B48 === decompiling should give something that looks more or less like this in sRPL : :: x<< ID myotherprog ; If the program is standalone, call it like: :: blah ID programname blah - http://www.ti.com/calc/docs/gmtrx.htm http://www.hp.com/calculators/products/buying_guide/index.html http://www.hp.com/calculators/products/product_info/buycomp_sci.html - Metakernel's editor does this. And I believe Stringwriter does as well since MK's editor uses the stringwriter kernel. - HP48 E-zine, the first on-line magazine related to HP48 HP48 news, daily updates ! http://hp48ezine.tsx.org . ____________________________________________________________________ HP48 E-zine, the first on-line magazine related to HP48 I find your easter eggs. Hopefully it is the real parts of the new HP. For those who haven't found it yet, you'll have to wait. Just kidding. Look in the copright section on index page! Enjoy! Just a note to let people know that the HPDATAbase has been updated to include the HP6S and HP6S Solar. The database can be found at: http://www.finseth.com/~fin/hpdata.html For those of you who are not aware of this resource, it is a collection of information on all HP calculators. http://www.finseth.com/~fin -- I have entered this exact program, and it compiled into (d == lowercase greek delta) PRG d SIX d CKN ~ d NULLLAN d SIX d NDUPN d DOBIND d 6GETLAM d 5GETLAM d 4GETLAM d 3GETLAM d 2GETLAM d 1GETLAM d ABND END And when I try to run the program with 1 2 3 4 5 6 on each respective level of the stack, it errors, saying XLIB 2512 45 Erro: Bad Argument Type Why is this? Remember that if you start your command line with PRG, your program wil be executed as soon as you type ENTER. You have to use the PROG word instead.... So your program will be: PROG blahblah END PROG is the equivalent to : ' :: ; While PRG is just :: ; Or in the command line extended syntax: PROG is : ~ PRG END Also, trust me, you didn't select the easiest way to program in sysRPL, use the MASD compiler instead rather than the normal comman line compiler... greek level of cummings - I didn't enter the program into the command using PRG. I entered this: pressed and got and that program, when ran, returned this: and left this: Why is this? Jeremy - Oh, and I forgot to add: It leaves d SIX on the stack. Try Transfile, which is available somewhere in hp Ring. I have downloaded recently hp explorer and it doesnÇt work, and I donÇt know why. noticias 19990518.204233.657165.NETNEWS@TWSUVM.UC.TWSU.EDU... - It's frustrating not to find the equations I find most useful in the equation lib. I would like to edit it. Does anyone know how to do that? It seem to me HP did not intend the user to mess with it. I hope I'm wrong. you're so right; I don't find the equation I need either. And as far as I know there is no simple possibility to add equations to the built-in library. However, there are quite good equation librarys out there. I personally use EQLIB 3.3 by Matthew Willis and think it's really good. Unless I am doing something wrong, both StringWriter and Miniwriter remap most of the HP keys, so I am still unable to access the MTH/VAR/PRG etc. keys as usual. Does anyone know any editor that uses a small font but lets me access the PRG (etc) menu just by typing the PRG (etc) key? For what I know, there is only one that allow you to continue to work with line with a new one much more powerful. http://www-miaif.lip6.fr/geraldenglish/mk.html I know that MK does. How much this helps you, I don't know, but MK allows access to the PRG and VAR menues for sure. ever cutting me in on it, other than an occasional thank you The world is a vast array of interactions, in which many of these go unmeasured (how much do we owe our mothers and fathers?) and all of our HP-related offerings are among them. If we get too caught up in thinking negatively and ungenerously, focusing either upon what we don't get or upon what someone else gets, then we deprive ourselves of some opportunity to gain a degree of fulfillment for what we put into the world, and we cut ourselves off from realizing the element of truth in the old saying that virtue is its own reward (and other similar bits of eternal wisdom, which some of our short lives have not yet encompassed). Have you missed the first of April? ;-) HP states in RPLMAN.DOC section 2.2 (on GD4): A calculator is distinguished from a computer in this context by: + very compact size; + ``instant on''--no warm-up or software loading/bootstrapping; + dedicated keys for common operations rather than qwerty keyboards. + ``instant action'' when a function key is pressed. If HP hasn't changed their definitions you likely won't see a qwerty Do you know the great Frequently Asked Questions? http://www.engr.uvic.ca/~aschoorl/faq/ Being an owner of the TI-92, I have become quite fond of the qwerty style keyboard. I do agree with the disadvantages people have listed, but after using it for a while I found that it made typing on the calc a lot easier and faster, especially since I'd rather make programs on the calculator rather than sitting at a computer. I think my biggest dislike of a qwerty style keyboard is the attention it draws from other people, but overall I felt the qwerty keyboard was worth the trouble. Either way, I have to admit I might consider buying the new HP. When I had bought the TI-92 I wasn't all that aware of the HP48 since no one at my high school owned one and virtually no one at my college had one. After borrowing one for the first time a couple of years ago and visiting web sites and discusion groups on the internet, I was quite surprised by the support for the HP48 by its user base. I think I would be concerned for my safety if I accidently walked into a room full of HP users with my TI-92. I'm not ready to throw away my TI-92 yet, , but if the new HP lives up to its hype, I'll probably manage to forget what a qwerty keyboard is and start practicing a little RPN. Dean - 0 HP wins over another soul!!! using faster, sitting the keyboard was the new no one After sites and the ready to probably RPN. I am currently using M. Kalinowski's tutorial to learn User RPL and I made the exercises in there to practice stack manipulation. But it is not quite enough... Is there any more exercises to practice that somewhere ? Any suggestions ? I guess it depends on your definition of reasonable. RPL, the user language, is very different than C. System RPL, the more powerful language that RPL is written in requires a compiler but gives you much faster programs. It's similar to RPL but a lot more complex. Does anyone know if the new HP calculator? will support 100 based T?? I have my room all ready to go for it. If not I guess I will have to settle for the 10 based 2 that it probably will support : Has anyone had any luck with the HP48/58 ether link hardware?? I can't seem : to get more than 20kbytes/sec on my network. I read somewhere that you have to put the batteries in backwards and pour superglue into the reset hole under one of the feet! :-) I can confirm that the component is capacitor. There is actually a 100K resistor connected between the HC174 reset line and VCC, and this capacitor connected between the reset line and ground. This timing circuit resets the HC174 when the calculator is powered on. I opened several HP48GX's, and all had a capacitor. And I have upgraded more than 20 HP48G calculators to date, always using 100K and 100nF, with no problems at all. I talked to the author of the document that described upgrading the HP48G to 384K and above, stating that hos document was wrong at this point. The mistake was confirmed, but he said he never had a problem in any of the many calculators he had upgraded. This is probably because two 100K resistors will set 2.5V fixed at the reset input of the HC174, which will be accepted as a HIGH or inactive state. As the calculator probably resets the latch during the boot-up sequence, the reset produced by the timing circuit is not really required. Note that this 2.5V level will cause excessive current consumption by the 74HC174 when it's on, according to 74HC series electrical characteristics. Of course, anyway the right way is to solder a capacitor. The exact value is not critical, and 22nF-100nF should do. And now that we are talking about hardware, I'll post some information I learned from upgrading so many calculators, and which I think is very IMPORTANT: In most of the upgrade documents around there, the card detect pins (pin 37, counting VCC as pin 1) are connected, through a switch to ground or the card slot VCC in order to disable/enable memory protection. When switches are not used, these pins are connected permanently to the card slot VCC. Note that the card slot VCC is not powered when the calculator is off. Also note that the card detect pins are inputs which recognice three states: low, high, and open. In order to detect the open state, the HP48 chip contains a voltage divider using two resistors which set these pin's voltage to 2.,5V when they're open. If you connect these pins to the card slot VCC, then when the calculator is OFF, you're powering ALL the circuits connected to this VCC from the internal calculator VCC (which is always on to retain RAM contents) through the voltage divider. The internal resistors that form the voltage divider have very high values, and so the current that can flow from permanent VCC to the card slot VCC is very low, but still enough to allow about 1.5V in the card slot VCC WHEN THE CALCULATOR IS OFF. The consequence is excess of current consumtion when the calculator is OFF. The current is still very low, and will result in still very long battery life, but it is high compared with current consumption if things are done properly. The SOLUTION is obvious: connect these pins to permanent VCC instead of card slot VCC (no mather if you do it directly or through switches). As I mentioned before, I've upgraded more thatn 20 HP48G to 384K and 768K. When the card detect pins are connected to card slot VCC, and the calculator is off, measured current consumption varies from 10uA to 100uA. When the card detect pins are connected to permanent VCC, current consumption was not measurable with my instruments, which means it was below 1uA. - http://home.nordnet.fr/~bdarcy/perso/buronux.jpg The key layout looks like a TI! It has a Y= button and the good ol' extra large ENTER button is gone!!! In fact, the top row is exactly like the TI-89, except that it has an extra 2D/3D for the extra key. I never thought those HP6's would be a precursor to the new HP's color scheme...which personally I'm not greatly fond of. However, maybe this is just a reaction to change, and I suppose changing from the conservative isn't all that bad, but just not shiny sky blue! If the new HP's operating system is as poorly written as the TI-89, I will be very sad...because from what I gather they apparently are promoting the equation writer and an APPS and MODE button. And BASIC for programming? So far, it appears that HP is trying to gear this calculator more towards the educational market than the 48 was, which makes it somewhat less appealing. - To me the HP49 looks great. I think it is a very nice improvement over the HP48. I will be very tempted to buy one. I own a HP48 which I unfortunately do not use often. Too sophisticated as a tool for my present needs!!! No time to play with it at the moment. I think the HP49 will fit the bill as an advanced math tool very well. The user interface we are told is significantly faster. Face it, for most users it will be pretty irrelevant that ML programs still run at the same speed. For what most users are going to do with it, it sounds plenty fast, after all the HP48 is not too bad either. But thanks to JYA it seems also more open to programming out of the box and there will be enough memory to go with it. There are some necessary concessions to sell as many units as possible. Costs is also among them. On another note, we are getting close to run Mathematica on a handheld. But even when this is going to happen, it will initially be very expensive ($1000 ++). Possible that you are going to replace your trusty HP49 with one of those puppies a few years down the line. - OLD: 3 batteries. 131 x 64 screen. RPL. RPN. New: 512K ram! 1MB flash! Languages Assembler and Basic Intereseting: Supporte la majorite des programmes tiers RPL, jeux, logiciels, allicatifs connectez-vous a www.hpcalc.org It means what you think it means. So how many colors will they make it in? http://home.nordnet.fr/+AH4-bdarcy/hp49.shtml What does it say about cable connection on page 2...? +ACI- CONTENU Adapteur et cable de connexion HP 49G a HP 49G ou modeles de la Serie HP 48 +ACI- well, i think my french is good enough so that i understood almost everything in the new hp49 pdf spec file. first off, it seems that with the exception of the exterior color Sa couleur bleu glacier apporte bune touche de fantaisie, hp hasn't messed anything up...looks like the calc was pretty well designed. Nouveau design industriel-does that men that the metal case is gone??? rpn is still there :), and the option to switch btwn rpn, algebraic and text modes seems really useful. 1M base ROM and 512K base RAM looked pretty good to me...but what about expandibility...will there be a 49GX? algebraic and symbolic math functions seem to be more abundant...but little mention of specifics big questions: how fast? how much? backwards-compatibility (programs) with 48g/x? specific symbolic and algebraic abilities? at first glance, the 49g looks like a real winner...but we will definitley need more specifics. The top 10 things I didn't read about in the PDF: 1. Is IR still there? They mention a cable for 49G-49G. I assume IR is gone. Oh well, as long as the cable is included. 2. Assuming no expansion card, are they making an expandable one? 3. How fast is the PC connection? Is it still 9600 baud? 4. Is the ram bank switched? 5. Which algebra routines added? I see Jordan form... How are symbolic matrices handled? As lists? 6. How much fun is the spiffy stack? Configurable? 7. Can I download the flash ram onto a floppy, and install some other, completely different operating system? 8. Do libraries live in ram or flash ram? Are libraries still supported? 9. What Mhz does it run at? How many multiples faster is this as compared to a 48g/gx? 10. Why basic? Whose opcodes is assembler in? What about sysrpl? - I've been reading the post's about the new hp49G and it's memory, somthing called flash, what is flash memory? Is it better than what the hp48 has? In my GX I have a memory card, is this what flash memory is and if so, how can rom have flash memory? ry - the flash rom allows the user to upgrade the rom in their calculator without having to buy a new calc. For example, if HP found that there was a bug in some part of the rom (a common occurance in newly released calculators), they could put a new rom image on their web page, and users could download it and upgrade their calculator so that it didn't have the bug anymore. In addition, HP could also use this feature to add new capabilities to their existing (flash capable) calculators. TI has been using flash roms lately, and while I haven't ever found a huge problem with my calc that had been fixed with a subsequent rom release, I think it is a very useful feature that, if used to its full potential, lengthens the life of a calculator substantially. jack - Flash ROM would also be a useful medium in which to store libraries. - It also says to look for the Other Interesting Sites link from http://www.hp.com/calculators, which is non-existant. I agree. The calculator looks similar to the TI-89... so similar that it make it hard to believe that HP would copy the TI design so much. In addition, as I said in a different post earlier, there is a Y= button and a whole bunch of other TI type functions. The enter button is small, which I've never seen on an HP calculator... In fact, has HP EVER made a calculator with a small enter button?? On top of that, it has a catalog button, an APPS button, a MODE button, The cursor has the same symbols as the 89, it even has TBLSET next to TABLE on top of the far right soft keys, which is exactly like the 89. Would HP really not utilize a larger screen? I am still not sure, if this is fake or not. Examining the picture shows, that at least the display is faked/edited. Prove: - the grey of the display is of uniform color (RGB 140/162/141) - this won't ever happen, if it would be a real scan. - the _very_ lousy border of the display plastics (white shimmering) I have no idea. It's 22.5 on my 48G :) - As for the sqrt decompile, I have yet to figure this out either. A fix for result. You will get the code for the sqrt itself then. As for the sqrt(real), it is %sqrt (that is % and the sqrt symbol as one token). I too have bee annoyed by the way it handles reals and the fix is to temporarily disable the table when doing sysrpl programming. I do this by simply renaming STARTEXT to STARTEX so the kernel cannot find the table. THen userrpl works fine. Granted these are annoying characteristics but I have learned to live with them. The one thing I miss in MK is a sysrpl debugger. -Al - FYI: Miniwriter doesn't leave the menu-keys available. - http://bofh.via.ecp.fr/~sam/openhp -- Sam. - 003.jpg: Am I reading this right? Copyright HP 1999 006.jpg: one has a blue screen, one looks black. ??? five-line stack -- which one?? are there hard plastic covers underneath these? 008.jpg: is that an IR window? 009.jpg: are the buttons on the sides? general: keys - plastic of rubber? when available? cost? - Matt - I have to agree that the lighter case would have more market appeal but I think a darker case such as we currently have is really superior. It's far less reflective and glaring. Also, a light case would make the screen seem darker. I find myself wondering if this isn't the real thing from HP simply because it seems to be more designed by marketers than engineers and that's the direction HP has been moving in for a while. And, if it is for real, will there be an expandible 49GX to follow? - Maybe the arrows do make sense. They seem to indicate that with the blue shift they have the same function as unshifted, but with the orange shift they mean top of page, left of page, right of page and bottom of page. That could be for the text editor. I did a translation of the brochure using AltaVista translations and it didn't translate very well. But it seemed to be saying that there was asm programming support built in. Any ideas on that? Wouldn't it be great if Jazz was in rom! - Well, ignore that last post. Seems I should have been using 3PICK3PICK...sorry about that :-) Jeremy If i ever forget to capitalize a proper noun, forgive me. i have ee cummings in my ancestry. Have a good day, and good luck in your endeavors! - The specs are: *powerful symbolic calculus module *step by step functionality to see the steps in solving linear systems, (for example) * Working modes RPN and algebric * good text editor * good symbolic performance (manipulation and solving of symbolic matrices) * Statistics * Finantial math * Intelligent caracteristics that allow the user to customize the work environment *it works with 3 aaa batteries Memory 1.5 Mbytes - 1 Mega Flash ROM 512 KB user memory Operating modes *3 modes RPN, algebraic and alpha-numeric *4 screen types *good and flexible text editor Mathematics (besides what has been said already) *factorisation, substitution, development, simplification and solving of linear systems, matrices... (in fact, all the things the hp48 does) * analysis (limits, derivatives, integrals, differential equations, taylor series, etc) * Trig * Exp and logs Statistics *basic descriptive statistics with good entering facilities *Aproximations, series interference measures, hypothesis tests, confidence intervals Enginnering and sciences *127 units *units conversions *database with 40 physics constants Graphics *15 different graphics parametrics, functions, polar, conic,diferential, statistics * function analysis, radix, zeros,extreme points, etc *2d and 3d with zoom Physical Caracteristics *industrial design *protection cover *high resolution screen 131x64 *dimensions: 18.0 x 8.1 x 2.9 cm * weight (with batteries) 266 grams) Accessories * Conection kit PC/MAC * carrying case in leather *Projection Table * datalogger Customisable * Object oriented programming * RPL, assembly, basic * supports most programs done on RPL. check www.hpcalc.com. - in this group - I will *NOT* believe that the piece being toted on sam's site is real until it is confirmed by J-Y, G-S, or Dave Arnett. Please, one of you three, save us from the immaturity of some people in this group. - http://bofh.via.ecp.fr/~sam/openhp/ I will stick with the 48GX Indifference until it us - I can't believe that HP would go with Basic as a supported language on the HP49. That would be too much of a good thing. If you just remember that the French brought us Ada, I'm sure they are pulling a prank on us. It's like their way of putting us down - a misinformed, misguided sense of snobbery. Like most snobbery is. Which is actually to bad since the old HP Rocky Mountain Basic is one of the finest programming languages ever to come along. It has all the best features of many of the other languages that support well structured programming. For a scientific programming language it was great. Many people would argue that any language the supports 'GOTO' can't be great (dykstra), but my take is you don't have to use it. I don't! No one like to debug spegetti code. One of the great things about that Basic is you could run it compiled or interpreted. You haven't lived until you write code and execute it immediately (no compiling it, linking it, and loading it), then when you get it debugged you can compile it to run faster. Plus you get things like code profilers for almost free. All languages should allow for compiled and/or interpreted execution. The Boeing 757/767 flight controls were developed on HP mini-computers running HP Basic. We would run the already know elements as compiled code in conjunction with control laws that were still under development as interpreted code. That way we could change controls law algorithms very quickly - no compilation etc..... - just RUN! Sooooooooo what does the real HP49G look like - http://pobox.com/~gentzel/ - URL? Specs? - I must point out that this will cost a lot more than a calculator and the software will be sold separately, so the final price will go up even higher. As far as I know, Maple CE won't be ready until Y2K (that's Year 2000 for you non-abbreviated-friendly folks :) It will also ONLY work on HPC and HPC Pro machines. Quite a mistake since the big ticket is in the Palm sized PC form factor (hint, hint; hint of things to come). Carlos. - I agree. If the new HP is still based on the old Saturn architecture, then the TI89/92 remains by far the most powerful calculator available. Based on what I have read, the HP49's symbolic capabilities are not much better than those of the HP48. It seems HP has lost interest in the calculator market, and is not prepared to spend on REAL innovation. Alfredo Todini - If this is what HP has become, from the company that actually made good stuff a few years ago, i would rather go TI. If HP can make this calc, think what will happen to the user base. Think for a second what will happen to tech support and things like that. This calc looks much worse than a TI-89 in my opinion. TI-89's are at least black, not metallic blue!!! I don't mind the rest of the calc, just the case. Its the blue. I'm sorry, but after this release, HP is going to die out as a major calc maker. However there is a silver lining. It hasn't been made official yet. Once it is, one way or another, then I will make the decision - The question in my mind is this: given that the perpetrators of this joke are HP48 users, fanatics, etc - *why* would they do this given that it can only hurt HP & HP calculators? The overwhelming response in this newsgroup so far has been disgust, which might push people to give up on future HP calcs altogether. [0] I might be proven wrong by an announcement on hp.com, but I doubt it. Here's why: a. the Alleged HP49 (AHP49)'s LCD is flush with the top of the unit, *unlike* all other handheld HP products I know of (including the Logic Dart) b. in http://bofh.via.ecp.fr/~sam/openhp/006.jpg, it looks like one has the old-style HP48 LCD and one has the new-style HP48 LCD (ie, one is more black, and one is more blue). Supposedly on the HP48 they changed for a reason, why try both in the new one? c. Why would it use the same LCD as the one engineered 10 years d. In every HP-designed calculator, the numeric keys have had the same position relative to the basic operation keys (+-/*). This is not the case on the AHP49. (eg, / is not next to 9). e. Having left-shifted functions (ie, blue) on the Fn keys conflicts with the ability to have left-shifted commands in the CST menu. f. In 1990, it was reasonable to design new Si for a calculator - perhaps the right low-power CPU didn't exist. There's *no* reason for this now, there are lots of low-power, fast, CPUs available. g. In http://bofh.via.ecp.fr/~sam/openhp/009.jpg, it appears as if the bottom half of the case's top corners are not as rounded as the top half of the case's top corners (ie, by the top of the LCD). Why are these pictures so fuzzy - I can take better with a $7 disposable Kodak. h. in http://home.nordnet.fr/~bdarcy/mag/hp49.jpg, it appears as if in the HP logo, the tail of the p angles too sharply, ie, it goes too left. -z -- # Zak Smith, MScEE, Unix Geek, INTJ. zak@computer.org http://apollo.demigod.org # Ft Collins, CO. More laws are *not* the answer. - e: ( left everything in its place although you should read most of this sub thread) stop judging me and *think* if it is possible. speed=3.75Mhz or ~3.9Mhz ( mine is 3.75 :( that is 4M cycles per second , which suites you fine. ok. given that cos(f) & sin(f) are computed once and are constants, we need only : read each point's coordinates, ('a' cycles) multiply with cos(f) & sin(f), ('b' cycles) add the corresponding terms, ('c' cycles) write the coordinates ( probably *not* doing perspective) ('d' cycles) write the (14-1)*14*2=364 line segments ('e' cycles) let's say that the CPU works only for this program( no interrupts etc) if you can squeeze all a+b+c+d+e cycles in 1/6 of a second, i declare you master of assembly. especially for the line drawing thingy but : think: the fastest line drawing algo i've experienced is the one by detlef mueller , which might be lightning flash, but cannot possibly do 2184 lines/sec... if you think that they ( the ACO ) did something faster on a machine that is not capable of 690 mults/sec, u need something white and tight =] maybe a trick , because for 6-7 pixel long line segments we might not need to draw a *line*, maybe something other :/ p.s.: *anything* insulting/offending is provided for amusement ONLY no offence is meant, and i'm sorry i used the f word in the beginning.son of a bitch , i'm good!!! - Motorola has all registers (8 address and 8 data) full 32 bit. It's arithmetic unit is 16 bit wide. It's data but is 16 bit wide. It's adress bus is 32 bit wide (altough model 68000 uses only 24 bits) but the memory is flat and continuous without bank switching. Now those are some example timings: Multiplication of two 32 bit into two-register wide 64 bit: 74 cycles Division of two register long 64 bit by 32 bit register: 144 cycles Addition of two 32 bit registers: 10 cycles Comparison of two 32 bit registers: 10 cycles Clearing 32 bit register: 10 cycles Negation 32 nit register: 10 cycles Branch taken 16 bit 18 cycles Branch not taken 16 bit 20 cycles Jump 32 bit 16 cycles Subroutine call 32 bit 32 cycles Load or store 32 bit register from/to absolute address 8 cycles Load or store 32 bit register from indexed or rel. addres 16 cycles We are talking here about 32 bit (8 BCD digits) timing. - A couple of us already tried telling him about the emulator, last night on IRC. He refuses to believe it exists. I'm afraid this one will deny the existance of the machine until he owns one (and even then, I suspect he'll crack the case to look inside). - http://bofh.via.ecp.fr/~sam/openhp/014.jpg The LCD isn't flush -- there's a shadow. on the right hand side. Fro a development model, I would expect this. My father once purchased by pieces an Atari 1450 XLD (remember that?). The case was a re-engineered case from a 1200. I wouldn't blink to see variance in the parts, purely for demonstration purposes. *THAT* is a good question. Look at the space shuttle's computer systems. Very un-aesthetic, by my standards. It's just Evil and Wrong. I would really dislike HP for doing something like this. Perhaps. You're assuming the same functionality in any CST menu. Irrelevent. Form does not determine existance. Considering this whole style is uncharacteristic of HP, we can't use odd style in a particular area to say it's not for real. Ref my response to (g) - You will be, more than likely. I am not seeing this. This has already been explained. Again, see a previous post. Obviously they changed some things! They may have changed the menu you are referring to. HP's mistake. Give me a break! You can't tell that nor your next comment from a picture! The person who took the pictures is not a professional photographer, and the lighting conditions were not perfect. I think that's part of it, but I think there's also a lots of things that aren't just right in how this whole thing has gone on that makes us suspicious. I mean, for example, what business reason would HP have to publicize a new product before its release? Tried and true HP differentiators being thrown by the wayside (ex, enter key, IR, etc)? Lack of features expected for so long? If it ends up being true, then I guess lots of us will still wait, hoping this is another HP38 and they are still working on a new scientific, engineering calculator. I'm still using my SX, actually. If I could magically change my SX, I'd probably just do the following with a wave of the magic wand: . N times faster . better resolution, backlight display . M times more memory . physically smaller & more durable, but with the same keyboard usability . more battery life . more data types . perl interpreter (*just kidding*) The problem I have with the 48SX/GX is technological and practical obsolescence: first, if designed now, many of my above wishes would be fulfilled just by the technology available now, and second, support, parts (replacements), and software will be increasingly hard to find, besides marking the end of the focussed scientific engineering calculator (The market of *my needs* doesn't exist anymore.). -z -- # Zak Smith, MScEE, Unix Geek, INTJ. zak@computer.org http://apollo.demigod.org # Ft Collins, CO. More laws are *not* the answer. Fine. The Space Shuttle has requirements that the HP48 does not: . Verification/certification of systems must be rigorous enough to ensure system will not fail - it is mission-critical . must interact with other systems in a 3/5-redundant system (if I remember correctly). . must be rad-hard . I'm sure there is lots of red-tape involved in upgrading those systems Consumer systems don't have these requirements. Would you be satisfied if you had to use a PC with a 286 with 640x480 16-color VGA? Of course not. Sadly, no. There were apparently too many problems with the HP48 being forbidden at some exams because of the IR port. It is a compromise. -:: o this is clearly absurd...it would mean that you could no longer use it with the Thermal Printer... just how would/could you use it with a printer...??? could one of the smart people that have started a list of why the 49 is a FAKE, A DIRTY ROTTEN FAKE & HOAX & FAKEY-HOAX...please post the first 200 or so reasons why it is FAKE AND FAKE, and then the rest of us can add items to it... i'm really very surprised that anyone that uses a programmable calculator thinks that it is real...??? It has approx. the same dimensions as TIs and Casios, if it's what you want to know. So no, it can't be a real hp48 hidden in it :) -- Sam. - He was speaking of the numerical algorithms in the 89. They were mostly programmed in C and are *very* poorly optimized. For instance, the HP48 takes < 2 minutes to invert a 30x30 matrix, while the TI89 takes in excess of one hour, if it finishes at all. - Yes. It also took less than one second to compute 200! with all digits (integers are now a different type than real) on the hp49 (don't know about TI's speed, though). Then, less than 5 seconds to do a prime factorisation of the result. Sam. -- :wq! - This has to be a mistake here... I would be surprised if it took that many bits! - You're right, I shouldn't have said that without knowing. Perhaps someone with some knowledge of the TI's inner working might enlighten us about how much memory an integer takes ? -- Sam. - Integer values in memory are stored using up to 614 digits. It means, an integer can have any size up to 614 digits (307 bytes BCD). - Well, you'll see that. I'm not a CAS _expert_ but I studied it a lot, I know what the Saturn is able to do and what HP managed to do with it, and I know what the MC68k is able to do and I saw what TI did with it. Sam. - Why don't they just make it the very best to begin with? Suppose nobody erable, etc...) and make them even better and build them into the ROM. This way the calculator is superior out of the box. It can't just be better hardware, it needs far improved software as well. At least have a sort of 48 emulation mode or something like that. - And what was this thing The design of this new HP goes against almost all the things that have made HP calculators so great in the past. In fact, we have come to think of these features as the defining property of an HP calc, features that the new one lacks. The new HP seems to lack the coherent design that the traditional HP's were built around. Components seems to have been just thrown in as was convenient. If I'm not mistaken, the new hp doesn't use the RPL language by default. What a mistake! This has been one of the main reasons -if not the sole reason for some- HP's customers have purchased their calcs! Also, it appears they have taken the liberty of shuffling the HP48's great keyboard layout ( yet another feature that attracted HP customers! ) to such an extent that it has lost all its previous functionality! Come on HP! How could someone be so stupid as to totally screw up something that used to be a great selling point of their products? If it ain't broken, don't fix it! As for the hardware the new HP has (or lacks) I'm disgusted. In almost every recommendation about the features the new calc should have a *bigger screen* and significantly *faster processor* have been the constants, everything else has just been icing on the cake. With the new TI calcs (TI-89 and TI-92) being on the market for over 3 years now, and the readily apparent demand for speed and graphics that runs through the *whole* graphing calculator user base, how could HP possibly foul up on these 2 points? That's right. The new HP has the SAME SCREEN RESOLUTION and the SAME PROCESSOR SPEED as the old ones! What a joke! How long can you possibly keep decades old technology alive especially in the face of your competitors' vastly superior hardware and the general level of today's technology at comparable prices? The new HP seems to be a strange chimera indeed. WhatI mean is it appears HP is trying to satisfy 2 users bases each of which bought their calcs for completely different reasons and uses them for completely different reasons! They're trying to go with the current trend of calculators designed for high school to beginning college students and at the same time they're attempting to satisfy their previous (you could almost say sole here) customer base of engineers, scientists, and upper college students. It's obvious these groups just don't mix! You can't put features designed for both groups on the same calc because they're mutually contradictory. What you get its a piece of crap that makes everyone unhappy! It's as if they've (unsuccessfully) tried to merge some of the features from TI's calcs with the features of their own, which as i said wouldn't be successful because they were made for 2 different types of people. I don't know of any self respecting engineer who would want to put his fingers on the new calc. Likewise, I don't know of any high school student who would rather purchase the new HP given the existence of other calcs such as TI's that were designed specifically for them and that have much greater functionality and usability then the new HP. It looks like HP is in jeopardy of losing hold of their niche in the calculator market ignoring the fact that their grip was already starting to slip. How could HP and the people at the ACO let down HP's loyal and trusting user base that's been around for so long ( and that given the steadily growing rate at which technology is advancing, won't be getting any smaller either )? Shame on you! As for me, I'm staying with my trusty old HP48GX that has served me well these many years. As for the new hp, I refuse to buy such a disgraceful piece of garbage and I encourage you to keep your distance to. Anyway, I've just recently become serious about HP48 programming and have been working on an ML replacement for the internal 3D plotter. At first I was disappointed that the knowledge and skills I had gained would go to waste just when they had come to fruition due to the advent of the new HP, but not any more! There's no way that I or probably any other HP48 programmer would want to waste their skills and their time trying to program such a piece of trash only to see their work fade into oblivion because nobody else wanted to buy the new calc! - This is the first time any info at all, including release date, has been released by HP in advance. Normally we find out the day it ships. That makes me kind of suspicious. - Can you be more specific what is akward in myriad of 'C', 'Pascal,, 'Basic' or 'ML' tools for Motorola 68000 available from old MacIntosh, AtariTT, Amiga up to proffesional compilers for industrial CPU that are in overhelming majority based on Motorola 68000 family CPU's. That also include tons of ready to pick up scientific and advanced math libraries on the market. The only thing missing is the TI specification of it's memory maps, OS API and header files but it is changing. TI web site already published primitive memory ROM map for TI92/89. HP didn't do it for 48 for the last ten years. - As soon as Microsoft can release a decent op sys that isnt a GUI based on a dos shell(dos the wave of the future) then windows might win agaist windows. But none of that will even matter when OS X is implemented. Oh...what group am i in? - Sorry, but that's not the right place for OS advocacy. Especially if you're trying to defend Windows :-) Robert - Not really. You can load Motorola 68000 ML compiler into TI92. PC is needed when you want to program Motorola 68000 ML using Intel CPU or you are really using high level languages with a lot of optimization and features, that usually require high power CPU, lots of RAM and disk space. I don't belive, one can 0.5 MB RAM and even much worse HP keyboard and it's tiny window. For serious proffesional work you do need PC. - Welcome to comp.sys.hp48! I hope that you have the opportunity to try one of the fine calculators designed by the Hewlett Packard Corporation, their products truly are wonderful! Although the stack-based system may at first seem a little unfamiliar, I am sure that it will grow on you. Quick calculations are just SO much easier in RPN! I was a little afraid of it at first, too, because everyone around me was using _that_other_system_, but after a few days I realized that, yeah, RPN was just so much more natural, and just plain _RIGHT_! It is true that the TI92 may be a little bit faster at graphing and some other things, but I think that you will find that once you learn a little bit more math, graphing with the calculator will not be that important to you. Instead, you will find that the intuitive and uniform interface of the HP just plain enables you to solve your computational problems more quickly. The built in programming language and list processing are just the thing for performing quick calculations calculations on small-ish data sets. You know, the type of stuff that is too tedious to key in, time after time, yet does not warrant the effort involved in writing an assembly language routine. All of the HPs built in functions and extensive library of physical constants really shine in this situation. Speaking of Physical Constants, you may also find that the HPs built in handling of units is very useful. I find it a very handy tool for performing elementary dimensional analysis, and as a _sanity_check_ on my calculations involving measured quantities, i.e., if the mass comes out in s/m, I KNOW I did something wrong! (I AM sorry that HP patented this technology, All calculators should handle units, and software patents only serve to stifle the growth of ideas.) - Yes, HP memory managment is superior to TI. Also true, that most of the software written for TI are games. It seems, TI did not developed proffesional users group yet, that would give more scientific software to the market. On the other hand I write myself 99% of the programs I use on my calc my so I really don't care about what is on the market, but I do care, haw fast my own programs run. It is not true that you need a shell for TI89/TI92+. Both those calculators have built in object ASM that allows to store assembly program in a variable and run it within OS as a regular programm. The problem is, that TI for a long time did not publish the format of the ASM header, and never published OS API, nor memory map. Therefore most of the programms written are using shell that was created in the past for the TI92 model. The 92 model did not allow ASM objects and did not have command that jumps to the assembly level. All that was changed with the release of the plus module. TI89 has already plus module functionality built in. Lately TI for the first time publishd on it's web site sketchy details about TI89/92 internals with the promise of more (and book) to follow. Jack - As far as I'm concerned, WarioHP is 350Kb source code and it was written on the calc ** itself ** using powerful and fast String-Writer program. Now I write my ZeldaHP source code on the PC using HPDEV because this project is just a bit bigger (just a bit :p) than WarioHP. -- Julien http://sunhp.zoy.org Samuel Hocevar a 216crit dans le message ... You obviously haven't used MK, EQStack, Java, or any other AGROB enabled stack replacement. Obviously :-). That was response to so called crap ROM of TI. If TI ROM is crap then HP ROM is mountain of shit. You need to BUY tons of software to make HP match TI. On the other hand TI really is a lousy OS. It uses BCD coding for floating point that is rouhly 10 times slower for Mototrola 68000 that binary coding, it uses 256 bytes (in extreme case) to code single integer number, it always treats expressions as a symbolic text. Despite such arrogance and abuse of the CPU cycle time TI usually outruns or matches HP. Yeah, trying to match TI by using extreme ML software for HP give false impression of how good HP almost is. You seem to forget, that TI is like that for it's base price out of the box and still leaves room for the user who is willing to code in Motorola 68000 ML to leave HP in the dust no matter what. Ja I don't think this statement is supportable. Numerically the HP48 will beat the TI-89/92. When ML is used, the HP just wins even more quickly. http://promotesite.webjump.com - I personnally uses the card for exchanging programs between 48GXes : it is much faster than IR transfer ... moreover, the cards are a safe place to back-up programs.... One positive point :) RPN is not easy to use ? Wow, that's a good joke : of course, i admit that at first, it is a bit hard to understand how it works, but when you're are accustomed to it, it is more powerfules and easier to use than Algebraic system. now. I would like to address someone who could answer those questions you can't. My apologies for my first messed up message, confused for all the information ... Even though, I'm glad with its new specs, + the processor speed performance (while keeping 48 compatible), + this new CAS*, + that the screen size went 10% up (for what I saw), and + the overall calc size still gets into the 48 case. I would really have liked the thickness and length reduced, and the resolution increased. Perhaps on the next project ! I thought that was the purpose of algebraic entry, so the calculator could do the work, and not you. It makes more sense to me to let the calculator handle all the order of operations. The only reason parenthesis don't exist in RPN is because you're working them in your head. Also, it seems that the logic of RPN is often justified with English sentences. By the same rule, I can make prefix notation appear logical: add four and five (+ 4 5), when as far as machines are concerned, it makes no sense, Okay, what do I do with this plus sign? For relatively simple problems, you're probably correct. I was a TI devotee until the HP-41 came out back in 1979 because of their (at the time) overwhelming memory superiority. After all, the TI-59 had 100 data memories back in 1977 while the HP-67/97 had only 26. It stood that way for two whole years. However, for complicated problems, I don't think you ever rely on the calculator to do the work for you. I teach business math and it's painfully obvious that it's easy to get lost in the mound of parentheses. Just try the sample mach speed formula from the HP-67 manual on an algebraic machine and you'll see what I mean. If you're keying these things into a graphical environment and seeing the equation as you go, what difference does any of this make? None. However, to quickly work through a problem, RPN's approach has been proven to take fewer keypresses. - Thats why its an extra function - you dont have to use it but its there if you want How, exactly? It is supposed to have all the rpn functions of the HP48. No, you dont get erable, etc - you get a CAS which (reportedly) does all they do and more. That is what many people in this group have been after for ages. Your opinion (though you dont seem to be exactly alone in it). At least you can stick it in a 48 case. Ok. The spec is higher than any other HP calc, and but for the display seems to take the TIs apart. Really sounds like a calc for 8 yr olds???? Maybe the students can see the better product when the pro is too arrogant to see past its colour. Yeah, I guess a calc which does what you want with min effort yeally is no good for some people. The little beast is not perfect, but it does look like an extremely good calc. I really fail to see how adding so many feature makes it a bad calc. - I'm glad they used the Saturn. Sure, I'd like to see it running faster. But using a Saturn CPU and the same screen size means we can still run a lot of the software we're used to running. My disappointment is that there's no IR to use with my little HP calculator printer and to tranfer stuff with. I used the IR a lot. Also I'd like expansion capability. Barry - Cool. That's a big plus for me. References as to which operations will be 10 times faster and which operations will be 100 times faster would be helpful. You may also wish to mention (actually, you probably won't) which operations will be slower. You don't get owt for nowt. Does one have to make this choice manually every time one switches on the calculator? If so, it's a big negative for me. I just want RPN all the time. Your post makes me suspect that the HP49 may in fact be real. Oh dear. Perhaps I should stock up on a few HP48GXs and memory cards as spares before they become end of line. - thank you for your informative summary/description of the prototype. i was waiting for someone with real information from the openhp meet to confirm technical details (and authenticity). if you are still actually following this thread after some of these ridiculous debates, please comment on the cpu and speed of this new machine. is this really the saturn processor identical to the hp48? are there or will there be any other hardware enhancments to this new model? is it possible that the 49 is running at 4mHz currently but planned to be running at significantly higher clock rate for the production model? i am having a hard time dealing with what seems to be rom software and memory improvments on the 48. will hp really release a machine to replace the 48 taking no advantage of changes in the silicon and handheld device industry since '92? - Come on Jean, from the above & your original post, it almost seems that someone was dangling a pink slip in front of you as you decided on your post! RPN IS EASY TO USE! Taken, I've NOT seen one or used one in real life so I'll give the 49 a BIG benefit of doubt when it comes to usability (OS-Keyboard-Screen). BUT taking away the IR just to make it sell to schools? I for one use the IR both for printing and transfering data to another 48 and/or notebook. Memory card - I've had my share of crash while on the road, and the only thing that saved me was a backup copy of mem on a RAM card. And of course - Well, it seems my TI92+ was actually a good choice to replace 48G. It seems for the next 10 years TI92+ will be much better machine than HP. HP49 is already on the edge it's hardware can deliver. TI has tenfold room to grow with it's current hardware and flash ROM. - [snip] This is the bit that I love the best. BASIC, hah. The Trades Descriptions Act[1] would have something to say here. It is nothing more than user-RPL functionalised so that DISP etc. have a function form. Where's the good old print command and other Basic stalwarts? Still, putting the word Basic on the specification sheet and advertising flyers will make sure that it gets the necessary tick in the box under programmability to ensure that schools and colleges are able to recommend it. - P I concur. RPN is much more intuitive after only a few days of use. I still remember the bright, sunny, happy day in 1987 that I purchase my first HP and started trying to use RPN. I won't go back to AES, I refuse. -- Charles Perry PE eepe on #Pipes remove x in address to reply - That's plain bullshit ! I used rpn/rpl exclusively for about 18 years (HP41 & HP48, several) and about a year ago I went back to AES, convinced I would hate it (I hated the one on the HP38G and still do) but now I love it, when it is properly implemented ! I hope it is so on the HP49G, nobody could tell me or show when I was in Paris on Friday : there only RPN addicts there... Both modes have their own pros and cons. If both are available, then it's great and I'll surely buy an HP49G as soon as it is available. Bob Pulluard - AES will never take fewer keystrokes than RPN. Also, RPN is arguably more logical: How can one expect to perform an operation on two operands with only one operand inputted before the operation? only - Is my logic flawed? Granted that more complicated operations involving many levels of parenthesis would require fewer keystrokes on the HP, but I would quickly get confused. 3 Plus 3 3 3 (Now what the hell do I do with the numbers.) Plus Algebraic entry seems to me the way I would write a problem, the way I would think it out, and the way I would solve it. - Yes it is. 3 ENTER (or SPC) 3 + Four keystrokes. RPN will NEVER require more keystrokes than AES. Yes they do. That's alright. You can use AES. Personally, the parentheses confuse me far more than the RPN. But don't make false statements about RPN. - Indeed, one of the `complications' of RPN is the behavior of the be stacked. function keys (+, -, etc.) cause 'tmp' (if there is one) to be stacked, and then the operation is applied to the top of the stack. So, simple as it is, RPN isn't all _that_ simple... - Or: 3 [SPC] 3 + [four] But now let's add five to the previous result: Alg: + 5 = [three] RPN: 5 + [two] Ah, now RPN is fewer, no? The parenthesis-free nature of RPN also saves you from having to type parentheses; another free benefit is that you can see the value of all intermediate results with RPN, whereas with AES a parenthesized expression is often computed invisibly (on some calcs, no intermediate results are ever seen at all). The natural collection of results onto a stack also saves keystrokes: 3 4 * See the product (12) 5 6 * See the product (30) If you now want to add the previous two results, just press + (with AES, you have to go back into a recorded list of previous answers, if provided, and select the old one, also still need + and = ) With the HP48, several of your previous results always continue to remain visible as well; if you want to go way back, then you can start the scroller (stack browser); previous results can also be juggled in many more ways using this stack browser. All these benefits went into the choice of RPN for early calcs like the HP35, back around 1972, and the unfamiliarity of RPN led HP to publish booklets with titles like Enter vs Equals to explain things like the above. When formulas and equations, rather than simple calculations, begin to enter the picture, algebraic expressions are indeed handy; the HP28 and HP48 have always offered such expressions also, although they automatically compile them, as soon as entered, into a more efficient RPN internal form for execution, so that they never need to be re-interpreted in the slower manner while running applications. Previous posts seem to indicate that a new model will offer more flexibility than ever before, following the trend of past years (in HP financial calcs) to let the user decide whether to use AES or RPN entry (without the slight penalty of two extra keystrokes for AES which the HP28 and HP48 have); so why doesn't the Great Debate over RPN vs. AES start winding down now? know that this new calc will be released some day in the near future. But all the details we know now make me very very sad. Its hard to express my fellings in a language thats not my native language but I'll try. The Saturn CPU was invented more than 10 years ago, the HP48S/SX came out in 1990 but the HP28 and even earlier models used it I believe. It was a great phantastic incredibly cool new CPU for pocket calculators THEN. But nowadays we have tiny little PDA machines that use contemporary processors, *modern* processors. And those don't have problems with the trade off between power supply and power/speed. I could go on and on. What I want to make clear is: obviously the HP calc R&D folks had to decide: invent a new, a proprietary HP calc CPU or take an existing, modern CPU from Motorola or another company (that would have ment a complete new system design including the operating system) -OR- reuse the old fashioned, for these days unbelievably slow 4MHz Saturn CPU (then they wouldn't need to invent a completely new system design and can use the existing OS). Well, they chose the latter option which I don't understand. Of course, everybody new that the HP48 could be much faster if the OS was rewritten in ML. I'd like to know what the inventor of the RPL philosophy, Bill Wickes, would say if he knew (I'm sure he knows!) they have rewritten everything in ML just because there was no time (HA! What did they do all the decades???) and/or no money to really think very well about a new, modern, concept that holds for the furure. Well, the 49G is there and we can't change. BUT -- and this makes me happy -- they can't do it again! They can't be faster than ML. So the next HP calc that is to follow the 49 HAS TO BE a REALLY new machine. Display. Unbelievable! Still 131x64 pixel. Impudence. But, of course, just a logic consequence of the decision to just make the new calc be a Super-HP48 machine and not really a new one. It may be of incredibly great contrast. Well, it may be color TFT it may be whatever -- as long as its 131x64 pixel it's garbage. This is absolutely not adequate for today's demands in graphical mathematics! Keyboard. Rubber keys! Desaster! I can not believe that a company like HP that used to make real good quality products has decided to make calculators with rubber keys and go in one line with the really really very very cheap chinese mass product calculators. I never thought that I have to say that some day but: thank goodness, there is TI that still builds calculators where the keys are at least made of hard plastic. (The key pressing feeling on TI calcs is bad, I know, but at least the keys are not rubber!) this? Ok, maybe you say HP didn't build the 6S, they just printed there name/logo on it. But after that: it's a HP product. I can read HP on it's top. And it's keyboard is garbage. That much for latest HP quality! So, after all, I think this HP49G is really a very sad milestone in Hewlett-Packard's calculator history. What a pity that they jump on each train and make their new products like all new products: colorful, r&d under time and market pressure, production has to be as cheap as possible (therefore rubber keys), we can see at any edge that the big caption for this HP49 project was save money, save money, save money. Really too bad! I still hope that there will be a new HP calc some day in the future that really deserves the name HP on it's top and that is really a new, a great new invention and a new advance in calculator system design. A calc where we all can really be proud of. Does anybody remember those days when the HP-48S/SX came out? Did the HP fans scream and scold and were upset about the 48 like they are today about the 49? I can't remember! That should make HP think ... - On Sat, 22 May 1999 12:35:20 +0100, Bruce Horrocks Sounds like something similar to the data logger that you can use with the Upper Level TI calculators. Vernier Software http://www.vernier.com/ has an interface box that attaches to the calculator and various sensors( Light ,Pressure,Voltage,Current, etc) that can be attached to it. We use their Universal Interface Boxes and various sensors in our Physcis,Geology and Astronomy labs with PCs and MACs. - - Look at this page. http://home.nordnet.fr/~bdarcy/hp49.shtml IT's where WAS the Hp49G picture. Now, we only get an error 404. Perhaps they are updating the page.... I don't know. - After reading the spill about the elimination of the IR port from the new calculator, I went to the ACT website to see how they handle the 48GX. I imagined it was forbidden, as it can do as much as the forbidden TI-89 with the help of a few programs. The 48 series can be used as long as the IR port is covered with masking tape. This brings me to a few other questions? Why is it that the 89 isn't allowed on the ACT, yet the 48GX is? HP users have had the advantage of having a hand-held, symbolic (with software) calculator for years, yet the TI-89 users aren't allowed to use their calculators on the test. This is just ridiculous! Granted, I took the test with a scientific calculator, and I thought it was easy, but now I can see where a symbolic graphing calculator would have assisted me. If the HP 49 is allowed on the ACT, which is obviously not going to happen, the 89 would be allowed. As a TI-89 user, I feel like I've been let down, not by TI, but by teachers and test administrators. If HP targets the 49 at high schoolers, they may soon be in over their heads when the realize that they can't use the HP on the ACT. The SAT, on the other hand, allows the 89, so I imagine the 49 will be allowed also. - I have no complaints about being competitive in the market and doing what has to be done, I mean 1.5 M is a BIG improvement, better contrast is GREAT, increased functionality and all the way down the line...great. Even if it is an incremental improvement, so be it. No problem. But why do they have to make it look so RETARDED? I just can't see this philosophy of design that says this looks 'modern', it looks assinine IMO. I SINCERELY ENVY the people who can overlook the design of the case, or those who LIKE it, I would be perfectly happy with the new HP if it just LOOKED better. Some people might think this is superficial but I REALLY CAN'T HELP IT, this BOTHERS me!! I want the design to properly reflect the functionality. Here you have the most sophisticated math tool on the planet looking like something designed by a group of kids from Nickelodian. It's criminal. One thing I LIKE about it that i've seen some criticism about is the color. I like it. But I want to see it look square, straight, precise... after all, it is a math tool. It should reflect precision, not look like a blob. Once again I ENVY people that can overlook this or like it. PLEASE PLEASE TEACH ME HOW. THIS REALLY HURTS ME!!!!! CHANGE THE DESIGN OF THE CASE. Functionally it IS a vast improvement in most respects over the 48 AND it is in a better position to compete in the student necessary evil to live with. the fact that full RPN is still in there is GREAT. Nobody should complain about the algebraic capability as long as RPN is still present. (Of course perhaps the keyboard is better for the algebraic side but I for one understand that HP MUST COMPETE in this market. But PLEASE don't make it look so stupid.. At least let HP be a trend setter in this area. I would like to see a case that is clean and functional with color (I LIKE the blue!) used for aesthetic appeal but nothing gawdy and gratuitously distracting. No offense, but people with Down's Syndrome are not going to be using this calculator. - I thought you could run Erable from port 2? I was just about to ask if you can have tables in covered ports, but I guess i don't have to anymore... I am just about to order a cynox 2mb (David - how long did it take to ship?), but I am worried that I will not be able to run everything with it - I am wanting to run Java, Alg48, Erable, and JazzLight (for now at least)... I know that Jazz can't be run from a covered port, but what about Java or the others? I don't intend on getting a 128k card either. -- Slot2.. which is What - I am referring to the JAZZ tables and the MK tables. I have both because I still use Jazz for debugging my sysrpl. hem. - Sorry about the strange title but I was reading a post complaining about the default to algebraic entry mode and then I read another about the lack of expansion slots and then I decided to put my responses to both in a new post. The reason that HP made it default to algebraic mode is [I'm sure] so that when dumb ass salesman[1] in electronics stores turn it on and enter 2 + 2 = in front of a potential customer, it works. If the 49 defaulted to RPN then it would all go pear-shaped pretty quickly. It's a bit like watching a Windows-salesman trying to demo an iMac. Which, in a roundabout sort of way, brings me onto my second point regarding expansion slots. The 48SX had a demo card, which as it's name rather obviously suggests, ran a demo of the calcs' features. This got round the dumb ass salesman problem since he only had to bring out the shop demo machine and turn it on. The demo card did all the selling. Unfortunately, the 49 has no expansion card slots, but... ...(pay attention here HP) why not ship the 49 with it's lovely flash RAM stuffed full of a super, gee-whiz demo? The salesman turns it on and the demo runs. Easy. Early on in the quick start guide are the instructions on how to purge the demo and free up the memory so the customer is happy once the machine is home and being used. [1] A generic concept: sales people who, due to the rapid turnover of products, haven't really got a clue about what they're selling and only get a half-hour intro to each product anyway. 1276 686623 - Not mentioned in the PDF spec. sheet; mentioned loads of times in c.s.hp48 as a feature for the next calc. - Good luck. I've been wondering...how does an operator work without both operands? Just like you logically need all the wheels of a car to drive, you need all the operands to operate. - T Okay, so I haven't illustrated that RPN requires less keystrokes. Like I said, I haven't deviated much from algebraic entry. With the use of a calculator's memory, storing frequently used keystrokes makes up for both systems, correct? As for the above, you get the first number, learn the operation so you know what to do with the next number. - wrong here. you call that LOGIC?? what does a machine think when you say: 3 + 3?? machines are sequencial... if you do this in seuquential order, the machine sees 3 + <= what do I do!!! - - I suppose I see that. I'd still prefer that th calc do the work when I enter the equation. If there comes a time when every second is crucial and the fewer the keystrokes the better, I'll turn to RPN. Generally though, as I'm reading word problems, it's easier for me to write the equation in algebraic form, but I can see how RPN could come in handy there. - - I understand now that with RPN you don't have to hit enter if the operand is what directly follows, however, please don't try to make the HP something it's not. The above sequence is FOUR keystrokes, unless that's a different number in RPN. - that done'... That's a good attitude and I agree with you, but then all the 'bad' things that get done have money behind them, too. embodiment of I'm not so sure HE's the embodiment of it. Perhaps he just knew what compromises had to be made to establish mass appeal, and was ruthless enough to make them (at whoever's expense). voice, the the womb...and boils down to along with the everything they the smarter you Roseanne is just a confused individual who needs to explain her highly unlikely fortune in a way credible to her own limited intelligence. And I think there's something in all of us that would like to believe that all we say, think and do, no matter how assinine, has been sanctioned from above. Fame and Fortune make this illusion possible and it takes a certain amount of sincerity in a person not to start to believe it on some level. I like what Jason Alexander said when his salary went to $500,000 per episode: In the real world, no one deserves this kind of money except perhaps nurses and teachers, but Hollywood is not the real world, and when you see billions of dollars floating by you naturally try and get some. dumbness. transition. evil Yes, evil may be a necessary evil, but I think all evolution actually takes place by the play or struggle between 'good' and 'evil'. If everything were evil there would likely be total destruction and if everything were good there would be no struggle to improve, because there would be no necessity to. - began this 'discussion'? I just said that it was not the right place for OS advocacy and added a humorous (at least it was intended to be has nothing to do with the subject of the original message and HP48 at all. Robert BTW, I do care which OS is better and believe me, I'm not the only one who does. If you are still willing to argue, please write either to my private address or to comp.os.linux.advocacy (what is even better). I don't think that the rest of HP48 society is interested in discussions about superiority of some OSes or childish comments. - Ok.. starting to get the hang of this.. the MK Filer allows me to RUN stuff from other ports.. Ok. I have a game hah.. *ok... ok.. I gotta have SOME entertainment for English class* Anywho. It's the TetrisGX from HP48.org's site. It's in a directory format and when on another port but home, it just outputs the program. How can I RUN it from another port? Is there another version of Tetris I should try? I like this version, that's why I'm using it. There's also other programs that use DIRS instead of libraries. What about them? - Sorry if this is a stupid question. I have been using my 48GX as a number cruncher for years and am just now getting interested in testing its limits so to speak. Assume I want to do the following: use the MetaKernal card (slot 1 I think) and Alg48, Erable, and a few other programs. Assume I have the metakernal card and a 512K ram card. How do I accomplish this? Bank switching? I need to keep quite a bit of the internal 128K free for complex matrix arithmatic. Any help will be appreciated. I am sure lots (most?) of you will laugh at the simplicity of this question. I bought my 48GX when I started working with complex matrices that were too large for a 28 to invert. I have done little other than number crunching and a few userRPL apps on it. I suggest Erable 3.024 because the newer version is buggy, and 3.024 can be spread between port 0 and covered ports. Assuming you use Erable 3.024, put KERNEL.LIB in port 0, ERABLEG.LIB in port 2 (or 3, or 4, or 5) and put Alg48 in one of the covered ports (ports 2-5) The calculator handles all bank switching for you. This setup should keep at least 80k free in port 0. Hope this helps, Jeremy - :: ... FALSE grob1 grob2 #x #y GROB+# ... ; Comments... Change FALSE to TRUE for GOR (instead of GXOR) grob2 is put into grob1. You may want to make sure grob1 is in temporary memory (CKREF). dan Does over clocking the processor cause it to use more energy and thus produce more heat? If it takes more energy HP could solve an old compliant about the 48 only using three batteries. Design the 49 to use four batteries and I would not always have that extra battery left over from a four pak lying around in some obscure drawer. - -Make two separate families calcs, one aimed at high school, one at pros -For the high school, make it look nice (like the TI-89, but PLEASE don't give it a neon case), give it Algebraic entry, with the option of RPN, give supported ASM, fast processor (10-15Mhz seems right, but even an overclocked Saturn is good), good display, symbolics like the 92+. -For the pros, make the HP58, give it a new processor, (NOT SATURN), around 20Mhz and let it emulate the Saturn -Double the resolution of the display 262x128 and to ensure compatibility, create a Saturn emulation mode where each of the old pixels represents 4 old pixels. Due to the enhanced power of the new processor, this should be quite feasible for games and other progs -Build ALG48, Erable and full symbolics into the HP58 -Give it solid construction like the 48, plastic keys, you could change the color a little to black or maybe even navy blue -Keep the RPN and RPL, and set RPN BY DEFAULT -Give the G model 512KB of RAM and 2MB Flash ROM -Give the GX model 2MB of RAM and 2MB Flash ROM, along with the ability for expansion card which can raise the total to 16MB RAM -Do this, and you'll kick ass -Try to satisfy both, and you'll fail Go one way or another, but keep separate calcs targeted at separate groups DON'T TRY TO PUT THEM ALL IN ONE CALC -- - Indifference indif69@usa.net Children in the front seat cause accidents, accidents in the back seat cause children. - I own a 48GX and I want to download some programs from www.hpcalc.org . In describing the file sizes of the programs, the site says that the first size is the file size and the second size is the size of the program on the HP. As an example, the program MATRIX v1.2 written by Arnold Moy shows two sizes, 114KB/13KB. . My question is how much of my memory on my calculator will be used? I haven't upgraded my memory and I want to make sure that I have enough . - a) does the 'grey scale' on the 49 work like the grey scale on the 48, producing no more than four levels of grey? b) what does the 1.00 key refer to...??? over the EEX key...? - no more ? I already saw 12 grayscale pictures on the hp48. Ok, they were a bit flickering, but 8 levels are just fine. Anyway, the new screen is far better, and though I only saw 4 grayscale pictures, I can tell that the contrast is really excellent, and I am quite sure 16 color grayscales are possible. - is that hideous color of the 49? Now I know: it's frozen hamster butt! If I had anything to do with it (which I don't), THAT would be the official newsgroup description of the color. And yeah, you're right. This whole mess is very, very sad. I'm just hoping that this is the 'student version' or something, and we'll get to see the real version soon. But I doubt it. still laughing way too hard, Doug Burkett Eaton OH USA - are seriously out of whack. It certainly looks different than an HP 48, but I think hideous is quite a stretch. I'm far more concerned about rubber keys (especially if they really are the cheapo rubber membrane keys with conductive ink, those are terrible), and the screwup on the ENTER key. - I don't know, I think the G-series color scheme is hideous... I like my brown with orange and blue highlights personally... I'm assuming that it's high-quality rubber keys. (I'll visit a store and check out the keys on the 6S, I assume they're the same tech.) The 'ENTER' worries me the most though. I think that's what has most people freaked out, the fact that this new HP looks like a TI/Casio with it's keyboard layout. I, for one, like the perfect grid pattern on my 48. (And my Sony remote control, for that matter.) No wimpy + shaped arrows, gotta have that inverted T. - interesting way of putting it... what do you suggest? if HP goes away from the Saturn architecture, they have to start from scratch with the OS and all the software. this would mean they would have to rewrite any code that is related to the LCD.. has been done in the 49.. but I have no clue how.. J-Y said that bank switching was eliminated.. but that is impossible because of the way the Saturn (the same one used in teh 48) address memory. As I understand it... The address pointers are only 20 bits... this gives a maximum address count of 2^20 which comes out to be about 1,048,576 addresses. Each address is 4 bits so that is 4,194,304 bits... divide by 8 to get the number of bytes... 524,288... divide by 1024 to get the number of kilobytes... =512k... as is, the Saturn can address half the ROM (half is 256k) and 256k of RAM at any given time. If it needs to address any memory location outside of this, it has to bank switch. - AND HOW MANY YEARS HAVE THEY HAD TO DO THIS?!! AND HOW MANY YEARS HAVE THEY HAD TO DO THIS?!! AND HOW MANY YEARS HAVE THEY HAD TO DO THIS?!! Its 1999. ... where's my next calc coming from? -- Chris Fortin, cfortin@bbn.com [Ph.D EE, SC-Elder:parared] ó274Á`Á274ó [D ownQuestion],[UDoubleDot][UDoubleDot],[DownQuestion][CapitalUHat]2 74[Do wnExclamation]`[DownExclamation]274[CapitalUHat][DownQuestion],[UD oubleD ot][UDoubleDot],[DownQuestion][CapitalUHat]274[DownExclamation]` [DownEx clamation]274[CapitalUHat][DownQuestion],[UDoubleDot][CapitalUHat] 274[ DownExclamation]`[DownExclamation]274[CapitalUHat][DownQuestion],[ UDoubl eDot][UDoubleDot],[DownQuestion][CapitalUHat]274[DownExclamation]` [Down Exclamation]274[CapitalUHat][DownQuestion],[UDoubleDot][UDoubleDot ],[Do wnQuestion][CapitalUHat]274[DownExclamation]`[DownExclamation]274 [Capit alUHat] Just think - blessed SCSI cables! Do a big enough sacrifice & create a +5, blessed SCSI cable of connectivity. Lionel on ASR. - Power dissipation of a CMOS circuit is given by the formula:Vdd^2*Frequency*CapacitiveLoad where Vdd is the supply voltage and frequency is frequency with which the outputs switch. Do the math, if the clock frequency were increased by 2x or 3x, the power consumption would go way up. Because the Saturn CPU is really an ASIC made by NEC for HP, if the ASIC die size were shrunk and other changes made, then the CPU could run at a lower voltage. This would give a very significant increase in battery life and allow the CPU to clocked at a higher frequency (safely). However this was not done, and pushing the CPU beyond the rated 4MHz is overclocking. HP wouldn't put out an unstable product, although they don't seem to mind putting out one that's gaudy and worthless. What Cynox and whoever else does is pushing the CPU/calculator beyond its designed limitations; even though it works, the current CPU speed is the techinical safe limit. Vasant. remove nospam to respond. - I'd love to have the GameBoy Color TFT display in it, to bad HP didn't check the electronic market more carefully. --:: o i just saw an item in the technology section of CCN the other day, where 'somebody' (?) had come up with a full colour display 1 inch square with a resolution ( i forget exactly ) that was higher than my CRT, which is 640 by 480...! they were claiming that it was easy to view with a magnifer that came with it...??? wouldn't that be cute, an HP58gx Wristwatch...! HP isn't jerking anybody around! Has HP been in on _any_ of these discussions? Why would they, when this is a hoax made up by someone in France? - Ok.. finding the dev. of 3rd root of X.. HOW? It outputs all this strange garble*. *garble* = Anything not in French or Standard English. =))) It was something like 3*drgroot*X*3/2 or something.. not sure.. So.. is there another READABLE way to do this? - This means that the HP48 has no built-in derivative of the third root of X. Try entering the equation as x^(1/3). This will give the proper answer. Also consider Erable or Alg48 for calculus problems. - What it returns is the following: 'derXROOT(3,X,0,1)' - basically, when the 48 doesn't know the derivative of a function, it prepends it with der and doubles the number of arguments of the original function: given a user defined function (or non user defined) F(x), the derivative of F(u) as returned by the calculator will be derF(u,u') (I used double quotes because I used a single quote for the u-prime to add clarity). Any additional arguments will also have their derivatives (such as derG(u,v,u',v')). You can define derF(u,u') if you know what the derivative is. In the case of derXROOT, you could do a define for 'derXROOT(d,a,dp,ap)=ap/d*XROOT(INV(INV(d)-1),a)+LN(a)-dp/d^2*XROOT(d,a)' to keep it in XROOT form or 'derXROOT(d,a,dp,ap)=ap/d*a^(INV(d)-1)+LN(a)-dp/d^2*a^INV(d)' to make it transform to exponential form for a little more simplcity. Just type in either of those and hit the DEF button. - on second thought, i typed that in and it threw at me some ln(x) crap when i did Dx[XROOT(3,x)] so that's not quite it - almost tho - i think it won't work in some cases without the ln(a) bit, but i'll have to get back with you guys on that after thinking it over some more... -- http://home.tampabay.rr.com/kupopo/ icq: 5453914; aim: kupopo1 the and to strange - I do. In fact, when you were in elementary school, how did you do addition? beside that, and drew a line underneath. Sounds a lot like RPN to me. Now, for a good comparison between the two input forms, imagine that this equation (which is an attempt at ASCII graphic equation display) has to be typed in both an HP48 and a TI89. y 1 - x + y 1 - x 1 - -- x + y 2 x ( xy ) y - -- * ( - - x ) x ( y - x ) 1 + -- y - x Ok, given this equation, here are the keys I would press on a TI83 (for the sake of my typing here, I ignore shifts such as alpha and 2nd): ( 1 - ( 1 - Y / ( X + Y ) ) / ( 1 - X / ( X + Y ) ) ) / ( Y ^2 - X / ( 1 + X / ( Y - X ) ) * ( X * Y / ( Y - X ) - 1 ) ) Now, here is the same equation input in the HP48: 1 1 Y X Y + / - 1 X X Y + / - / - Y 2 ^ X 1 X Y X - / + / X Y * Y X - / X - * - / Now you can obviously see where RPN comes in handy...and how few keystrokes are needed to input the same equation. I find myself far more confused trying to determine exactly which parentheses go where, based on looking at a textbook style equation. In fact, what I had to do, in order to translate the equation into AES format, was to start in the middle of the equation and work my way out. In fact, this is the exact same thing I do with RPN. I found RPN easier not only because it required fewer keystrokes. Also, I had a running copy of the equation on my stack...I could instantly see if I made a mistake, rather than punching in an inaccurate equation and sorting through it when I got a Missing Parentheses error message. I hope this illuminates to people accustomed to an AES why RPN has such religious followers. Once truly tried, there is no going back, based on these characteristics. - I completely understand the sentiment, but here you *probably* won't find anyone willing to trade their trusty 48 for a calculator that TI threw out on the market right before the 89 in order to get a few more bucks out of those people who HAVE to have the best...before those same people decided HP was the best - I am surprised that the new calculator has had such a negative effect on the 48... it's still a great calc and I will continue to use mine until something better comes along.. I have yet to see such a product. - It's easy: Most of the Students who own an HP own an HP48G. HP is willing to hit the student market. So Students who own an HP48G won't be able to upgrade their calculators. If they just change the OS, TI users won't switch to HP if they just change the OS because they won't notice the change. So HP won't only release an ram card for GX. Inside the HP49G, they also have certainly change some composants. But I think, they should release on OS for HP48GX like Metakernel but a better one :-) - I don't doubt that the TI92 has a faster processor. But are there any ML programmers for it? If so, why is it still so much slower for numeric arithmatic? - But numerical math also requires loops and other things to make it useful. I have been saying this for a LONG time now. In fact, that's what this whole post was about. - But the hardware is only HALF the story... how well the software is written also plays a very large role in performance.. If you write the code in C, you can expect to see a little less performance than if you code in asm. -- Aaron. - You don't know your HP very well. the 48 displays with 12 digit precision, and calculates with 15 digit precision. Also, with either Erable or Alg48, you can find the exact precision factorial of any number (within memory constraints)...you cannot do this on a TI. Try finding the factorial of 1000 on your TI92. - This is the basis of most CAS (Computer Algebra Systems). Multiprecision integers and floats are needed to guarantee exact symbolic computation and not just numeric approximation. Dynamic data structures (even for numbers) are essential in a good CAS. This is actually a real FEATURE and not a waste or crappy coded algorithms. Carlos. This is only ** one ** impressive example on which HP49 is far better and faster than a Ti-89 with a faster CPU... Don't be childish, we probably have in the future a new calculator with a faster CPU but in fact all we want is just a big maths monster calculator, isn't it ? A faster CPU will be cool for games, but one said that GameBoys may be better for that job ! Let HP49 becomes that new HP monster calc who will eat TIs.. And wait for the final product please. You may go to see StarWars while, which movie comes in october in france :-( - - be still much too long! Also the power consumption will increase with a speed up. I personally want a longer and not a shorter battery life (about the half) for my calc. If you use the speed up module you also *loose* the clock, because it's running too fast. On my original GX I can see the same EQ with EQStk/TED/VV in 2 sec! So the CPU isn't that important. Much more important is the code! And it seems like the code on the 49 will be higly optimized! I'm looking forward to it! (maybe they'll change the keyboard layout a bit (btw. it's only a prototype till now!) and use a color which looks a bit more serious, but that's really not too important IMHO) RPN behaves exactly like on the 48 and there's no need to fear that algebraic mode would eat up the ease of use for RPN mode (due to a different keyboard layout). The 49G will not be the perfect calc (which can't be built to suit everyone's wish/needs), but for many people it will be a far superior calc compared to the 48. Those who need IR or other advantages of the 48 (built in EQ lib for example) are free to use their trusty old GX! BTW, some have complained that the built in EQ lib isn't too useful. Speaking for myself I used it very seldom, because those formulas are sometimes in a simple form, or they weren't in the lib at all. Maybe ACO team already plans to produce a 49GX version with IR and mem card slots? We just don't know! Also we don't know what the other plans of the ACO are! (I'm sure they have plans beyond the 49) Greetings from Cologne Peter E-Mail: karpfenteich@gmx.de _______________________________ Do you know the great Frequently Asked Questions? http://www.engr.uvic.ca/~aschoorl/faq/ and the superb HP48 Software Archive? on for the photos The large picture on hpcalc.com and the ones taken by an attender at OPenHP are not significantly different enough for J-Y to waste bandwidth disputing. They are both the same calc to some immensely large degree (so much so that I haven't noticed a difference yet). Surely they are enough alike not to matter in the face of all of these criticisms? I think that's very minor. - Yes, I'm somewhat disappointed in the keyboard layout, no expandable Your looking at the killer calculator. Like it or leave it. The bean counters will not not finance a new design if this calculator does not live up to their spreadsheet calculations. Business cycles are too short these days to have engineers around building good products. An engineer on unbillable time is an engineer not paying bean counter overheads and generating revenue. Go find out how many hours you must work to cover one administrative bean counter hour. It is disgusting. Joe Frisco - I was trying to say that ACO is likely to be working on the TI crusher as we speak. The smart thing to have done is to work on **both**, the TI crusher and the HP49 to keep customers until the TI crusher arrives. And that is what I think ACO has been doing. So... would you rather have NOTHING until around 2002? Or at the least a HP49? The fact that they called it the HP**49** indicates that it's an incremental improvement on the HP48G's. If it were a TI crusher, they would have given it a completely different number (HP58? of HP68?, etc.) As far as this mess goes, don't blame ACO, but blame HP HQ. They were the ones that transfered the Calculator division to Singapore, who were not interested in it. Before ACO was created, I thought that HP was going to leave the calculator business. Personally, I think that ACO is developing a new proprietary processor, say the Saturn II?, for the TI crusher. A saturn II could have the original Saturn buried in it someplace to run Saturn code, so that all that work writing a new OS is not wasted. Besides, the work on the new OS is valuable as a learning tool for future OS's On the issue of the Bank Switching, the HP48G's **copied** the code to system RAM and then ran it. Doesn't make a lot more sense I believe this is what ACO has done for the HP49. TI & Derive, The Derive people are sure a lot more experienced with a CAS than TI is! By bringing the Derive people in, TI probably cut work on a new calculator by about a year. - Well, with the release of this proprietary emulator for the HP (which I assume was not sanctioned by HP with the release of the ROM included in the package), some interesting issues have been brought up. 1. This emulator is tagged with Sebastien Carlier over the entirety of the program. My assumption is that this work is based either in part or in whole on Sebastien's version of the HP48 emulator that we have all used before. specifies many stipulations on how source may be aggregated, modified and released in binary form. 3. This release, whether intentional or not, may be in violation of the GPL in some places (read the GPL at http://www.fsf.org/copyleft/gpl.html): 2. You may modify your copy or copies of the Program or any portion of it, thus forming a work based on the Program, and copy and distribute such modifications or work under the terms of Section 1 above, provided that you also meet all of these conditions: Please note that the GPL mentions nothing about proprietary modifications to the source code for internal use only. These requirements apply to the modified work as a whole. If identifiable sections of that work are not derived from the Program, and can be reasonably considered independent and separate works in themselves, then this License, and its terms, do not apply to those sections when you distribute them as separate works. But when you distribute the same sections as part of a whole which is a work based on the Program, the distribution of the whole must be on the terms of this License, whose permissions for other licensees extend to the entire This is the important part of the issue. If the program is simply the GPL'd emulator with a few modifications to enable it to handle some of the new hardware, then the release of the full source code to this program is required by law. 3. You may copy and distribute the Program (or a work based on it, under Section 2) in object code or executable form under the terms of Sections 1 and 2 above provided that you also do one of the following: Basically, to comply with the GPL, the source code must be provided in some form or another. 4. You may not copy, modify, sublicense, or distribute the Program except as expressly provided under this License. Any attempt otherwise to copy, modify, sublicense or distribute the Program is void, and will automatically terminate your rights under this License. However, parties who have received copies, or rights, from you under this License will not have their licenses terminated so long as such parties remain in full compliance. 5. You are not required to accept this License, since you have not signed it. However, nothing else grants you permission to modify or distribute the Program or its derivative works. These actions are prohibited by law if you do not accept this License. Therefore, by modifying or distributing the Program (or any work based on the Program), you indicate your acceptance of this License to do so, and all its terms and conditions for copying, distributing or modifying the Program or works based on it. Essentially, by modifying the GPL'd emulator, Sebastien (or possibly HP) has agreed to the terms of this license and is required to make changes available. I believe that the GPL is designed to ensure that once a program is set free, it must stay free, so that no company or individual (even the author!) can bring it into a non-free domain. This is an interesting case, however, and I don't know the full situation. I do know, however, that Sebastien or HP *must* make some sort of statement in the near future saying that the source code for this program will be released. The ROM code is copyrighted by HP and should not be in this package, but the emulator itself is under the GPL and these terms: http://www.fsf.org/copyleft/gpl.html No matter what the purpose, basing a work on a GPL'd work carries certain consequences for derived software. I invite discussion on this point. - The HP49, in its current (admittedly prototype) state, offers very little to make me want to rush out to get one to replace my HP48GX. If my HP48GX dies, I may be compelled to get one--but not before looking at current TI offerings. It's depressing to think that most of the HP49G's target buyers (high-school math AES users) will never fully realize the power of that machine. What a waste... - To me the HP49 looks great. I think it is a very nice improvement over the HP48. I will be very tempted to buy one. I own a HP48 which I unfortunately do not use often. Too sophisticated as a tool for my present needs!!! No time to play with it at the moment. I think the HP49 will fit the bill as an advanced math tool very well. The user interface we are told is significantly faster. Face it, for most users it will be pretty irrelevant that ML programs still run at the same speed. For what most users are going to do with it, it sounds plenty fast, after all the HP48 is not too bad either. But thanks to JYA it seems also more open to programming out of the box and there will be enough memory to go with it. There are some necessary concessions to sell as many units as possible. Costs is also among them. On another note, we are getting close to run Mathematica on a handheld. But even when this is going to happen, it will initially be very expensive ($1000 ++). Possible that you are going to replace your trusty HP49 with one of those puppies a few years down the line. - I added the brackets to ensure that there isn't any misinterpretation. :) Basically, having the emulator (and ROM!) available for free download, along with the source offers many benefits. Some of these reasons are for having the ROM available with the emulator, some are for just having the emulator and some are for having the source: 1. It allows people to evaluate the HP49 before using it in the real world. Personally, I wasn't a big fan of the new calc until I had a chance to see it in action with my own eyes. The experience of running this emulator showed me how well it works, even though I don't own one. 2. I certainly wouldn't use this for schoolwork (it's just not as convenient), but if I was doing something on my real HP49 and it was too slow, I could just transfer the program to my computer and have it run in the background, without wasting my batteries or my time. 3. It allows developers to SAFELY test their programs. This is how most emulators benefit developers. I can't tell you how many times I lost my source code while writing with Jazz. :) 4. Having the source code available allows the end-users to extend the program in ways that the original developer hadn't seen before. Look at where EMU48 has been taken by its new author. The new program is more fully-featured than the original one. HP won't be making any money off this and the architecture of the '49 is almost 100% known, so does it hurt them at all? - Some obvious changes I'd recommend would be to keep the wide enter the menu menu a huge there's a HP aspects Well put! I would add, to the feature list above, the IR port. I think it may be unidirectional for the sake of allowance in school tests, but it should be there. Othervice what should I, and all other owners of 82240B IR printer, do with it? If I buy the new calc and want to continue using printer, then I have to use it with the older model, but why should I do it if it has the same and even inferior capabilities comparing with the newer one? - Very well put. I think this post really hit the mark and I hope we're all surprised by a wonderful product in the future. - It's much better to deal with government people... They are well known for there intelligence over the teachers.. :) - This new flag appears to only affect the menus which switched to choose boxes in this release. Those which were always choose boxes in the 48Gx seem to remain so after toggling the flag. Of course whether one can draw any conclusions from an unofficial emulator which may represent a relatively old snapshot of the code (though the ROM image is dated only a month ago) is obviously questionable. For example, the emulator defaults to RPN, not algebraic as the datasheet says the final version will. Some other notes after playing with the emulator for a little while. The arrow keys no longer double as alpha keys, but they retain their seen a prototype that has these functions labeled on the case. Significant funtionality seems to be missing from the emulator (things like the APPS and TOOLS keys don't seem to do anything) and there are many bugs. I have crashed the calculator many times through simple operations. This may in fact represent instability in the emulator rather than the ROM code, but if this truly represents the state of development then I'd say that there are a couple of manyears left to go before the software will be done. The 49 is supposed to feature user-flashable main system ROM, which is a good thing because it may take a few revisions to get all the bugs worked out. Overall, I think I like a lot of the software changes, though the calculator appears to be harder to use and more complicated rather than simpler. In many cases it is simpler, but when you stray from the simple path then you get lost in a maze of incomprehensible complexity much quicker, since you won't have had to learn what any of this stuff means to get started. The user interface (keyboard) changes are questionable. I miss the one-key clear function, and it now takes 7 keystrokes to change angular mode from DEG to RAD rather than two as on the 48. I assume that all of the 48 functions are there, but many of them seem to be impossible to find on any menus. I'm surprised that there is no help key. With the expanded ROM, I would thing that a calculator quick-reference would have been an obvious big win. (Though the datasheet indicates that the 49 will ship with a quick reference card as the 48s units did. I wonder if it will fit in the slip-on cover? It would seem silly to reintroduce the pocket guide and at the same time take away the case that has the pocket to keep it in.) skills. If anyone has any ideas for sysRPL and ML (sysRPL first, if possible) programs I could write, please, post the ideas so I can try it out and learn some stuff. - If I read JYA correctly, 128K of Ram is dedicated to System RAM, with another 128K of User RAM. I assume that most programs will be stored in the 1M of user Flash, with more volatile data in the User RAM. In any case, the 128K of system Ram and the 128K of user Ram can be in part of the 512K, with a window of memory overlaying the System Flash, and the User Flash. I assume that user programs will be copied from the Flash to the system RAM using a possible way to make the bank switching transparent to the OS. There are many more options, all revolving around caching or copying the run program (or some portion thereof) into the reserved system RAM. WilliamTStorey ech.net Reading this site has made me feel as if I trapped in an angry lesbian convention. Most of the users here have no basis for their arguments, or no real basis at least: Ewwww its purple... Waaaaa I cant find the enter button its too small... The buttons are rubber!?!?!?!? uh......I cant touch them... If your major concern is how your calculator looks then I suggest painting racing stripes on your calculator, wearing it around your neck, and shutting the hell up. Sure, they took away the IR. Sure, they are wasting ROM space by implementing a counter-intuitive and inefficient AES. These are the only things that I can see wrong with it, other than the fact that it has no expansion slots, but Im going to have to trust the people that have actually used the calculator and said that it doesnt need it. Listening to the bull shit that people are posting, people that have not even touched the calculator is really dicerning. C'mon you guys are supposed to be better than those TI-freaks - While true, it's also true that wheels won't help you cross an ocean. I haven't used my IR port in the last 8 years -- I can live quite well without it, even if it means I'll never be able to run Tricorder on the 49. The RAM specs are higher than my GX's (that's including current cards, which are mostly filled with stuff included in the 49 ROM). I'm still holding out pending a hands-on excursion, now that I've had a taste of the functionality. - I suggest you reread the posts as most of the arguments are very well founded. Lack of such features as an IR port, an RPN keyboard, and even a frivolous color in a professional environment are cause for concern. And it doesn't require hands on experience to evaluate each of these characteristics. I myself intend to give the HP49 a chance but I will not criticize others for voicing their concerns! At least that way there is a chance for some changes to be implemented. And need I remind you, if you wait for hands on, you won't get that chance. - Does this mean that it doesn't let you use menus over apps in some cases (i.e. L.S. STAT, L.S. SYMBOLIC, etc)? I rarely use the R.S. versions of many of them and this would be a sad thing indeed... - If this is true then why develop a miracle product with poor economic what I understand they are doing very well at it too. Besides that, the 49 should do fine (the shrill whining is from fanatics whose expectations were orbiting Jupiter). It's faster, it's got a ton of memory, I never used IR, I can adjust to the enter key in the lower right, the color is great. The wonderful pool of programs, books, tools, games, etc. is retained. - fields. - first exposure to HP is now more likely via a printer, or scanner, which are VERY consumer oriented (poor support, poor durability, cheap price) - the instrument side of HP is shrinking and the consumer side is growing (not sure about this but I understand the consumer side is creating more profits) But despite all those arguments I've decided the early complainers are a bunch of EMOTIONALLY CRIPPLED TECHNO WHINNERS whose propeller status badge *seems* tarnished. If the 49 really is as fast as they say, I am convinced it's a winner. Who cares if the CPU is 40 MHz if it is slow (take a Newton 120). ACO did a great job of refining an excellent product. In particular, the scope of software on a 49g is awesome. It is very important to have a standardized set of software so that developers and users are synchronized, and tool use is undiluted. This common collection elevates the 49 over the 48, even if all of it is already available on the 48. - If i ever forget to capitalize a proper noun, forgive me. i have ee cummings in my ancestry. Have a good day, and good luck in your endeavors! - Is it dangerous to reduce the size of an object, eg STRING object by simply decrementing the size , (2nd nibble) of string object? Will the discarded memory be cleaned up in a garbage collection? Can it happen later on that the object is skipped, and in that case would HP48 expect the memory following the reduced string to have a meaningful prologue? This may be a stupid question, but I'm new to ML and hate to learn everything the hard way... so I'd appreciate any advice. - It depends of what you call powerful. True that the TI89 has a more powerful C.P.U but do you know that the TI89 is slower than a 6 Mhz Z80 TI86 slowdowned by RAM bankswitching for number crunching and basic speed ? I have some benchmarks if you are interested. For example For k,1,1000:Endfor runs in 17 s on a TI89 in approx mode and in 5 s on a TI86. Perhaps that if TI completely rewrite and strongly optimise the TI-OS it will be able to produce a TI89 capable to blow away the HP49 but why TI will do that as its main target is high school students ? Unless someone else completely rewrite the TI-OS you will never see a TI89 capable to blow away the HP49. And about this utopic project let me just underline that TI plan to release a SDK and ASM support in January 2000. the expect sure a very the small desk. must The question is not that HP has not enough money the question is why HP will spend a lot of money on a no-profitable sector ? I am sure that HP was ready to screw up its calculator departement. I am also sure that HP has let a last chance to its calculator departement to show that they can produce a calc which can make HP earn some bunks there. Calculator like the HP48GX are very nice but they are very bad for the business. Do you know that even the bulky TI92 has almost killed the HP48 market in France ? Now about the C.P.U: How many time will it take to devellop a full new OS on a new C.P.U ? It has taken TI almost 5 years to release the TI-OS whom parts have been released by third parties companies. It has taken 5 years to TI to produce a craspy OS ! How many time would the A.C.O have needed to produce a quality OS on a whole new C.P.U very different from the SATURN 3,4,5 years ? Do you think that HP would have accepted to finance such a project ? Of course not especially when it would do no profits but perhaps lose money during this time. About speeding up the SATURN: Why for ? The goal is to reach the best performances possibles with it and it is what the A.C.O has reached with the HP49. About the impact of the actual work: Yes the OS will have to be rewritten for a new C.P.U except if this C.P.U is an evolution of the SATURN like the 68020 for the 68010. Also they will already something to begin with even if it is a completely different C.P.U. Because you think that all these attacks agaisnt the A.C.O and the new HP were nice ? No they were crasppy. They are a few ones who have done clever and constructive critics on this new HP but for the majority it was mindless bashing. Because of that i was very angry when i have posted this message so perhaps i have been a bit too far. - Increasing the clock speed would require redesign of the IC. This could be as simple as moving the part to a faster CMOS process, maintaining the same netlist and pinout, and thus we have a quicker chip. Of course, they would have to add a pin or a register bit to move from the current 4 MHz speed to the hgigher speed. The existing pin is defined to change between 2 MHz and 4 MHz. There would also have to be some rescaling and redesign of the analog areas of the chip, including voltage regulators and the phase-locked clock synthesis circuitry. Beyond the digital netlist, this would be a major redesign and requalification of the mixed-mode analog and digital IC. This would probably be about a 19 month project. I believe that was Preston Brown's estimate when he was asked. (Preston was the engineer who did the 2 MHz to 4 MHz speed-up.) I believe that ACO is grabbing up talented engineers as quickly as they can find and settle them. They haven't had time or staff to do a project of this magnitude. Now, some of you are probably getting a bit steamed at this point saying to yourselves, Whaddya mean??? These guys at Cynox are already running the existing chip at double speed. I can't believe that Dave is feeding us this line of Frozen Hamster Droppings (FHD)! We used to be able to trust that loser. Of course, nobody but me is sitting at the computer and thinking to himself or herself, I've seen the qualification data on the Yorke IC. I know that they have plenty of speed margin at the nominal process flow and at room temperatures. But I also know that, at the corners of the CMOS process window and at temperature extremes, the parts are marginal at 4 MHz. I remember how we seriously considered doing batch splits, sorting out the fast parts from the slow parts, and introducing the 48G as a 2 MHz machine. I imagine that most HP48 G-series machines could be overclocked as long as they aren't run through the complete HP environmental test suite in that mode. And I would have a major ethical dilemma if management pressed me to release a product with those caveats. Lots of people admire the design quality of an HP product. This is part of the cost of reliability. There are times we miss. But in general we manage to stick to the standards that established the HP brand as we know it today. ACO specifically approched me regarding overclocking of the existing Yorke IC for use in the 49G. I recommended against the idea, based on initial qualification data and the NEC CMOS process constraints. - I'd be glad to participate. Shall we make a team and organize ? -- Sam. - Sure.. Let's start with J-Y's post and go from there? It seems to batch up the first half-dozen questions, as well as a sketch of the history. - I want to upgrade my HP48G to 768K as per the instructions found on the web (http://studwww.eurecom.fr/~grundsch/big_upgrade.html) However, I am having a hard time finding a source for the 512K sram chips. Does anyone know where I can buy one? Also, is the www.jameco.com/cfm/detail.cfm?sub_class_code=1G5&name=Static%20RAMS - : The rubber keys have me concerned, : however, as does the keyboard. Reteaching my fingers the right motions : sounds about as appealing as learning the Dvorak layout (better or no, : it is still a process of relearning and therefore a hassle). learning the Dvorak keyboard layout was a highly worthwhile undertaking despite the painful ~3 month conversion process. Similarly, I found the comparatively easy task of switching from AES to RPN to be very rewarding. I sympathize with fellow RPN devotees who detest the repositioning of the ENTER key on the infamous HP49, just as I loathe the popularity of the QWERTY keyboard in spite the benefits of Dvorak. I feel that optimizing the keys for AES users at the cost of convenience to RPN users is a step in the wrong direction. Sadly, us RPN advocates (& Dvorak users) are in the minority, as HP has apparently yielded to the (arguably inferior) AES crowd. Fortunately for Dvorak users, most OS provide Dvorak keyboard support via software re-mapping. Hopefully the HP49G has user-definable keys like the HP48 that can be programmed to emulate the HP48 RPN-friendly layout, although the absence of the large ENTER key can never be compensated for :( - The issue is either buying a new CPU or DEVELOPING ONE.... Does everyone agree that a new CPU is desirable? I mean, yeah, I can use my old 8080's in a new design rather than a StrongArm, but who's going buy the product? I think the issue is make or buy new technology, not use 10 year old technology! -- Zebb-Hoff Machine Tool, 9535 Clinton Rd, Cleveland OH, 44144, +216 631 6100 Advanced Design & CNC Manufacturing - Molds, Prototypes, Tooling, Fixtures Info: http://www.zh.com or ftp parts to ftp.zh.com (Pro/E, IGES, DXF) - It may seem this simple, but I'm not so sure... The HP200 has had an amazing long life just due to the fact that, although it is old and slow, there is still nothing like it in the power consumption department AND it can run just about any DOS-based program. 3rd parties have recently expanded memory (to 64 mb, which makes discussions about 1-2mb in the 48/49 seem a trifle small) and doubled the speed, but still no one has produced a low-power chip (the 200 uses the 80186) that will allow anything like the 200's battery life in any 80286 or follow-on. Many of us keep actively _trying_ to find a better palmtop than a 200, but a surprising number of us keep coming back after selling or giving away whatever we bought to replace it. I think the issue is getting work done, not the age of the technology in the product. Obviously YMMV. - If it is possible please change the navigation. I want to be able to go to the solution with the right arrow key and back again to the original table (and from there to the categories) with the left arrow key. I haven't found a way to go back from the solution to the problem. If you press cancel sometimes the first stack level gets displayed. And purge the message from the hidden directory either after you have read it (via MSG) or when the next solution without msg is copied to the stack. Nice work though. cu Balazs Fischer Balazs.Fischer@studbox.uni-stuttgart.de Fingerprint: 8F EF C0 BA 22 B4 15 B3 56 27 07 04 37 7B D6 00 ... This probably doesn't help, but... - I will see what I can do about these changes. I am using XCHOOSE by Matias Mutchinick; I will ask him to make some changes in XCHOOSE for INT48 . bottom of every equation table. P.S. I would like INT48 to be the largest set of Integration tables that the HP48 community has ever seen! Also, I would like to add the add-on feature as soon as possible so that INT48 will be versitile too. - Actually ALG48 can be run from any port without noticable speed penalty. cu Balazs Fischer Balazs.Fischer@studbox.uni-stuttgart.de Fingerprint: 8F EF C0 BA 22 B4 15 B3 56 27 07 04 37 7B D6 00 ... Most people make sense, I'm not one of them. - In order to explain my 'sysRPL-problem' it is probably best to start with an example; Suppose that 'A*X+B' 'A' are on the stack and that there is a variable X containing a number. Calculating the derivative with respect to A gives the number that is stored in X, i.e. the result has been evaluated. ??? Is there a (sysRPL-)command that calculates the derivative WITHOUT evaluating the result afterwards, i.e. without substituting numbers for the corresponding variables? - Which would imply that the HP bean-counters would REQUIRE all future calcs to be like that of the 49 (education use targeted vs. the fractional 1000's of professional users) - ie, no IR, etc. :-( At any rate, lemme know where the line starts to get me one of 'em 49G! - I was referring at that time to JAZZ where (at least I do) most of the editing is in ALPHA mode - in that environment (ED), you hit ALPHA-RS-TAN (43.6 i believe) Well if it still is logical (ROW*10+COL+SHIFT*.1) then programs that look for the ENTER key on the 48 will end up finding something completely different on the 49 (don't have picture handy, so i don't know what) - I'm aware for JAZZ on the 48. I was curious about the 49, but figured it out independently. *IF* the arrow keys on the 49 reduce to rows 2 and 3, I'd expect ENTER = 105.1 (Row 10, Col 5). IF the arrow keys represent their own row(s), I don't know what they'd be. - Go to hpcalc.org and look at the official pic. The keyboard got swapped a bit. On that pic, the [#] and [BASE] are shifted [3]. - i don't think anybody the HP6s in mind when they said low-end, inexpensive HP... i think that we were thinking more along the lines of a reasonably priced, smaller 42 or 32 type calculator...and something that could be minimally programed...!!! what is the point of the HP6S...??? there are already a dozen calculators sitting by the cash registars at the supermarket that will do essentially everything that someone would use the HP6s for...??? if you're going to be using SIN and SQ on it, then you'd certainly have a graphical calculator or soemthing with a few memory registars in it... i really can't imagine anyone buying an HP6s for anything other than as jewelry...!!! - Many of the objections to me seem to be about the keyboard (Enter key placement/size, shared Z and / key, etc. I have read the photo of the keypad on hpcalc.org, it seems to me that the 2*3 keypad section to the left of the arrow keys (letters G-L) could have easily been replaced by a 3*3 keypad, allowing both the larger, centered enter key and the removal of the Z from the divide key. I would like some feedback from anyone who knows if there is a reason why that 2*3 section is separated and isolated where there is space you have some reactions. James Doebbler - PS: Saturn is an old processor? Hey, the 68000 used by the TI89/92 dates from 1979!! And the other TIs use a Zilog Z80 and that dates from about 1975. PPS: Am I the only one who is starting to find TinyWanda a little more than irritating now? PPPS: I am inclined to forgive stuff, because this is the first good HP calculator announced with some 100 million people connected to the internet. The info got to a lot of people very very fast. PPPPS: Since it's blue, do we get to call it Big Blue? - Well, 4*e^(2*PI/3*i*n), where n = any integer, are all correct representations for (-8)^(2/3). when n = ...-6,-3, 0, 3, 6... you can simplify it to just 4. - I was of course interested, and made a few tests yesterday. Compressing source code (30Kb), entries tables, Sokoban, xymodem etc I could not deduce which was more efficient in general for what kinds for objects. The biggest difference was 3K in favor of BZ (Dan Kirkland's chess source code, closer to 40K), the biggest in favor of FlashCmp was about 500 bytes for Sokoban (if I recall correctly). Overall whenever there were significant differences in the compressed size, BZ won 4-2, and there were 3 ties (+- 10 bytes). I intended to bring the data with me, but forgot my calc at home. Also, the BZ compressor seems to be a bit faster, but the FlashCmp decompressor is faster than the original BZ. My understanding is that some French guys have optimized the BZ decompressor in separate releases, in which case the code would look quite a bit like the FlashCmp decompressor (only the control bit decoding is essentially different in LZ77 class decompressors). However, I may not have exactly the standard stuff on my own calculator. It seems clear that FlashCmp is more efficient for some kinds for data, but I could not deduce what kind by disassembling the library. - - file transfer between a PC and my HP48SX via INFRA-RED. I only want to know if there is software available to do this, or the way to configure the hyperterminal. - In the hp48 the metakernel, erable, and possibly jazz, all these applications need to be in port one. Erable is the absolute address version. Anyway due to their use of absolute addresses they can not all be stored simultaneously, this is beside the fact that there would be no room in port one for all three, but in the hp49g will we have similar problems with an application not being able to be installed because it was designed to use a specific memory block that another application is already using? Kind of like conflicts in which both applications use the same memory addresses. Also I maintain the bean series of programs (checking account library). So as I understand it, I would disassemble the library with the metakernel, and transfer it to the hp49g and reassemble it there and convert the directory to a library, and it should work just fine? please forgive me if I misunderstand something. - Ho ho ho. When he said it was customisable I guess he _really_ meant customisable. Camberley, Surrey GU15 3PD. - programs in a directory, it just runs THAT program, if it links to another, I'm out of luck. And yeah, EVALing is the execution for it... but still the same problem remains -david - Reading about the HP49, I am still seeking a broken HP48 to replace my broken LCD. :) I am ready to pay a reasonable price for your broken artifact. I just want your body, ops, LCD. :) It's been 4 months now... I am desperate... :( - I am an Engineering Student/Engineering Technician. I've been using my -48GX for about 2.5 years now. I own a TI-92plus as well, it sit's in my desk collecting dust. Maybe it's just the size of the TI or the fact that I am used to RPN now, but I couldn't get used to the TI. It seems to the HP is much more appropriate for serious Engineering work anyway. I saw the HP-49 bitmap for the first time today. My first impression was YUCK!!!! Did they hire the guy who designed the IMAC to come up with this thing!?!?? However, it kind grew on me, I now think it's kinda cute. Superfluous, I know!! I went to hpcalc.org and read the spec's . If what they say is true I think I will definitely buy one. They've addressed almost all of my concerns with the -48 here. As for the loss of the IR port: so what! I always thought that was gimmicky anyway! All I used it for was to make a $150 remote for my Sony Trinitron, yippee-skip! As long as I can put the thing into RPN mode and compile my favorite -48 programs for it, it will be a worthy upgrade for me. I look at this new HP product as simply that: an upgrade! like win 3.1 to win 95, not perfect, HARDLY!!! But an improvement! I work in the engineering field and I recognize how hard it is to get a good product to market. Give it a chance guys (and gals)!