands and > commas for functions? Aw, no one uses that old programming language > anyway. On the 89 there is only 1 command that can take it arguments seperated by a ';' It is the augment command. augment([5],[6]) gives [[5,6]] augment([5];[6]) gives [[5] [6]] I thing ti engineers like to make things hard for themselves because this is accomplished by two different tags for augment, the tokenizer picks between them depending on how the arguments are seperated. > I own an HP-49G and haven't had any of the problems that you have > described. The equation writer (which you have to buy separately on > the TI) takes care of most of the formatting issues. Four links: http://triton.towson.edu/users/bbhatt1/ti/eqw.htm FREE http://home19.inet.tele.dk/ew/eqwflash.htm NOT FREE http://www.calvin.edu/~sstear70/exw.html FREE And for the 83+ http://www.softheiss.de/int_prj_prettypt.html FREE, The interface for this is an entry line that simulateously displays what you are writing as a pretty print expression. -Samuel S, My Yahoo acount is just for Spam. ==== I'm glad to see that you spent some time experimenting. Some of your comments are fair, others are the result of inexperience with the machine (notice I avoid using the word 'ignorance' :-). > So off we go. The 40 is a durable, powerful, low price offer with a > wonderful CAS and I like this. Bear it in mind that the calculator was aimed at a low price high school market. It's pretty damn good for the price charged! > Now, I start by a little program to > retrain my skills on the machine. No need to reread the manual (er, > User Guide), this is targeted at the educational market you know. Do I really need to comment on the statement No need to reread the manual???? > So let's switch it on and press [Program]. If you'd read the manual you'd know that's not necessarily the best way to go. The proper procedure, except for small utility programs, is to write an aplet. > My simple goal so far is to > display the date correctly formatted (DD/MM/YYYY) plus the day on > screen. Hum, no string manipulation. True and a bad lack. Hopefully fixed soon with the addition of a library currently being developed. > No free position display of data > on screen (only DISP ;). Not true. Use DISPXY or DISP with multiple vars output. You can also use MSGBOX to pop up a box with the result. > So let's display on > consecutive lines. Well well, how was this 'difference of dates' > function called by the way ? No catalog. No way to search for a > function. Nothing in possibly related function catalogs. The MATH button contains functions sorted by use. Unfortunately not linked to syntax which, I agree, is a pain. > By the way, > those menus never recall what was the last family used, so let's > scroll up and down every time... Simply press the first letter of the menu or function you want. For example, pressing the 7 button ('P') will jump to Polynomial functions. This trick works in ANY menu. > OK, I'm done, this program can't be written. The input form capability > can't display a screen for input of multiple values at once (while it > is possible, the system does this all the time - see the [Plot Setup] > for example). The program can be written but admittedly not using input of multiple values. The command: will produce the output you require. The only drawback is that 30 April, manipulation means that you can't have the extra zero. > [..]the history displays only two past calculations > at a time (one line for expression, one for result, this 2 times makes > for the 4 available lines). Not very good, since there is a wasted > line both at the top AND the bottom of the screen? Not a valid criticism. If you are scrolling back through the history then it is more likely that you will be seeking a result than the calculation that led to it. This is not an opinion but based on observation of students using the 38G and 39G. > the bad expression. Well, why make it easy ? Now, I > face an egg and chicken problem. I want to know the syntax, but I am > still editing the equation. Note that as it was incorrect, it was not > pushed on the history stack. Fair enough, but the help function is a > command-line function itself, hence a conflict. I agree with this criticism. > Strange that the 40G doesn't take the 'standard' > syntax used on every single graphing calculator I know : > integral(function, var, start, stop). Crap! I don't care what the rest of the world does, it makes sense to me to have the parameters in the order in which you speak the function. By this I mean, we say The integral from 1 to 3 of funtion dx and that's how it is entered: integral(1,3,function,X) Other functions, such as differentiation and summation are the same. As JYA says, HP did it first so the rest of the world are the ones not doing it right! > Moral : Syntax should be an interactive tool not interfering with > current entries, maybe even a catalog (!!). I agree. > Now is the time to do what I was aiming at : graph a function defined > by a 'complex' calculation. If you mean a calculation involving complex numbers then very few calculators will graph these in the way you seem to expect. Certainly not one for this price. If you simply mean a calculation which is 'not simple' then it can be done but admittedly not conveniently. In your program, create a list var (say L1) using MAKELIST that contains the function values for Xmin to Xmax step (Xmax-Xmin)/131. eg. MAKELIST(X^3-X,X,Xmin,Xmax,(Xmax-Xmin)/131) /> L1: If your function is too complex to fit into the MAKELIST then simply create an empty list and then use a FOR loop to calculate its values. These values correspond to the pixel points on the horizontal width of the screen. Now, in the Function aplet, graph the function F1(X)=L1(ROUND((X-Xmin)*131/(Xmax-Xmin),0)+1) I'm NOT, by the way, claiming that this is at all easy or convenient! Just that it can be done. I agree that it would have been nice to be able to define a user function and then graph it. However, again, this is not the sort of thing one expects in a mid-range calculator aimed at high school students. > My function takes one real number and spits out another. It's time for > [Program] mode again. I must enter a program name, only the 40 hasn't > the brains to put me in alpha mode on name entry. I agree. > I like mixed-case names, capitalizing words. So my > program will be called LimMoy. How do I type this ? Alpha lock is > not available on the keyboard (you must look up the soft keys), and > then switching between upper and lower case is a pain. Finally I get > it right. I think it is better on the soft keys so that it is only available when required and does not take up a permanent hard key. A large group of people spent a LOT of time thinking about what keys should appear as hard keys. I don't see any easy way if you're going to insist on mixing cases. > You all know that all variables are global on the 40, and you only get > 26 of these (ok, 27. Or 28. 29 ? Right. Changes everything). conflicts on>, . var of any required length and use it to store your vars. I agree it would have been nice to have longer names for vars though. > I don't recall if the A..Z > memories can be used for complex numbers, and at this time I don't > really care any more. Time for quick results, my patience is growing > thin. They can't. Use the complex vars Z1, Z2,... Z0. > Another probleem arises immediately : how do I type the FOR loop I > need ? [...] how weird ! HP chose to use the comma to separate > arguments in a function BUT the semicolon ; for arguments to > commands. Why, Oh Mamma why ? Every language has its small ways of doing things. This is quite similar to C as someone pointed out. > Don't tell me the machine can't decide > between these based on the context. True but it would have meant more complexity in the OS. More bugs, more size..... > Do I have to say that verbose programming style is the norm on this > machine (take a look at INPUT) ? Did you try using the short form of PROMPT : ? INPUT is specifically set up to be as flexible as possible to as to allow easy 'user friendly' programming. Remember (again) that this is aimed at high school students - we want the input forms to be very verbose. > Also, where is the [STO>] key on the > keyboard ? Oh yes it is a soft key, but to me this function is so > fundamental that it should be available at any time. Ummm... it is. Just go through the CHARS button if it is not a soft key in the current context. Four key presses - SHIFT CHARS LEFTARROW ENTER. > At long last, my program is correct (albeit pitiful). To run it, you > have two possibilities : either use the [Program] menu, but you can't > see the result, RTFM. Just include a FREEZE command at the end of the program. > or use the command line, but you can't get a typing > aid for the program name. HP, you did it again ! > (***Correction : as I reread this, I recall a solution on the command > line : just use the [Vars] menu. Still 5 keypresses at minimum, but > this is much less than retyping the name !) Actually 4 keypresses: VARS 'P' RIGHTARROW ENTER (if it's the first in the list), otherwise one more to press the button of the first letter in the name. This seems to me quite reasonable. > The results are correct, now I want to plug this program into the > grapher. But this is not possible : RUN is not allowed > in the grapher, neither is (X). No Graph(Something) > capability from the command line either. So finally I'm stuck !! Yes > they finally got me !! Covered earlier.... > Too bad this toy is a little too expensive for > the trash can. To be honest, I'm not convinced other graphing > calculators (apart the TI92/89) are capable of defining user functions > (the wonderful HP71B had them in good ol'times...). For the record, I > finally did this and a lot more on the TI92 with great success, the > TI92 has even a script capability which allows you to create > worksheets mixing comments and calculations with the capability to run > these on a split screen, what a nice feature (teachers must love this > !). As was commented by others, I don't think it is fair to compare it with the TI92. I think a better comparison is with the TI83+. For a detailed comparison of the HP39/40G with the TI83+ by Tim Wessman see http://www.hphomeview.com/timwesmn/ti83-hp39.pdf > What strikes me too is the bad use of the available keys. The 6 Aplet > management keys are of minor importance and could have been put as > Shift- or Alpha- functions of the soft keys (these key combinations > are unusable now as the soft keys have no secondary function > labelled). Now this shows your total misunderstanding of the philosophy of the calculator. The whole HP39G revolves around the aplets and the keys APLET, HOME, PLOT, SYMB and NUM are by far the most commonly used on the calculator. Hence their prominence. > Also, why put the alpha characters BELOW their > corresponding key ? I typed the wrong key an uncountalbe number of > times. It takes a short time to get used to and then you never do it again. > Well, this is the end of the road. You bring up some interesting and valid points in a few places but you are too quick to make a final rejection without considering that you may be mistaken in some places. ==== > > HP chose to use the comma to separate > > arguments in a function BUT the semicolon ; for arguments to > > commands. > > Isn't this the way C++ works? Semi-colons for loop commands and > commas for functions? Aw, no one uses that old programming language > anyway. > > On the 89 there is only 1 command that can take it arguments seperated > by a ';' It is the augment command. > > augment([5],[6]) gives [[5,6]] > augment([5];[6]) gives > [[5] > [6]] > > I thing ti engineers like to make things hard for themselves because > this is accomplished by two different tags for augment, the tokenizer > picks between them depending on how the arguments are seperated. > > > I own an HP-49G and haven't had any of the problems that you have > described. The equation writer (which you have to buy separately on > the TI) takes care of most of the formatting issues. > > Four links: > http://triton.towson.edu/users/bbhatt1/ti/eqw.htm FREE > http://home19.inet.tele.dk/ew/eqwflash.htm NOT FREE > http://www.calvin.edu/~sstear70/exw.html FREE > > And for the 83+ > http://www.softheiss.de/int_prj_prettypt.html FREE, The interface for > this is an entry line that simulateously displays what you are writing > as a pretty print expression. > > > > -Samuel S, My Yahoo acount is just for Spam. my HP and are impressed. They will be happy to know that they can get one for free. I also know a lot of TI-83 owners that will be happy about the other link as well. --CS ==== > As JYA says, HP did it first so the rest of the world are the ones not > doing it right! HP most certainly didn't do it first. There have been dozens of computer algebra systems before the HP48. -- ==== > As JYA says, HP did it first so the rest of the world are the ones not > doing it right! HP most certainly didn't do it first. There have been dozens of > computer algebra systems before the HP48. Were they integrated into any handheld calculators? ==== > I think you should DETACH the library after purging it, but I am > not shure. No, this job is done by the Filer automatically. In the worse case, you should detach the library BEFORE purging it otherwise you may get an error: Object in use ==== When i use variables in a matrix and invert the matrix i often get huge expressions inside the matrix. I have tried to do eval and simplify. The later works sometimes, but not all the time. If it's not possible to simplify, then i have a followup question. Is there an easy way to put the denumerator outside the matrix if the denumerator is the same for all elements? If you want to help me you can try this matrix: [['x+2' -4]['L1-2' 'L2+X+5]] Then invert it and you see the hell i'm trying to explain =) My machine is a HP49G with rom 1.19-6. // Jocke B ==== > When i use variables in a matrix and invert the matrix > i often get huge expressions inside the matrix. I have tried to do eval > and simplify. The later > works sometimes, but not all the time. > > If it's not possible to simplify, then i have a followup question. Is there > an easy way to put the denumerator outside the matrix if the denumerator is > the same for all elements? > > If you want to help me you can try this matrix: > > [['x+2' -4]['L1-2' 'L2+X+5]] > > Then invert it and you see the hell i'm trying to explain =) > My machine is a HP49G with rom 1.19-6. > > // Jocke B Multiply the inverse matrix by the determinant of the original matrix and then EXPAND that result. The final result will be the matrix of numerators, and the determinant will be the appropriate common denominator. In general, a 2 by 2 matrix, M = [[a b ][ c d ]], with nonzero determinant D = a*d = b*c, has inverse INV(M) = [[d/D -b/D][ -c/D a/D ]] = [[d - b][ -c a]]/D ==== > Multiply the inverse matrix by the determinant of the original matrix and > then EXPAND that result. The final result will be the matrix of > numerators, and the determinant will be the appropriate common > denominator. In general, a 2 by 2 matrix, M = [[a b ][ c d ]], with nonzero > determinant D = a*d = b*c, has inverse INV(M) = [[d/D -b/D][ -c/D a/D ]] > = [[d - b][ -c a]]/D > Okidokie... Thx! // Jocke ==== > Multiply the inverse matrix by the determinant of the original matrix and > then EXPAND that result. The final result will be the matrix of > numerators, and the determinant will be the appropriate common > denominator. > > In general, a 2 by 2 matrix, M = [[a b ][ c d ]], with nonzero > determinant D = a*d = b*c, has inverse INV(M) = [[d/D -b/D][ -c/D a/D ]] > = [[d - b][ -c a]]/D > > Okidokie... Thx! > > // Jocke Sorry, typo in last paragraph, D = a*d = b*c should be D = a*d - b*c ==== in a SysRPL library, how can i call a xNAME inside another xNAME like a subroutine call (GOSUB or GOTO) ? Jorge Luis ==== Please, see the examples in ML (machine language) for call to subrutines, in the Eric Rechlin Web site (www.hpcalc.org). The examples are written by Peter Geelhoed for HP-49G. Later you can call this subroutine from SysRPL. Miguel Angel CAPORALINI HERK (M.A.C.H.) **************************************************************************** **** > > in a SysRPL library, how can i call a xNAME inside another xNAME like > a subroutine call (GOSUB or GOTO) ? > > > Jorge Luis ==== normally, when you create a new xNAME, you also create a EXTERNAL x??? entry at the top of your source file (or in a .h file). for example: EXTERNAL xA EXTERNAL B xNAME A << .... >> NULLNAME B << .... >> then, any call to xA or B will be compiled in ROMPTR calls. > Please, see the examples in ML (machine language) for call to > subrutines, in the Eric Rechlin Web site (www.hpcalc.org). The > examples are written by Peter Geelhoed for HP-49G. > Later you can call this subroutine from SysRPL. > Miguel Angel CAPORALINI HERK (M.A.C.H.) > **************************************************************************** **** > > in a SysRPL library, how can i call a xNAME inside another xNAME like > a subroutine call (GOSUB or GOTO) ? > > Jorge Luis Is it possible to access the CAS built in to the HP-39G ROM? My understanding is that the HP-39G has the same ROM as the HP-40G with CAS, but the CAS is not accessable on the HP-39G. The CAS is nearly identical to the HP-49G version 19-1. Any help here would be appreciated. ==== Last time I suggested someone to disable the IR, in order to gain access to the 40G CAS, the poor guy did it physically thus permanently damaged his 39G. You just have to figure out how to do it in software AND all by yourself... )-: > Is it possible to access the CAS built in to the HP-39G ROM? My understanding > is that the HP-39G has the same ROM as the HP-40G with CAS, but the CAS is > not accessable on the HP-39G. The CAS is nearly identical to the HP-49G > version 19-1. Any help here would be appreciated. ==== > You just have to figure out how to do it in software > AND all by yourself... )-: A dilated fear. HPCC #1046 ==== > So does mine, as I expected that it would. It did occur to me, and I > (thought that) I had tried exact mode before posting. I can only guess > that maybe I released the RightShift key before pressing the ->NUM key, > and failed to notice that the ~ in the status area hadn't changed to You do not have to switch your calculator to approximate mode in order to get a numerical result. RightShift + ENTER pressed simultaneously will switch the mode between approximate and exact. However, pressing just RightShift then ENTER will call ->NUM which gives a numerical result no matter what mode you're in ==== > So does mine, as I expected that it would. It did occur to me, and I > (thought that) I had tried exact mode before posting. I can only guess > that maybe I released the RightShift key before pressing the ->NUM key, > and failed to notice that the ~ in the status area hadn't changed to You do not have to switch your calculator to approximate mode in order to > get a numerical result. > RightShift + ENTER pressed simultaneously will switch the mode between > approximate and exact. However, pressing just RightShift then ENTER will > call ->NUM which gives a numerical result no matter what mode you're in Yes, I'm well aware of that. What I was trying to do was get a symbolic result. -- James ==== Some questions about hp48 C++ cross-compiler: Did anyone find it useful ? (MinGW32). xgcc: No such file or directory (but I have it in /usr/bin) Under MinGW32 (when trying to make samples): Making examples... make[1]: Entering directory `/saturn-local-hp48/samples/float' xgcc -Wall -O2 -c -o mandel.o mandel.c 0 [main] make 2468 proc_subproc: Couldn't duplicate my handle<0x3C> for pi d 2488, Win32 error 6 xgcc.exe: No input files make[1]: *** [mandel.o] Error 1 make[1]: Leaving directory `/saturn-local-hp48/samples/float' make: *** [all] Error 1 I have read somewhere that it requires Cygwin (but i think that it should also work under MinGW) . -- Vojtech Sazel --- Checked by AVG anti-virus system (http://www.grisoft.com). ==== How about learning RPL? Maybe it would make some things easier;-) Raymond Vojtech Sazel schrieb im Newsbeitrag > Some questions about hp48 C++ cross-compiler: > Did anyone find it useful ? > (MinGW32). xgcc: No such file or directory (but I have it in /usr/bin) Under MinGW32 (when trying to make samples): > Making examples... > make[1]: Entering directory `/saturn-local-hp48/samples/float' > xgcc -Wall -O2 -c -o mandel.o mandel.c > 0 [main] make 2468 proc_subproc: Couldn't duplicate my handle<0x3C for pi > d 2488, Win32 error 6 > xgcc.exe: No input files > make[1]: *** [mandel.o] Error 1 > make[1]: Leaving directory `/saturn-local-hp48/samples/float' > make: *** [all] Error 1 I have read somewhere that it requires Cygwin (but i think that it should > also work under MinGW) . -- > Vojtech Sazel > --- > Checked by AVG anti-virus system (http://www.grisoft.com). ==== I had to find the derivative of tan(x) on my HP49, and while it was on the stack, I idly pressed INTVX, thinking I'd get tan(x) back again. Instead, it returned 'tan(x)-atan(tan(x))'. I could see by inspection that this evaluates to tan(x), but simplifiers like SIMPLIFLY, TSIMP, EXPAND, COLCT and so on just made the expression more and more complicated. They couldn't seem to grasp that atan(tan(x)) = x. I tried the same thing on my HP48 and it returned yet another large antiderivative, but ALG48's ASIM converted it to tan(x), right away. A flag setting, maybe? (I had 'Principal value' selected.) Bill Curious Markwick Toronto, Canada ==== > I had to find the derivative of tan(x) on my HP49, and while it was on > the stack, I idly pressed INTVX, thinking I'd get tan(x) back again. > Instead, it returned 'tan(x)-atan(tan(x))'. There's no reason you should expect that the integral of a derivative to give you back exactly the original function. At the very least the integral of the derivative can differ from the original function by any arbitrarily chosen constant. > I could see by inspection that this evaluates to tan(x), but > simplifiers like SIMPLIFLY, TSIMP, EXPAND, COLCT and so on just made > the expression more and more complicated. They couldn't seem to grasp > that atan(tan(x)) = x. Well, that's certainly not true for all x; it would only be true for -pi/2 < x < pi/2. Even assuming you can make that simplification, tan(x)-x is rather a different function than tan(x). Since I get d/dx tan(x) = tan(x)^2+1 (which could also be simplified to sec(x)^2 = 1/cos(x)^2), it is interesting that int(tan(x)^2,x) can be simplified to tan(x)-x, which makes it look as if INTVX is forgetting to deal with that '+1'. > I tried the same thing on my HP48 and it returned yet another large > antiderivative, but ALG48's ASIM converted it to tan(x), right away. > > A flag setting, maybe? (I had 'Principal value' selected.) My HP 48SX with ALG48 and the INTGR library also produces the less confusing (and arguably more correct) result. -- Steve VanDevender I ride the big iron http://jcomm.uoregon.edu/~stevev Little things break, circuitry burns / Time flies while my little world turns Every day comes, every day goes / 100 years and nobody shows -- Happy Rhodes ==== > > I had to find the derivative of tan(x) on my HP49, and while it was on > the stack, I idly pressed INTVX, thinking I'd get tan(x) back again. > Instead, it returned 'tan(x)-atan(tan(x))'. You must have a defiective calculator then. Mine returns 'TAN(X)-ATAN(TAN(X))+X' That extra +X on the end is important. ==== > > I had to find the derivative of tan(x) on my HP49, and while it was on > the stack, I idly pressed INTVX, thinking I'd get tan(x) back again. > Instead, it returned 'tan(x)-atan(tan(x))'. > > There's no reason you should expect that the integral of a derivative to > give you back exactly the original function. At the very least the > integral of the derivative can differ from the original function by any > arbitrarily chosen constant. > > I could see by inspection that this evaluates to tan(x), but > simplifiers like SIMPLIFLY, TSIMP, EXPAND, COLCT and so on just made > the expression more and more complicated. They couldn't seem to grasp > that atan(tan(x)) = x. > > Well, that's certainly not true for all x; it would only be true for > -pi/2 < x < pi/2. > > Even assuming you can make that simplification, tan(x)-x is rather a > different function than tan(x). > > Since I get d/dx tan(x) = tan(x)^2+1 (which could also be simplified to > sec(x)^2 = 1/cos(x)^2), it is interesting that int(tan(x)^2,x) can be > simplified to tan(x)-x, which makes it look as if INTVX is forgetting > to deal with that '+1'. > > I tried the same thing on my HP48 and it returned yet another large > antiderivative, but ALG48's ASIM converted it to tan(x), right away. > > A flag setting, maybe? (I had 'Principal value' selected.) > > My HP 48SX with ALG48 and the INTGR library also produces the less > confusing (and arguably more correct) result. When I did it on the 49 ROM 1.19-6, I got 'TAN(X)-ATAN(TAN(X))+X'. It wouldn't simplify it down any more than that. Like Steve said, that would be presumptuous on the part of the calculator. --CS ==== X > As I turn the contrast up - the lcd begins to show (progressively), a kind > of rippled pattern of sorts (not a great description... but the artifact > is more or less abstract: hard to explain). [I could upload a photograph if > need be] X I would not worry, but a link to your web-page where we could see the photos would be very nice OR if you don't have a web-page then maybe Father Joe would allow you to use his page, he seems to be collecting pictures of the HP 49G displays in various conditions of light sources etc. (this is my idea) ask Joe: http://www.holyjoe.net/hobbies.htm > Side-note for those interested: The CN1500... that I own has noticeably more > comfortable buttons (easier to push) than another 49G that I have tried > (didn't check serial). Buttons are responsive and *very* seldom go without > registering a push. No rainbow effect. Screen window scratches very easily > (though I can't compare... perhaps earlier models were even more easily > scratched ^^) Nice to here that you are very satisfied (-: ==== I've posted an image to http://home.mweb.co.za/mw/mwpooch/49LCD.jpg. Not the clearest photograph - but it shows the problem (?) effectively. My question now is this: Is the same thing present on all 49G calculators... ie. is this normal? Should I be worried enough to contact my local HP support? The affected pixels appear to always be the same pixels: ie. - the problem doesn't seem to move around. Viewing the LCD from different angles also makes no change. Lastly: left calc without batteries for 15+ hours; also to no avail. I've had the calculator now for only 4 months: am going to be seriously annoyed if something is already wrong with it. This especially considering they'll not likely be able to switch it out right now if something is wrong. Anyway, there's the photo. I await your opinion ^^ eshylay ==== > I keep it with its plastic >protector clipped on; inside an HP48 soft case; inside another padded carry >case. I always have the serial port covered with tape. gee.... and I thought I was fussy with mine.... ...unless you live by the sea, or in a dessert or jungle... why such extreme precautions ??? these calculators are supposed to be RUGGED.... and almost indestructible... for some weird reason... I always find that when I get too overly fussy, something goes wrong.... or else some accident occurs. I certainly dont fuss like I used to..... and I enjoy life a whole lot more.... Just my 0.02 worth.... ==== Sorry! I cannot access that URL, are you sure about spelling? Can you use somebody else's PC to ensure it works from the outside world, too? I've posted an image to > http://home.mweb.co.za/mw/mwpooch/49LCD.jpg. X ==== The link worked 12 hours ago but not now. ed > Sorry! > I cannot access that URL, are you sure about spelling? > Can you use somebody else's PC to ensure it works > from the outside world, too? > > I've posted an image to > http://home.mweb.co.za/mw/mwpooch/49LCD.jpg. > X ==== >Sorry! >I cannot access that URL, are you sure about spelling? >Can you use somebody else's PC to ensure it works >from the outside world, too? OTOH, I haven't seen that pattern in my 49. SOmetimes there are vertical lines that are a little darker, but not this kind of diagonal pattern. -- Beto ==== Sorry about the trouble with the URL, it seems you had some bad timing: my ISP (who hosts the file) is having some difficulties... erm... working :) I've uploaded to another address: http://chaosproject.netfirms.com/49LCD.jpg If there is any other problem, just let me know and I will 20kb). eshylay ==== It worked now! I've never seen that before?! Can't help you on this one. Read the other answers and go figure if you can (I surely can't) ???? > The link worked 12 hours ago but not now. ed > Sorry! > I cannot access that URL, are you sure about spelling? > Can you use somebody else's PC to ensure it works > from the outside world, too? > > I've posted an image to > > http://home.mweb.co.za/mw/mwpooch/49LCD.jpg. > X > > ==== Well thanks for the help. Contacted HP today and they said they'd swap it out once they have stock. As expected, that is not currently the case. Apparently a world wide 49G shortage. Ahem. We'll see when it happens - the sales woman advised that stock may be arriving in a week or two. I'll believe it when I see it - either way I'm somewhat peeved that something is alreay wrong with my lovely calculator. So long as they sort it out - I'll forgive :) Anyway, - again: appreciated you taking a look. eshylay ==== Dear Eshylay. My almost new'n'out-of-the-box HP49 has exactly the same problem (I would discard the theory that you did anything wrong to trigger the problem.) Serial CN148... so produced a couple of weeks before yours. If you want, I can upload pics to compare. Problem may be slightly less severe in mine than in yours, but it's obviously the same artifact. You mentioned you contacted HP for a replacement (eventually...). Maybe I will, too. I'm shaky about the LCD quality at this point. Eduardo ==== Sorry to hear I'm not the only one then :s... No need to post pictures (unless you specifically want me to look and compare?). I'll see what my local HP says about the problem when I hand the calculator in - see if they want to exchange or repair something in particular. I'm just worried that the problem will get worse. Otherwise, even, the amount of money that I paid for the unit was considerable (especially considering my... erm... budget ^^)... I want to make sure I've got something that isn't even mildly faulty. Speak to your local office and see what they suggest. eshylay ==== I am a little allerted because the command LIBS gives on my calc the following List on the stack { 257 2 257 2 2... It looks like lib257 is double-attached? lib257 is not visibible in the Filer, so what could be wrong, and is there someting to fix and how? ..Heiko ==== > LIBS > gives on my calc the following List on the stack > { 257 2 257 2 2... > > It looks like lib257 is double-attached? That's very weird. First of all, the 257 library is the MASD library, so when attached you should after calling LIBS: { MASD V5 257 } Try removing all the external libraries and see what it displays, then re-install each library one at a time to see which is the faulty one (probably not attaching the library correctly) ==== > It looks like lib257 is double-attached? That's very weird. First of all, the 257 library is the MASD library, so > when attached you should after calling LIBS: > { MASD V5 257 } I have found the initiator: Informbuilder v1.6 which is I think obsolete, because v1.7 is abailable. I will make my experience with the new version of it and will report on that. But nevertheless I got still { 257} when calling the LIBS-command not MASD V5! And lib 257 is invisible with the filer ?! I would like to mention, that I have had no really problem with the mis-attached lib, I have just wondered about the double-attachment. Sorry for my nosy questions. Best wishes Heiko ==== I noticed that HP no longer has the HP49G on its web site. The HP48GX is there, and several non-professional calculators. Looks like Carly is not interested in pro products. GR ==== > I noticed that HP no longer has the HP49G on its web site. > > The HP48GX is there, and several non-professional calculators. > > Looks like Carly is not interested in pro products. > > GR I noticed the same thing yesterday when I was trying to look up the specs of the HP-49 compared to the HP-40. In order to find the 49, I had to go to google and search for it. I couldn't even get there from HP's search engine. Once I got there, I was able to find the path (support I think). Wierd that the search engine at HP's website doesn't even recognize the HP-49g. It said no matching documents. --CS ==== As the subject says... since I have been to this group... I remember reading it using Unix as a matter of fact... the old 286 day's... Anyway where can I gat a copy of the FAQ and download sites? JB Originator: mschaef@io.com (MSCHAEF.COM) ==== >As the subject says... since I have been to this group... >I remember reading it using Unix as a matter of fact... the old 286 day's... >Anyway where can I gat a copy of the FAQ and download sites? These three sites should be a pretty good start, particularly the first if you're still an active user. http://www.hpcalc.org/ http://www.hpmuseum.org/ http://www.hp41.org/ -Mike -- http://www.mschaef.com ==== > >>As the subject says... since I have been to this group... >>I remember reading it using Unix as a matter of fact... the old 286 day's... >>Anyway where can I gat a copy of the FAQ and download sites? > > > These three sites should be a pretty good start, particularly the first > if you're still an active user. > > http://www.hpcalc.org/ > http://www.hpmuseum.org/ > http://www.hp41.org/ Also, the FAQ is here: http://www.engr.uvic.ca/~aschoorl/faq/ -- darren ==== I would guess yes. But would like to know for sure before trying to upgrade. Sincerely, Kevin Waite ----- Original Message ----- ==== Kevin Waite schrieb im Newsbeitrag I would guess yes. But would like to know for sure > before trying to upgrade. > The revision 'R' is VERY stable. There are nearly no software bugs, and those very few are very well documented. What do you want to upgrade? If you want to add more RAM, the only thing to know is that you should add less than 4MB. However, 1 or 2MB should be sufficient in most cases (IMHO;-) Raymond ==== > I would guess yes. But would like to know for sure > before trying to upgrade. > The revision 'R' is VERY stable. > There are nearly no software bugs, > and those very few are very well documented. I could publish a not at all small list of not-documented bugs of the HP48G (JYA also knows several such bugs), but I have no time to do it. Fortunately, most of these are not noticed by the normal user. I used my free time to perfectionize my Timeman for the HP49. Presently only on my site. It obsolates both, the traditional method of setting date and time on the HP49, AND most calendar, time, and clock calibration tools available on hpcalc.org. The paging calendar is in two formats, depending on your setting of flag -42. One can virtually browse through the month, years, and even centuries and scan any wanted day. The calendar preserves its correct shape. - Wolfgang ftp://ftp.math.fu-berlin.de/pub/usr/raut/HP49/time/Timeman/ ==== Wolfgang Rautenberg schrieb im Newsbeitrag > >[..] > The revision 'R' is VERY stable. > There are nearly no software bugs, > and those very few are very well documented. [..] > I could publish a not at all small list > of not-documented bugs of the HP48G > [..] > Fortunately, most of these are not noticed by the normal user. > And this is the point. They don't affect the normal user's work in a negative way;-) Up to 1998, I used a rev. M w/o problems, the only thing I had to be aware of was the well known STO port bug. > I used my free time to perfectionize my Timeman for the HP49. > You're the *man Maybe you should make a version for the HP-48, too, so many more users could use it;-) Raymond ==== > > You're the *man > > Maybe you should make a version for the HP-48, too, > so many more users could use it;-) Yes, Wolfgang, pleeeeeeease. ==== > Maybe you should make a [Timeman]version for the HP-48, too, > so many more users could use it;-) > > Yes, Wolfgang, pleeeeeeease. why don't you port Timeman to the 48 instead of loosing your time in suggestions? Timeman uses only three supported commands not in 48-SysRPL: $>grobCR, AskQuestion and ViewStrobject. No problem for you, Raymond, to reprogram these three useful commands for the HP48, using either 48-SysRPL or ML :-) In this NG is discussed much nonsense, but very rarely some useful project. Such a project would be to make a library which ports the most important additions of ACO to SysRPL back to the HP48. Apart from the mentioned examples the Virtual Stack tools, etc. Clearly, it would be a child's play for JYA to present such a lib to the 48-fan community . But he's probably interested in that his HP49 sells well, not the HP48 :-) - Wolfgang. PS. Interesting to know which of the well-known and less known bugs of the HP48 have been fixed by JYA in the HP49 and which not. ==== I do not know if this should be ROM level R or not. Please advise. Sincerely, Kevin Waite ==== R is the latest ROM... none has been produced since. ==== If so, how is this done? Or are the GX rom files out on the various sites just for the emulator? Sincerely, Kevin Waite ==== I'm no expert but I know the 48GX doesn't have flash ROM, hence isn't upgradable (as the 49G is). You are right to assume the ROM files out there are mostly useful for emulators. Eduardo > If so, how is this done? Or are the GX rom files out on the various sites just > for the emulator? Sincerely, > Kevin Waite > ==== Salut a tous les posseseurs de 28, Je viens de recuperer une 28 en super bon etat dans un vide-greniers pour 4 euros. C'est une super machine, la premiere que j'ai eue qui se programme en RPL (avant ma 48SX puis ma 48GX). Du coup ca m'a donne l'envie de faire tourner un emulateur de 28 mais la probleme, il faut une image de la ROM. Comme je suis un peu faineant et que je n'ai ni recepteur infrarouge ni 48, ca serait super si quelqu'un pouvait m'envoyer une copie des ROMs, je pense pas que HP s'en sente lese pour autant ;) A vot' bon coeur m'sieu dames! Serge ==== > Salut a tous les posseseurs de 28, > > Je viens de recuperer une 28 en super bon etat dans un vide-greniers > pour 4 euros. C'est une super machine, la premiere que j'ai eue qui se bonne affaire ! > programme en RPL (avant ma 48SX puis ma 48GX). Du coup ca m'a donne > l'envie de faire tourner un emulateur de 28 mais la probleme, il faut > une image de la ROM. > Comme je suis un peu faineant et que je n'ai ni recepteur infrarouge > ni 48, ca serait super si quelqu'un pouvait m'envoyer une copie des > ROMs, je pense pas que HP s'en sente lese pour autant ;) > A vot' bon coeur m'sieu dames! > > Serge* J'ai la rom sur une D7 mais h.8elas le lecteur de D7 qui pourrait la relire est HS Si jamais tu vois un integral PC d'HP (station unix portable) dans une brocante... Paul > ==== Does anyone know how to get the 49 simplify 'sqrt(15*X*XROOT(3,X))' into something that the 49 will integrate? I was able to do it by hand for this example, I was just wondering if the calculator would do it for me. --CS BTW, how do I get to the radical symbol on the computer keyboard? ==== > Does anyone know how to get the 49 simplify 'sqrt(15*X*XROOT(3,X))' > into something that the 49 will integrate? I was able to do it by > hand for this example, I was just wondering if the calculator would do > it for me. You have a TI-89, don't you? It'll do this integral. > BTW, how do I get to the radical symbol on the computer keyboard? Character code 0xD6 in the Symbol font, at least on Windows. Try the Character Map (charmap.exe) -- ==== > Does anyone know how to get the 49 simplify 'sqrt(15*X*XROOT(3,X))' > into something that the 49 will integrate? I was able to do it by > hand for this example, I was just wondering if the calculator would do > it for me. The 49 seems to like radicals more than powers since it converts everything to powers... anyway, if you want the calculator to do this integral for you, a command I would use is TSIMP (EXP&LN,NXT,TSIMP or TRIG,NXT,NXT,TSIMP). It simplifies it to a form that the 49 can integrate with ease. I am not sure why the 49 can't directly integrate it, and why TSIMP would be required. Maybe a more knowledgeable person will answer that. Albert ==== im searching for a real good User-RPL Programming Guide. OK there are still existing many documents but i'm still searching for the best one. But I found really nothing about HP-Basic for the HP49. Marc. ==== Marc Kamphausen schrieb > im searching for a real good User-RPL Programming Guide. OK there are still > existing many documents but i'm still searching for the best one. But I found really nothing about HP-Basic for the HP49. Search for HP48 on www.hpcalc.org for usrRPL it is for ?90% valid. HP-Basic is something which is slightly different to calculator language. ..Heiko ==== > Did you know that you can use the Vernier Universal Lab Interface to > connect to the HP48GX? A graduate student in Physics at The University > of Wisconsin at Oshkosh did it in 1997,I think. Use a search engine own collecting device! ==== My default modulo value is 13. If I use the [MODE] [CAS] options to go to the CAS MENU, and move the cursor so that the modulo value is highlighted, the bottom of the screen says Enter modulo value. If I enter 3, or 3. for that matter, my calc says BAD ARGUMENT TYPE. If, however, I use the file manager and go to the CASDIR I find a integer veriable MODULO. Using edit, I can change the value to 3. IF I then go back to the CAS MENU my MODULO value is indeed 3. But after a warm start the value is back to 13. Why is the default value on my calculator 13, while all the books and manuals descripe that it should be 3? Also, why does my calculator not keep my modulo setting after a warm start? Should these CAS variables be in my HOME directory? I use the beta rom 1.19-6. Also, it happend that when I entered a vector , e.g. [LSHIFT][*][2][SPC][3][SPC][5][ENTER], the result is [2 3 5]. All numbers entered without a dot afterwards is used as integers and not real values. I suppose this is some flag or setting that has been changed, but I do not no which one or where to change it back again. Interestingly, if I have a vector in the form [2 3 5] and try to convert it from rectangular coordinates to say for instance spherical coordinates, it remains unchanged, although the same procedure alters a vector in the form [2. 3. 5.] to spherical coordinates. Kaspaas ZA ==== I have the Emu48 program and the hp49g-1.18.flash file. How can I get Emu48 emulating an HP-49g? ==== > On the 89 the sign() functions is defined as following: sign(positive) = 1 > sign(0) = +-1 > sign(negative) = -1. On the Hp49, sign(x) evaluates to x / |x| or x/abs(x) if you will. > Solving sign(x^2 - 3)*x =0, for x can only give one solution x=0, as > the sign() function will never be 0. A simple graph of the origional > function (abs(x^2-3)-10)/2 shows that at +-sqrt(3) the graph is > pointy. Taking righthanded and left handed limits of the derivative as x goes > to sqrt(3) gives two different results. Yes, but it still looks like the graph of the derivative has valid roots at +-sqrt(3) and 0. In a Ti89 the sign() function can never be zero, but how come when I run it through Maple or other math software, it gives the same solution as the Hp49? For example, solving signum(x^2-3)*x for x in Maple 8 gives {x = 0}, {x = 3^(1/2)}, {x = -3^(1/2)}. So, which is the correct answer? The limits from both sides are different, but the solutions the HP49 gives are the minimums of the original function. Using the same methods as you would by hand, the HP49 gives the solutions. The sign() convention of derive/ti-68k is different from both maple and the hp49. Ti89: sign(0) = +-1 Maple 8: sign(0) = 0 Hp49: sign(0) = ? Which is right? Albert ==== Yes, but it still looks like the graph of the derivative has valid roots at > +-sqrt(3) and 0. In a Ti89 the sign() function can never be zero, but how > come when I run it through Maple or other math software, it gives the same > solution as the Hp49? For example, solving signum(x^2-3)*x for x in Maple 8 gives {x = 0}, {x = > 3^(1/2)}, {x = -3^(1/2)}. So, which is the correct answer? The limits from both sides are different, > but the solutions the HP49 gives are the minimums of the original function. > Using the same methods as you would by hand, the HP49 gives the solutions. The sign() convention of derive/ti-68k is different from both maple and the > hp49. Ti89: sign(0) = +-1 > Maple 8: sign(0) = 0 > Hp49: sign(0) = ? Which is right? Wolfgang may have an opinion... ==== > The sign() convention of derive/ti-68k is different from both > maple and the hp49. > Ti89: sign(0) = +-1 > Maple 8: sign(0) = 0 > Hp49: sign(0) = ? > Which is right? > Wolfgang may have an opinion... The signum function is originally defined for reals only. Any expansion to other objects, be it complex numbers or zints or algebraics or matrices is not normed. Hence, the author of a tool using an expanded signum function should clearly say in his docus how his system handles the expansion. This holds in particular for computer algebraic systems. The question which is the right one is as meaningless as the question Is 0^0 = 0 or 0^0 = 1 ? - Wolfgang ==== > Ti89: sign(0) = +-1 > Maple 8: sign(0) = 0 > Hp49: sign(0) = ? You missed one for the HP49: SIGN(0) = ?, but, to be compatible with the HP48 series, SIGN(0.) = 0. ==== I partly agree with your judgement about the sign function. If you want a definition for sgn(z) for all complex z<>0 that makes any mathematical sense at all, it better be z/abs(z). I doubt anyone ever has used an extension of the usual sgn(x) for x<>0 and real which isn't this. On the other hand, what sgn(0) is should be well documented since it isn't standard, if even you want to think of it as defined. About the 0^0 question, it is undetermined in the strictest sense but there are very good reasons to make 0^0=1. There's some discussion about this in chapter 1 of Knuth's Art of Computer Programming and, recalling from memory, he presents these arguments: (i) 0^0=1 if the binomial theorem '(a+b)^n == SIGMA(k,0,n,COMB(n,k)*a^(n-k)*b^k)' is to hold when a or b =0. (ii) If 'LIMIT(F(Z),Z=W)==0' AND 'LIMIT(G(Z),Z=W)==0' where F(Z) & G(Z) are *analytic* functions of Z (infinitely differentiable isn't sufficient) near W, then 'LIMIT(F(Z)^G(Z),Z=W)==1' (I think this was proven in some old book.) Nowadays I tend to be annoyed when a calculator complains if I ask it to compute 0^0. (I have yet to see a calculator that makes 0^0 anything other than 1, but I'm sure there must be some out there that give the value of 0^0=0). Eduardo > The signum function is originally defined for reals only. Any expansion > to other objects, be it complex numbers or zints or algebraics or > matrices is not normed. Hence, the author of a tool using an expanded > signum function should clearly say in his docus > how his system handles the expansion. This holds in particular for > computer algebraic systems. The question which is the right one is > as meaningless as the question Is 0^0 = 0 or 0^0 = 1 ? - Wolfgang ==== > (ii) If 'LIMIT(F(Z),Z=W)==0' AND 'LIMIT(G(Z),Z=W)==0' where F(Z) & G(Z) are > *analytic* functions of Z (infinitely differentiable isn't sufficient) near > W, then 'LIMIT(F(Z)^G(Z),Z=W)==1' (I think this was proven in some old > book.) I should be more precise. IF the limit of F(Z)^G(Z) exists as a limit in the *complex* plane (aka, Z is complex and approaches the (real or complex) value W) then it has to be 1. I realize this is really off the thread's topic, though. Eduardo ==== { HOME HOME. }@ note the dot after my own HOMEdot It is so small in many fonts. PS: There must be a better SysRPL/ML solution... > written it! So {HOME HOME} is your home away from home? ;-) Eduardo That brings up another related thing: > I have a 'HOME.' just below the real HOME > so that it is less cluttered > no matter in which do you prefer to think as your HOME > but the IOPAR and all such OS things goes to the real HOME > PS: Could this be the chosen one? > http://www.hpcalc.org/details.php?id=2395 > ==== :-) Eduardo ==== > no matter in which do you prefer to think as your HOME > but the IOPAR and all such OS things goes to the real HOME Unless you have renamed your HOME ... http://www.hpcc.org/V21N3/home.html HPCC #1046 ==== Oh, much too nice! Could we please have that in the 49G, too?! no matter in which do you prefer to think as your HOME > but the IOPAR and all such OS things goes to the real HOME Unless you have renamed your HOME ... http://www.hpcc.org/V21N3/home.html > HPCC #1046 ==== > > Writing like INTEGRAL(min,max,function, variable) is by far the most > logical > way of writing an integral, as it's how you write it on paper. > So then you agree that on the hp 2+3 should be written 2+3 > Enter not as 2 Enter 3 +. Ok thats good to know. ;-) > Colin's an HP38/39/40 fan so what would he know about 2 Enter 3 + ? :-) I am neither ignorant nor an idiot (contrary to some opinions in this forum ;-). I have owned and used RPN calculators in the dim prehistoric past. Personally I agree that it is far more efficient once you get used to it. However, years of experience in the classroom have shown me that only the top 25% of students ever understand it well enough (and quickly enough) to make it a desirable option for the medium to low ability student. That's my ONLY objection to it. > Here's a little test though. Suppose you want to evaluate 6x + 5x^2 + > 4x^3 + 3x^4 + 2x^5 + x^6 where x=7. How many keystrokes do you think > that it takes under algebraic versus RPN? On the 39G I entered it as read e.g. 6*7+5*7^2+4*7^3... etc. for a total > of 30 keystrokes. Actually only 29 since ^2 is one keystroke. :-) > How many keystrokes do you think it takes on an HP-12C, say, which is an > RPN only machine? Answer: 19. A piddly 19. That's less than two-thirds > what it takes on the 39. Wanna know how? I'll make that a mini-challenge and see who's up to it. No special tricks needed so you don't have to use a 12C. > :-) Actually, I can do it using only 27 keystrokes on my humble little HP39G! Still not as good as 19 but getting better. Just use: 7 + 2 ENTER * 7 + 3 ENTER * 7 + 4 ENTER * 7 + 5 ENTER * 7 + 6 ENTER * 7 ENTER I thought this might be your method, using factorising as 6x + 5x^2 + 4x^3 + 3x^4 + 2x^5 + x^6 = x(6+x(5+x(4+x(3+x(2+x)))))? Actually.... 22 keystrokes is enough if we use the 'intelligence' of the HP39G efficiently! Just enter the keys below and it will fill in the * and )'s as required ;-) 7 ( 6 + 7 ( 5 + 7 ( 4 + 7 ( 3 + 7 ( 2 + 7 ENTER Now I'm sure you'll correct me if I'm wrong but wouldn't this also require 22 keystrokes in RPN as below? 7 ENTER 2 + 7 * 3 + 7 * 4 + 7 * 5 + 7 * 6 + 7 * Maybe I got the wrong method because I can't get 19. Finally, I'm probably going to expose my ignorance here but bear in mind that it has been 20 years (!) since I've used RPN other than for 5 minutes here and there so be forgiving! Doesn't doing it as a straight forward expression using RPN without factorising or tricks require 37 keystrokes as follows? 7 ENTER 6 ^ 7 ENTER 5 ^ 2 * + 7 ENTER 4 ^ 3 * + 7 ENTER 3 ^ 4 * + 7 ENTER 2 ^ 5 * + 7 ENTER 6 * + Maybe I've got this wrong but if not then I think 29 (the straight forward method on the 39G) compares very well with 37 in RPN!! ==== I have never seen a Too Many Arguments error on a 49. On a Ti89 I see it quite often, sometimes when the brackets are not in the > right spots or there are some missing. > Why don't you use the Equation Writer? I do, but only for more complex equations... Sometimes I do not know the arguments needed for a function so I have to 'guess' what order to put the arguments in and how many are needed. When I was learning the 49 I got the Too Few Arguments error quite often ==== > Here's a little test though. Suppose you want to evaluate 6x + 5x^2 + > 4x^3 + 3x^4 + 2x^5 + x^6 where x=7. How many keystrokes do you think > that it takes under algebraic versus RPN? On the 39G I entered it as read e.g. 6*7+5*7^2+4*7^3... etc. for a total > of 30 keystrokes. How many keystrokes do you think it takes on an HP-12C, say, which is an > RPN only machine? Answer: 19. A piddly 19. That's less than two-thirds > what it takes on the 39. Now compare symbolics, on a 49 (in RPN mode) it takes 35 keystrokes to type in 6x + 5x^2 + 4x^3 + 3x^4 + 2x^5 + x^6 (with the x's), while in algebraic mode, it takes only 26 keystrokes. That's less then three-quarters of what it takes in RPN mode. Albert ==== Well said. Superior technologies are rarely mass technologies. > Writing like INTEGRAL(min,max,function, variable) is by far the most > logical > way of writing an integral, as it's how you write it on paper. > So then you agree that on the hp 2+3 should be written 2+3 > Enter not as 2 Enter 3 +. Ok thats good to know. ;-) > Colin's an HP38/39/40 fan so what would he know about 2 Enter 3 + ? :-) > I am neither ignorant nor an idiot (contrary to some opinions in this > forum ;-). I have owned and used RPN calculators in the dim prehistoric > past. Personally I agree that it is far more efficient once you get used > to it. However, years of experience in the classroom have shown me that > only the top 25% of students ever understand it well enough (and quickly > enough) to make it a desirable option for the medium to low ability > student. That's my ONLY objection to it. ==== > I am neither ignorant nor an idiot (contrary to some opinions in this > forum ;-). I have owned and used RPN calculators in the dim prehistoric > past. Personally I agree that it is far more efficient once you get used > to it. However, years of experience in the classroom have shown me that > only the top 25% of students ever understand it well enough (and quickly > enough) to make it a desirable option for the medium to low ability > student. That's my ONLY objection to it. > I agree with this objection, having observed the same. I have another objection against RPN, it can not handle function with variable arguments number (except if you precise the number of arguments at stack level 1), which for example is the reason why you have INTVX, RISCH, and the integral symbol, 3 differents function for the same integration function. And RPN programs are, if not commented correctly, very hard to understand, and even if commented correctly harder to modify (except if you are not using the stack intensively). Algebraic languages (like C++ or higher-level scripting CAS language) are much easier to code and modify. One of the big advantage of RPN is the memory footprint which was essential 15 years ago and (for a CAS) the fact that it must be a non-typed langage because of the stack. It is also excellent for programming parsing expressions if you don't want to use tools like flex/bison. <662e00ed.0304300747.7c3f355b@posting.google.com> <3EB1FFDD.8000705@iinet.net.au> ==== > Colin's an HP38/39/40 fan so what would he know about 2 Enter 3 + >I am neither ignorant nor an idiot (contrary to some opinions in this >forum ;-). I have owned and used RPN calculators in the dim prehistoric >past. Personally I agree that it is far more efficient once you get >used to it. However, years of experience in the classroom have shown me >that only the top 25% of students ever understand it well enough (and >quickly enough) to make it a desirable option for the medium to low >ability student. That's my ONLY objection to it. I've heard it suggested that HP should put RPN into their low-end models so that children can be exposed to it from a much earlier age i.e. from when they first start to use calculators in the classroom rather than by the time they need something more advanced and are ready to buy a 49G level machine. Does this sound reasonable, practicable? > Here's a little test though. Suppose you want to evaluate 6x + 5x^2 + >4x^3 + 3x^4 + 2x^5 + x^6 where x=7. How many keystrokes do you think >that it takes under algebraic versus RPN? > On the 39G I entered it as read e.g. 6*7+5*7^2+4*7^3... etc. for a >total of 30 keystrokes. >Actually only 29 since ^2 is one keystroke. :-) Nope it's 30. Unless I can't count key-presses while scrolling through the expression on a 39G. :-) > How many keystrokes do you think it takes on an HP-12C, say, which is >an RPN only machine? Answer: 19. A piddly 19. That's less than >two-thirds what it takes on the 39. > Wanna know how? I'll make that a mini-challenge and see who's up to >it. > No special tricks needed so you don't have to use a 12C. > :-) Actually, I can do it using only 27 keystrokes on my humble little >HP39G! Still not as good as 19 but getting better. Just use: 7 + 2 ENTER * 7 + 3 ENTER * 7 + 4 ENTER * 7 + 5 ENTER * 7 + 6 ENTER * 7 >ENTER Getting there... >I thought this might be your method, using factorising as 6x + 5x^2 + 4x^3 + 3x^4 + 2x^5 + x^6 = x(6+x(5+x(4+x(3+x(2+x)))))? Red hot! >Actually.... 22 keystrokes is enough if we use the 'intelligence' of >the HP39G efficiently! Just enter the keys below and it will fill in >the * and )'s as required ;-) 7 ( 6 + 7 ( 5 + 7 ( 4 + 7 ( 3 + 7 ( 2 + 7 ENTER Getting colder. Not bad for the algebraic machine though. >Now I'm sure you'll correct me if I'm wrong but wouldn't this also >require 22 keystrokes in RPN as below? 7 ENTER 2 + 7 * 3 + 7 * 4 + 7 * 5 + 7 * 6 + 7 * Maybe I got the wrong method because I can't get 19. Finally, I'm probably going to expose my ignorance here but bear in >mind that it has been 20 years (!) since I've used RPN other than for 5 >minutes here and there so be forgiving! Doesn't doing it as a straight >forward expression using RPN without factorising or tricks require 37 >keystrokes as follows? 7 ENTER 6 ^ 7 ENTER 5 ^ 2 * + 7 ENTER 4 ^ 3 * + 7 ENTER 3 ^ 4 * + 7 >ENTER 2 ^ 5 * + 7 ENTER 6 * + Maybe I've got this wrong but if not then I think 29 (the straight >forward method on the 39G) compares very well with 37 in RPN!! 9/10 for effort :-) The bit you've forgotten during that 20 years is that the traditional HP machines have a 4 level stack (X, Y, Z and T) where T retains its contents rather than dropping them when items are consumed by an operation. Thus the 19 keystroke version is: 7 ENTER ENTER ENTER 2 + * 3 + * 4 + * 5 + * 6 + * The saving comes from not having to enter the 7 repeatedly. This is one of the very few examples where the old-style stack has an advantage over the unlimited stack of modern RPL. It's also pretty idiot-proof to evaluate since it's a repetitive enter the coefficient, press +, press * and repeat operation. ==== > However, years of experience in the classroom > have shown me that only the top 25% of students ever understand it > well enough (and quickly enough) to make it a desirable option for the > medium to low ability student. That's my ONLY objection to it. > > I've heard it suggested that HP should put RPN into their low-end models > so that children can be exposed to it from a much earlier age i.e. from > when they first start to use calculators in the classroom rather than by > the time they need something more advanced and are ready to buy a 49G > level machine. Does this sound reasonable, practicable? My first instinctive reaction was that it wouldn't make a difference - the top 25% would still be the only ones to really benefit. But then I stopped and thought about it: to be quite honest I don't know. I've never heard of such an experiment being conducted (getting them really young I mean) and I don't know whether it would make a difference. Does anyone know of a serious study having been done comparing achievement using RPN vs algebraic, differentiated by mathematical ability and/or age at time of first exposure? 9/10 for effort :-) The bit you've forgotten during that 20 years is > that the traditional HP machines have a 4 level stack (X, Y, Z and T) > where T retains its contents rather than dropping them when items are > consumed by an operation. Thus the 19 keystroke version is: 7 ENTER ENTER ENTER 2 + * 3 + * 4 + * 5 + * 6 + * The saving comes from not having to enter the 7 repeatedly. > You're quite right - I was not aware of that. What a neat idea! I had been assuming a three or four level stack as was the case on the machine I used for some time in the early eighties. A very early HP that my brother bought - the first programmable one I seem to remember. Don't know what the model was. I can see why the idea would be handy but I suspect that, despite your comment of it being 'idiot proof', you'd really have to have a fair bit of practice before it became easy to use without having to stop and think about it. I'd have to have a play before I could make a guess as to whether it would raise or lower that 25% who'd get it well enough. ==== X > 9/10 for effort :-) The bit you've forgotten during that 20 years is > that the traditional HP machines have a 4 level stack (X, Y, Z and T) > where T retains its contents rather than dropping them when items are > consumed by an operation. Thus the 19 keystroke version is: 7 ENTER ENTER ENTER 2 + * 3 + * 4 + * 5 + * 6 + * The saving comes from not having to enter the 7 repeatedly. This is one of the very few examples where the old-style stack has an > advantage over the unlimited stack of modern RPL. It's also pretty > idiot-proof to evaluate since it's a repetitive enter the coefficient, > press +, press * and repeat operation. A) Since I use NDUPN quite often, I have it on keyboard, - one can spell it: 7 [SPC] 6 hold [ALPHA] press N D U P N release [ALPHA] then [ENTER] continue with DROP 2 + * 3 + * 4 + * 5 + * 6 + * - call it from the CATalog: 7 [SPC] @ or [ENTER]@ 6 [CAT] [ALPHA] [N] [DownArrow] [ENTER] continue with DROP 2 + * 3 + * 4 + * 5 + * 6 + * - use [TOOL] softmenu keys: 7 [SPC] 6 [TOOL] |STACK| (F3) and [LeftShift] PREV (Lor NXT) then |NDUPN| (F1) continue with DROP 2 + * 3 + * 4 + * 5 + * 6 + * - or have it on some key like I do 7 [ENTER] 6 shift&hold [LeftShift] ANS @ assigned key continue with 2 + * 3 + * 4 + * 5 + * 6 + * (I have my own NDUP which has the DROP build-in as ; ) B) The hard way: 7 [ENTER] [ENTER] [ENTER] [ENTER] [ENTER] [ENTER] continue with 2 + * 3 + * 4 + * 5 + * 6 + * OR 7 [SPC] 7 [SPC] 7 [SPC] 7 [SPC] 7 [SPC] 7 [ENTER] continue with 2 + * 3 + * 4 + * 5 + * 6 + * OR 7 [SPC] 7 [SPC] 7 [ENTER] [LeftShift] [CMD] [ENTER] [ENTER] continue with 2 + * 3 + * 4 + * 5 + * 6 + * Ayway the HP 28/48/49 style is not as good in keystrokes as the old system. there is always a trade-off... PS: Is there yet another way to do it? ==== > Ayway the HP 28/48/49 style is not as good in keystrokes as the old system. > there is always a trade-off... > PS: Is there yet another way to do it? Yes. [p/r] + 7 * [p/r] 7 [enter] 2 [r/s] 3 [r/s] 4 [r/s] 5 [r/s] 6 [r/s] For a total of seventeen keystrokes... dd. ==== > I agree with this objection, having observed the same. I have > another objection against RPN, it can not handle function with > variable arguments number (except if you precise the number > of arguments at stack level 1), which for example is the reason Expect one list of arguments. ==== > >Here's a little test though. Suppose you want to evaluate 6x + 5x^2 + >4x^3 + 3x^4 + 2x^5 + x^6 where x=7. How many keystrokes do you think >that it takes under algebraic versus RPN? >On the 39G I entered it as read e.g. 6*7+5*7^2+4*7^3... etc. for a total >of 30 keystrokes. >How many keystrokes do you think it takes on an HP-12C, say, which is an >RPN only machine? Answer: 19. A piddly 19. That's less than two-thirds >what it takes on the 39. > Now compare symbolics, on a 49 (in RPN mode) it takes 35 keystrokes to type > in 6x + 5x^2 + 4x^3 + 3x^4 + 2x^5 + x^6 (with the x's), while in algebraic > mode, it takes only 26 keystrokes. That's less then three-quarters of what > it takes in RPN mode. Albert How about (where E = enter) 7 E E E E E (puts 6 copies of x on the stack) 2 + * (x^2 + 2x) 3 + * (x^3 + 2x^2 + 3x) 4 + * (x^4 + 2x^3 + 3x^2 + 4x) 5 + * (x^5 + 2x^4 + 3x^3 + 4x^2 + 4x) 6 + * (the whole megilla) 21 strokes and the values of x only has to be entered once. btw, isn't there a command on the 49 for evaluating a polynomial given the values and a vecor of coefficients? ==== > Now compare symbolics, on a 49 (in RPN mode) it takes 35 keystrokes to type > in 6x + 5x^2 + 4x^3 + 3x^4 + 2x^5 + x^6 (with the x's), while in algebraic > mode, it takes only 26 keystrokes. That's less then three-quarters of what > it takes in RPN mode. I got 34 for both entry modes. You didn't count the 2 enter pushes per term in the middle, right? If you do, you get 2*4 more pushes, and that's not including the final Enter, which would bring the EQW version to 35. I do have to admit, the entry was much more straightforward in EQW mode though... -Aaron ==== X > btw, isn't there a command on the 49 for evaluating > a polynomial given the values and a vecor of coefficients? [LeftShift] [ * ] gives [ square brackets ] then I use [SPC] [1 2 3 4 5 6 0] the [RightArrow] 7 hold [ALPHA] P E V A L release [ALPHA] and press [ENTER] PS: Maybe one could use SEQ or DOLIST ==== I had to find the derivative of tan(x) on my HP49, and while it was on > the stack, I idly pressed INTVX, thinking I'd get tan(x) back again. > Instead, it returned 'tan(x)-atan(tan(x))'. Or write SIN(X)/COS(X) instead of TAN and do DERVX then INTVX. ==== I have a 48GX too and its display does *not* present the same problem. In > fact, it seems to be of excellent quality. Also, judging from previous > responses, the effect is either not noticeable or absent from other people's > calculators. Maybe this flaw is to be considered normal in calculators > produced using a certain type of LCD (perhaps provided by a > less-than-optimal third-party manufacturer of parts during a certain > period?). It does seem like your 49G's display is worse than mine but my > display looks worse than the pics reveal (at relatively high contrast levels > only). In any case, even if nondefective, it's poor quality. At least I'm > somewhat at ease that my calc doesn't have the *capacitor* problem... With > the shortage of replacement calcs and the *arguable* normalcy of these > flawed displays I'm debating whether asking for a warranty replacement would > get me far at all. If you are dissatisfied with the quality and feel you are getting less than > you expect, go for it. It's not like these are giveaway products! It is HP > we're talking about here (specially HP prices, if anything's left from the > original company.) Eduardo Now I'm confused. Visited an electronics store today and (for interest) took a look at the > one > calculator they actually had in stock: a 48GX. When adjusting the contrast high, the same thing - again - seemed to occur > with the LCD. > The 48GX produced the error far less visibly than my own (something more > like Eduardo's > pictures present). > Now - honestly - what are the chances that the single 48 in store; the > only > other LCD I have > physically looked at in a good while - would present the same problem? Please - I would like to hear some more thoughts on this. *IS* it normal? > I'm only going > to get a chance to drive to my HP dealer in a couple weeks since I'm now > one > week from > exams... so there is no other way for me to be sure. Please take a look at my images and Eduardo's and simply see if the same > thing presents > itself on your own calculator. also > alright? - > or is this kind of a random draw thing? If something -is- faulty, I would insist on a replacement - I am a student > and put a great > deal of money down for this calculator. Even the peace of mind that this > is > normal would > make me happy ... I just don't want to have to worry that 2 months after > my warranty > expires - I'm suddenly stuck with the LCD that decided to finally kick it > altogether. Opinions -much- appreciated! In the least, please let me know if your > display looks > ==== I wonder whether a new HP-49G is still worth buying if I see al the quality issues (e.g. Capacitor failure, bad screen, ...). I read in this newsgroup that the warranty is of no use. I also know that the keyboard is not as good as the keyboard of the HP-48. Could you please give me some advice? > ==== I used to have two 48GX and one 49G plus one TI 89 No I have one 48GX plus one 75C and one 49G plus another 49G Neither of the 49G are the new Chinese models, but the keys are OK NEVER try to compare the keys to other brand of calculator! The HP 48GX keys are much better, but there is no FlashRAM in the 48 That alone makes the 49G my primary calculator on the road. At home on numeric computing on the stack I use the 48GX The old 41CX serves as the back-up alarm clock. I would still get a 49G as the only calculator if that was the question. Maybe 48G+/GX is better for you?! The warranty works beautifully in Finland. Some capacitor failures (due to the supplier of those beauties) where immediately replaced. Where do you live? > I wonder whether a new HP-49G is still worth buying if I see al the quality > issues (e.g. Capacitor failure, bad screen, ...). I read in this newsgroup that the warranty is of no use. I also know that > the keyboard is not as good as the keyboard of the HP-48. Could you please give me some advice? > Arnold > Belgium > ==== > you can not see this lib with the filer because it's in ROM, and it is not a > real lib in the user lib sense of the term. Of course it will appear! all libraries with an ID > 256 will appear in the > LIBS menu. > The the 257 library doesn't have an empty name but MASD V5 With ROM 1.19-6 on my HP49G lib 257 appears in a list, with an empty string after calling LIBS. It does *not appear in the *filer. Hope this clearifies. best wishes Heiko ==== > Of course it will appear! all libraries with an ID > 256 will appear in the > LIBS menu. > The the 257 library doesn't have an empty name but MASD V5 Cyrille, I'm very disappointed! Well, actually, if you look at the source for the LIBS command, you will see (line 862, CCVR/covports.s): *_PrepLib A=0.A A=DAT0.X LC 100 ?A<=C.X -> .q1 % If rom lib, do nothing LC 7FF ?A=C.X -> .q1 % If not a lib, do nothing LC(3)788 ?A=C.X -> .q1 % AntiBugBonux for erable LC 700 ?A=C.X -> .q1 % command library so, any libs of rom id <=256, and different from 788 (Erable) and 700 (command library) will not be displayed. I wonder why the second erable library and MASD libs have never been added there... This is a bug... Cyrille ==== I wonder if you can help me. I was a happy owner of a Hp 49g till it die. It doesnÇt tur on or off. It doesnÇt cancel, ando sometimes it reboots alone. I am from argentina. I am trying to get Hp Argentina to replace my calculator for new one (or repair mine) but, with out paying a cent. The want me to pay u$s 130 for a rechange, they keep my calculator and they give me a new one. It is a lot of money for me. I canÇt afford it. Did Hp of america accept the fabrication errror? Did they change th calculators for new ones. I wonder i f you can help me sending me all the information you have about the problems of the hp 49g wich were made in china. I know of the capacitor, but i donÇt know if it is that. Please help me. Esteban Suarez Marenzi Argentina, Buenos Aires ==== Do you have installed a program in the flash of the hp49g because it can cause the calculators to reboot all the time ( you can't do anything it just reboot after 3 sec ) the problem can be solved by pressing the drop key when it reboot and delete the faulty program ( i had already this problem because i had installed hp48 program on the 49 ) I hope that i have been helpful Fran.8dois MULLER > I wonder if you can help me. I was a happy owner of a Hp 49g till it > die. It doesnÇt tur on or off. It doesnÇt cancel, ando sometimes it > reboots alone. > I am from argentina. I am trying to get Hp Argentina to replace my > calculator for new one (or repair mine) but, with out paying a cent. > The want me to pay u$s 130 for a rechange, they keep my calculator and > they give me a new one. > It is a lot of money for me. I canÇt afford it. > Did Hp of america accept the fabrication errror? Did they change th > calculators for new ones. > I wonder i f you can help me sending me all the information you have > about the problems of the hp 49g wich were made in china. I know of > the capacitor, but i donÇt know if it is that. > Please help me. Esteban Suarez Marenzi > Argentina, Buenos Aires ==== Arrhmmm! I surely could also be dead batteries, faulty library causing reboot, etc... > Give us all the contact information and we will surely ask HP > why the HP Argentina is acting like this. > I HAVE NEVER BEEN TOLD BY HP TO BUY > A FAULTY CALCULATOR AGAIN. > There must be a some sort of mistake. > Ofcourse you should have a replacement unit for free!!! > I wonder if you can help me. I was a happy owner of a Hp 49g till it > die. It doesnÇt tur on or off. It doesnÇt cancel, ando sometimes it > reboots alone. > I am from argentina. I am trying to get Hp Argentina to replace my > calculator for new one (or repair mine) but, with out paying a cent. > The want me to pay u$s 130 for a rechange, they keep my calculator and > they give me a new one. > It is a lot of money for me. I canÇt afford it. > Did Hp of america accept the fabrication errror? Did they change th > calculators for new ones. > I wonder i f you can help me sending me all the information you have > about the problems of the hp 49g wich were made in china. I know of > the capacitor, but i donÇt know if it is that. > Please help me. Esteban Suarez Marenzi > Argentina, Buenos Aires ==== Give us all the contact information and we will surely ask HP why the HP Argentina is acting like this. I HAVE NEVER BEEN TOLD BY HP TO BUY A FAULTY CALCULATOR AGAIN. There must be a some sort of mistake. Ofcourse you should have a replacement unit for free!!! > I wonder if you can help me. I was a happy owner of a Hp 49g till it > die. It doesnÇt tur on or off. It doesnÇt cancel, ando sometimes it > reboots alone. > I am from argentina. I am trying to get Hp Argentina to replace my > calculator for new one (or repair mine) but, with out paying a cent. > The want me to pay u$s 130 for a rechange, they keep my calculator and > they give me a new one. > It is a lot of money for me. I canÇt afford it. > Did Hp of america accept the fabrication errror? Did they change th > calculators for new ones. > I wonder i f you can help me sending me all the information you have > about the problems of the hp 49g wich were made in china. I know of > the capacitor, but i donÇt know if it is that. > Please help me. Esteban Suarez Marenzi > Argentina, Buenos Aires ==== Mybe what your problem is the common problem what affect the HPs.49G made in China (serial CN1..........). The better capacitor is of ... 470 microfarads and 6,3 v.d.c. See the next page ... http://www.geocities.com/deachp/duda8.html Miguel Angel CAPORALINI HERK (M.A.C.H.) **************************************************************************** **** > Arrhmmm! > I surely could also be dead batteries, faulty library causing reboot, etc... > Give us all the contact information and we will surely ask HP > why the HP Argentina is acting like this. > I HAVE NEVER BEEN TOLD BY HP TO BUY > A FAULTY CALCULATOR AGAIN. > There must be a some sort of mistake. > Ofcourse you should have a replacement unit for free!!! > I wonder if you can help me. I was a happy owner of a Hp 49g till it > die. It doesnÇt tur on or off. It doesnÇt cancel, ando sometimes it > reboots alone. > I am from argentina. I am trying to get Hp Argentina to replace my > calculator for new one (or repair mine) but, with out paying a cent. > The want me to pay u$s 130 for a rechange, they keep my calculator and > they give me a new one. > It is a lot of money for me. I canÇt afford it. > Did Hp of america accept the fabrication errror? Did they change th > calculators for new ones. > I wonder i f you can help me sending me all the information you have > about the problems of the hp 49g wich were made in china. I know of > the capacitor, but i donÇt know if it is that. > Please help me. A140 ==== hp49g 1.19-6 try this: 'INTVX(COS(X)^5xSIN(X))' EVAL TRIGC i don't get the right answer do you? the right answer is: '-(COS(X)^6)/6' the answer given is: '-((COS(X)^6-1)/6)' also i got on flag 116: prefer cos() but and i still got the result on sin(). ==== > hp49g > 1.19-6 > try this: 'INTVX(COS(X)^5xSIN(X))' > EVAL > TRIGC i don't get the right answer > do you? the right answer is: '-(COS(X)^6)/6' the answer given is: '-((COS(X)^6-1)/6)' also i got on flag 116: prefer cos() but and i still got the result on sin(). > Oh come on - remember about the constant of integration????? ==== <639cb51.0305021125.66dd0907@posting.google.com>, > hp49g > 1.19-6 > try this: 'INTVX(COS(X)^5xSIN(X))' > EVAL > TRIGC i don't get the right answer > do you? the right answer is: '-(COS(X)^6)/6' the answer given is: '-((COS(X)^6-1)/6)' Both answers are correct or both wrong because their difference is a constant! ==== it also happens with: INVTX(COS(7X)^5xSIN(7X)) and in all other similar expressions. the constant is not taken away like in the other integrals (like invtx(sin(x)) ==== now try this one and try to dig out the *constant* INVTX(SIN(7X)^5xCOS(7X)) good luck! ==== oops i got it: sin(7x)^5 and cos(*x*) [not 7X] OK time for me for a nap (JHM) ;) sorry for the lost bytes but it was fun, wan't it. peep ==== > now try this one and try to dig out the *constant* INVTX(SIN(7X)^5xCOS(7X)) good luck! Which constant? There are infinitely many possible CORRECT constants! If you want some *particular* constant, you must require some additional condition, such as having the antiderivative equal to zero at X=0. This condition, 0 at X=0, is probably the most common condition to used to pick a particular antiderivative in general. Note that the simplest antiderivative wrt X for 'COS(X)^5*SIN(X)' which is 0 at X=0 is the one given by the HP49G! ==== > This doesn't solve the problem. While painting, ie. permanently > using DOT+ or DOT-, your device amounts to permanently switching > + to slow down the cursor's speed. > PS. I would add a second coordinate viewer for painting, showing > the genuine pixel coordinates, i.e., 0,0 in the left upper > corner. > For an kind of graphic editing, you should use my GTOOLS libe that > comes with EDG apart from that I don't know what EDG is, I hate loading big libs for doing something occasionally only. Very seldom I've to correct a grob as in the case of Timeman which shows among other things a world map with a moving noon meridian. Correction is easy on the 48 but painful on the 49. You contradict yourself in some sense. If you still support the 49 why then should the 49 in some details be worse than the 48? What is the problem slowing down cursor speed in the PICT edit mode a little bit? And what would be the problem to add a graphic modus PAINT in which the (x,y) menu key shows the true pixel coordinates? As you know it is impossible to set PPAR with normal means to display these coordinates. Not only by big things but already in some very small details you could demonstrate the superiority of the 49 over the 48 :-) - Wolfgang ==== I'm not sure if this coincides with the rules for the forum so if it does not let me know. If anyone is interested I'm selling my HP to PC serial cable on ebay. http://cgi.ebay.ca/ws/eBayISAPI.dll?ViewItem&item=3022269725&category=20335