B429 === Subject: Re: 20th battery set La 50g se puede alimentar por USB por lo que dura bastante m.87s la bater.92a conect.87ndola al USB cuando la usas en casa. === Subject: Re: Perhaps a Bug Yeah, happens in my hp50g too === Subject: I can use rechargable batteries on my HP50G Hi everybody it's posible to use rechargable batteries on my HP50G... If you understand Spanish you can read this http://www.adictoshp.org/modules/newbb/viewtopic.php?topic_id=6594&forum=13 there are two positions between to use rechargable or AAA batteries alkalinas... === Subject: Re: I can use rechargable batteries on my HP50G Hi! I've been using NiMH as well as alkaline batteries in my 50g since I first got it (Sept '06) with the only difference that alkaline's give you more a little more use with the ((*)) annunciator on. Other than that, the amount of use seems to be the same. I tried both just as an experiment just to see if there was any dramatic difference between them and didn't notice any. === Subject: Re: I can use rechargable batteries on my HP50G I posed a question concerning the use of rechargable batteries in my HP- 50G to HP Total Care and received this reply dated 2 March 2007: We do not recommend using batteries other than 1.5 volt AAA batteries with your 50G, as it may cause the calculator to malfunction or suffer circuitry damage. Bad news for me, but I have decided NOT to use rechargables based on this recommendation from HP. Charles A. Huffer === Subject: Re: I can use rechargable batteries on my HP50G Load of CRAP! They just don't want to open themselves up to any legal issues if your rechargeables leak or anything. You are completely safe using rechargables in your calc. Only issue is that the power doesn't last quite as long. I have been using rechargables in my calcs and in all the data collectors we sell. Probably several hundred thousand hours of usage and there has never been one reported case of battery issues. TW === Subject: Re: I can use rechargable batteries on my HP50G What about using the calculator with the USB cord plugged in, in order to not use the batteries?? As I study near the PC, I always connect the USB cable to the calc to save my batteries. And remains connected for long periods of time. Could that represent some risk to the calc's circuitry? Is there any maximum time for a USB connection? === Subject: Re: I can use rechargable batteries on my HP50G That's wise. No. You can even REMOVE all *5* batteries while running off USB power (although that would sorely tempt Murphy to interfere with your USB power source!). No. It won't overheat or anything. It's a great feature, isn't it? -Joe- === Subject: Re: I can use rechargable batteries on my HP50G Now I feel safer using this method, because I heard some things in my U, but guess that is just some kind of weird envy from 49g+ owners haha :) You're right, using this battery saving trick is a very nice thing indeed! But, I wouldn't want to tempt old Mr. Murphy, so I don't take out my batteries while doing this. And although I know that's impossible, I kinda hope that it might charge my batteries a bit. While we are at it, I recommend the Batstatus prog for checking the battery life of our calc! Download at hpcalc.org === Subject: Re: I can use rechargable batteries on my HP50G http://www.gaak.org/user/percymc/archivos/batstatus20.lib ( there aren't a answer that said me why I can't use rechargable batteries on my HP50G exactly. bye === Subject: Re: I can use rechargable batteries on my HP50G Taking a guess (or helping to bail out HP :) It could be a problem if you could overcharge any batteries to the extent that the charged voltage exceeded the tolerable maximum. This doesn't seem likely with NiMH or NiCd, but what about alkalines and other new types -- oxyride? With an advanced substance for the cathode or negative (-) side, called oxy nickel hydroxide the batteries maintain higher voltage http://www.wisegeek.com/what-is-an-oxyride-battery.htm Did JKH suggest recharging these (even tho not sold as rechargeable)? Let the buyer beware http://industrial.panasonic.com/www-cgi/jvcr21pz.cgi?E+BA+3+AAC4007+4++WW http://industrial.panasonic.com/KM/instructions_E.html === Subject: Re: I can use rechargable batteries on my HP50G Not at all. YES you can use rechargeable batteries. They just won't last as long. TW === Subject: Re: I can use rechargable batteries on my HP50G HP: My past experience is that Alkalines are more likely to leak (Ray-O-Vac just happening to be the worst I've ever encountered, especially their old rechargeable alkalines, as it happens), and when those leak, it's very corrosive stuff. So is HP legally liable for having recommended alkaline batteries, if those leak? If so, how long is the Statute of Limitations? (for my HP48G[X] and HP49s with corroded batt. contacts, if not yet dead). Does anyone remember N size cells? (for HP28/19B/18C) There were some 12v batteries of exactly the same size; did many HP calcs get fried by inadvertent substitutes? === Subject: Re: I can use rechargable batteries on my HP50G AFAIR N size were HP41 batts === Subject: Re: I can use rechargable batteries on my HP50G They were also the size used in the 28 series (that opened up like books). They fitted nicely into the inside pocket of a sports coat or suit jacket, and had some graphing capabilities. === Subject: Re: I can use rechargable batteries on my HP50G Virgil grunted: Confirmed. I have a 28S here that is out of action merely because I can't find the batteries easily (I live in central .fr) and can't afford them when I do find them :( -- === Add [NANAE] to the Subject: of any correspondence or said correspondence will be deleted unread. === Subject: Re: I can use rechargable batteries on my HP50G Try http://www.batterywarehousedirect.com/m5_view_item.html?m5:item=138840 === Subject: Re: I can use rechargable batteries on my HP50G Places to purchase 'N' Cells: www.radioshack.com 1.5 Volt Alkalines, 2-pack $3.99. Model and Catalog # 23:023 Also, http://www.batteriesamerica.com/newpage3.htm Cat. # N-150N. Rechargables, 150 mAh Nicads $2 each. Many unusual size batteries for sale here. Charles A. Huffer === Subject: Re: I can use rechargable batteries on my HP50G Charles grunted: central France. radioshack.com and batteriesamerica.com aren't much use to me here :( -- === Add [NANAE] to the Subject: of any correspondence or said correspondence will be deleted unread. === Subject: Re: I can use rechargable batteries on my HP50G format=flowed; reply-type=original I found N batteries for my HP-28S in a shop in Italy without any troubles: 3? for 3 batteries (expensive). They were Duracell batteries manifactured in EU. On the web you can find NiMH elements, too. Look for N or LR1 cells. Massimo Santin === Subject: Re: I can use rechargable batteries on my HP50G I don't know. I just know how lawyers tend to look at things, and if they can introduce more and more loopholes that can be claimed to invalidate your warranty. . . TW === Subject: Re: I can use rechargable batteries on my HP50G When I have to change my batteries, about every two weeks, unfortunately, sometimes I have measured the voltage of the replaced batteries. It usually runs about 1.25 volts. The rechargables are rated at 1.2 volts. Does the bad battery indicator remain on continuously with rechargables? rechargables in the HP-50G, please post that information for all to see. Charles A. Huffer === Subject: Re: I can use rechargable batteries on my HP50G No. They go down quick, remain constant for quite a while, then dip down and the battery indicator turns on. Once it comes on, you have less time to change. I generally just run it until it powers itself off before replacing the batteries. I've never yet lost any memory. TW === Subject: Re: Multiple Undo Program / Command Logger I did some reading up on this topic and fiddled and fiddled and I now I think I have a very nice unlimited stack undo that seems to work flawlessly so far. For anybody who may be interested in using this themselves here is the code I ended up with. I would welcome any comments/suggestions regarding improvements. As I said, I'm but a novice grasshopper SysRPL programmer. :: DEPTH #0=?SEMI DROP AppMode? ?SEMI :: SuspendOK? NOT?SEMI :: ' xHALT EvalNoCK ROMPTR A3 10 ; ; ; That does it. Assign CONT to HIST button and KILL to UNDO. And everything works very nicely, both in and out of menus/apps, and the stack. Ryan === Subject: Re: Multiple Undo Program / Command Logger [about a multi-level UNDO, originally in old thread] [a program using the above] Excellent improved program! You don't need inner secondaries after ?SEMI, so you can shorten to: :: DEPTH #0=?SEMI DROP AppMode? ?SEMI SuspendOK? NOT?SEMI ' xHALT EvalNoCK ROMPTR A3 10 ; @ -92 SF ASM2 'GbENTER' STO Note: On HP48G[X/+] use ROMPTR A3 E instead. Or perhaps CONT to UNDO? (for original behavior of UNDO, except now multi-level when flags -62 and -63 are set) Nice work! === Subject: Re: Multiple Undo Program / Command Logger On Sat, 05 May 2007 18:43:13 -0500: [about a kind of push and pop for Last Stacks, allowing multi-level UNDO] Flags -62 and -63 must both be set for the following automatic programs to work. It appears that AppMode? ?SEMI can also be omitted, since HALT itself is concerned only with SuspendOK? HALT errors if suspending isn't OK, which is the problem that Ryan Edwards had encountered with the original version at the start of this thread (and subsequently solved). So the bare remaining essentials are (in SysRPL): :: DEPTH #0=?SEMI DROP SuspendOK? NOT?SEMI ' xHALT EvalNoCK ROMPTR A3 10 ; @ -92 SF ASM2 'GbENTER' STO @ Beta is Alpha RS B The first ?SEMI above protects against manual execution with an empty stack, the second ?SEMI protects against trying to HALT when HALT is not allowed; if allowed, it then halts (which creates a brand new Last Stack) ready to later resume and UNDO the original stack upon execution of CONT. If keyboard UNDO is attempted instead, it tries to undo the *new* Last Stack, which tends to have no effect when BetaENTER (or HALT) keeps starting a fresh new Last Stack each time. Note: On HP48G[X/+] use ROMPTR A3 E in place of ROMPTR A3 10 FWIW, you can have a different BetaENTER program in every directory, so you can make different directories for using different useful BetaENTER programs for different purposes (any directory not having its own BetaENTER program will look to higher directories to search for one). Whenever the following (for all HP48/49/50) performs an algebraic RPN function from the keyboard, it first leaves on the stack a list of the original args and function, then the result, so it resembles what ALG mode does, but for RPN: DUP TYPE 18 == { + SWAP } { DROP2 } IFTE To re-evaluate any list from above, you could use The following, with flag -3 clear, builds an algebraic expression using RPN operations; use DUP EVAL [or REDO] to get a final numeric result: @ HP50G/49G[+]/48Gii only @ (needs 256 ATTACH before entering or transferring) The next version extends the above by *automatically* showing the numeric result at every step: @ HP50G/49G[+]/48Gii only @ (needs 256 ATTACH before entering or transferring) Does the above resemble this TI RPN interface? http://www.paxm.org/symbulator/download/rpn.html The following resets flags and modes after any operation, thus undoing forced mode changes for CAS commands (though it can't restore deleted variables, if any); you should first set your preferred modes and do one PUSH (turning off USER mode while you do this), so that there will be something to POP the very first time: << PATH POP PUSH EVAL { -62 -63 -120 } SF Faster replacement for the combination of POP and PUSH: << { HOME CASDIR ENVSTACK } RCL DUP SIZE Replace SWAP EVAL in the above with NIP to avoid restoring the current directory, thus allowing POPU to replace PATH POP PUSH EVAL (restore flags only, allowing UPDIR and/or entering a subdirectory). Other uses of BetaENTER could be to automatically display level-1 (if numeric) as degrees+minutes+seconds, or in some other special format, or to automatically simplify any algebraic results, or to automatically do all sorts of other stuff (but can it slice bread?) === Subject: Re: Multiple Undo Program / Command Logger On Sun, 06 May 2007 12:18:32 -0500: To include CAS functions (and advanced 48G functions), change that to: OVER TYPE POS { + SWAP } { DROP2 } IFTE I should have used that to re-do the original post :) === Subject: Question Hi all, I have a simple question. When setting up a plot on the HP 50g with the F1 key; that is, press and hold the left-shift F1 keys, there is a menu selection CHOOSE (third label from the right). All it seems to do is beep and never lets me CHOOSE anything. Am I doing something wrong? (It wouldn't surprise me)... === Subject: Re: Question Create a couple of functions in the F1 key mode then clear them (use the NEXT key to see the CLEAR option). Then CHOOSE will show you a list of those functions which have been cleared from the EQ list, which you can append to that list by choosing them. Thus you can select which of the defined functions to graph, and the order in which they will be graphed, among other things. === Subject: Re: Question For some reason, I can't replicate this. It seems like when first loaded, the Y= screen contains the content of the EQ variable in the current directory. At this point, hitting the choose button displays a list of the functions that have been defined in that directory using the DEF key. I tried clearing the list,adding new equations to it using F2 (ADD), and then clearing it again using NXT F3 (CLEAR). After clearing the list, hitting CHOOSE doesn't do anything at all (unless there are already defined functions). So in short, from my experience CHOOSE only allows someone to add equations defined as programs in the current directory. By defined as program I mean: Is this what other people get? Berk -- === Subject: Re: Question Yep. Note that in the F1 graphing mode when you ADD a new equation in the IF graphic mode, it will disappear if you immediately CANCEL, but if you have graphed it, and then CANCEL, its name will be in the EQ list and it will be defined as a function in the current directory. With a little practice, you can select or define those functions you want to graph in any combination, and, once graphed, they will be in your current directory when you leave the graphing mode. And various other graphing choices can be found in the F2, F3 and F4 graphing modes, though for some graphing options you are better off accessing them through using 81 MENU and its dependent memus. === Subject: Re: Question === Subject: Re: Question I like them too. === Subject: Where to buy 50G in San Francisco I will be for one week in San Francisco downtown. As I saw the 50G is much cheaper in the US than in Germany. Does anybody know a good place to get a 50G there? Tom === Subject: Re: Where to buy 50G in San Francisco Go to Fry's Electronics. They probably carry it in stock. You'll like Fry's. http://sanfrancisco.citysearch.com/profile/924336/palo_alto_ca/fry_s_electro nics.html Scott Guth === Subject: Re: Where to buy 50G in San Francisco Yeah, Fry's has them in stock, but it's a bit far from downtown SF, which is why I didn't mention them. === Subject: Re: Where to buy 50G in San Francisco Circuit City sells it. http://www.circuitcity.com/ccd/productDetail.do?oid=173592 http://www.circuitcity.com/rpsm/storeid/242/zip/94109/oid/173592/rpem/ccd/ge tStoreDetails.do === Subject: Poll for other languages: Your HP in Spanish, French, Italian, Portuguese ... As some of you might know the 49G is capable (to some degree) of three languages. I do not believe that the built in French and Spanish messages will be extended in the future because the ROM is almost full and there are still errors in the ROM which should be fixed first by HP. Also HP doesn't seem to have any interests to make localized version of the 49G available. The system RPL word SETMSG allows exchanging of built-in messages arrays but is restricted to non covered memory. Furthermore the infamous STO bug resets the built-in messages if something is stored in a port. However, I have developed an engine for allowing translation of all built-in messages arrays (around 2100 to 2200). The engine is written in ML (around 600 Bytes). For acceleration it is loaded to a protected memory area at start and circumvents the STO bug. The messages can reside in any port. Once loaded, exchanging the message is fast, it slows down each key-press at around 0,2 milliseconds and has gone through extensive testing and proven to be stable. Now, since it doesn't looks like HP is going to release complete language packs (not to mention to move the messages which are still hard coded in some built-in application to messages arrays) and since I already have enough trouble with my mother tongue and do not speak Spanish, French, Italian, Portuguese, I'd like to start a poll if you - would like to have a native speaking version of your 49G/49G+/50G (assuming that you're not a native English speaker)? - if you would like to have one, how much are you willing to pay for it? Developing the engine and translating all messages takes a lot of time and effort. - willing to help to translate the message to your preferred language (of course you'll get a copy of it ;-). A template source file for this is available on request. If you want to translate the built-in messages please keep the length of the translation in mind, if the translated text is to long for a message box it will be truncated, but a message longer than 32 characters will crash the calc if used as a help string for the new IfMain2 engine, for example. Also the message might be to long for a field in an InformBox. So some creativity for translation is needed and also the use of some abbreviations... An example of a 49G running completely in German can be found on my website www.software49g.gmxhome.de. Andreas === Subject: Re: Poll for other languages: Your HP in Spanish, French, Italian, Portuguese ... Andreas M.9aller a .8ecrit : Personnally I will always stick to the original english language. Herve, Paris, France === Subject: Re: Poll for other languages: Your HP in Spanish, French, Italian, Portuguese ... I'd like to help you translating messages in Italian or, at least, giving it a try! ;) Ciao, Stefano === Subject: Re: Poll for other languages: Your HP in Spanish, French, Italian, Portuguese ... My mother language is spanish, I dont have any issue using the calculator in english (I hate the way the Texas calculators supports multilanguages changing functions and commands, for example: Integrate is Integrar in spanish) but I can help translating the resources of the HP. === Subject: New version of nonpareil for Mac-OSX available Just in case anyone interested missed it: http://versiontracker.com/dyn/moreinfo/macosx/28330 ____________________________________________________________________________ __ Dr Chris McDonald E: chris@csse.uwa.edu.au Computer Science & Software Engineering W: http://www.csse.uwa.edu.au/~chris The University of Western Australia, M002 T: +618 6488 2533 Crawley, Western Australia, 6009 F: +618 6488 1089 === T-Online does not support Usenet spammers === Subject: Re: 50g borderline dumb questions Can somebody please explain to me what the syntax for the above two lines is. I can't quite understand what the 'R|^' is supposed to be. How do you type it in? I found the ^ symbol in the CHARS menu, but I couldn't find the backslash anywhere. Is the '|' the same symbol found on top of the TOOL button? Also why are there have so far ignored them. Any explanation would be much appreciated. Berk Birand -- === Subject: Re: 50g borderline dumb questions Actually you didn't ignore them; if you ignored them, you would have typed into your calculator two consecutive but most likely you mentally translated each of those *pairs* of posted characters into *one* program delimiter character, which doesn't appear on anyone's computer keyboard. So the backslash means that the next *two* characters (from a set found on most computer keyboards) represent *one* byte in the calculator (usually a character not found on most computer keyboards); of course, there's also an alternative form nnn (decimal character number, always three digits). The calculator hasn't quite as good a mental capability as a person, so a precise and unambiguous system is used, enabling the calculator itself to translate the internal complete 8-bit character set (almost identical to ISO-8859-1) using a 7-bit universal computer/email/BBS/newsgroup ascii set, as summarized by the various posted charts, which were originally included with all HP48 series calculator manuals (such as the one you can download as a PDF from www.hpcalc.org, in which will be found a lot of knowledge that was entirely dropped from HP49 series documentation, even though the HP49 series calculators still include all the original capabilities, even if undocumented for new purchasers of HP50G/49G[+]/48Gii). Although many posts rely on human judgement to interpret, in a form that no calculator could digest by itself, everything I post is in calculator-digestible form, using translate mode 3 (leaving no untranslated 8-bit characters), in part because I always either test it by actually downloading to an emulator, or have exported it directly from an emulator, posted, eliminating all chance of human error (except for mine :) http://en.wikipedia.org/wiki/ISO_8859-1 === Subject: Re: 50g borderline dumb questions Here it is: The calculator itself can automatically translate these when files are sent via cable between calc and computer. It's also possible to translate automatically on emulators or from SD cards: === Subject: Re: 50g borderline dumb questions The backslash is used to signal that what follows is an ACSCII representation of a character does not appears in ACSII In this case the |v is , I believe the code for the down arrow character, numbered 143 in the CHARS menu. Somewhere, probably at www.hpcalc.org, there is a listing of them. === Subject: Re: 50g borderline dumb questions ASCII values had different meanings in HPs). Berk -- === Subject: Re: 50g borderline dumb questions http://holyjoe.net/hp/tiotable.png -Joe- === Subject: Re: 50g borderline dumb questions This is overkill, but it's certainly clearer than that ancient photocopy: http://holyjoe/hp/tiotable.htm -Joe- === Subject: Re: 50g borderline dumb questions Hi Joe. Seems like the link is broken.... May you check please? TIA. Giancarlo === Subject: Re: 50g borderline dumb questions He forgot to type the '.net' part of his domain name. The corrected URL is: http://holyjoe.net/hp/tiotable.htm come to expect! -- Dave Boyd That's sucker talk. -- Raven, _This_Gun_For_Hire_, Universal, 1942 === Subject: Re: Spherical Triangle Hans, Nothing wrong with being stubborn, I have been accused of this myself, but consider this: In any plane or spherical triangle, the greater side is ALWAYS across from the greater angle and vice versa. In Case 2, a = 120 Deg., A = 55 Deg., and B = 150 Deg. are fixed values. Since angle B is greater than angle A, side b MUST be to be a possible solution because it is NOT greater than a = 120 Deg. Charles A. Huffer === Subject: Re: Spherical Triangle Charles, you are right. Being stubborn, I am also sometimes too quick to draw conclusions (these properties might go hand in hand). The angle B of my second solution is 30 Deg, not 150 as it should. I should have seen this, if not before, in the vector analysis, which is essentially correct ( except for that values for vB and vC should be swapped for the 2nd solution). I became aware of my mistake some time back. Found the need to improve the method. Here are six examples which may serve as test cases (for others and myself). They pass the Gauss analogies. Sides lowercase. All in decimal degrees. Given Solutions 1st 2nd a = 70 c = 19,163875 96,040746 b = 60 A = 117,273153 62,726847 B = 55 C = 18,089581 109,843528 Given Solutions 1st 2nd a = 70 c = 123,682742 351,521879 b = 75 A = 52,835561 127,164439 B = 55 C = 135,115556 352,817553 Given No solutions a = 70 b = 50 B = 55 Given Solutions 1st 2nd a = 70 b = 83,445404 96,554596 A = 55 c = 132,619845 155,275377 B = 60 C = 140,09858 158,616735 Given Solutions 1st 2nd a = 70 b = 41,145987 138,854013 A = 55 c = 92,66203 219,424363 B = 35 C = 119,450279 213,614053 Given No solutions a = 70 A = 55 B = 65 Hans M === Subject: Re: Spherical Triangle Hans, One other thing. The original Case 2 problem a = 120 Deg. A = 55 Deg. B = 150 Deg. These are fixed and not part of a calculated solution. But making a new problem with a = 120 A = 55 B = 30 there is only one standard solution, b = 31 Deg. 54' 42'' c = 143 Deg. 20' 45'' C = 145 Deg. 37' 14'' which is your Case 2, 2nd solution. Also, Peanuts was the Unwort des Jahres Sorry about the error. Charles A. Huffer === Subject: Re: Spherical Triangle Hans, Seems your new method is working just fine! Just a few small differences. As a German banker, who managed to coin the 'Word of the Year' several years ago said, just 'Peanuts'. Solutions having values over 180 degrees are non-standard and never get calculated in any of my programs for Spherical Triangles. The 4th example, 1st solution, b = 83,445 404, I get ...401 c = 132,619 845, I get ... 841 The 4th example, 2nd solution, b = 96,554 596, I get...599 c = 155,275 377, I get... 382 C = 158,616 735, I get... 739 The 5th example, 1st solution, C = 119,450 279, I get...278 Everything else if fine. Perhaps Scott could run these on the 50 digit machine to see what that produces? Charles A. Huffer === Subject: sysrpl: serial receive routine help please I'm trying to get a routine to send and receive modbus messages via the serial port on the hp50g. The comm parameters are 9600 n 8 1, and I've set the translate code to 0. I have the transmit routine working ok, and can verify that remote data is being returned, but I am not getting serial data correctly. Returned messages are garbage after about the third or fourth byte, and the overall message is incomplete. My attempts have used variations of this routine: * XMIT packet to the ModBus Device xXMIT ( Send function call packet ) * Receive the response packet from the ModBus Device % 1 xWAIT ( Wait for response message ) DOBUFLEN ( Number of received chars in response message ) DROP ( Result Code ) DOSRECV ( Get Message of BUFLEN chars from buffer ) DROP ( Drop success code ) Can anyone help me get this to work? -Dot- === Subject: Re: sysrpl: serial receive routine help please Those are correct. It sounds more like you are having a setting issue (parity or something). TW === Subject: Re: sysrpl: serial receive routine help please Dot, I found over the weekend that some of the first cables did have a problem at low baud rates. Only on Rx data, as you describe. If you contact Sampson, you will get a new one. We will notify all, but since you are having a problem, I'll make sure you get a new one first. Sorry, Jim D. === Subject: Re: sysrpl: serial receive routine help please Jim D, the problem I seem to have and hopefully they can make arrangements that will help me get past this problem. -Dot- === Subject: Second TI-Nspire report I have published a second report on the TI-Nspire (with a few comparisons with the hp49g+ as to strengths and weaknesses of algorithms), focusing on students' and parents' observations and feedback to working with the handheld calculator. The report is available (in German) at http://www.noemanetz.de/folgeseiten/artikel/CIMS-SH_Nspire.html Michael Kuyumcu === Subject: Re: Second TI-Nspire report Fairly interesting report from what I could understand, it would be nice to have it translated in English! On your CAS algorithms remarks, I can give you a few hypothesis * there is most probably nobody at TI working on improving the maths algorithms. My guess is that they have concluded the CAS is good enough for the nspire, and they are most probably right. Competition will not come until PC CAS will be widely available on handhelds. The math soft of the nspire is therefore most probably exactly (with a few improvements in the non math areas). * Points 1, 2, 3 of your Qualitat der Algorithm are not math-related improvements. * point 4 is because the size of an integer in base 2^8 is coded on 8 bits: try 256*log(256.) * point 5 would require something like MPFR on PC * point 6 is because they use the Taylor formula instead of arithmetic operations on expansion. That's what I find the main weakness of the TI, but it's not apparent before student are at the University, hence TI does not care (same for arithmetic functions on polynomial or Bezout identity for integers). * point 10 is because it's hard to know exactly if a combination of square roots is not 0. If you compute with sqrt of 2,3,5,6,7 and -1, you must work in an extension of degree 2^5=32 and most importantly you must check at each step that it does not reduce further. As for the comparison with the 49g, the CAS software is not 15 years old, 10 (or even less) would be more correct. === Subject: Re: Second TI-Nspire report Hi Mr Parisse, Unfortunately it won't be anytime soon. The other problem is that handheld need to become almost as cheap as calculator. Though since symbian and windows mobile have become the dominant O.S of this market,the price of handheld won't significantly decrease anytime soon. They will just get more and more powerful. The 32 KB limit is indirectly math related. Especially with algorithm as poor as the taylor one. This limit should have been removed since at least the TI92+ and the TI89.Though the conversion of bigger integer to string(for display) would be awfully slow unless strongly optimised with low level C or even assembly. Or just use this capability from Derive. This is in fact a weakness of Derive not specifically of the A.M.S. I have been quite disapointed when i find it out. It is distrurbing to see a decades year old C.A.S having such limitations. Well considering that the determinant of this matrix can be easily compute with the formula: m[1,1]*m[2,2]-m[2,1]*m[1,2],assuming m is the matrix i think like Hassan that the problem come from the computation of the determinant. However note that the C.A.S also has trouble with other operations(REF,RREF) on this matrix. i have chose the square to avoid any easy simplication from the calculator and to clearly underline the trouble it has with such matrix.Note that i have been kind,i could have used symbolic values instead of number. What do you think for example of this: [1/(sqrt(n01)+i*sqrt(n02)),1/(sqrt(n03)+i*sqrt(n04));1/ (sqrt(n05)+i*sqrt(n06)),1/(sqrt(n07)+i*sqrt(n08))] === Subject: Re: Second TI-Nspire report parisse@domain.invalid a ?crit : focus exclusively on Derive code. When you think about it,it would have been smarter to use Derive as the CAS base from the start. But the first point as an impact on math performance as it can strongly reduce the problem solving capability of the calculator and significantly slow down some computations. If they had used Derive code from the start,they wouldn't have this limitation anymore. i sincerily don't understand them. The NSpire will be incompatible with the TI89T/V200 and will be be available on both the handheld and the PC.It would have been simpler and smarter to start from the Derive code. Well,i thought that the TI89T/V200 also target colleges(thought certainly not engineering schools and advanced math/sciences college classes). At least i bought a TI92 when i was in college and i should thank Jean Michel Ferrard for its excellent book related to the TI92(i don't remember the title) which helped me significantly increase the math capabilities of the calculator and thus make it much more useful for my studies. Especially if the CAS like to use autosimplification for everything. Anyway,in this case the C.A.S use inappropriate algorithms for small matrix. The C.A.S seems to use the comatrix divided by the determinant formula to compute the inverse of a matrix thus i think that in this case the problem come from the computation of the determinant. And the C.A.S is incredibly slow to compute the determinant of this matrix. in function for such cases and which used the exact formula for 2*2 and 3*3 matrix and reducing the matrix to a pseudo diagonal matrix for bigger matrix. Well,it was available 8 years ago,so it is even younger from a customer point of view ;-) === Subject: Re: Second TI-Nspire report to do and can't come up with a translation, though I agree It would be 1) I think that if you can't use more than 614 digits for integers, this very well constitues a weakness of the underlying software architecture, and in turn of the algorithms. 614 digits may be enough for most school purposes, but if a student wants to experiment further, s/he will run into a wall pretty soon, and this can well be attributed to algorithm design (as the algorithm is tightly interrelated with the underlying data structures like integer numbers). On the hp, you have really long integers, up to 524288 digits, if I am correct, at least in theory, and equally long floating point approximations to real numbers (as an in-built data type!). And all of this with a pretty old software! 6) Yes, this was Konan Yao's idea, too. I think it will become apparent already in grade 13 (which we teach at German Gymnasiums when we will be dealing with Taylor developments. 10) I don't agree. Mathematica doesn't need a single second to print the correct answer. It must be a matter of (non-existent!) algorithmic quality. Michael Kuyumcu === Subject: Re: Second TI-Nspire report It might be because Mathematica does not check that the det of the matrix is non 0. How is the inverse returned by mathematica? Is it relatively compact or huge and without sqrt in the denominator of the matrix elements? How much time do you need to check that A*inv(A) is indeed identity? I have done the experience with Xcas (fixing a few bugs on the fly), and it requires intensive computation because you have first to find the minimal algebraic extension of Q containing all the sqrt, then each computation will be translated on polynomials of degree 16 with complex rational coefficients (or degree 32 with rational coefficients). After that you must translate back the polynomial as a symbolic expression in terms of the generator of the extension (e.g. sqrt(2)+sqrt(3)+sqrt(5)+sqrt(7)) and depending how you rewrite it, it can be time costly as well. I would guess that computing the inverse on an arm 70 Mhz should take around 10 minutes with the current Xcas algorithm. On the 49, I used a recursive representation for sqrt, but it works only for sqrt, not for more general algebraic extensions. Or maybe mathematica has a better algorithm, if that is the case, I'm curious to know how it works... === Subject: Re: Second TI-Nspire report In fact, I just read that Mathematica assumes a non-zero determinant. To check whether A*inv(A)=id took around 1 second, maybe 1.5 Michael Kuyumcu === Subject: Re: Second TI-Nspire report Mathematica gives the following symbolic answer including sqrts: http://www.noemanetz.de/bilder/Nspire/matrixform_inverse.gif Michael Kuyumcu === Subject: Re: Second TI-Nspire report It is indeed not simplified. Checking that A*inv(A)=id is probably not a good way to test the sqrt simplifications algorithms of mathematica, since using their inverse this can be checked using only rational simplification (i.e. replace sqrt by unknowns, without using sqrt(x)^2=x). It would require some identities involving sqrt. === Subject: What is an HP 48GS? At Calcpro's website (www.calcpro.com), you will find the following description in the Calculator - Graphing section (under HP 48GS & 48GII); just click on More Info: Science and math like never before. Like the HP 48G+ and then the 48GII, the new 48GS has a generous memory of 128K for greater expansion with a faster processor. Both; the 48GS & 48GII comes with; user's manual, serial cable, batteries, leather case and CD containing the serial interface software and an 864 page user's guide. It lists the price for a 48GII at $94.95 and the 48GS at $99.95. So just what is an HP 48GS anyway? And would you pay the extra $5 ;-) mk === Subject: Re: What is an HP 48GS? What is especially curious; is that there is no mention of this calculator on the www.hp.com site...??? === Subject: Re: What is an HP 48GS? Not really. Remember how the 50g was arriving at people's doors months before it ever showed up on the webpage? It is almost a distibutor that spills the beans first. TW === Subject: Re: What is an HP 48GS? First I've seen of it. Assuming this is correct, I would guess it is 50g magical transformation) Then you'll be sure it has a good keyboard. That would follow recent history quite well. TW === Subject: Re: What is an HP 48GS? Yes, and an '48gs' still has not much to do with a real HP-48G series calc. Why don't they simply call the '48gII' or even the '48GS' by their real name: 49g-/49gs... Oh, I forgot: The number 49 is burnt due to some, errr, misunderstandings;-) Slightly OT: Reminds me of the 'hp invent' initiative, where they pretended to go 'back to the roots'. which somehow implied quality, until it came clear it was only a marketing gag... Now they seem to abuse the '48 series' label again. And the naming is a dumb decision IMHO. Consider the following: Someone downloads a software package from somewhere, where the readme states: Runs on all HP-48 G series calcs. That person will be at least very astonished if he realizes that the '48gII' and '48gs' are something completely different, and not 100 percent compatible even at the user interface level. Even more OT: Just bought another HP-48GX with B/W display, nearly reaching my lifetime supply:-) === Subject: Re: What is an HP 48GS? and flash with a USB port JY === Subject: Re: What is an HP 48GS? I'd like to see the complete spec's on this, but I think this is a very good move in hitting the student market. I've been hesitant to recommend the 48GII to students because inability to upgrade the ROM. CAS, exact/approx modes, Alg/RPN modes, flash rom, mostly (?) compatible with 49g+/50g, and for well under the cost of a TI-84+. If so, it's going on my recommendation list as soon as it's widely available. -wes === Subject: Re: What is an HP 48GS? Jean-Yves Avenard a ?crit : Well HP is looking more and more like TI in a close past. === Subject: Re: What is an HP 48GS? Nice observation, yes... -but then TI looks more and more like :-) Sorry for bad language -it came all by itself :-) manjo === Subject: Re: What is an HP 48GS? I don't think so they just are more and more focused toward education and especially in high school. Assuming the overtaking of laptops in colleges and engineering schools they haven't much choice left. However their strategy with the NSpire could really pay off in the future. If they can manage to make the NSpire powerful/features rich enough they can insure their future in the education market for numerous years. Something HP can't. When you think of it,it is frightening how much more advanced the NSpire is compared to competitor products such as the HP39G and the Casio fx-9860 G. === Subject: Effects of number formats on numerical integration I was doing some research on the different number formats in order to ultimately find a setting that I'm comfortable with. As I was browsing happened to be quite ancient (the poster who was in high-school then probably has kids by now). The problem was that of calculating the definite integral of a function that looks like 1 - 9/x^(2/3) Between the bounds 0 and 2. In order to see how the number formats would be used, I typed this in in the equation editor (including the integration bounds that I input using R-Shift TAN). I first set the number format to SCI 1, and obtained a result of -2.8E1 after a few seconds of calculations. I then set it to SCI 2, and did the exact same calculations. This time, I got a -3.16E1. I was gonna try SCI 3, but the calculations took a lot longer than I could wait. Now the question is, how can this number format have such a big effect on the outcome of this integration? Am I using the wrong function for doing this? How can one know which result is right (I'm guessing/hoping that the SCI 2 one is more correct)? What does the argument to the number format do, since it clearly doesn't correspond to the number of significant digits? Any explanation on this issue would be very much appreciated. Berk PS: The post that I mentioned is here -- === Subject: Re: Effects of number formats on numerical integration For numeric integration, it's probably intuitively clear why integrands which are infinite at the endpoints (or anywhere) need more iterations, since most of the final value is contributed near these unbounded locales. It's like adding up the grains of wheat in this tale: http://www.mainlesson.com/display.php?author=fabre&book=science&story=dervis h http://ddrive.cs.dal.ca:9999/page/lvl3/8 http://ccat.sas.upenn.edu/~bvon/pages/chess.htm As you go square by square in the above, adding up the total amount of wheat thus far, the quantity nearly doubles at each step, not drastically unlike what happens as further initial integrand samples probe nearer and nearer to the unbounded values of such a function (fortunately the convergence of the Romberg method to a finite limit tends to go just a little faster than this :) === Subject: Re: Effects of number formats on numerical integration This is described in the manual. SCI set the accuracy. SCI 1, means 10^-1 accuracy SCI 2, 10^-2 etc.. For the most accurate answer do 12 SCI or STD in STD mode calculations will be made up to 12 digits accuracy. JY === Subject: Re: Effects of number formats on numerical integration I couldn't find anything in the manual pertaining to the use of the number formats when doing integration. I know what SCI is supposed to do as far as rounding up and using the right mantissa goes, but it seems to have a different meaning with integration (or I just fail to see how it is applied). How can the same integration produce -2.8E1 with SCI 1, and -3.16E1 with SCI 2. It doesn't look like these two numbers agree with the 10^-1 and 10^-2 accuracy. As a side note, when I use the Exact mode to perform the same integration, I get the exact solution within a few seconds (versus the tens of seconds that the SCI 2 takes in approx mode). What is done differently when the approx mode is used? Wouldn't it be smarter to get the approx mode to Berk -- === Subject: Re: Effects of number formats on numerical integration The Romberg method of numeric integration is a sampling method, which makes successive passes through the interval of integration, each time evaluating the integrand at points which are between the points previously used (or closer to the endpoint at the two extremes). The points are selected by applying a third-degree polynomial to what would otherwise be interval halving, causing the sampling points to be more bunched up near the endpoints and most sparse in the middle of the interval, which is good for functions whose complete integral is more determined by regions near the endpoints, and not as good for functions which are ill-behaved in their interior, instead of at their endpoints. For any given function, the computed value of the integral will depend on how many passes are made, and this in turn depends upon when the algorithm decides to stop; key factors, then, are (a) how it decides when to stop, and (b) how much unpredictable variation occurs with each successive pass, and between those passes and the unknown true answer. The algorithm knows nothing more than the value of the integrand at those points where it was evaluated, and by using a set of weights for each pass, based on a certain ideal assumptions, such as polynomial fitting, which rarely agrees with the actual function's behavior. Deciding when to stop is based on making an estimate of probable error bound, using the current number display mode as an indication of the uncertainty in each computed function value; the final value of this estimate is stored in variable IERR; the original specs (when first introduced for the HP34C) said that when the separate estimates of the integral for three successive passes were all within the error bound from each other, then that was when to stop. I don't recall how much the 48G series user guide went into this, but back when there was an Advanced Functions Handbook for the HP15C, there were very extensive chapters on both the numeric Solve and numeric Integrate functions, illustrating each matter in great depth. The original error bound for fixed number display mode was also defined differently than now, to be an absolute constant, rather than a fraction of the function value, but it's conceivable that this might even have been abandoned solely because its meaning was not fully grokked by users, and may have caused too much misunderstanding. In attempt to condense it all into one pithy sentence, it's like the Three Blind Men and the Elephant fable, where, depending upon exactly which part(s) of the elephant are sampled independently by each observer, they all get entirely different impressions of what they think the entire elephant is actually like :) http://www.wordinfo.info/words/index/info/view unit/1/?letter=B&spage=3 http://www.kheper.net/topics/blind men and elephant/ http://en.wikipedia.org/wiki/Blind Men and an Elephant If the function can be integrated *symbolically*, then yes, you'd be able to get an *exact* answer in that case, whereas merely taking samples and making a numeric *estimate* from those samples produces as exact an answer only in relatively simple cases (such as for low-degree polynomials). As to comparing the numeric results for different number display accuracies, in most cases, just like in imaging, the more points (e.g. pixels) are sampled the more accurate an overall estimate emerges, and more samples are made for higher-precision settings of the current number display mode. It is assumed that the evaluated function is correct to the number of digits specified, and that's what's used to estimate an error bound for the summation of all those individual errors combined. Look at the value in IERR to get an estimated error bound for the entire integral. In a method using only sampling, with a limited number of samples, the final true error depends on how well the method suits the integrand function which you are using; it's often possible to transform badly chosen functions to make a much better choice for this method, by a change of variable, but this is left entirely to the intelligent user, because this method is a purely numeric algorithm, utterly divorced from anything symbolic (it made its first appearance in the HP34C, circa 1979). The number format specifies your confidence in the accuracy of the value of the function at each point; this only indirectly determines how many iterations the algorithm performs before it stops, and then the final *estimate* of error bound is stored into IERR. This makes a lot of sense when applied to formulas which approximate real-world facts, such as drag on a glider as a function of airspeed; when it comes to evaluating theoretical integrals which you take to be of infinitely precise functions, you must simply be aware that you are using a tool which isn't made for the purpose, and which can not guarantee any particular accuracy you might want, because it's only essentially a *sampling* method, not an algebraically exact method -- it's only the latter, involving deterministic computation, for which a number display format can be relied upon to show the answer to exactly as many digits as you asked for, give or take some rounding and other numerical errors which can creep in anyway. By now it's clear that it's an issue of exact *symbolic* integration vs. *numeric* integration by the *sampling* of many function values (with a total of 2^N-1 samples after N Romberg iterations). It's very much like the story about Carl Friedrich Gauss, whose teacher purportedly assigned his class some dull and boring busy work as an exercise -- to add up all the integers from 1 through 100. It's claimed that most of the class did it the hard way (by actually adding all 100 numbers), while Gauss perceived how to do it immediately, by using only the endpoints of the range, as it were: http://library.thinkquest.org/4116/History/gauss.htm http://www.jimloy.com/algebra/gauss.htm Yes, it would be smarter for the *user* to try exact mode first. This calculator does not try to be Mathematica; in this environment, for the integration function, Approximate mode *implies* numeric integration, and exact mode implies symbolic integration, and it's left to the user to decide which to use -- in fact, the educational comparison of the two modes wouldn't be possible unless the calculator let the user make the choice, and the entire target market of these calcs is education, isn't it? ;-) === Subject: Re: Effects of number formats on numerical integration On Thu, 10 May 2007 15:59:44 -0500: And the value returned in IERR (estimate of error bound) correspondingly decreases, to suggest the likely smaller uncertainty in the numeric result. The HP34C and HP15C had some bug, in which the returned integral values could differ, even for the same number of iterations (at exactly the same set of points sampled) which might be made for some different number display mode settings (only the error estimate should differ in these cases, but not the integral value); it was probably not merely coincidental that those were the very same cases in which the error estimate tended to be a bit too low (or to put it differently, a wrong answer was returned; not *too* wrong, but wrong enough to spoil potentially better results). AFAIK this was all fixed in HP48/49/50. In HP34C and HP15C, a current estimate of the integral (as of the last completed iteration) was always available in LastX, every time that the integrand function was called; this feature doesn't seem to be formally available in HP48/49/50, HP34C/15C also lacked multiple integral ability (only one block of memory locations was available for intermediate storage). When the integrand is a formula for which the CAS can find an anti-derivative, this whole thing sure does get simplified, as with other things in life, once deeper insight and experience is gained. === Subject: Re: Effects of number formats on numerical integration Simply that in this example, when adding all intermediary numbers you add too much inaccuracies with 10^-1 (SCI 1) Two completely different algorithms. What you suggest may work well in some cases, but not in others whereas the Taylor algorithm would. === Subject: Q:Editing on PC and transfer to HP49g Hi. Is it possible to edit text on PC and transfer to HP49g. Can I use HP4XDEV2??? If not, which program is suitable for that. Edin === Subject: Re: Q:Editing on PC and transfer to HP49g Also, you can use the Conn4x program. Double-click on the object, and it'll appear in a Notepad window. You can edit it, Save, then Exit. On Exit, it'll do the actual save to the calculator. The problem I have is that the arrows don't translate in Notepad. You just see the little box. However, you can cut and paste the boxes. === Subject: Re: Q:Editing on PC and transfer to HP49g Hi. I found WinHP program for PC, in which I can edit text and then transfer to HP49g. Edin === Subject: Re: Q:Editing on PC and transfer to HP49g Hi Has. I'd use a plain text editor (Notepad) on the PC, copy the files onto the SD card and then use this little marvel from John H. Meyers': Hope this helps. Giancarlo === Subject: Serial Interface: GPS OK, Question I got the 50G to receive position and time data from a GPS receiver (eTrex Legend). I want to add some more features before posting it to hpcalc.org. In the meantime, does anyone have any ideas about toggling (easily) between Serial and USB mode? === Subject: Re: Serial Interface: GPS OK, Question flag -78 determines whether it uses USB or the serial port is that what you need ? manjo === Subject: Re: Serial Interface: GPS OK, Question Do you think your program will work on a Etrek Vista? Harold A Climer University of Tennessee at Chattanooga Dept of Physics,Geology,& Astronomy Room 309 Grote Hall 615 McCallie Ave Chattanooga TN 37403 Harold-Climer@utc.edu === Subject: Urroz books i am back again after an absence of nearly 2 years. And now my question: www.greatunpublished.com. Now, 4 years later and with a new computer I tried to open them. (unfortunately my printout has been lost during a move) But my acrobat reader 7.0.8 could not read them. (WIN XP SP2) But i still have the .exe-files. :-)) They can be started. A plug-in is loaded. .... But the pdf-files can not be read. If I try to open the pdf with acrobat reader is closed. (acrobat shows the same behaviour) I believe to remember that something had to be changed in the configuration of acrobat reader but i did forget. Are there any suggestions? greetings Stefan === Subject: Re: Urroz books I remember that a code came with them. You have to unzip them again and enter the code. The copying the PDF files won't work. You have to go back to the original archive and expand it again. === Subject: Re: Urroz books Have you tried Acrobat Reader 8? === Subject: Re: Urroz books Welcome back :) Try opening them with a different reader. I've used Foxit for years with no problems. http://www.foxitsoftware.com/pdf/rd_intro.php === Subject: How to execute downloaded HP49G programs I've been searching for quite a long time how to connect the HP49g to the computer but the usb connection seems to be definately buggy : ( (Sometimes the connectivity kit allow me to connect then leaves me when i try to do something saying 'transmission problem'). So I decided to try to put the programs I downloaded from http://www.hpcalc.org into the sd card then read them from the calculator, which works. But I have no idea how to run these programs. Some contains several files (gif and .bin), some contain 1 file with no extension, some contain 1 .hp file and some others with .lib extension. I'm really confused, can someone help me ? === Subject: Re: How to execute downloaded HP49G programs On Thu, 10 May 2007 09:37:25 -0500: Only calculator objects should be transferred; attempting transfer of other computer files and/or documentation is likely to cause some error, so if the package comes with any explanation of what the files are, it is likely to say exactly which files to transfer, and often even how to install them. These calculators can't use .gif images, and probably not some other of the files which you may have been trying to transfer, which might have caused the USB transfer errors. If the package does not say what the files are, you could open each file using WritePad (by dragging it to a shortcut for WritePad which you can make on your Windows desktop, or by using an Open with context menu (right-click) choice. Any *binary* calculator object file *must* begin with HPHP49- (for this series); the only other files which can be for calculator use would be plain text that is recognizable as UserRPL that could be acceptable if typed into the calculator manually; if a file happens to begin with %%HP: then that's also a text calculator file -- both those files and calculator binary files (and *only* such files) can be usefully transferred to the calculator via USB cable. Once you get valid calculator object files into the calculator, recalling each one to the stack will immediately make obvious what kind of object it is (library, program, etc.); libraries are useful only when moved or stored into a port (usually flash port 2 is preferred on HP49 series, because it has more room and is more permanent). Text files representing UserRPL can be automatically adjusted for backslash sequences representing 8-bit when you transfer via cable the transfer software does the adjusting automatically in translate mode 3, but no such adjustment will occur by itself for files transferred on an SD card, so a little bit of calculator software (small program) will be needed for such files (only readable UserRPL *text* files, not binary files). The most recent version of such small programs that I've posted (universal for HP48/49/50) can be found here: === Subject: Re: How to execute downloaded HP49G programs Hi. I personally fully agree with your decision to go with the SD - much quicker and safer :-) Now: installed on the 50G. That means that you've gotta recall them from the SD on the stack (using the built-in Filer). You'll see that the library is identified by You are supposed to store the lib in a memory port (usually port 2 is OK), so please key: [LEFT SHIFT][.] @putting a pair of :: on the stack [2] @number of the chosen memory port [RIGHT ARROW] @going past the 2nd : [6][6][6] @the lib ID number [STO] @storing into port no. 2 [ON]+[F3] @warmstart the calc Now you've attached the library to port 2, and the library will be visible pressing: [RIGHT SHIFT][2] (press [NXT] if you need to scroll many libraries) To execute the lib, just press the soft-key it's associated to. If the latter applies (i.e., if after recalling the file on the stack you can see a lib ID number), then see above. Otherwise, you may want to save the program in a variable (say 'MYPROG') to run it anytime you want. To do that, after recalling the .hp file on the stack, key: ['] @putting a pair of ticks (variable name delimiter) on the stack [ALPHA][ALPHA] @locking ALPHA mode [M][Y][P][R][O][G] [ALPHA] @unlocking ALPHA mode [ENTER] @putting the name MYPROG on the stack [STO] @storing the file on the 2nd stack level into the variable name on the 1st stack level Now you'll be able to run MYPROG by pressing [VAR] (press [NXT] if you need to scroll many variables) and the soft-key associated to MYPROG. Hope this helps. Giancarlo === Subject: Re: How to execute downloaded HP49G programs *.hp is the default extension for HPGCC programs. That's what they are most likely. In this case, search this newsgroup for how to run hpgcc programs. Claudio === Subject: Re: How to execute downloaded HP49G programs Debug4x projects generate .hp too === Subject: Re: How to execute downloaded HP49G programs I didn't know that. We may as well then change the default extension in the next version of hpgcc to avoid future confusion. Claudio === Subject: Re: How to execute downloaded HP49G programs Why not call it .arm or something that would clearly indicate it's ARM based software ? -that way everybody should understand it's ARM based software and it won't wun on older models. manjo === Subject: Re: How to execute downloaded HP49G programs That's a very good thought! At the same time... What's in a [Windows file extension] name? as Juliet the Geek once asked: http://filext.com/ ... ARM Acorn ROM/Disk Image File ARM Armadillo Software Protection System ARM Alphacam Router Parmeter Macro (Planit) ARM Unknown Apple II File (found on Golden Orchard Apple II CD Rom) ... GCC Gnu C++ File GCC VISUAL EPR COSEUL.EXE Result ... HGL HP Graphics Language (Plotter File) HGP HighGrow (Legal Marijuana Growing Game) Plant File (Slick Software) ... HP HP Graphics Language (Plotter) HP Thor Database Primary Hash File HPD Harry Potter Related :) HPG HP Graphics Language (Plotter) HPP C++ Program Header ... HGC seems to be unclaimed there, FWIW. But no matter -- after all, wasn't Java already in HP48 orbit *before* it fell into the Sun? ;-) http://www.hpcalc.org/search.php?query=java http://ei.cs.vt.edu/book/chap1/java hist.html http://java.sun.com/features/1998/05/birthday.html === Subject: Re: How to execute downloaded HP49G programs .hax (Hp Arm eXecutable) or .atx (Arm Toolbox eXecutable) -indicates that atx file is to be used with Claudio's ARM Toolbox (just to add some examples) manjo === Subject: Re: How to execute downloaded HP49G programs I'd recommend hpc? TW === Subject: Re: How to execute downloaded HP49G programs On Fri, 11 May 2007 18:14:00 -0500, TW: [re default Windows file extension for hpgcc] .hpg ? .gcc ? .hpgcc ? (anyone remember .flash for 49G? :) Where's the nominate and vote page? ;-) -- === Subject: Re: How to execute downloaded HP49G programs I see those and think: hp graphic gcc compiler hpgcc the software itself I still think I like hpc because I read it and think HP C program. I really don't care that much though. TW === Subject: Re: How to execute downloaded HP49G programs How about .hgc ? Here === Subject: Re: How to execute downloaded HP49G programs You could try .hpgcc :-) Windows has supposedly been capable of supporting file extensions longer than 3 chars for ages now but no one uses them. I'm not sure whether this is because long extensions don't actually work in Windows or because Windows users find them too alien. :-) -- Bruce Horrocks Surrey England (bruce at scorecrow dot com) === Subject: Re: How to execute downloaded HP49G programs I'm going to try this a.s.a.p.! === Subject: Re: How to execute downloaded HP49G programs program. library or binary file Other files are probably just help files and don't get downloaded to the calc. === Subject: Using Serial Port - New Programs I'm putting together some programs that use the serial port capability of the 50G. They're going to be incorporated in the astronomy and satellite prediction stuff on http://members.cox.net/above. The added features will be geared to amateur radio satellite and APRS operations. 1) Grab time and position from a GPS receiver using 4800 baud NMEA format. 2) Trigger an APRS transmitter when an appropriate satellite is visible. 2a) Turn the APRS transmitter _off_ when the satellite goes below the horizon. (2 and 2a) are done via alarms on the HP-50G. I'm open to suggestions. It should be posted after some testing. === Subject: Easier Sub-menus This program makes the creation of sub-menus much easier. Let's say you have the following menu list: { HERO DMS BASE GAUS } If BASE and GAUS both contain lists, the following output is produced by CNVT: { HERO DMS } (Get it from www.hpcalc.org). This makes the sub-menus active even during program entry. I begin all the names of my sub-menu lists with chr 134 so they are easily identifiable. This also groups them all together at the back of the VAR menu when it is sorted alphabetically. @ CNVT %%HP: T(3)A(D)F(.); << @ A list should be on the stack 1 @ Operate on one item at a time << IF DUP TYPE 6 == @ If it's a global name THEN IF DUP VTYPE 5 == @ and it contains a list THEN END @ takeover program END DOLIST === Subject: Re: Easier Sub-menus As soon as I posted I saw a couple of bugs: 2) There should be a way to allow a list variable to pass uninterpreted Now you can use &name to opt out of the list processing, by using Here you go: %%HP: T(3)A(R)F(.); << 1 << CASE DUP TYPE 6 == THEN IF DUP VTYPE 5 == THEN END END DUP TYPE 2 == THEN IF THEN END END END === Subject: Debug4x Build 121 Released New version at: http://www.Debug4x.com/Debug4x_b121.zip 1. User RPL debug points (inserts hidden 41 MENU HALT). 2. Verb Completion can now find commmand that CONTAIN a string (alt+ctrl+space) and not just begin with a string (ctrl+space). 3. Save as in *.rtf (rich text formats), also copy to clipboard includes this format. 4. ctrl+/ makes/unmakes comments 5. New command: Align comments at right edge marker. 6. New command: insert date/time 7. Editor View Options Keystrokes now allows assignements of many commands and remembers changes. 8. Many changes to Text Execution window including HP Character table, save stack area (also *.rtf format). 9. Window captions have short file name at beginning followed by long name. Makes identifying items easier. 10. Fixed multiple copies of Debug4x running problems (I hope for the last time . . .) Many, many changes, see VersionInfo.txt -- - - - - - - - - - - - - - - - - Bill Graves RKBA! bgraves@ix.netcom.com === Subject: Current Debug4x Ver 2.2 Build 121 ... William Graves... For version 2.2, the listed current build number of 121 does not agree with the help about screen which shows build number as 120. Which is the correct build number? -Dot- === Subject: Re: Current Debug4x Ver 2.2 Build 121 ... Looks like the final beta 120's .exe is in the package and not 121. There is also a bug or two - the project files get sorted and this changes the project's ROMPTR numbers. I need a few days to investigate and fix. Sorry for the trouble. -- - - - - - - - - - - - - - - - - Bill Graves RKBA! bgraves@ix.netcom.com === Subject: Re: Current Debug4x Ver 2.2 Build 121 ... I sure do appreciate all your dedication to this very useful tool! -Dot- === Subject: hp-50g bad keyboard I was very excited to learn about the new hp-50g and the improvements that were made, particularly with the keyboard. I bought a 49g but didn't like it. I avoided the 49g+ because of the keyboard problems. Well, wouldn't you know that when my new hp-50g arrived today that the keyboard is defective. When you press the 9 key, you hear and feel the mechanical click, but it doesn't register the number. If you press the key much harder, a nine will register. Has anyone else had problems like this with the hp-50g? BTW, HP support was quick and very helpful with getting this fixed. Mike M. === Subject: Re: hp-50g bad keyboard its good that this has been fixed for you. personally, i think the 50 keyboard is ok, not as good as their keyboards used to be but definitely acceptable. the one point i had with the 50 is that i found the keyboard delay unacceptable. for example, if i typed 1200 fast, it would come out as 120. turns out this is a software too long. it's a pity this cant be in flash since currently, it reverts to factory default on reset. good luck, === Subject: Re: hp-50g bad keyboard This is where 'STARTUP' comes in; if you store a program into that name in HOME, said program is run automatically, right after every warmstart; as well as anything else which you'd like to set up, e.g. flags, key assignments (in lieu of relying on whatever was previously saved, which might have included temporary assignments, and will always be lost during any Recover memory), go to your favorite directory, display your name for identification of lost/stolen calc, play tones to simulate Windows startup, etc. === Subject: Re: hp-50g bad keyboard however, i keep hard resetting my unit, so it probably wont survive that. the hard reset is due to my own programs which frequently crash whilst in development :-) === Subject: Re: hp-50g bad keyboard That's what backup is for, e.g. :2:BK001 ARCHIVE @ backup to flash port 2 (3 for SD card) [crash!!!] :2:BK001 RESTORE @ complete recovery! In the emulator, [crash!!!] ARCHIVE/RESTORE backups may also be made in emulator (and then restored to a real calc, or vice versa, thus keeping an emulator in sync with a calculator) === Subject: Re: hp-50g bad keyboard [re backup/recovery via ARCHIVE/RESTORE] By the way, a RESTORE automatically warmstarts at the end, so your restored STARTUP program will automatically execute. So now you have no excuse not to make backups :) === Subject: Re: hp-50g bad keyboard Looks like a job for an emulator! 8^) Tom Lake === Subject: Re: hp-50g bad keyboard No! So, KEEP it! It's a massively unique collector's item!!! ;-) because that's the only way that we can keep tabs on HP's *outsourced* quality control. I hope that incidents like yours will be rare. So far, HP's been doing extraordinarily well with the HP 50g, but it would be unreasonable to expect them *never* to happen, because there ain't nuthin perfect in this here world. -Joe- === Subject: Should EMU48/49 backup on startup by default? Every time I start Emu48/49, just to get started, so that I can always restore back to my initial point. was not so available as now -- was that why backup was optional, and even why Delete was offered? Would it cause anyone a problem (or save everyone some effort) (it could also be another option in some file, if desired)