A110 I've had an HP48 since the early 90s and used it throughout college. One of the most endearing features on it was the solve which I used extensively in my Engineering Economics class. (As well as Joust and Tetris). Now that I am out and making some $, I want to take those equations I used from Engineering Econ and put them to use on my Palm III PDA along with some RPL scripts. Has anyone else desired this? I'm picturing that there must be a way to get the basic scripting features and equivalent of an HP equation solver onto a Palm III. Heck, this kind of emulator could be a sizable revenue stream for whoever makes it. HP, TI, or John Doe. I'm willing to pay for it, even if it doesn't do Joust or Tetris. Is anybody willing to build it? --Chuck Jaeger PS: Any whining about being a traitor and going over to the dark side will be features. Plus I have 2Mb of RAM all for ~$200. I have a solution that probably will work. -You can drill two tiny holes in the LED-cover right in front of each LED. -Then stick two fiber-optics cables at the top of each LED. - Then mount these optics-cables close to each other on a pen or somthing. Then you probably can find a way to read the barcodes. I have a fiber-optics that i use from my HP48sx to my VCR (Which also have a hole in the IR-LED cover) . 1. Turn off the ticking clock on the HP display. 2. Maybe just get this version of kermit for windows: http://kermit.wwarthen.com/ It can properly enter server mode, for one thing... Also, there is a program to do the dumping job on hpcalc.org.... . 1. Turn off the ticking clock on the HP display. 2. Maybe just get this version of kermit for windows: http://kermit.wwarthen.com/ It can properly enter server mode, for one thing... Also, there is a program to do the dumping job on hpcalc.org.... Good luck... - Rob These were intended to be used with the old plug-in printer, but should work with the infrared printer (82240B) as well. Raymond Improve and replace, the very least that can be done with a flash-upgradable hp48, would be to allow users of older rom versions to upgrade to the latest, and with so many known entry points, analyzing some (not all!!) of the system routines, and changing them into ml instead of system-rpl, should be feasible to those who have developed better replacements as external libraries. That's true and absolutely NOT the goal of the project. I hope so! :-) Remember that there was a lot of people who claimed that ml programing the hp48 could not be done, to dificult, not enough info, etc! Where to Get the HP48 FAQ ========================= This document briefly describes where to get the latest version of the comp.sys.hp48 FAQ. The main site is: http://www.engr.uvic.ca/~aschoorl/faq/ All formats of the FAQ are available at the above link, including the text, PostScript, HTML, and SGML versions. ============= and comp.sources.hp48 as needed (usually every two weeks). Also, remember that all official FAQs (including this one) are mirrored at rtfm.mit.edu. Specifically, this one is at: ftp://rtfm.mit.edu/pub/usenet-by-group/comp.sys.hp48/ following key for authenticity, and provide MD5 Checksums for the remaining files. You can get a copy of this public key through the URL above. Type Bits KeyID Created Expires Algorithm Use sec+ 1024 2CFAA0BB 1997-06-21 ---------- DSS Sign and Encrypt sub 2048 F940E148 1997-06-21 ---------- Diffie-Hellman Alternate Sites =============== Alternatively, you can download the FAQ at one the following locations: North America: -------------- Latest: http://www.hpcalc.org/docs/faq/ Older: ftp://ftp.hp.com/pub/calculators/hp48g/docs/ Europe: ------- Latest: ftp://ftp.stud.fh-heilbronn.de/pub/systems/hp48/incoming/ Older: ftp://ftp.stud.fh-heilbronn.de/pub/systems/hp48/info/FAQ/ Translations: ------------- Espanol (Spanish): http://www.alumnos.utfsm.cl/~aarrieta/hp48.html Portugues (Portugese): http://members.tripod.com/~area48/ Contacting Me ============= I rely on your input in order to keep this FAQ up to date. If you have any is: aschoorl@engr.uvic.ca Charset: noconv iQA/AwUBNoqLNehzXOws+qC7EQJ4OACg/0aYBeXvteh7Hq8h9NRm8ttKPbgAoKSH 8ZbNg6DFVr1Hbn+R5ZoQ5AQs =Ss4t -- http://www.engr.uvic.ca/~aschoorl/ Correct me if I'm wrong, but you apparently didn't set the line, so kermit doesn't know on which port it must send packets. If your cable to the HP is connected on /dev/tty1 (COM1 under DOS?) then you must type set line /dev/tty1 set speed 9600 under kermit before sending your file. UPS - I should have written it - Ive started kermit with the parameters for line and speed... As mentioned in another post I think my problem is the new kermit v7 - if someone has a working .mykermrc or similar for this kermit please post... What do you mean by losing? If you choose correctly your functions, you should win, not loose!! -------------- (1/cos(x))^(1/2) whereas the TI-89/92 does. Basically, if the integrand involves a radical, Erable is helpless. I was just wondering how hard the remainder of the risch algorithm would be to implement. You have to help Erable to solve this one. Enter the undefined integral int(0,x,tan(t)^3/sqrt(cos(t)),t) call TAN2SC to apply tan=sin/cos then 'cos(t)=y' EXEC this make the change of variables Then EXPAND should return the antiderivative that is 0 for x=0. The aim of Erable is to help you do maths, not to help you cheat at exams. Here you have to think how to solve the integral and Erable will do the computations for you. sqrt(1/cos(x)). I don't know if that changes anything. Im my opinion, with the arrival of Computer Algebra systems, teachers should focus more on proving theorems than the actual practice of them because the solving can be left to computers. I agree! The teachers should really try to focus on making the exam examples simple to calculate, either with or without Erable/TI-92,-89 and hard to figure out what to do. Because mathematics is NOT raw applying of algorithms or methods, but more of understanding, what is going on. So banning any kind of a calculator is stupid. Of course no lecture notes is to be allowed in the calculators, but if the exam is well formed, it should be the same if you had a Calculator Ultimatum or just an abacus. Previously posted: Proving theorems can also be left to computers, as I was given to understand in my Methods of Mechanical Proof course; in fact, isn't the only proof of the 4-color theorem actually a computer-generated proof? (which some people were slightly uneasy about, because no human mind had quite completely comprehended and checked it?) Certainly. What about the once-popular idea of practicing basic knowledge and technique? Although acting is about something much more than memorizing lines, it is still customary to learn the lines, isn't it, for if one hasn't even learned the text, how is the great interpretation of that text to follow? What great violinist, pianist, dancer, etc. has not begun with some basic practice of some scales, steps, or other routine exercises which are designed to establish familiarity with the underlying basic experience and technique? It would be a misfortune for these to be the only activities one performs, or to be out of balance with other forms of creative development, but I think it's also a misfortune to think that a skyscraper can rise without a deep foundation beneath to support it. IMHO if a student can not do something very basic to the subject without a calculator, some foundation is missing which will lead to a lesser ability to build up to higher accomplishments, and that is why I continue to opine that there could (and should) be some place for closed book, no calc exercises, along with those which assume some mastery of basics and allow the use of tools which assist with them; in fact, for the highest-level tasks, such as those which resemble practical new challenges of a profession, it would seem suitable to have no restrictions upon the use of tools and any other resources, much as it is in professional practice. If you ever had to undergo some operation, say, I wonder how many would be happy with a surgeon who had to keep looking up the index to the anatomy book (or even a more modern CD-ROM version) during the procedure? If we have really learned any subject, we ought to be able to do something very basic to that subject without help; otherwise I do not think that we are very well prepared to even understand what our tools do, which leaves me wondering about the user's judgement and competence to use them properly. I couldn't have said it better !!! With or without the tools, a student should be able to go over the steps of a calculation to the final answer. Math is one of those subjects that you really learn it by doing it (since there are so many different ways to get to the answer, but with experience you can always find the easier ones). This is appallingly wrong, though tempting. --tim Computer algebra systems are a terrific aid to learning math. They should be blended into the traditional curriculum, for in the real world, engineers and scientists will be using them to solve problems. There should however, be a minimal proficiency by hand requirement so that students understand how the machine algorithms have been derived. - hpman With the arrival of handheld computer algebra systems instructors either (1) ban them outright (2) insist that all work, step by step, be shown, or no credit for that problem on an exam. I much prefer the second one! Technology is here, use it or go home. No more slide rules. No more endless tables and charts to look up and interpolate just to approximate an answer. HP's and PC's in the classroom and in the workplace. Go pro or go home. Technology, use it or be a Physisist. On our faculty in math classes we're not supposed to use any kind of aid to calculate faster. You can just have paper, pencil and an eraser and (if needed) some rules. When I asked if I can use the slide rule they said NO. On my question can I smuggle it because it's just like rule they said You can try. I'm a child of calculators and learning to use the slide rule just for showing of to my friends is depresing. But if we're talking about pure (numerus) calculations then I must agree : USE YOUR HP :) -- So that means Finalnd collaborated with Nazi Germany, right? That's what I remember from my history lessons... and if that's not something for you to be proud of, heck, I dunno what is! Woo-hoo, Finland rules! BTW, I am not trying to stoke any jingoistic flames here, so let's let this tread die it's silly death now, 'kay? And if the sarcasm of this post has eluded you, just back away from the keyboard... slowly... the material, let them use a calc. I never understood the purpose of not allowing calculators in calculus. In my many calculus classes, I had one professor who didn't allow them, and on his tests, I spent more than half of my time just doing basic math! Calculus class should be spent learning how to do calculus, not multiplication or trigonometry! Just my $0.03 (Inflation, don't 'ya hate it?) Until you see what a step-by-step computer algebra system can do and then your only option is the first one :( Carlos. A step by step CAS? Sounds like a dream true for students... his computer that does step by step stuff Contests? Bad idea. When compared with HP48GX in symbollic math, TI-89 is not measly... Por Dios, Carlos, Àc227mo sabes tanto? If you check http://www.poliplus.com you will see that we do develop Computer Algebra System (CAS) Software. The field is not that big, so everybody knows everybody (hence why I know so much). When it comes to Computer Algebra, the University of Waterloo (UW) has excellent courses taught by one of the original Maple creators Professor Keith Geddes. UW also has tones of CAS material in the library. to answer jemfinch02@aol.com (Jemfinch02) question : The thing is that the Risch Algorithm has extensions and the basic extensions for exponentials, logarithms and trigs are just minor adjustments to the basic algorithm for the integration of rational functions. Now the extension for Algebraic functions (Radicals) uses a different mathematical framework so things get complicated. To answer your question : It would be hard... but then again, you can use heuristics for this type of integrals. Carlos. A heuristic algorithm involves the use of tables. It is how the Derive software built into the TI-92 and TI-89 works. It basically takes an input, matches the pattern with a database of tables, for example, an integral table. Then, it takes that pattern that most closely matches the input, and fills in the blanks. Works quite well, actually. Indeed, I hope the new HP has this enhanced pattern matching capability as far as integration goes. begin 666 Catd.dat JYA says Right! also Sparcom Mathematics (PN 10071-1A) and Calculus Pacs (PN 10111-1A). both are in new condition in the boxes with manuals. best serious offers considered for one, both or entire package. Calculus Pac PN 10111-1A This pac is a collection of tools and equations that will significantly extend the calculus capabilities of the HP 48SX and HP 48GX. Access the function library for curve fitting routines and root-finding. Integrate by left, right, and midpoint rectangles, trapezoids, and Simpson's rule. Access integral tables, plus add your own. Graphics include 2D and 3D, including hidden-line removal. Symbolic vector and differential equation solving. Features: [Eth]Function Library [Eth]Integration Analysis [Eth]Plotting Tool kit [Eth]Vector Tool kit [Eth]Integral tables [Eth]Limit Analysis [Eth]Reference Data [Eth]Constant Library Function Library [Eth]Fit one point/slope [Eth]Fit two point/slope [Eth]Fit many points [Eth]Piecewise functions [Eth]Build polynomial form roots [Eth]Find roots of polynomial [Eth]Taylor expansion calculation Integration Analysis [Eth]Left rectangles [Eth]Right rectangles [Eth]Midpoint rectangles [Eth]Trapezoidal rule [Eth]Simpson's rule Solve and plot differential equations using the following methods: [Eth]Euler's method [Eth]Modified Euler's method [Eth]Runge-Kutta method [Eth]Slope Field The plotting Tool kit contains commands oriented at enhancing the graphics capabilities of the HP 48SX. 2-D Graphics features include a trace function. 3-D Graphics features include plotting of parametric curves and surfaces along with hidden-line removal. Now Symbolic Vectors are supported by the Calculus Pac! Access over twenty-five programmable commands. I have a HP48GX that is new and I have a need to copy all the data from 1 128K mem card to a blank 128K mem card. Does anyone know the exact key sequence ? Rick first scientific calculator. Please point your browser to http://cgi.ebay.com/aw-cgi/eBayISAPI.dll?ViewItem&item=74211417 whoops, that's the wrong equation. I meant. x^2 (1 + 2x^2)(e^ ) The only thing you have to do is get the 48GX ROM image because it is not shareware or freeware. There are directions to help you get the ROM image off of a 48. Jeremy The FACTORIAL command does the job. The new HP Jornada 420 is the first Windows CE palm-size PC to incorporate an easy-to-read, rich-color display. The brilliant, 256 color, 240 x 320 pixel screen improves readability dramatically over previous monochrome models. And this... We were impressed with the clarity, sharpness and brightness of the Jornada 420's DSTN (Double Super Twisted Nematic) screen, which outshone the FreeStyle 540's reflective thin-film transistor screen indoors, although both units were quite readable in bright sunlight. I know that some of you are pretty scared of battery life in the future new HP calculator if a color screen is included, but let me tell you that I'd be more than delighted in using a ton of batteries in order to have a full-color-screen calculator. An guess what? HP has the technology to do that TODAY! Hey you, ACO guys, would you mind to include that in the HP58 features? I PAY THE BATTERIES!!! - Roberto Perez-Franco icq# 14607303 Crazy... i use calculators for science and engineering, not childish entertainment, and would hate battery life affected by useless color gimmicks Do you know something? I also use calculators for science and engineering. In my whole life I've never played a single game in my HP and in fact I never will. I also use paper and pencils for science and engineering. I am the best student in my class. And I use color pencils in my calculations. Why? Because color is a GUIDE. It guides the eye and the mind through the whole mess of numbers and formulas. In my paper, I use one color for formulas, another for given data, another for obtained data. And although my tests look as colorful as the notebook of a 5 years old child, my grades are the very best. I think that color is useful in science and engineering, my friend... and if it works in a notebook and a sheet of paper, it will work in a calculator. Sorry I can't help it. The fact that you use colored pencils doesn't mean that that is the reason for your excellent performance in class. Perhaps your excellent performance occurs in spite of your use of colored pencils! Lance Because for colorful if If the new HP58 (I want one, I want one!) would have a colour display, it would be good. About battery life, an AC connecting would solve a lot. I think most people use theirs at a place where AC is available, anyway. Maybe it could be fitted with an Lithium Ion battery, such as the ones in the better mobile phones, thus re-chargeable. Additional batteries can be bought to cater for different users. I have 3 batteries for my mobile, tiny ones for an Ericsson GF788. They are about 4 cm by 3 cm. I'm sure the geniuses at HP (or whatever the calculators are going to be called) can incorporate such a feature. I wouldn't even mind if the whole thing became a little bigger, or heavier. People now say the HP48 is too big, but users don't care. Maybe HP should consider bringing out 2 new types, one lean mean calculator only, and one model with all calculator functions, but also incorporating other options, such as easier text input, database program, not to mention games! I know this wouldn't be acceptable for students, but I managed to pass college with my trusty compatriot the HP48GX. I for one would be interested to purchase a PDA/calculator machine. Bart van der Plas like a damn photograph...better than colour television...! i was however VERY VERY VERY annoyed that they were using MicroSoft Software to operate it...!!! and that it was something like 540$ (???) to get around the battery sucking problem...which the tag next to the picture mentioned, it uses a plug in modual (?) that you keep it in at night, that recharges it...which sounds very reasonable... while its sitting there, it could display the time, or tie into the internet ( my speculations ) and display all sorts of things...??? or maybe you could ( of course ) install a sci-calculator program and make the Jordona the new '58'... the colour though...is fabulous...256 colours is ( when well chosen ) are all the colours the human eye can reasonably discern, you use the million colours tables to pick the ideal 256 colours for any given picture. ) ----------- :: o .---..-..-..-..-..-..-. . .-. .-. .-..-..-.. .-. `-' `-'`-'`-' `-' `. ^ .' `--^--'`-'`-'`-'' `--^--' Thinking Is The Enemy; Dada Your Way To Happiness! Can any friend from Germany explain me the text of this page ? It talks about a new solar HP6s calculator. not 100% efficient Carlos Marangon HP48 for beginners! http://www.geocities.com/~hpfb http://members.xoom.com/hp48fb (mirror) perhaps not the finest style, but The HP6S / HP6S Solar For smart people: The intelligent (solar powered) calculator with fractions calculation The HP6S and the HP6S Solar are two new scientific calculators from Hewlett-Packard, which are used by pupils in math classes, but also assist adults who are concerned in scientific calculations. The HP6S is available in two different models: battery-powered (the blue one) or solar- and battery-powered (the silver one). One of both is free with every order you'll do during the CeBit fair ! Rubbish. 256 colours is totally inadequate. 256 *grey levels* is totally inadequate. Go and look at a decent photograph some time, especially something taken with a medium-format camara. 256 colours from a large palette is adequate for a calculator though, I should think (:-). --tim I think what he meant is that it is great if you can pick the best 256 for any picture, which you can. However, more colors will always look better for more pictures at once. Glad to see there is so much response to the colour items. However, I am sure that for a scientific calculator maybe even 4 colours would be enough. I don't want my integrals to come up in 16 million colours. It takes long enough to compute now, let alone then. If HP would manage it though, maybe a bundled copy of PhotoShop would be nice... Maybe even a full version of AutoCAD too? All joking aside, I haven't really read anything on my PDA/calculator idea. One base calc only model, one combined one. HP58 and HP58PDA? Bart van der Plas What is PDA, Bart? What is PDA, Bart? Sorry, some advertising slogan I picked up somewhere. Personal Digital Assistant. I meant to imply that I would be interested in buying a HP, with all the calc features, and features such as US Robotics Palm Pilot, or Philips Nino. features, and features such as US Robotics Palm Pilot, or Philips Nino. Sounds great! But how much would that cost? all colours As a graphics artist and photographer, I disagree strongly. 256 colors isn't even enough to truly represent monochrome images in their full splendor. For color images 65536 colors is a good start, and for high quality, 16 million colors are a neccessity (combined with high resolution and dithering it is enough to fool the eye into true continuous tone (non-dithered 16 million color is pretty smooth tone, but there are some noticable bands) Ken If it didn't significantly affect the price, I'd think it was a nice feature, but it's certainly not on the top of my must-have's for a calculator... Ken On Fri, 5 Mar 1999 12:27:01 -0500, Ken Alverson It would be nice if you had a choice of re-chargable battery pack or non-rechargeable batteries. Similar those used for the HP41 series. ( Not necessarily N cells) Harold A. Climer Lab Instructor Dept.Of Physics & Astronomy U. Tennessee at Chattanooga If the new calculator didn't use standard batteries like AAA or AA then I don't think I'd buy one. Barry BATTERIES!!! The only way I would not buy the new HP is that it used a nuclear reactor as power source. I remember my old Casio CFX (oh sorry but this was my first color calc). The screen was not really readable and It was useless for maths.. excepting for the Plotter. So I can hardly imagine a new model of HP CG(X) based on 4 colors screen. Julien Meyer I had a Casio CFX-9800G and I loved it color display. It was great to plot several functions at once and gived the calc a great look. I think color has a lot of applications in math software. For example, one color for the entry, another for the result. Or one color for variables, other for directories, other for libraries and so on. It is a matter of imagination. And batteries? My CD player eats more and is less usefull. Speaking of batteries, I think the triple A batteries kick ass. I used to own a Casio with lithium batteries and it cost approximately $15 to replace them!!! They didn't last much longer either although that probably has more to do with the excellent power usage of the HP48. I think the HP48 paid itelf off simply in battery usage alone! Steve I don't agree with you Steve. When you use your HP at school or during you are working for a week, the batteries will be to change every two week.. and it costs about $6 each time. I remember when I was programming my HP48, then it was every week I needed to change batteries.. everyweek for two years :) And when you use other kind of re-usable (sorry I have not the word..) batteries then it is every 2 to 3 days. I think this is better in TI calculators, but I'm not sure. Julien Meyer own a them!!! with the in other imagination. Like I said, it's useful, but not high on my list of priorities. Speed and erable level (better?) symbolic math capabilities would probably be the most important for me. Ken I wouldnt like to see color in the new one if it meant short battery life. But I'd be ok if the life was half what it is now. I change batteries every 3 months although they still have enough life in them at that point to play a some0 tapes in my walkman. Even changing once a month would be ok. Color would be nice but if they don't do color what I'd most like to see is a brighter high contrast screen like the newer TI model's have. Barry since till now many claimed a color display: I agree! A new calc should be improved in calculating purposes. I don't agree here. Keep in mind that the HP48 is a calculator! I want to use a calculator everytime and everywhere. My HP32SII runs *forever* on batteries. The HP48SX needed less batteries then the GX and I would like to have an increased battery life!!! Especially the PC-transfer eats up to much batteries :-( Sure! But that shouldn't have a high priority. Much more important are the size, battery life, cost!, ease of reading the display and last not least the capabilities of the calc! (I'm very old fashioned: If I want to play a game I grab my wooden chess board or go out and play Frisbee ;-) A higher contrast screen would be great! If the background would be brighter/lighter (what's the right word here?) that would do the job. In addition the LC (liquid cristals) should be closer to the background to avoid those very annoying shadows when the calc is viewed under a strong directed light (coming from the front and not from the point of view). Just my 0.02 Euro (I try to get used to the new currency ;-) Peter _______________________________ Do you know the great Frequently Asked Questions? http://www.engr.uvic.ca/~aschoorl/faq/ and the superb HP48 Software Archive? http://www.hpcalc.org to look for *old* HP48 postings see Why not having both? Could someone explain why my 48GX returns -195948537906 for TAN(pi / 2) and not an Infinite Result error as TAN(90) degrees would? What am I doing wrong here? ----------------------------------------------------------------- Isaac Tipton isaac@spamless.shadetreeweb.com remove spamless from address when replying Shadetree Web Design http://www.shadetreeweb.com/ ----------------------------------------------------------------- Nothing. The HP48 must convert PI/2 to its numerical value before calculating TAN - and its numerical value is NOT the same as PI/2. -195948537906 is actually the best possible 12-digit value of TAN(1.5707963268). Before you ask why SIN(PI)IS zero, and SIN(3.14159265359) is -2.0676..e-13, that's because SIN(PI) is automatically simplified - I guess the AUR contains a list of all automatic simplifications, and naturally TAN(PI/2) is not included - because it is a little hard to simplify to 'Infinite Result'. : Checked AUR. Seems that all the automatic simplifications are not included there, because there is no reference to SIN(PI) ... Best wishes from, -- Robert Tiismus http://www.physic.ut.ee/~robert I guess my question has to do with the relation of 90 degrees and PI/2 radians. For example, when I enter these two values in the TAN function on a TI86 I get the same message Error Domain, which is the same as Infinite Result, that TAN 90 is undefined. In other words, it seems the TI can understand the relationship of the two inputs and then give the same result but the HP needs to simplify PI/2 first. Is there some way to set the HP so it interperts PI/2's radian relation to 90 degress as an expression rather than its numeric value? Does this make sense or should I be looking at this differently? - I don't know about the TI-86, I know the 92 has an 'exact' mode and an 'approximate' mode - meaning that in exact mode it doesn't calculate PI/2, but rather uses a lookup list (whan evaluating) to see that TAN(PI/2) is undefined. And of course in DEG mode 90 degrees is exact.. If the TI86 doesn't have an exact mode, than it may test the input value to see whether it equals PI/2 to 14 digits, what I believe the accuracy to be. That would be questionable, to say the least. Hash: SHA1 Can the 48GX be set to return exact values in this way? an PI/2, is value accuracy to be. ==---------- Own iQA/AwUBNuN6MxSyxevsOQAjEQIt1gCfbiHcnei6cT+MENJUJ72BhlIN52cAoKnz Ef40cXNyl0L5KrKe6hTZU1E4 =7LZM How do I include calling labrary functions in SYSRPL ? Peter Michael sods@control.auc.dk Thanx for this free advert :-)) Enjoy it at : http://home.nordnet.fr/~bdarcy and please send feedback to bdarcy@nordnet.fr bye Benoit Darcy - HP48 E-zine : http://home.nordnet.fr/~bdarcy I entered that integral at www.integrals.com and it gave the following result: Log[Y] - 1/2 * Log[A^2 + Exp Y^2] Actually, the answer contained the variable X, since that site is set up to integrate wrt X, so I substituted. Give www.integrals.com a try; it's a helpful site. I'm not sure why Mathematica didn't work for you, since www.integrals.com is powered by that tool. -Dan Guess you made a typo: 1/(x*(1-Exp[-(x/A)^2])) Cannot be solved by www.integrals.com, and your solution does not derivate to the integrand. Werner Huysegoms remove the x before replying The HP-42S cannot print bar code to the IR printer. End of story. There is no way to do this. Period. Now...regarding the HP-41. The PPC rom contains NO routines that enable or help bar code printing per se. There is a Barcode analyzer in the rom, but that only analyzes bar code scanned in by the wand...it does NOT help printing it. There is NO way to print HP-41 bar code on the dedicated 82143 thermal printer without modifications to the printer itself (I'm 90% sure of this one). There are programs that enable the HP-41 to print bar code to the 82162 HP-IL thermal printer AND to the HP-IL thinkjet. Several of these programs require the use of other plug-in modules. You can find these listed on my website at the URL below. Gene Http://members.aol.com/hpgene Does anyone have any thoughts as to how the purposed breakup of HP into separate entities will effect the calculator division? Will the weaker divisions be sacrificed just to make the bottom line look better for the computer related divisions? Harold A. Climer Lab Instructor Dept.Of Physics & Astronomy U. Tennessee at Chattanooga calculator. Now it works back normally but aren't there any consequences. And how come that my hp turns off? calculator. Now it works back normally but aren't there any consequences. And how come that my hp turns off? --------------------------------------:: o a) it's broken b) the batteries are going dead c) an alarm went off with a program that told the calc to turn off. d) evil gremlins, or the dead trying to communicate with you e) quantum anomoly f) bug in some program you were running g) ??? ----------- :: o .---..-..-..-..-..-..-. . .-. .-. .-..-..-.. .-. `-' `-'`-'`-' `-' `. ^ .' `--^--'`-'`-'`-'' `--^--' Thinking Is The Enemy; Dada Your Way To Happiness! ^^^^^^^^^^^^^^^^^^^^^^^^^^^ Dumb. Really dumb. X. Yeah, but at least it protected him from getting a concussion. -Joe- How did you accomplish this? I have tried using my laptop (which is also a Toshiba Satellite) but have been unsuccessful. Richard Alimi richard@ralimi.hypermart.net HTMLSoftware ----------------------------------------------- Want great free software? Go to HTMLSoftware's site at http://ralimi.hypermart.net/ Does anyone know a quick way (program ) to have an object copied to the stack multiple times. For example I need to have 5.343 entered 50 times, but I don't want to have hit the enter key 50 times. do the trick. I always do that when I need multiple copies of a object - it doesn't take more than a couple of seconds and surely beats hitting ENTER many times. If you find yourself doing this a lot of times, you can just write a small (that program takes an object in level 2, a real number N at level 1 and menu if you like. Ice Brad skrev i meddelelsen MST) with y in position 2 on the stack and n in position 1, will leave you with n copies of y in positions 1 to n on the stack. -- Virgil vmhjr@frii.com this seems to be a silly question. What is it good for? OK. Here is my USR-RPL-solution: This gives You : 4: 5343 3: 5343 2: 5343 1: 5343 Easy, isn«t it? I've prepared a circuit simulator for those begginer students that are starting to know circuits. It is extremely easy to use and very user friendly. It's name is Assistant II and it is based in the Assistant and powered by Per Stenius' CSim. Get it at HPCALC www.hpcalc.org, in Electrica Engineering area. If you wish a more advanced circuit simulator, try ZAC or CSim for numeric simulators and SVF and SCF for symbolic simulators. - Roberto Perez-Franco http://pagina.de/Perez-Franco icq# 14607303 I've tried to install erable about 20 times using setup. In the past I've had no problems but in the last couple of days I've had no success. It frequently generates an error Too many retries, then when I go to check my ports I find that port1 has automatically merged with port0. Have no idea what's going on but is driving me nuts. Any help would be greatly appreciated. Thx Terry Bowman I don't like the automatic installation if Erable, I do it manually. But first you need a working comm program. I am using HP Comm found at http://over.to/hpcomm it is more flexible than Kermit. My user memory (0 port) stays empty when downloading Erable. My port 1 must be empty (clean). Then I download Erable1 or Erable (from the Absolute version) to the stack. Then I do the install thing to port1 purging the variable left in the stack. Then I download the other needed components (setup, gxkeys). Once I have SETUP and GXKEYS as variables in the stack I can install Erable. Before that, I put UFL in a different port since it will work from anywhere. I do install Java in port 1 after Erable since it will run from port 1 too. After finishing all that hard work, I do a complete backup of user memory with that wonderful utility if problems occur. The latest beta release of Erable seems very stable, but RainEQ is not which I also use. I also make sure my port1 is read only after I complete the installation. J.R. Martinez anadigit@pacbell.net jmartin@calstatela.edu ideas of the program structure, mnemonics etc. What I really need is more information about using the memory, storing of objects(lists,matrix,etc), algorithmics for data compressiom and some built-in sub-routines like Bel216n thanx! ----------- :: o .---..-..-..-..-..-..-. . .-. .-. .-..-..-.. .-. `-' `-'`-'`-' `-' `. ^ .' `--^--'`-'`-'`-'' `--^--' Thinking Is The Enemy; Dada Your Way To Happiness! Salut 210 tous, Je vends plusieurs accesoires pour l'hp48 : - Livre HP 48 Fa224tes vos jeux en assembleur (216dition DUNOD) = 150 F - Livre HP 48 Math216matiques en pr216pa (216dition DUNOD) = 100 F - C211ble dans sa bo224te + son logiciel = 200 F - Explications + puce 128ko pour G = 50 F - idem pour GX mais pour passer de 128ko 210 256ko = 60 F Pour tous renseignements : dumaisp@club-internet.fr Richard How do I install it, since it has not enough memory to copy it to port 0 ? Try this: Press VAR and them press the button corresponding to your library name: The library will be pushed to the stack. Purge your library from the memory Now the memory is free and your library is on the stack. Enter 0 then STO to store it normaly. original variable from the calculator. For example if the library is called BIG, you would enter 'BIG', then PURGE it. Now enter the port number and store it, and turn the calc on then off. Your in business!! Dennis During a program I want to show a provisional result on the display only for two seconds. Then the calculator should continue the program itself without any key stroke. How to program? Use the comand WAIT, found in [PRG][IN] Jo213o Pedro P220go This is the code you need. Copy in a file and transfer to your hp. %%HP: T(3)A(R)F(.); << @ ... YOUR PROGRAM HERE THIS CALCULATOR WILL SELF-DESTROY WITHIN 2 SECONDS... @ YOUR MESSAGE HERE 1 DISP @ DISPLAY MESSAGE FROM ROW #1 2 WAIT @ WAITS 2 SECONDS @ ... YOUR PROGRAM HERE In order to learn the syntax of these commands, and many more, please refer to AUR (HP 48G Series Advanced User's Reference Manual), hp part no. 00048-90136. Browse http://www.hpcalc.org to find a list of international vendors of hp48 products. ciao enrico. has anybody noticed the small image at http://www.hewlett-packard.de/kaufen.html which shows two calculators that you'll get for free if you web-order during CeBit time? If my eyes don't cheat me it's the HP logo in the upper left corner of these little - obviously solar driven - calcs... Any more specific infos available? Stefan. I wonder if it's an HP original, or just a re-badged unit ... Peter Khor it's a new calculator. It's the HP6S, and is available in blue or silver, with either battery power or dual (solar+battery) power. Apparently it's a compact scientific calculator aimed at students, and it does not seem to use RPN (though I hope I'm wrong). I'd be interested in hearing more about it as well. Hopefully this will be a sub-$25 calculator to compete with TI's 30 series. Eric Rechlin Bismarck, ND eric@ hpcalc.org http://www.hpcalc.org/ No, there's no official press release yet so I guess HP-Germany decided to leak the information ahead of time to drum-up support for CeBIT. Why on earth have the option of battery or battery+solar? Next to no one is going to buy the battery-only version (except that it seems to be a nicer colour - the impression I got from the web page is that the battery-only one is the blue one, the solar+battery is the silver one). It just ensures that HP will have totally unnecesary, self-inflicted manufacturing and distribution headaches. Doesn't look like it has RPN. Would be nice to have included as an optional mode - even if not the default. It's a pity that HP don't seem to realise that RPN is their Unique Selling Point that distinguishes them from the rest of a very crowded and competitive market. -- Bruce Horrocks (...speaking for myself) EASAMS Ltd, Waters Edge, Riverside Way (Watchmoor Park) Camberley, Surrey GU15 3PD. Judging by the photograph, there *is* no battery-only version. The blue one clearly has the same solar panel in the top-right as the silver one. I think it's solar-only (blue) or dual-power (silver). -- Rick Duh... [slaps forehead with hand] ...of course. I've always considered solar-only as about as much use as a chocolate teapot and so the thought never occurred to me. -- Bruce Horrocks (...speaking for myself) EASAMS Ltd, Waters Edge, Riverside Way (Watchmoor Park) Camberley, Surrey GU15 3PD. Don't be too harsh on yourself - the web site does *say* the blue one is battery-powered, after all. You're right about solar-only, though; I bought a teeny TI for about 4 quid for the amusement factor. It needs quite strong light to work. -- Rick Hellicar Somewhere in Basingstoke That's a good question. Maybe the point was to make people choose between a nice color and a useful feature. :) There could be a legitimate reason not to have a solar cell, but I don't know what. (The solar cell makes the calculator flakier? You live somewhere that has no sun?) As someone pointed out (not sure who), it's entirely possible that HP didn't make this calculator and is only selling it. Personally, I looked at the photo and thought, Nah, this can't be made by HP. Or if it is, HP's standards just went down. This is an ironic development. I've always wondered why HP never made solar-powered calculators (it seems pretty logical to me). I've always wondered why they keep their prices so high. *Now* comes the answer to my questions, and I and not likely to buy it! (That's why I put answer in quotation marks.) I wonder how good the keyboard is? -- Derek This is almost certainly a made-to-order calc from a no-name Taiwanese/Korean/somewhere else Asian manufacturer. No doubt an analysis of the rounding errors and trig. accuracy versus a proper HP will be illuminating. -- Bruce Horrocks (...speaking for myself) EASAMS Ltd, Waters Edge, Riverside Way (Watchmoor Park) Camberley, Surrey GU15 3PD. I'm rather sure it is: the text says that you get one of them for free if you order at the CeBit and that it is designed for use in school. Sounds not very expensive. CU Stefan This calculator shown in the german site is not similar to previous HP calcs, specialy the keyboard. I think a third party firm is making the calculators and HP pasting the name HP6S on it. It seems to be a model for general purpose in high and elementray school, maybe also for some promotion a way the firm is giving it for free if you order a product. ================= The writen above is my opinion olny. I dont know nothing of official about these new calcs. Carlos Marangon HP48 for beginners! http://www.geocities.com/~hpfb http://members.xoom.com/hp48fb The following URL http://www.hp-shop.de:80/hp/gratis.asp?mscssid=169199 shows a bigger picture and names the model as a HP 6S (scientific non-RPN model with two solar or dual-power versions) I found nothing about this at www.hp.com, does anybody know how old this model is ?, or its price ? I search for a cheap 128k RAM card for a hp48SX. Could someone help me? JM Ps: i'd like to find it in France if possible. Strange. Can't believe that there was no official announcement from HP telling that they introduced a new calculator model. Must be brand new - all official calculator web pages at HP don't have a link to the 6S ... Barry just out of curiosity, does anyone know what microprocessor is used in the hp 48g? kapil On Sat, 6 Mar 1999 21:03:22 -0600, Kapil Sakariya Benoit is right... Apologies Topshape It's a NEC Saturn cpu at 4Mhz 210+ Benoit Darcy (zdar on IRC) - HP48 E-zine, PCteam, Performance Calcul. ____________________________________________________________________ HP48 E-zine, le premier magazine 216lectronique des passionn216s de HP48 L'info HP48, mise 210 jour quotidienne http://hp48ezine.tsx.org øøøøøøøø [O S lash][OSlash][OSlash][OSlash][OSlash][OSlash][OSlash ][OSlash] ø [OSlash][OSlash][OSlash][OSlash][OSlash][OSlash] ø[OSl ash][OSl ash][OSlash][OSlash][OSlash][OSlash][OSlash][OSlash] [OSlash] ø [OSlash][OSlash][OSlash][OSlash][OSlash][OSlash] ø[OSl ash][OSl ash][OSlash][OSlash][OSlash][OSlash][OSlash][OSlash] [OSlash] ø [OSlash][OSlash][OSlash][OSlash][OSlash][OSlash] ø[OSl ash][OSl ash][OSlash][OSlash][OSlash][OSlash][OSlash][OSlash] [OSlash] ø On Sat, 6 Mar 1999 21:03:22 -0600, Kapil Sakariya It is a Saturn processor, manufactured by HP.... Topshape The microprocessor used in the Hp48g is the Saturn microprocessor. If you shoyo@intervista.com.br and I will see what I can find about it. The intern program of the TI is very nice..... I think it is better than the HPs progz..... CU Talmente semplice che 217 come un gioco, un gioco fantastico che pu230 per230 avere il potere di cambiare la vita, la nostra vita, che 217 preziosa! Il tempo perso non si pu230 pi235 recuperare! Per ci230 quando mi 217 arrivato questo messaggio ho pensato: e[AHat] l'occasione che aspettavo! Vuoi saperne di pi235? Rispondi Subito !!! Julien (but http://home.nordnet.fr/~bdarcy works always fine) 210+ Benoit Darcy (zdar on IRC) - HP48 E-zine, PCteam, Performance Calcul. ____________________________________________________________________ HP48 E-zine, le premier magazine 216lectronique des passionn216s de HP48 L'info HP48, mise 210 jour quotidienne http://hp48ezine.tsx.org [OSlash][OSlash][OSlash][OSlash][OSlash][OSlash] ø[OSl ash][OS lash][OSlash][OSlash][OSlash][OSlash][OSlash][OSlash ][OSlash] ø [OSlash][OSlash][OSlash][OSlash][OSlash][OSlash] ø[OSl ash][OSl ash][OSlash][OSlash][OSlash][OSlash][OSlash][OSlash] [OSlash] ø [OSlash][OSlash][OSlash][OSlash][OSlash][OSlash] ø[OSl ash][OSl ash][OSlash][OSlash][OSlash][OSlash][OSlash][OSlash] [OSlash] ø [OSlash][OSlash][OSlash][OSlash][OSlash][OSlash] ø[OSl ash][OSl ash][OSlash][OSlash][OSlash][OSlash][OSlash][OSlash] [OSlash] ø This project requires all your attention !!!! I've made a banner in order to put in your WWW pages with this link : Thanx to put this in your page and please send me feedback about this at : bdarcy@nordnet.fr 210+ Benoit Darcy (zdar on IRC) - HP48 E-zine, PCteam, Performance Calcul. ____________________________________________________________________ HP48 E-zine, le premier magazine 216lectronique des passionn216s de HP48 L'info HP48, mise 210 jour quotidienne http://hp48ezine.tsx.org [OSlash][OSlash][OSlash][OSlash][OSlash][OSlash] ø[OSl ash][OS lash][OSlash][OSlash][OSlash][OSlash][OSlash][OSlash ][OSlash] ø [OSlash][OSlash][OSlash][OSlash][OSlash][OSlash] ø[OSl ash][OSl ash][OSlash][OSlash][OSlash][OSlash][OSlash][OSlash] [OSlash] ø [OSlash][OSlash][OSlash][OSlash][OSlash][OSlash] ø[OSl ash][OSl ash][OSlash][OSlash][OSlash][OSlash][OSlash][OSlash] [OSlash] ø [OSlash][OSlash][OSlash][OSlash][OSlash][OSlash] ø[OSl ash][OSl ash][OSlash][OSlash][OSlash][OSlash][OSlash][OSlash] [OSlash] ø I have a Toshiba Satellite laptop with an IR port. I would like it up so that I can transfer files between my HP48 and the laptop. I have already installed the drivers for the IR port on the Toshiba, but it doesn't recognize when the calculator is within range of the IR port (a detection utility that does this is included with the drivers). Does anyone have any ideas on what I am doing wrong? Or has anyone successfully set this up who can tell me what settings they are using? Richard Alimi richard@ralimi.hypermart.net HTMLSoftware ----------------------------------------------- Want great free software? Go to HTMLSoftware's site at http://ralimi.hypermart.net/ x-no-archive: yes Didn't work on my Libretto either. I guess the HP48 does not send back any info the MS software recognizes as valid IR device. Use the EMU48 emulator (the patched version by Ch Giesselink http://privat.swol.de/ChristophGiesselink/emu48.htm allows both wire and IR transfers, cool), it works as if you had two physical HP48 exchanging data. As far as I remember, I did not have to change any of the default settings for the Microsoft IR driver. rgds Thomas Perfect on my IBM ThinkPad with 4MBps IRda port., hehe you're Toshiba aint works You have to SET flag -57! Did you know that you also can type in a program instead of an alarm string? bye, Sven http://beam.to/hp48network x-mac-creator=4D4F5353 http://www.hpcalc.org/ Bookmark this site! I have used 48 Explorer and HPShell and find them both to be excellent. Some bugs in the new 48 Explorer. The link program from Hewlett-Packard is pure ca-ca. Mike Kent, Go to the following link and download either HPExplorer or HPComm they work nicely in windoze http://hp48.ml.org/pc/link/ Yves-Marie Gagn216 Could you tell how much cost a 512 Ko ram card? pgdo@club-internet.fr 3com palmtops. I thought about my calculator and remember about tools you can get from the internet. While looking I noticed different emulators. I what the features of a Windows CE or at least the database functions. Is there anything I can do the have my HP48 performs there tasks or should I sell it and invest in a palmtop Well, I have seen lots of database aplications to the 48, try to look at the site: www.hpcalc.org . If you don't find what you're looking for there, then there will be a very little possibility that this thing exists ( it's a really good site, lot's of programmers send their creations to there. About the Windows CE, it will be a shame to have an HP48 resources spended away with a Microsoft operational system ( 33% operational ) :). Felipe Carvalho shoyo@intervista.com.br The HP48 isn't really suitable for your database needs. The palm sized computers all have a flat data base built in which is quite adequate for the need, and you can always transfer info to your PC. Many of the calculations that you may still find useful can be done with a few small packages that may be downloaded. The software for 3Com devices is quite extensive, and you might find MathPad and RPN useful: MathPad is a solver, and RPN is a RPN calculator for which scientific, probability and other software may be attached. Several new Windows CE machines are due in the first part of this year. They will have a lot of memory (16MB usually), color screens, and processors running at around 100MH. The HP entry, called the Jornada* 420 is especially interesting. They will include scaled down versions of some standard Microsoft programs such as Excel and Word, which may partially overcome their lack of third party software. Unfortunately, you will not be able to program them in a simple way. None of them, to my knowledge, has a simple user programmable language. Programmers have to deal with a Windows type API, which has a steep learning curve. I expect that this will be remedied either by the manufacturers or by ingenious programmers. They started out as organizers and remote data gathering devices, and the idea that someone would want to program them easily wasn't considered. * Jornada is a Spanish word without an exact translation, but it roughly means activities done in a day. ECHIP, Inc. powerful, at least it runs on WinCE platform... but I found not that much use of it. so I returned both item. I got a laptop, so I won't use it for internet or for typing... yeh, I could type, like 10 words per min. I only Both items are much better than the 48GX. I currently own a 48G and I am out of memory already. Mostly because of the calculus programs I put in for my engineering courses and math. If you decided to sell it, I am very interested in buying yours. Lastly, if I am you, I will go with an used HPC... you could do more on it. Dave Lau well, well, well, HP48GX is 1.28MB with standard memory card, and multiple 4MB banks with thrid-parties... using anything but HP for science/engineering is insane, but since he mentioned sales.. what a strange deviation from physics... and want to use the calc, or for other reasons need to be in a dark place while using the calc. Is there any posible way to backlight the display? (Actually one of those little booklights works nicely, but the holy grail is backlight.) I don't need color but backlight would be great. (though batteries...) -b My Game Boy backlight/magnify works fine w/ my 48... 210+ Benoit Darcy (zdar on IRC) - HP48 E-zine, PCteam, Performance Calcul. ____________________________________________________________________ HP48 E-zine, le premier magazine 216lectronique des passionn216s de HP48 L'info HP48, mise 210 jour quotidienne http://hp48ezine.tsx.org [OSlash][OSlash][OSlash][OSlash][OSlash][OSlash] ø[OSl ash][OS lash][OSlash][OSlash][OSlash][OSlash][OSlash][OSlash ][OSlash] ø [OSlash][OSlash][OSlash][OSlash][OSlash][OSlash] ø[OSl ash][OSl ash][OSlash][OSlash][OSlash][OSlash][OSlash][OSlash] [OSlash] ø [OSlash][OSlash][OSlash][OSlash][OSlash][OSlash] ø[OSl ash][OSl ash][OSlash][OSlash][OSlash][OSlash][OSlash][OSlash] [OSlash] ø [OSlash][OSlash][OSlash][OSlash][OSlash][OSlash] ø[OSl ash][OSl ash][OSlash][OSlash][OSlash][OSlash][OSlash][OSlash] [OSlash] ø instrument, you diminish it. Backlighting has nothing to do with being a toy. Are laptops considered a toy because they have backlit screens? of users who don't calculate in the dark? would it be worth the extra power consumption? gonna add a light power switch? it's a calculator, not a laptop. i'm all for a more legible display with better resolution. but i don't want to pay extra for a lighted display that i will never use. in all my years of using a calculator (41CV from '82 until about '90 and a 48SX since), i've never said, i sure wish this thing lit up. however, for those of you who feel strongly on this issue, he's your chance. design and build a snap-on accessory light with it's own power source. toy Heck, my little Sharp 256kb Wizard has a backlit screen... it simply enhances the readability in reduced lighting (or reduced eyesight) situations... Megan Gentry Former RT-11 Developer +--------------------------------+-------------------------------------+ | Megan Gentry, EMT/B, PP-ASEL | Internet (work): gentry!zk3.dec.com | | Unix Support Engineering Group | (home): mbg!world.std.com | | Compaq Computer Corporation | addresses need '@' in place of '!' | | 110 Spitbrook Rd. ZK03-2/T43 | URL: http://world.std.com/~mbg/ | | Nashua, NH 03062 | pdp-11 programmer - some assembler | | (603) 884 1055 | required. - mbg | +--------------------------------+-------------------------------------+ like playing tetris under the covers? why does everyone want to turn a calculator into a gameboy. backlighting, color, etc. Why? We want to be smart as (put some well respected scientist) and at the sam time be just kids. Even MS windows comes with some games (gee-wiz look there's a Minesweeper just like the one on my HP :)) I don't think the original poster even mentioned having the backlighting to play games ... careful, otherwise your reply would almost make you look like a disgruntle TI user ;-) (exposition for the uninitiated - hop over to the TI ng, and most all post usually revolve around how can I play games on my TI and that genre). - then again, granted, there *might* be hints that the back lighting could be for gaming purposes ... even so I'd say that it's really a non-issue. I for one have at times wished for side-lite screen as my 48 sits atop a hi-rizer and I'm behind the wheels. Back to the issue ... has anyone who's opened up their 48 found room for LED's on the side of the screen? Peter Khor so far. to ask, but has anyone got a better idea? Ed -- _____________________ edDELfasula@bigfoot.com If you catch yourself before you do any other command which takes arguments, you can undo any keyboard-initiated STO, PURGE, or left-shift var-key (so long as only one variable was involved and flag -55 was clear) via: LASTARG STO LASTARG (lastarg is right-shift ARG) You *can* use BZDIR with variable hiders (as I do myself), by just changing the string BZ VARS PG within BZDIR to something like BZ VARS2 PG, where 'VARS2' is the name of a program which acts like VARS but includes all hidden variable names as well. If you happen to have Hack.lib also installed, then you are all set, because it already includes BZ, PG and VARS2; otherwise you need these additional stand-alone programs: BZ was on Goodies Disk #10/Utils, you may use Bill Wickes' XPURGE from Goodies Disk #7/Utils in place of PG, and you may use FASTA (below) in place of VARS2; hpcalc.org also has all of this stuff. HIDEVARS - Yes, another hider! 1997/11/18 BZDIR improved even further (v 3.2) 1997/11/11 Lightning-fast VARS command 1997/11/15 [use FASTA as VARS2] This is exactly what DPORT does, see: User-RPL Port Tools [Storing variables on Ramcard (Slot 2)?] DPORT can be used with hidden variables in directories (note that PATH RCL recalls the complete current directory, including any hidden variables, and CLVAR clears even hidden variables, in any directory except HOME). You can hide objects in any directory (see above). You could store all your favorite common programs in HOME, and hide them all. In my HOME directory, *everything* is hidden except sub-directories, into which you can not accidentally store anything, because UserRPL STO performs a built-in no-clobber check for directories (Error: Directory Not Allowed) You could keep complete programs in ports, and in user directories just keep stubs which execute the real programs. You could make your own libraries. Re-write for TI calcs? (don't they have a protect bit?) How about chicken soup? Yet another approach: MENUADJ -- protect variables from STO 1997/02/16 ----------------------------------------------------------- Here is a solution I use for important objects: Store the object in port memory, which is protected against overwriting, (and a bit cumbersome to delete).: :0:name STO for (executing of object): left-shift LIBRARY PORTS :0: name or include the object in a CUSTOM -menu, remember to include the :0: in the object name, e.g.: However, I suspect if you have many of these assignments, it will slow down the calculator Gjermund Skailand algebraic 128k port2 that RS it 'programName' use to use? It works like this: Assume TMP is a directory with a subdirectory TMPSUB with a global variable test ' TMP RCL :0:TMP2 STO to recall the variable test :0: { TMP2 TMPSUB test } RCL or perhaps use EVAL if it is a program (object) to be evaluated. With additional RAMcards you may want to use some utilities to handle this. Gjermund try parsing the url thru altavista's translator http://babelfish.altavista.com/cgi-bin/translate? Peter Khor that bad at all! :) Does anyone else want matlab for the hp48? Yes, for years. Matlab is a great math environment. If you ever need to do math that might even slightly use matrices, try matlab. It was originally written to be a user-friendly front-end to the EISPAK, etc. routines from NETLIB. It would be wonderful if the new hp[5]8? included matlab-like notation. I'm not sure how the glue to rpn would work, but it could be done, perhaps with a special stack mode. There is even a FREE (GNU) matlab clone out there, called octave, that HP could use as a starter for ideas. The original source for matlab 1.0 is public domain, if HP is phobic about the GNU licence. (Must be released open-source) In many ways, Matlab is easier to use than the HP. These are mostly convenient notational issues. For example, in Matlab you can create a vector from 1 to 100, by 0.25, like this: a = 1:0.25:100; Another huge strength is the dot notation. HP has a similar thing, misleadingly called parallel processing, which works on LISTS (not arrays). Let's create a function b(a) where b(a) = a^2 ... b = a.^2; This saves so much time. No loops. There is no 1d vs 2d vector distinction (always bugged me) to slow you down. The plotting routines are all based on vectors, unlike the hp48. You would plot this like plot(a,b) There are obvious ways to plot vectors in the hp48 - e.g. using stat routines. For the most part I can get along without matlab on my hp48. But if I could get a spiffy rpn hp handheld that did matlab, I would like it better. A lot. There are several libraries that give you Matlab-like functionality. Stat03 - libid 1042- C. Meland MATFUN - libid 837 - Jon duSaint - Matt Willis I think I know what you want. First, convert both 1D arrays into 2D arrays. The first column is just an index (1, 2, 3, ... , n) and is the same in both arrays. The second column in each array is the data. Scatter plot the first array, then without changing any of the plot options and without erasing the PICT, scatterplot the second array. Lineplots aren't available without third party software, although it wouldn't be very diffucult to rough out a program to plot a series of LINEs on the screen. (at least it wouldn't be hard if it weren't so late... ) Daniel Brooks Jeremy That's modular computation. Put 666666667 in the MODULO variable, then 10 10 POWMOD 100 POWMOD. Seems it is not divisible by 666666667 after adding 5. I see, you mean 10^(10^100). Seems POWMOD does not work with 10^100 as a string embedded in a symbolic (found a bug!). You can prove it anyway: 666666667 is prime (ISPRIME returns 1), hence a^666666666=1 modulo 666666667 therefore by Fermat theorem 10^(10^100)=10^[(10^100) modulo 666666666] modulo 666666667 Hence by doing 666666666 'MODULO' STO 10 100 POWMOD then 666666667 'MODULO' STO 10 SWAP POWMOD you will get -5. (BTW, you must install the MODULO.LIB for POWMOD to work) God! I don't know a sh*t of math! Jeremy You might want to check out the Dallas DS1820 temperature sensor at http://www.dalsemi.com/DocControl/PDFs/1820.pdf -- <-<-<-<-<-<-<-<-<-<-<-<-<-<-<-<-<-<-<-<-<-<-<-<-<-<-<-<-<-<-<-<-<-<-<-<-<-< And it keeps going and going and going and going and going and going and... jkivijar@cc.hut.fi http://www.hut.fi/~jkivijar Ralph If you have GX, then use ERABLE 3.201. If you have a G, then I don't know. Maybe Neopoly. If it exists, it is in www.hpcalc.org cards. Then I bought one and I thought I had a problem, and that caused further discussion. But the problem was my own mistake. Everything is working well. I want to be careful that I haven't left the impression that Cynox cards are a problem. It's working perfectly. It came packaged very nicely and gives me more room (2 meg) than I really have need for, and that is nice. The price was low. The packaging was very nicely done. The card was nicely made. It's obvious these people take pride in what they do. my questions promptly and thoroughly and let me make up my own mind whether I wanted one. In short, I think this is a good card and I reccomend Cynox to anyone who is looking for an alternative to the HP cards. Barry a) the resolution is such, and the viewer is standing back far enough, just far enough, so that they can not percieve a granulation of the pixels, if there's no granulation, then 256 well chosen pixel colours is plenty. the average eyeball (?) will NOT be able to discern a difffernce in picture quality when compared ( in a fair test ) side by side with the same picture displayed with more colours... ( or even more so ( less noticable ) with a different picture ) ----------- :: o .---..-..-..-..-..-..-. . .-. .-. .-..-..-.. .-. `-' `-'`-'`-' `-' `. ^ .' `--^--'`-'`-'`-'' `--^--' Thinking Is The Enemy; Dada Your Way To Happiness! Britts are funny... colours, not colors hehehe Actually, they are all I'd ever dream in a calculators display. I'll even be happy with 16 colors. :o) Yes, perhaps this is true. However, this assumes that one has 256 well chosen colors for a particular image, which is quite different than having only the same 256 colors available for all images. Furthermore, navigating through color-space to pick the optimum 256 colors for a given image is not a trivial task. Of course, none of this has anything to do with a calculator ;) -- Steve Please reply very quickly!!!!! Jo Gijbels. Why don't you try chemlab27 avaialable at http://hpcalc.org ----------------------------------- How about you Jimmy, you an oakman? P.F.Geelhoed@Student.TN.TUDelft.NL new screen and menu, etc. This is a useless example but illustrates the basic problem. I would expect the loop to terminate once right-shift key is pressed, but instead the loop terminates with an undefined variable name, and only when on is pressed. Methinks I do not understand the concept. Prithee, your assistance! DEFINE kpNoShift ONE DEFINE kpLeftShift TWO DEFINE kpRightShift THREE DEFINE kcLeftShift THIRTYFIVE DEFINE kcRightShift FORTY DEFINE kcOn FORTYFIVE :: 0LASTOWDOB! CK0NOLASTWD RECLAIMDISP ClrDA1IsStat POLSaveUI ERRSET :: TRUE FALSE { LAM LUP LAM Running } BIND BEGIN LAM LUP WHILE ' :: DA3OK? ?SEMI DispMenu.1 SetDA3Valid ; ' :: kpNoShift #=casedrop :: DUP#<7 casedrpfls kcLeftShift #=casedrpfls kcRightShift #=casedrpfls kcOn ?CaseKeyDef :: TakeOver TRUE ' LAM Running STO FALSE ' LAM LUP STO ; DROP 'DoBadKeyT ; kpLeftShift #=casedrop :: DUP#<7 casedrpfls kcLeftShift #=casedrpfls kcRightShift #=casedrpfls DROP 'DoBadKeyT ; kpRightShift #=casedrop :: DUP#<7 casedrpfls kcLeftShift #=casedrpfls kcRightShift #=casedrpfls kcOn #=casedrpfls DROP 'DoBadKeyT ; 2DROP 'DoBadKeyT ; TRUE TRUE { { KEY { :: TakeOver Primary DISPROW3 VERYSLOW DOCLLCD ; :: TakeOver Left-Shift DISPROW4 VERYSLOW DOCLLCD ; :: TakeOver Right-Shift DISPROW5 VERYSLOW DOCLLCD FalseTrue ' LAM LUP STO ' LAM Running STO ; } } } ONEFALSE ' LAM Running ' ERRJMP POLSetUI POLKeyUI REPEAT ; ERRTRAP POLResUI&Err POLRestoreUI ClrDAsOK ; Dennis Hallo, I need ,,, better I give an simple example ( it could be more complex): S^2+(2*D/Wo)*S+1/Wo^2=0 and S^2+(a+b)*S+c=0 with them I want to get a=f(D,Wo) b=f(D,Wo) c=f(D,Wo) Which Programm is suitable (Symbolic) to do this? I have Alg48, Erable, AlgB and Neoply. Hallo, I need ,,, better I give an simple example ( it could be more complex): S^2+(2*D/Wo)*S+1/Wo^2=0 and S^2+(a+b)*S+c=0 with them I want to get a=f(D,Wo) b=f(D,Wo) c=f(D,Wo) Which Programm is suitable (Symbolic) to do this? I have Alg48, Erable, AlgB and Neoply. Get it free at http://www.hpcalc.org or at http://i.am/HPlus Can anyone please help. Scott MacIntosh ScottMac99@aol.com you can use the emulator :( Dennis I have completed the ROM Dump, and it was successful. Emu48 appears to be working properly. Specifically, I am trying to run some survey programs that I have for the HP48 through emu48. I can't figure out how to load these survey programs into emu48. Scott MacIntosh ScottMac99@aol.com The devil is in [not knowing] the details Where did these programs come from? What are they called? Did you ever load them into your actual HP48? Are the program files binary or ascii? (this would be quickly apparent after trying to import them using the load object function of the main emulator window) Are they a library, a directory, or single programs? Are you trying to transfer programs from a physical ROM plug-in card? What's up, Doc? - Elmer ----------------------------------------------------------- I use those (such as Erable 3.201 [erabsolu.zip on hpcalc] } Also, why is Emu48 limited to 128kb? How do you add on ramcards? Jeremy 200.245.102.137 Well, I'm starting to program in ML, and I'm looking to everything and anything that I can learn. A friend of mine told me about data compression, and now I'm really curious about how to do it. Can anyone help me out? Felipe Carvalho shoyo@intervista.com.br I can't believe my good Hp48 can't do Conic relations! Even my classmate's cheap TI-83 has a nice program that solve the problems.. put the equations into standards forms etc. I wonder if there's any Conic program out there for Hp48? I looks in www.hpcalc.org, search engines...irc... no luck finding anything..help me.. yourself! ( o o ) -------------------------------------------------oOOO--(_)--OOOo------ Thorsten Lueck --------------------------------------------------- (----( )------- _) ) / (_/ --------------7BE4F78FBE41589666307597 Vends HP 48 GX, Excellent Etat + kit de connexion PC + manuel d'utilisation + diff216rents manuels de programmation + nombreux jeux (tetris, columns, drivers ...) + editeur de texte + Toutes les principales formules des cours de math, physique et chimie de Math Sup et Math Sp216 ( env 100Ko) Prix : 1000 F --------------7BE4F78FBE41589666307597 --------------7BE4F78FBE41589666307597-- HPShell 3.12 Integrated Developing Environment FREEWARE for all HP48 on DOS PCs. ==================================================== * You program your HP48 and have never seen a real developing environment ? * You look for an easy tool for file transfer ? * You need someone managing your backups ? * You would like somebody's doing all the stupid char conversion stuff ? * You don't want to spend money ? * And you want it all in one ? Well, have a look on the HPShell ! After 5 years of continuous development it includes now a huge list of features one would like to have while programming the HP48 or simply using it. This is a (not complete) list of it's features: ----------------------------------------------- * Multi functional Editor with Syntax highlighting, Line Wrapping, HP48 Displays, ... * File Transfer (Kermit included, ANY other protocol is also useable!) * Backup Management * DOS and HP48 Directory Management * Macros, Hotkeys, Toolbar * free defineable Toolprograms * and much more !!! * IT IS FREEWARE ! The HPShell can be found online on the Web at Here you'll find a more complete list of features, screen shots and a list of download-links (including previous versions) spread all over the world. I have also collected a huge list of HP48 related links on the internet.Go and check it out ! Differences to the previous verison (3.11): ------------------------------------------- This release is a pure bug fix. Source Code of the HPShell: --------------------------- With this version the source code of the HPShell is released and for more infos. -- Tom Wellige http://www.wellige.com http://hpshell.wellige.com I'm looking for a set of manuals (owner's and reference) for the HP 28S calculator. I've got a new Cynox card and it is working very well, but i've seen that it is very hard to change the battery, in fact I don't know how to change it without losing memory, because I have to do that with the card into the slot. How do you do it? I lost my HP 48GX manual, if there is anyway I can find one on the Internet? Also, is there any books written dedicated to teach people how to program under HP48GX? I wan to do some small programs under my Jason I recently bought the HP48 to try out (I have 30 days to return). I definitely like the calculator. I was using an HP28S. My question is should I spend the extra bucks for the G+ or GX? I know that the G+ has more memory and the GX takes expansion cards but those cards seem to be pretty expensive. I am assuming that with the G+ I could download third party software to increase the functionality of the calculator. Is the available extra software ware or expansion cards really that good? I am an Electrical Engineering student. Also is there a program or book that has lots of examples to help in learning all of the features? HP48GX + 2 x 128KB RAMCards. In my opinion, you should spend the extra bucks for memory. I am an Electrical Engineering student, too, and there are so many, many good programs... Therefore, I'd suggest a HP48 with more than 128 KB. This means, either a GX with at least 1 extra card or a blown-up G. For a list of manufacturers (and prices) see http://www.hpcalc.org/buying.html I have a GX with 2 memory cards (total memory 512KB). I'm very satisfied with this, but today I'd surely buy a G with 768KB. The G MAY EVENTUALLY cause problems because of the upgrade, but first, memory cards may do so, too, and second, the G is SURELY much, much cheaper. CU Stefan programs I wanted on it. Go to www.hpcalc.org and look at all the programs you want; add up their total memory cost and leave yourself probably 20kb for large calculations, and you'll see what I mean. I bought a new HP48 recently and I had to decide between the G+ and the GX. I chose the GX and I am happy I did because soon I'm getting a 2 mb ram card for it; comething I couldn't do with a G+. Anyway, Good luck and happy hunting! Jeremy a simple bitmap into a GROB for the 48G? QXW0300a) IR transfer between PC and HP48 ------------------------------- There're several questions about tranfering data from a PC to a HP48 and vice versa using the Infrared port. Some theoretical aspects: Pulse with: HP transmitter : typ = 52us, min = 46.8us, max = 57.2us HP receiver : typ = 52us, min = 40.0us, max = 80.0us IrDA: 3/16 of Baudrate = 1/2400*3/16 = 78us Wavelength: HP: 940nm IrDA: 850..900nm Remember: You can only use this interface to connect with the 2400 Baud serial protocol. Tested distance: 5..30cm I tested an Infrared connection with a Tekram IRmate IR-220 for motherboard connection supported by the BIOS on COM2 and with the IrDA interface of a Compaq Armada 1570D Laptop (fixed a resource problem with the factory setting of the IrDA port first). For the first test I used Emu48 V1.10beta5 as software on the PC. I transfered an UserRpl file in binary mode. Both hardware works. As second test software I used MS-DOS Kermit 2.32/A. I tried to transfer the same UserRpl file in binary mode as before. With the Tekram adapter this work, the Compaq transfer fail! In general, the biggest problem is that you get IR reflections from the transmitter to your receiver while transmitting data. Before you receive the answer, the software has to clear the receive buffer. Some IrDA receiver hardware does this without software support, so most cummunication software will run on this hardware. Emu48 has no problems, because the original HP48 romcode is running, I think the HP48 clears the receive buffer after transmitting data. Most popular transfer software doesn't clear it's receive buffer after transmitting, so it mostly depends on your hardware if the IR transfer works. I hope that this posting will answer most of the Infrared questions in this NG. Christoph Gie[Section]elink, cgiess@swol.de http://privat.swol.de/ChristophGiesselink/ I've been using Link 48 (which I prefer to Kermit) successfully on a couple PowerMacs with OS 7.6 and 8.1 to do wire transfers and have no problems. However, recently I tried to use it with a G3 minitower running OS 8.1 and a G3 laptop running OS 8.5 and have had Link 48 timeout on both machines. My GX display flashes processing command a few times before the timeouts occur, which suggests to me that at least some initial communication is successful. It puzzles me that this problem occurs only with G3s. Does anybody have any insights into this? == With the HP-49G, you don't need TGV. They are a text editor like Eden. message : 37E2818C.B8DE9E10@netcourrier.com... -- could CRLIB be made the same? One significant use of libs in my experience was to expand the allowed algebraics. Ah, no I wasn't in USER key mode. I'll have to try it again now that I'm staying in USER mode. (I'm not completely happy with it, but I hope I'll find ways around my problems.) Yes, I've seen this behavior as well (with 1.05). It seemed to have something to do with the inifinite line edit flag. I am not able to reproduce this at will either. I either did a warmstart to get out of it, or a RESTORE, as you did. I tried the following integral on the Hp49: INTVX(1/(1+x^4)^(1/4)) The Hp49 isn't able to compute it, and doesn't give an answer. After waiting for a long time, I had to stop it. The Ti89 correctly calculates it, in a short time. Is there anything wrong in the way I entered the integral on the HP49? This bug is also in 1.10. By the way, how did you get DOT() to do anything in algebraic mode? Here's a different one: With 'X' on the stack (RPN, exact modes), hit LS ABS. 'X' is returned to the stack unaltered. To get '|X|', I need to go into the equation writer. I think I remember this working correctly in 1.05, but am unsure. is is a very useful command if you don't want to use the shifted comma. Cross() command (does this make sense?!) Weird, but on my v1.10, (x-1)! returns the correct result x!/(x-0.) why does the calculator add the 0? and why is it a real not an integer? Also, entering (x-2)!, and (x-9)! (i'm assuming this works for all interger (x-n)!), returns the same answer: x!/(x-0.) now, that is definitely a bug. This is amazing! Very impressive. At first, I was thinking that it had something to do with reals versus integers, but it doesn't. It just gets the right answer, with infinite precision. A new version of the integer powers has been included into the ROM v1.11. Much faster than anything you've seen before VERSION and VER Was just in the Palouse for MosCon about 1.5wk ago. Say hi to the marching band for me, eh? How do I merge the memory of my 1 mb RAM card with the memory of my HP48? I tried the Merge command (purple-shift 2) but it just says Merge1 error: Port Unavailable. Is there something I have to put on the stack to get it to work? I tried doing a PINIT before Merge and I tried it without doing PINIT, but with no effect. I might be using the command incorrectly. Would someone please explain the memory structure of the HP49G to me please? More specifically, what is FLASH memory? How much user memory do I have? I read in advertisements that it has 1.5 Megs of memory. I was under the impression that it was all user memory... My 'FILE MANAGER' says: IRAM 130KB ERAM 255KB FLASH 980KB Home 130KB The main reason I ask is if I press MEM I get 138,197.5. Another question, I tried to back up a program to my computer, and it got to packet no. 2 and stopped. I tried this a few times with the same result, and the last time I noticed an 'Insufficient Memory' message flash briefly on my 49 screen. Has anyone else had this happen to them? You do not have the 1.11 ROM, only a beta which does not include code for numeric evalutation of PSI (only for Psi the digamma function). PSI is the polygamma function. We use the same notation as MuPAD PSI( expr , integer ) where integer is nonnegative. The polygamma function PSI( x, n ) is the n -th derivative of the digamma function Warning for Maple users: Maple use reverse argument ordering for the polygamma function (integer first) He probably means that he wishes the calc would show all messages and menus in French, not just error messages. This is done for software like MS Windows or Office in some countries in Europe, Asia, Africa and South America. If you only get some of the messages in your language it may be easier to just go with the full English version, and that way you don't need to understand conflicting names or terms for the same object or operation. : He probably means that he wishes the calc would show all messages and menus : in French, not just error messages. This is done for software like MS : Windows or Office in some countries in Europe, Asia, Africa and South : America. : If you only get some of the messages in your language it may be easier to : just go with the full English version, and that way you don't need to : understand conflicting names or terms for the same object or operation. [confirmed by Sean himself] We know how the error messages are incorporated in the HP48 (and presumably 49) architecture. I wonder how the menus and so on are incorporated, and how hard it would be to write a HYDE setup for them as well as for the error msgs? (Everybody remembers HYDE on the 48, right?) Does anyone have suggestions on how to send files from the 48 to the 49? I have tried everything I could find,but the 49 manual seems to tried several things on my 48GX: LS I/O, manually setting up Kermit (I may be doing this wrong) and all I get is connecting, retry 1, retry 2... on both calcs. Could I have a defective cable? (It came with the 49) Any suggestions would be much appreciated. The number in the FILER is amount of memory free in that port. HOME and IRAM is the same, or, HOME is _in_ IRAM. That's why you have 980+255+130+(size of vars stored in all ports)=1,5MB memory. MEM reports free memory in port 0. When you enter a complex number and have it shown in polar form on the stack, you can't edit it properly. You can press downarrow to edit and do what you want, but when pressing ENTER to finish the edit, the calc complains about invalid syntax at the angle symbol. ROM v1.10 OBJFIX (posted some time ago by JYA IIRC) fixes binary stuff for me. -- No, put the object into a variable and put the name of the variable on the stack. Then the RCL will work... I'm still having trouble with the transmission, though, JYA (below) tells me I should be transmitting in ASCII, and when I do I get an error in the transmission that I didn't get in BIN, so I actually have taken a (hopefully small) backward step at the moment... independant variable for a parametric equation. I seem to have corrupted it. Now when I try to use it I get Bad argument type. If I type T on the command line and enter I get a number whereas if I type anyother letter I get that letter. How do I get T back? Frankly, I'm not sure WHO the target audience was supposed to be. The calculator design makes it look like the target is the first-time HP buyer. But, the fact that it comes with a 49-to-48 cable (which is useful only to 48 owners) instead of a 49-to-PC cable (which is useful to nearly everyone) makes it look like it's aimed at 48 owners. So it's anybody's guess who the documentation writers were thinking about when they produced the manual. It seems that it knows :-) TI92+ return an answer after 2 seconds, and the answer is: -(ln(abs((x^4+1)^(1/4)-x))-ln(abs((x^4+1)^(1/4)+x))+ +2*arctan((x^4+1)^(1/4)/x))/4 I strongly agree. I'm surprised there haven't been more posts mourning the loss of the MES in the 49G. I don't mind the loss of the 48GX's expansion slots or IR port (the 49G's memory and cable more than compensate). And I would love to have the 49G's speed and CAS. But I rely heavily on the MES, along with supplemental software (like User Equation Library and Equation Browser) to store related equations. As far as I know, there is no comparable feature on any other calculator from any manufacturer. How difficult would it be to port the MES from the 48 to the 49G? If HP doesn't regard the MES as an important feature, couldn't they make it available as an optional download for those who do ? My hope is that the loss of the MES is only temporary, and that an improved faster version will appear in a future 49G ROM upgrade. But I have no idea if this is realistic. SC Beadle (remove qqqq from address when replying) use. It's my understanding that the relevant UserRPL commands in the HP48 (like MINIT, MUSER, and MSOLVR) are simply not there in the 49 (please correct me if I'm wrong). And I'm afraid that my programming ability is too limited to build a new MES from scratch. So it looks like I'll be sticking with my 48GX for now, unless HP, or anyone else, can come up with an equivalent MES for the 49G (hint, hint). Repeat until FIND = { }, or SIZE (= # elements in FIND that are in VARS) Note the HP's MSOLVR made a distinction between between KNOWN vars that were given at the start, or were calculated. With the 49, also, you can solve systems of equations simultaneously, though I'm lacking knowledge on the limitations of the solver. If SOLVESYS gets ported by Bredhel, that might be better. The problem with MSOLVR was that it was very possible (easy) to get situations that led to inconsistencies. The stack can be doen in pretty print -- check your flags. As for MTRW -- would be nice, but I'm not sure how useful it would be, or what impact it'd have on the interface. reading this, it was actualy flag -81 the one that did the trick). Any way of reducing the font used in VIEW? I've got two little programs in my home directory: RPN (entered in RPN mode) ALG (entered in ALG mode) So now, I can play Jekyll-Hyde with my 49. I leave it as an exercise to the reader which is which, YMMV. Impress people on both sides of the aisle. Assign a program to a key, and use it to set the flag. This is what the USER mode is for. Just added these to the EP library (now L1791), and sent then to hpcalc.org. Figured they were useful enough to warrant inclusion. I have the 48 AUG, but not the 48 User Guide. I must admit that it would be very helpful if HP could have released the 48 User Guide in PDF for those of us new to the 48/49 series (at least until they release the other 49 AUG parts)! As has been often said, the 49 manuals aren't really adequate at all. I suppose I might have to buy a 48 User Guide as well :-) The 48 AUG is proving quite useful in understanding the 49, but in places it refers to the User Guide for an explanation (slightly annoying when you don't have a copy :-) ) What was puzzling me with the TVM oddity was the implication that there must be two sets of code (at least) for producing the soft menu display - one (new style) that gives the standard inverted display, and the other (old style) that displays as black on a blank background. I would have imagined that all code in the 49 would be using the same menu display routine - I assume that the TVM routines must instead be using a left-over from an earlier machine. Talking of oddities, anyone know how the 49 has a two colour display - mainly black, but with blue annunciators? I've never come across an LCD like this before (apart from full colour ones, of course!). I was wondering if anyone else had noticed this? I kind of think it is neat. It would really be cool if the RS key annuciator was red. And maybe the alarm one too [ ((0)) ] Todd Still looks definently blue too me. Even when I turn the contrast all the way up, the rest of the screen is completely black but the announciators still look blue to me. Maybe it's just my eyes? Todd The difference is that the main display is multiplexed, and each horizontal row is only driven a small portion of the time (probably 1/64th). The annunciators are either non-multiplexed, or multiplexed by a smaller factor. This results in the different color. There's no real way to take advantage of this, because it is not possible for software to change the multiplex rate, and even if you could, it would cause other display anomalies. (Actually, if the display controller works like that of the HP-28, it might be possible to cause the annunciators to be driven less of the time, which would presumably make them black.) I was wondering if there was a nifty way to increment a storage variable. For instance: I want to be able to enter a vector/matrix and push store on my custom menu and have it automatically store it in 'A', then I will enter another vector and push store and have it automatically store it in 'B', and so on. I thought of adding strings then making them objects etc. but it will be quite a bit of finagling and was wondering if someone here may shed some light on the subject. There is no way in UserRPL to increment a storage variable, you'll have to go around characters. It's pretty simple though - strings are easy to handle. Here's an example; This program stores vars with names 'A001', 'A002', ......and so on in the path specified by path in the program. It uses a global var Co in the directory holding the vars. Another program could be used to reset the directory; You mean ... << 64 'n' STO Note that this will automatically stop after 'Z' :) If you would rather just accumulate one list of matrices/vectors (or any objects whatsosver), place an initially empty list into after the system apps, with numbers just after them. So this would keep the number sequence *after* the system apps. Mind you, I forgot to mention that this code only works if you have a string label in not care to test if the first entry was a string, that should maybe be added to the code for it to be a bit safer ? SWAP + and + perhaps ;-) Now to mod the above (*grin*)... If EP were visible, you could use it thusly: So you can have pre-defined CHOOSE menus, and not have to worry about editting some of the hardcode. You could also use this to provide a BACK selection (just calls the id of the previous menu). You just need to store this ID somewhere (EP should be able to do it). The lowest LID free is still 4 (probably it must be stored in P0...)... most the If the programs crash, that is, produce a screen with a traceback, then it means the file cannot be converted. Due to a feature of the MS-DOS compiler (or perhaps a feature of the programmer), under MS-DOS exceptions are not working, and instead of a message like Unexpected end of file you get an error. This will be solved in the next version. However, if the programs hang, that is, they don't do anything for a long time, then it is a bug. I need more information to solve it, especially the file you are trying to convert. They hang. Can't even bring up a --version on them. I'm using Virtual PC, which works very well for such things as Quickbooks, so I feel fairly confident in its reliability. I'm trying to do this via the MSDOS prompt. http://www.hpcalc.org/utils/menu/ If you're able to get them to work, mind cc'ing me on the command line you used? I am posting this here in hopes that anyone has direct knowledge of the development and/or pending release of any calculator based ml/sysrpl development tools. Obviously compiler/decompile for sysrpl stuff but it is far from optimized and dreadfully slow compared to the two leaders. And I really detest doing it on a pc for other than large projects. Which leads to another question: would anyone proficient in C/C++ be willing to port the new version of rplcomp,sasm, makerom etc to *nix? I have been able to use the old tools with the new entries but I also have not attempted to use any 49 specific commands which are flashpointers. I feel sure the 48 version of the tools would not handle them correctly. I much prefer to do it longer contains the windows virus. Not sure about anyone else but I am thoroughly enjoying my 49 so far. Regardless of the little problems I have spent hours playing, peeking, poking and otherwise exploring it. Not to mention I have soent a lot of time with my old college books seeing what the 49 can do that the 48 could not at the time I needed it to. Very fun... -Al Forgot to mention that the RPN-command stops with a line looking like: RPN() Then you have to press ENTER, and get some rubbish on the stack: 1: :`RPN` 2: NOVAL I do not know why, any help is welcome. Discovery on the HP49G is fun ! an other way to switch in RPN mode after a TRM is to hold the keys B, C and D down during boot time. This will turn on the RPN mode, Simple Alpha and attach the lib 256 Unless you would use emulator, that exacly mirrors Saturn and HP48 architecture. Keep in mind, that that requirement is not supported by HP49 due to the different memory layout, different keyboard layout and (maybe) by addition of new type of variables. Therefore, by your requirements HP49 is already incompatible with HP48 and most of the user written ML tools :-) HP48 would be dead now. Ok, a TakeOver in front makes it work in the command line, :: TakeOver DoDelims $_'' AlgEntry? case SetDA1NoCh SetAlgEntry ; but it is not switching to Algebraic mode though... I dunno, I will keep playing when I have some more time. Got a program, RPN~ALG, which toggles the state of flag -95. Unfortunately, the state change doesn't appear to have effect until after the program exits. In particular, when switching from ALG to RPN, the (previously mentioned) stack contains: 2: :`RPN~ALG` 1: NOVAL I've attempted to remove this using IF -95 FC? THEN DROP2 END after the flag switch, but that of course fails, because RPN~ALG hasn't finished. Is there some way to do this such that the history and result are *not* on the stack when finished? No. The final C means and after testing, Clear it. And that's what it does. Just got my 49G two days ago and noticed something odd. When I'm in Algebraic mode I can enter the Plot Setup screen by pressing and releasing the blue left-shift key and then pressing 2D/3D. When in RPN mode I have to push and hold the left-shift key while pressing 2D/3D in order to get to the Plot Setup screen. If I push and release the lef-shift key I get the little left arrow at the top of the screen, but only an error beep when I subsequently press 2D/3D. Is this a bug or a feature? I'm running version 1.05 This is a feature. On the HP48, LeftShift plus F1 to F6 have already a meaning. In order to stay backward compatible we had to this. OK, but is there a way enter the F1-F6 plot drawing functions in RPN mode ? Try this : You can use this in an expression. This '?' feature is rather handy. I'm wrong! (embarrassed :- | ) As pointed out by Werner, the 41 and 42 returned 1 for SIGN of 0. The 0 for 0 result apparently didn't start until around 1986 with the HP28C. Still the 41 and 42 seemed to encourage the use of SIGN for non-mathematic programming purposes. It returned zero for alpha arguments! A neat way to check data type before the TYPE command. Anyway Bernard has pointed out that the planned change to SIGN is for exact decimal integer zeros, not reals. Transfer the HP48 programs in approx mode and all is well. I should never have doubted Bernard and the rest of the ACO team! has this been asked before? I'd like to enter the FlagBrowser and the Display-Setup screens without going via the MODE screen. What are the correct entrypoints? Forgive me if it has been asked and point me to the answer. You are using integers! Use floats instead; e.g 0. 1. and 5000. Then it only takes 31,5s. on my HP49G. There's also a function named TEVAL to time evaluations on the HP49G. Just consider 1 is true and 0 is false. No problem : meaning) Seems that SOLVE, SOLEVX or ISOL are not able to handle this. This is quite strange, are there special flags to reach the solution or is this unimplemented ? Trying to solve this expression gives the # DE23h error message : Non isolable operator. Yes. But true is everything that doesn't evaluate to zero. And false is everything that evaluates to zero. Also the = operator is == because = is used for equations. That means your exambles would be: The third examble is solving '|x|=3' for x. Now if you have '|x|=3' in stack level 2 and 'x' in stack level 1 and you press solve, you get the error Non isolable operator. But you can multiply '|x|=3' with '|x|=3', expand the result and solve the resulting equation x^2=9 for x. (|x|*|x|= x^2). You get the result {x=3, x=-3} which is also interpretable as x=3 OR x=-3. OK, I played around with the emulator and found out that the HP49 will evaluate X==X, to solvevx(abs(x)=3) (should be x=3 OR x=-3). so on give only two results: true or false (1 or 0 for the HP49). Since the HP49 doesn't know what X is it leaves the term unevaluated. The time. It *asks* if they are both true. Now asking expects an answer which the HP49 can't get because X can be anything. (Even a potato ;-)) So the HP49 leaves everything as is waiting for a better time when X can be determinated. That means if you make further calculations and you get X= now understandable (?). = tells the HP49 that lhs *is* equal to rhs of an equation. == asks only is the thing lhs equal to the thing rhs?. Sure, it works too. No problem! Just another way of doing the same thing. I'm just remembering the FC?C xx SF xx (xx being the flag number) of the 41, so maybe it influences my style sorry replying to my own post : left shift (violet) is 'SHIFT' right shift (red) is 'CTRL' alpha (green) is 'TAB' Hope this helps (also, it's faster) I am trying to get a numeric return for DERIV() according to the instructions in the pocket guide but have gotten nowhere. What am I doing something wrong here. 1: 2 STO x ' DERIV(x^3,x)' ENTER 1: DERIV(x^3,x) If I understand correctly, I should have gotten a return of 12 at this point. I have tried every variation and combination of keystrokes I can think of and can't get a numeric result. I'm in RPN mode using ver 1.05 and flags 105 and 3 are clear. btw I have the clock displayed and I noticed after accessing the ZFACT dialog box the clock is displayed on the graph until screen is refreshed. I wouldn't think this should be normal behavior. The only way, that I know of, to print a graph from the plot screen is to press the ON key and the I/O key at the same time. Make sure you have the ON key pressed before you press the I/O key and the graph will be transmitted to the printer. It's sort of like a print screen feature. However, you will not be able to print the equation screen or the options screen this way. For some unknown reason it prints junk for the equation line etc... This bug has been posted some weeks ago and we have fixed it. As was explained, you can not expect a calculator to invert a 4*4 matrix with 16 variables in a reasonable amount of time anyway. Let me challenge that :-) I can think of problems for which this would be useful indeed, while I'll admit that there are often symetries and/or simplifications (zeroes or recurring coefficients for example) in those. However, if it was to be hardcoded, it would make the HP look like a Porsche, no ? The main work left to compute would be the simplification 'real' computation... Just to stop the speculations: I have a PhD in maths (at the University of Paris XI in 1992) and my job is half-time teaching maths half-time doing research in maths. Not to say that you need to have a PhD to judge one CAS w.r.t. another one. But I raised the question of the math skill because I saw many posts saying that the 49 CAS was inferior to the TI89 CAS (and I believe you said that once too). Ask yourself: * what is the percentage of the commands described in the CAS advanced user guide that you understand? * how far do you understand them? For example, I'm sure many of the people posting here know what integration and factorization means. But how far? e.g. can you explain why the 49 can integrate (4x^3-2)/(x^6+2x^3-x^2+1) as the 89 fails on this one? And they might probably right. You can always ask someone you trust that has the maths knowledge. Maths is not like computer science, it's a very old science, more than 2000 years old. I believe that a highschool student can make a significant discovery in computer science (means a software probably) but that's highly improbable in maths. Have you ever programmed in SysRPL ? Sounds like a stupid question, but: since you say a SysRPL programmer doesn't have to care about pointers and memory allocation the question automatically arises. You forgot that many, if not most, of the objects a SysRPL programmer uses are pointers to the underlying code or data. Try with the word HARDBUFF. It's a pointer to the currently displayed GROB (at least on the HP48;-). Try with {}N and/or INNERCOMP. INNERCOMP returns pointers to the elements in a composite object, not the objects themselves. Here comes into sight the memory allocation theme. One of the important points for an advanced SysRPL programmer is to reduce the time used by the program. In this case it's the garbage collection which should be taken care of. You can do a GC before a memory critical condition (like decomposing a list) or after, but maybe it will be done automatically because the programmer did not care;-) As you might know, the GC may take a relatively long amount of time. So this point should be considered as important for speed. Especialy the CAS routines seem to consume very much memory, which could cause many garbage collections which take a respectable amount of time. The bankswitching doesn't make it faster, too. Additionally, there are 'BANG' type SysRPL routines, which do not work on a tempob copy of an object (or pointer to an object), but insead they manipulate the original object... Examples are PUEL, PUTREALEL, GROB!, and many more. If you give one of them wrong parameters, you are likely to get serious problems with your allocated memory. Just try it (but have a paper clip aside;-) Don't get me wrong, this is not a critic of the CAS. I made my baccalaureat before I had the occasion to get an HP28:-( But after all, I didn't need the symbolic capabilities of any symbolic calc in those days. I don't even need it today...so I don't use it and therefore I can't criticize about it. It definitely is a great piece of abstract software, maybe made for the theorethical math professor, but in real life (I'm working as a professional software engineer for industrial solutions), the symbolic capabilities seem to be not that important. In these days, it's seems to be only a marketing argument: ('Buy the XY, it does more symbolic things than the XX from brand B';-) --------- At least the same things, but remember : there is Flash ROM on the 49, so that in a few month, even with all the best programs on the 48, the 49 will be really much powerfull with new stuffs you will never see on a 48. Do you have such a serious problem with the HP right now ?? or ? I am still using ROM 1.05 (the cable has not arrived yet), but I havent really had any big problems with the calculator, though I use it for diff., stats, integ., graphs... The only thing I am truly annoyed about, is the fact that the docs are very poorly made and the damn cable still hasn't arrived :o/ GETKEY or KEYOB on Joe Horn's Goodies Disks (search hpcalc.org) might be what you're looking for. KEYEVAL is only available on the HP48, when you've installed Erable. Funtion: negative argument (of the key number) yields the standard (non-USER) key definition, while a positive argument will give you the USER key definition (if existing). What do you try to accomplish exactly? Then we might be better able to help you better :-) Peter _______________________________ Do you know the great Frequently Asked Questions? http://www.engr.uvic.ca/~aschoorl/faq/ and the superb HP48 Software Archive? http://www.hpcalc.org to look for *old* HP48 postings see http://www.deja.com Seeing all these complaints, it should be a good idea for ACO to switch the units menus to softkey-only mode, like they did for the PRG/RUN menu in revision 1.10. I assigned a toggle header program to a key. I use this frequently now, as I find it very useful. I'd like to do the same now for the font display. Here's what I'm attempting to do: Write a small UserRPL program that will toggle between Minifont / System Font 6 / Sys Font 7 / Sys Font 8 that I can assign to a key. I have been able to test for and switch between minifont and the other font sizes as there are flags that test for this; (Flags -72, -73, and -80, which determine whether or not the minifont is used for stack display, editing, and EQW display, respectively) I can simply set all three flags to toggle on minifont, and then clear all three flags to return to one of the other font sizes. However, I am stumped as to how I can easily toggle between the 6/7/8 point fonts. I located a command called of something more easily used in a program (like the current font size). the third char, which indicates the font size, with SUB and then switching to the next higher font size (or if I've already hit font 8, to switch back down to minifont). Does anyone know of a simpler method or a command I've missed that will simplify things? This seems like a pretty inelegant solution and I am wondering if anyone out there knows of an easier way to do this. I guess this is a challenge to the group: what is the smallest UserRPL program one can write to toggle between the mini, 6, 7, and 8 point fonts? I don't understand this sudden hurry for removing the cover. Isn't the LCD under that even more easily scratchable? If it gets damaged it will be a lot difficult and more expensive to repair. Maybe you have to buy a new calculator. Now dust and spills can go easily into your calc. Why? Ignacio Do you want the algorithm, or the program itself ? In case your question is how do I calculate the CRC... *If* CRC16 is the CRC used everywhere on the 48, then this might be what you want. I had written one in C to build libraries on my computer In C, it reads as : where you iterate on 'i' through an array of nibbles (4bit) I don't have time to translate it to RPL yet. Anyway, if this is what you really need, just ask. In case you don't know C. ^ is bitwise exclusive or & is bitwise and 0x is the prefix for hexa numbers. correct Unfortunately, I have to agree with you; the 'manual' is worthless, and many people are going to have a bad experience with their first HP because of it. Combined with the bugs, HP will lose a lot of face with critical customers. As a student now in college, I have always depended on my HP and it's exceptional quality. I feel that my money was well-spent when I bought my 48GX; because I knew I was getting the best. My 95LX is ten years old, and still providing faithful daily service as my organizer and constant companion. After having HP and RPN, it is inconceivable to be stuck with a machine of any less calber. My 49 was a sad experience with documentation, but my have made the transition tolerable. Besides, who can say no to 1.5MB of memory in a calculator? Not to mention the speed gains in pull-down menus and all the built in software. My only real problem with the 49 is the screen. The software will be fixed, and already people have found a way to remove the offending screen. How long will it be before HP or a third-party company like Digitalis or Da-Vinci release a screen replacement? I will continue to buy nothing but HP as long as they continue to produce products of their legendary quality. Just my .02 Is there anyway to get JAZZ on the HP49? Yes, I know about hpconv, but I don't have C on my computer so hpconv doesn't help. Does anyone know where I can download a copy of C/C++ that will run on a P133 w/16MB RAM? Please be specific, I'm a first year CS major and just learning C/C++. I am trying to learn assembly and user/sys RPL for the 49. I'm reading Eric Rechlin's book: Introduction to Saturn Assembly Language, is there anything similar for RPL? The manual is useless in this regard (like everything else) And even the manual from my 48 lacks answer: CHANGETYPE will simply replace the first five nibbles of the object with the new type; since Library Data is a data object with a length field following the type, you can only change similar objects to Library Data this way (such as Code object, Array etc). Now, almost the same question has come up a while ago, it was a request to change a list into a Library Data object and vice versa, using the format also internally used by the HP: The # xyz is an *embedded* bint to indicate the library that's supposed to use it, ob1 .. obn is what you really want to 'store' inside the LibData. It is very important that you not use a ROM-based bint such as TEN or TWENTY. Use # A and # 14 instead. Here's the two programs that perform the conversion: (.ASC at the end) * * L2LD * * change a list to a Library Data object * Out : libdat |DOEXT0|Length|---------Body----------| * * beware - the list type is checked, but NOT that the first ob is an * embedded bint * :: CK1NoBlame CK&DISPATCH1 list :: CKREF CODE GOSBVL =SAVPTR A=DAT1 A D1=A D0=A B=A A GOSBVL =SKIPOB D0=D0- 5 AD0EX A=A-B A LC(5) =DOEXT0 DAT1=C A D1=D1+ 5 DAT1=A A GOVLNG =GETPTRLOOP ENDCODE ; ; * * LD2L * * Change a Library Data object back to a list (see L2LD) * * In : libdat |DOEXT0|Length|---------Body----------| * * Beware: no checks are being made to seee that the contents of the * LibraryData object is indeed a list, starting with a bint * * bytes: 47.0 * check: B37Ch * Werner Huysegoms 19981016 * :: CK1NoBlame CK&DISPATCH1 # AF :: CKREF CODE A=DAT1 A AD1EX LC(5) =DOLIST DAT1=C A D1=D1+ 5 LC(5) =DOBINT DAT1=C A D1=A GOVLNG =Loop ENDCODE ; ; L2LD.ASC: %%HP:T(1); D9D20D29512BF8112040D9D2044B73CCD20D30008FB9760143131130D88F9103 0184132E03488B201451741418D34150B2130B21303053 LD2L.ASC: %%HP:T(1); D9D20D29512BF8111920FA000D9D2044B73CCD20C20001431333447A20145174 34119201451318D465D2B2130B2130C73B And we all *know* that problem; and, really, that horse is now glue -- the adhesive on JKH's tape? It is the solutions (both pragmatic and lasting solutions) we should concern ourselves with: provide our own documentation as we find things (as we have always done), and make sure HP gets the message to include this information to save us the bandwidth. Considering the ACO's feedback on here, I suspect that at least *some* pressure toward better documentation is being applied. === diciembre.) ÀY de d227nde te sacas esa informaci227n? (Creo que tenemos tantas ganas de ver algo nuevo que cualquier rumor se da ya por cierto...) And now for the English-speaking people in the world: After quoting some sort of (rather humorous) traditional song, which seems to imply that there will be no new HP calculator for a zillion years or so, Perez-Franco states his belief: he thinks a new calculator should be available in december. Is that a founded rumor? P.S.: Take my advice. Learn Spanish. You'll like it. Ivan (Áviva Bruckner!) Great stuff yes, but pretty cryptic for the masses I think. If they didn't like FORTH... Different thought: For their 58, I wish HP would create a emulation to market with it. You could even try a limited freeware version. Have it run would help bridge the gap between calcuator and computer. Oh, I would also make the ROMS downloadable to the 58. Different thought: For their 58, I wish HP would create a emulation to market with it. You could even try a limited freeware version. Have it run on all pla... -----------------------------------------:: o ooooooooh! i like that...in fact, all the calculator manufactures should come up with emulated calculators to run on the various big-box platforms, so that we can try them all out...and they should be designed to take into consideration the clock-speed of the computer that they're running on, so that the results that they return will be comparable to the actual calculator...!!! if they wanted to be really terrible about it, they could install a hidden time-lapse file in the comptuer that it was initially installed on...so that you can only play around on it for a certian amount of play/and/real time* and then stop working, no matter how you would try and reinstall it...??? -------- * play/and/real time : define a certain amount of time that the user could play arround on the emulation, like 200 hours, OR 6 weeks, which ever comes first. And then, that's it, you would never be able to use the emulator again, unless your either found and deleted the hidden file, or reformated your hard-drive...( OR like a virus, the emulator would copy the time-file to any system disks or back-up tapes that you allowed to come in contact with the original or copy of the time file...??? (!) ----------- :: o .---..-..-..-..-..-..-. . .-. .-. .-..-..-.. .-. `-' `-'`-'`-' `-' `. ^ .' `--^--'`-'`-'`-'' `--^--' Thinking Is The Enemy; Dada Your Way To Happiness! Make it also for the Mac, please. -- Virgil vmhjr@frii.com And the new next generation Amiga as well! Henrik Mikael Kristensen Make it in Java. Then (most) everyone will be happy. Dan -- Just because some of us can read and write and do a little math, that doesn't mean we deserve to conquer the universe. A masquerade can cover a sense of what is real to deceive us; to be unjaded and not lost, we must, then, determine truth. Kurt Vonnegut. (rearranged?) I am refering to Nicholas Wirth's version of Pascal. Not as the language may have mutated to during the passing years. Many of the `Pascal' programmer learned the language on those early versions and still do their variable initializations using assignment statements, for example. That is the nature of the poor programming styles to which I refered. I did not confuse them. You did. Yopu seem to believe that Pascal programmers have adjusted their programming style. I do npt see evidence of such progress. The languages still encourage poor programming practices. Yes, those are much improved languages. That is why I did not offer as examples of poor language models. Since I am not comfortable with them at all, I could not justify using them as examples of good language models. Fortran 90, Cobol 85 and PL/1 also have ``global gotos;'' depending on your exact meaning---Fortran 90 and PL/1 do not permit branching to compound statements. Cobol does not even have a solid idea of a procedure or compound statement. That is not syntax, that is semantics. And, there is a very good reason for making the distinction --- Using C as the example: int A = 1, B = 1; print(Same variables? %sn,(&A==&B) ? Yes : No); print(Equal variables? %sn,(A==B) ? Yes : No); There is a strong need for being able to make both tests. Java is OK. If you are following a linked list, for example; then you might want to test whether you have returned to the same object during your search. Even the first edition of the K&R C language book deprecated the use of the goto statement. And, I have never found them to be necessary. What I do miss (and which I had encountered in PL/1 is labeled loops which permitted such (in a pseudo C like form): for(j = x; j; j &= j-1) { if(i == j) break fred; } } Initialized variables are A & B above, particularly if they are statically allocated and already have their values when executions. Tcl behaves like the original Basic and Pascal languages, variables are initialized by assigning values to them. Notice that I said ``models.'' The language needs to be tailored to the envionment in which it is used. I did. You said, and I quote more exactly this time, both are missing reasonable data types. Pascal, even Wirth's, had integers, reals, enumerations, strings and records - at a time when C didn't have enumerations. How much more reasonable do you need? And spare me the testy responses. If you didn't want a discussion of the topic, why did you post? Then you are referring to a language that hasn't seen the light of day in two decades. _I_ refer to the ANSI and ISO _standards_ for the Pascal language, not some cockamamie proprietary extension. Complaining about features missing in Pascal because they were missing in Wirth's initial implementation is like complaining that Fortran lacks complex number support because you only used the IBM 1130 Fortran II compiler. If you aren't familiar with the standard, fine, criticize ancient history if you like but don't expect to be taken seriously. Regardless of what your favorite feature might be, using assignment statements to initialize a variable is _not_ among the bad programming habits people like Knuth or Djikstra have enumerated. Nor is variable initialization a make-or-break feature of a language. Wirth, at least, continues to disagree with you, even Oberon does not have initialization. If you want to talk taste, fine, but don't try to caste it as one of the eternal verities. As for the utility of the feature, see below. Basic, perhaps. There is only one implementation that respects the standard I am aware of, and the field is loaded with non-standard mutants of immense popularity like Visual Basic, which think of themselves as their own standard. Pascal, however, has done quite well by comparison - but even in its original version, the P4 compiler encouraged good programming by the standards of the time, and modern versions do as well or better by the standards of today. Yet you offered C and C++ as exemplars of good program- ming practices - languages where even experts have trouble writing portable programs. You complain of Pascal being unorthogonal yet C is a mass of special cases, from inconsistant handling of null pointers (which sometimes crash standard library procedures and sometimes don't) to inconsistant handling of data types (where structs can be treated as values in para- meter-passing but arrays cannot). And C++ is just unspeakable, I suspect Stroustroup created it just to make people realize how simple and elegant Ada was. All of them ancient history and none offered as superior to Pascal. But Pascal having a goto is a Bad Thing and C having one is okay? That is the message you are sending. Yes, there is. But there shouldn't be. It's an implementation detail shining through the compiler. In Java, yes, that need was built into it. But it is hardly necessary to make the distinction and few other languages bother. It's an artifact of trying to adapt C-style coding practices to a language that is trying to look like C but not support explicit pointers. So did Wirth's book about Pascal. Nor have I. Useful but still just syntactic sugar. As I said, this is not a make-or-break feature of the language, and they complicate implementation and linkage. And I've dealt with far too many programs that managed to clobber their initialized global space to have much something else entirely under Windows, where even con- stants may not be all that constant. Initialized var- iables are a valuable aid to documentation, but they make little difference in the reliability of software, and absolutely require memory management hardware to be worth the expense of implementing them, IMHO. -- complex numbers spinors quaterions variable length strings of of values including characters of sizes appropriate for ASCII, ideographic languages and Unicode such that _any_ character can be present in the value. (The C language null terminated string do not meet this last requiement.) bits fixed length bit fields variable length bit strings ragged arrays sparse arrays pointers Yes, I know one can ``shoot oneself in the foot with pointers.'' I would settle for controlled pointers similar to PL/1's or C++ references. areas which could be mapped to shared memory I wanted a dicussion of the topic. I did not want any testy responses, such as yours. Be careful of whom you can ``son''. You are not one of my parents. ANSI and ISO _standards_ are mutations of the original language. According to you, those are not ``some cockamamie proprietary extension.'' They are an effort to correct the design errors of the original language. It is the design errors of the original language which are the source of my complaint. Those design errors caused many students to learn poor programming practices. The newer variants of the language, to preserve the usability of the `dusty decks,' permit the use of the older constructs in the old way. The result is no pressure to learn better programming practices, even if one is using compilers which support the newer variants of a language. At least get the system correct. Fortran was developed on the IBM 709 and 709x (x=0 or 4) series of IBM computers (mainframes of their day). I have programmed in FORTRAN II on an IBM 7094-II, which is the first version of Fortran to be released officially by IBM and have done extensive programming in FORTRAN IV on an IBM 7044, which the lower performance and slightly difference variant of that architecture. Ancient history is exactly what I am addressing. The habits formed by those languages are hard to break. The students which used those language to learn how to program are the teachers of today. The sins of that generation is being passed onto the current generation of students. I would like languages designed to encourage strongly not to use those poor programming practices, which is where I feel this thread started. I do expect to be taken seriously. I have had problems with TeX because of its design which came in part because of its Pascal heritage, in particular because of Pascal's failure to have compile time initialization. Tracking down uninitialized variables in TeX on a AT&T 3B1 was nightmarish. C is quite consistent. People using it to implement `standard libraries' and other code were not careful using it. The result is the crashes that you mentioned. The difference in handling structs and [] was and is quite clear in first and second edition K&R and in ANSI/ISO C. It is just that many, such as you, find the specification to be surprising. This is far away from my original topic. It may be a topic close to your heart. You seem to want a language which is verbose, operator and data type poor, which is simple to program in because of those restricted and restrictive resources and obtain the results desired in comparatively simple programs (which is the purpose of student languages). I prefer languages which are terse, operator and data type rich even if this means that one must be careful when using the language. The HP48 user RPL is terse and operator and data type rich. It also, as I remember, lacks a goto statement (I do not have my manuals at hand to check). You are incorrect. I received my first C++ compiler by asking Bjarne Strousup, himself, for it. I have had the pleasure of several conversations with him at Bell Laboratories and of hearing several presentations on the history and design of the C++ language. These included dicussions of his purposes. Ada, to me, was an attempt to correct the design problems of Pascal. It is interesting to note that the United States Department of Defense has dropped officially any requirements to implement systems in Ada. Ada 95 seems to be the last Ada. All are superior by design and features to Pascal. Apparently, I ``shot your hobby horse'' (Pascal). They are all individually more heavily used in production code than Pascal. No, I only was pointing out that other languages also have the goto statement. The question is not whether a language has a goto statement. The question is whether a language encourages a programmer to use the goto statement that a language may have. Fortran until Fortran 90 encouraged the use of the goto statement. Cobol has a `feature' that is even worse than Fortran's assigned goto---the ability to modify where any paragraph proceeds to upon completion in a manner that leaves the fact of that modification quite invisible. PL/1 shares with C and C++ the strong discouragement of using the goto statement that the languages have. Try 2. You do not seem to understand the need to have both the ability to determine whether two references to variables are to the same variable or to two different variables with the same value. Some language attempt to define away the need and the problem by stating that programs which have such a need are incorrect programs because of that very need. INTEGER A /1/, B /1/ C The following is acceptable per the Fortran standards. CALL SUB(A,B) C The following is `undefined' and `implementation dependent' C per the Fortran 66, 77 and 90 standards. CALL SUB(A,A) SUB may need to test for and behave differently depending which of the two calls it receives. I do not disagree with you about it being ``syntactic sugar.'' But, the reasonable, efficient, effective implementation of this useful, frequently occuring construct is a goto statement. The presence of such labeled loop and corresponding labeled breaks and continues eliminates one of the strongest pressures to use goto statements. Once one starts using the goto statement to implement this construct, one is susceptable to using the goto statement elsewhere in even less appropriate uses. I suspect that many of the goto statements int the game in a different thread) are mostly implementations of these constructs. Yes, I have seen Fortran compilers which did not create temporary variables in the following situation: ... CALL SUBR(1) ... SUBROUTINE SUBR(A) IMPLICIT NONE INTEGER A A = A+1 RETURN END That is not a language design issue per se. The language specification should specify that only variables may be passed and that, if necessary, temporary variables be created or the language should provide for restricted, read-only access to arguments so that the compiler can generate the appropriate code for the case that the procedure does not change the argument and checks that the procedure in fact does not change the argument. Of course, a clever programmer can always find a way to circumvent should checking. The early FORTRAN language specifications did not make that requirement clear. Never, ever underestimate the power of stupid people (PERIOD). It only takes one to cause a major disaster---April 26, 1986. Rev 8:11; one of the little villages nearby was Polyn (Wormwood). `One thing that pessimists and optimists agree on is that this is the best of all possible worlds.' Maybe RPL is the best language for HP calculators. Randolph J. Herber, herber@dcdrjh.fnal.gov, +1 630 840 2966, CD/CDFTF PK-149F, 60510-0500, USA. (Speaking for myself and not for US, US DOE, FNAL nor URA.) (Product, trade, or service marks herein belong to their respective owners.) Larry, I could have given. The main reason I suggest it or Basic, is for the s