A8 Silly me. I figured those words were in there (%1+,etc). Aparrrrrently not. I ASSUME that this is used in the dark side, too (that is: ALGebraic Mode) and it returns the argument and thus is accepted as a function It was really a joy to help you Ray. That way I also learned a lot. I didn't expect that flag -128 has also a side effect on REALASSUME, but you found out that it does. :) So I also have to say a big thanks for that. X IF Mika Heiskanen still reads this NG: THEN I would be very pleased for a version of ISPRIME? with an adjustable time limit and commands to handle it: END But pictures stay. And that lasts a few seconds every time, that's really annoying? On Goodies Disk #5, I put a shortcut called ANTMENU. Here's an excerpt from ANTMENU.DOC: Sounds like what you want. Warning: ANTMENU was written in 1991 for HP48SX version E . It may not work on your calculator. If you need an update , post your HP model and ROM version, as well as what version of ANT you're using. A prog' that pops up to remind you of the key function ?!!!! What are you a govt employee ?!!! The 49 keyboard sucks , you know like terrorism , govt , and toilets that don't flush . ====================================== (this is from a comic strip: an exited cat uses it :-) My favorites: Commands: STOVX, RCLVX, DISPXY EQW: EXP(X) displays nicely as e^x, LIMIT is now a beautiful lim MASD: a lot lot, but I like this one: support for symbolic arrays FILER: XSEND and CHDIR (which occationally works in a strange way) Menus: CHOOSE Boxes will finally have *all* the needed flag settings eg. BASE will start with HEX DEC OCT BIN choices where your current choise is hilighted these flag settings will have a star (multiply) mark appended They surely work nicely. The calculator is now perfect To get the more bandwidth clip out the garbage from the replies Try using scissors... like this: Toshiba has a __55V_____ SRAM that is the lowest i've seen but DigiKey no longer has it . 128KB and 10.8 milwatt/mhz . Many of the so called low power are 50 mw/mhz ! This is outrageous high power for a batteyr powered calc' . The lower the standby current, the longer your HP48's memory will be preserved whilst it is in the off state. The lower the operating current, the longer the batteries will last while it is on . Since most people usually spend more time with their calculator off than on, i'd vote for lower standby current. My 48GX has AA batterys . I've seen many people say this, but not one has said how they did it. Care to post some pictures? ================================ Tech wannabe's ! Since the stdby is equal to shelf life of battery it don't f****** matter ! I carved out the plastic above and below the box . I carved a bit away from the sides and bent the plus/minus metal tabs to allow them to back away from the longer batteries . I made 2 brass links to go between the batteries . BTW Any one know if you can fudge using 3.3 SRAMs ? on my 48GX i go to chars and it says KEY : <. (FM , ) What is this ? How can i get this char' from the K.B. . I think this means that to get the ';' sign you have to select the comma ',' as fraction mark sign. This can be done from the Modes menu (right shift + CST), in a check box near the bottom (or faster, seting flag -51). Then, when you press left shift + '.' you'll get the wished ';' sign. But if you have the dot '.' as fraction mark sign (flag -51 cleared), you'll not be able to get the ';' directly from keyboard (unless you use some program like the great editor ED from Jazz). This is one (relatively fast) possibility to save also the key-assignments. But does it pay? Clearly, if STARTUP is present in HOME (and it should!) it must eval a safe program from Port 2 (to set at least your flags after a crash :-) 2:SU restores my standard flags, sets the VAR menu and shows the free memory in a MSGBOX. Clearly, what STARUP may do depends to a high degree on the individual user. However, this all does not yet answer Heiko's question to make a fast HOME backup in Port 2. This is quite simple: Store the program (or its shorter and faster SysRPL-equivalent) in Port 2, under the port name ARCH, say. That replaces the last version of the HOME backup by the current all alarms and key-assignments), or it creates the Port 2 name Home, if not yet present. So, saving your HOME even against taking out the batteries is quite simple: Navigate with Filer to Port 2 and eval ARCH. I also have a file REST in Port 2 which does simply the Filer. No need to search the very hidden commands ARCHIVE and RESTORE in the menu system. PS. The Filer should perhaps have some ARCH and REST already in its application menu. What's your opinion? god damn i only have like 8! what kind of things do you use them for? About 25 keys are assigned for program developement, Emacs alone needs about 10 keys on its own :-) You'd better asked how to assign so many keys without interfering the standard functionality of a key. Of course, I use almost all shift-hold keys, and many keys have a double-function as to whether longhold or not. Only one key is assigned with a double-click. I just measured how much time RCLKEYS needs to put the 10 KB assignment list on the stack: One minute and 10 seconds! Therefore, I never use this not optimally programmed command (inheritage from the 48 where it was sufficient because one cannot afford for memory reason to assign so many keys on the 48). You must have an amazing memory. I'd need a cheat-sheet to remember that many key assignments. 2: YourAlgebraic 1: n n can be 0, 1, 2 or 3 looks like an expression in text book appearance. It doesn't do exactly what you want as the used font isn't the minifont, but at least you get the looking of the EQW. Yes, but fortunately the command set for manipulating graphics is very powerful, even if you only use User RPL. A small command sequence that displays an algebraic in pretty print somewhere on the screen would be: @of algebraic on stack level 1 @stack level 1 { #x #y } @Put coordinates where the grob @of the algebraic should appear @on stack level 1. #x #y are @binary integers ROT @Bring the grob of the algebraic @to stack level 1, so that the @order of the arguments is OK REPL @Replace part of the grob returned @by -LCD (current screen) with @grob of the algebraic, starting @at coordinates #x #y @Put result at stack level 1 @current screen 7 FREEZE @Just for the display to persist @until a key is pressed. Can be @ommitted if display should @persist only during program ==== i have installed the new ROM without any problems, but i don't know how lim works correctly. in RPN&exact mode: if i put an algebraic on the stack, than the equation for the limit and i press lim it works correctly, but now comes the strange part: ==== I would have to say that this would be something with your flags or perhaps a corrupted ROM install. I'm not sure if the latter is possible with this You are the problem JYA . You say Joe has done this and Joe has done that (Joe who has been helping people ) but you don't see that Joe made a mistake and can't admit it . This is the problem . Joe can't get over it . Then you flamers jump in and protect Joe . You are the flamers . You igo maniacs are the problem . And you flame me 'cause of your big head ! You did not need to post that nasty i see no usefull..... ==== I deactivated my 'spam' filters yesterday. Now I see it was the wrong decision TinyWanda's postings were funny, sometimes even very funny, sometimes not, but it was fun to read most of them and I never thought about setting a filter to flush his postings. ==== Does SASM, RPLCPL, etc. have a problem with words like %1+ and %1-? I ask because compiling ASSEMBLE CON(3) 0 RPL xNAME TPDF :: CK2&Dispatch BINT17 :: %2 %1+ ; ; for the 39 won't work, whereas if I replace %1+ with %1 %+, it works just fine. Is this a problem with the compilation programs, or have I found a strange bug in the 39? TW ~The enemy's gate is down. fine. bug But have you checked in the .LR file created by the linker that all the references are resolved ? Oh yes, you are absolutely right. I only thought that ATTACH should be mentioned, as it provides the capability of attaching libs without a ==== Yes, The update will take your Hp48SX to an Hp48GX version R or something. Just send it in to an Hp authorized repair station and with a fee of approx $50-60 and you will get a new 48GX. Short of that you will have to hack into your system with and solder a new ROM chip robbed out of a hp48sx with the rom you want. This will involve removing a SMD from one 48 and re soldering it onto another 48 without damageing the SMD chip. GOOD LUCK! Its not impossible, I have done it, (Years ago with printer SMD chips, not a 48), but I was ==== Anybody know what happened to TinyWanda? Certainly was one of the most colorful characters in this group. Wrote killer software too. ==== I got a similar project to work very well on my HP48GX. I used a KPC-3 TNC and a Yaesu VX-5R 2m ham radio. On the other end, I had another TNC little wire that went from my calculator to my backpack, and I basically Lately I've been working on making a smaller TNC that would work better with the calculator. One of the problems I encountered was with linefeeds and when the KPC-3 would send a packet. The new TNC will use an 8051 microcontroller and an MX-614 modem chip. If anyone's interested, I can post the design to the group (disclaimer: I'm not an EE - yet - so the design won't be all that great. I know enough to make things work, but not enough to really design things well, so I welcome criticism and comments). Another issue is password security. Anyone with a radio and TNC on the scroll by. Encryption is not legal over the air (in the US), so that's machine change based on an algorithm, and have the same algorithm on the calculator. I never could get a perl script to work to change the password though, for security reasons the passwd command doesn't take the input from STDIN, from what I understand. If anybody out there has necessarily in perl), that would be great. I got the project to work well enough. The ANSI 1.5B terminal emulator is great! It works well, and was well thought-out. The screen characters per line, etc. for the output to the calculator. The When using the text editor the calculator will occasionally warmstart. I don't know why; I don't do anything special or unusual when this happens; it doesn't happen with any kind of regularity that I can see. When using the calculator, it will occasionally freeze, in the same manner that it warmstarts -- inexplicably. When it freezes the only responses I get out of it are from the shift and alpha annunciators at the top of the display. They turn on for a very short moment and then turn off. The busy indicator is on, but for minutes nothing happens, and after a while I warmstart it myself. I have ROM 1.16, eql49, and no other libraries. Neither of these are repeatable; they just happen. I checked the HP49G FAQ, but it's section on perpetual warmstarts doesn't seem to apply to this situation. When I use PINIT, I get the times of the warmstarts, but no information on their cause. Does anyone know what's wrong, or how can I figure that out myself? Has anyone experienced this before themselves? I *really* don't want to restore the calculator to its original state through the hole in the back, but if this is the only way to stop the troubles I'll do it. Will I still have ROM 1.16 if I do this? When I first started SysRPL and ML programming, before I became very wary and more knowledgeable, I made a few bad SYSEVALS and lost my filesystem. I think that this might be related to my current problems. Could it be? ==== I'm seeing this thread in mid-discussion but if I'm understanding what you're talking about then, at least on the 39/40G, there's a better way than using IFTE. The simplest way is to divide by the domain. For example, suppose you have f(x) defined as: sin(x); x<-1 x^2-2; -1<=x<=3 and then enter this as three separate functions... F1(X)=SIN(X)/(X<-1) This works much better than IFTE because IFTE joins the end points of the functions with diagonals, producing a less than perfect graph. This alternative method causes the functions to be undefined outside the domain (divide by zero outside, by one inside) and so the functions are only plotted for the section of the axis for which they are defined. The result is generally very good. Being a 38/39/40G user rather than a 48/49 user I can't guarantee that it will also work for them but I suspect that it will. -- Colin Croft Old mathematicians never die; they just lose some of their functions. ====================================== ==== Ok.. I've identified the bug.. Damn.. :( Will try to fix it .. function... in get before arg ==== Could you please log a bug in the HP49 bugzilla: http://bugs.hpcalc.org function... in get before arg (please restarts. of from 2. Without cutting away unnecessary text. you not yourself obtain a little netiquette? You answer this way (as do many) all the time. succeeded It all page for many of us, I think. It does not, however, say anywhere that reponding at the top of the post is bad. So why is it?? I usually adjust myself to the current thread, but when I'm the first to reply, I reply at the top. Why is this a sin? Isn't this just a personal view of how you like it to be? Why is one better than the other?? Raymond P.S I'm not throwing petrol on the fire here, Im just curious. ==== IMHO, Rule 2 should be reversed. Just cut out all text which is not directly related to the answer. There is a third very nice rule to be found in ==== Well, I've been an internet and newsgroup user for over 10 years. And: 1) I don't like to waste time, and I way prefer to read what is most interesting at first. Especially since all newsreader present messages in thread. If you want the original message, either read the message above or click on the reference message at the top. Netscape do that VERY well. If you read your news from goggle for example, then the original message is also just above. Sure the usual netiquette says otherwise 2)Not everybody has a huge bandwith, even at present. You snip what has nothing to do with your answer. That's what the original message is for. If you want to read the original message, then do it. You caught me on this one, I was in a hurry and answered quickly. What I want to do with my HD space is up to me as far as I could tell. You can always filter my messages, but you would miss ROM 1.19-6 :) you junk. Is this the rule that have prevented you to answer the Terrorist -thread? It didn't I see. all over the net, and that's the way it's supposed to be. I can't use an answer to anything, if I do not know what was answered. It is not logical to open a message, read the answer at the top, scroll down in the message body to find anything that could fit on the answer, go back to the top and read another line, go back down in the message body... A simple question - are my posts here easier or more confusing to read, than the one JYA posted that I've just answered? (don't use the example one I posted yesterday as a benchmark, as I clearly stated that I do not usually post that way.) There is a limit to how many posts you can go back. Often the very first post in a thread is gone from the news server. Of course - that's what I'm doing. Hehe :-) ==== Ok. Just to make everybody happy, for once I will follow the rules : The netiquette is very well defined. Have a look at RFC 1855 http://www.cis.ohio-state.edu/cgi-bin/rfc/rfc1855.html The interesting part is on Page 8: - If you are sending a reply to a message or a posting be sure you summarize the original at the top of the message, or include just enough text of the original to give a context. This will make sure readers understand when they start to read your response. Since NetNews, especially, is proliferated by distributing the postings from one host to another, it is possible to see a response to a message before seeing the original. Giving context helps everyone. But do not include the entire original! [..previous quoted messages deleted..] You sure _you_ read it ? Try http://www.faqs.org/usenet especially ==== Sure I can and will do ASAP, that is after supper :-) I already registered myself with the bugzilla system for this purpose. -0400) to Think of a newsgroup thread as a conversation; replies follow the question or comment that they refer to. James newssvr17.news.prodigy.com!Richter0.news.prodigy.com!newsmst01.news.prodigy. c om!newscon06.news.prodigy.com!prodigy.com!enews.sgi.com!news-out.spamkiller. n ==== You'd better. Otherwise it'll be 40 lashes for you and then you'll walk the plank. (we'll give you some of those little orange inflatable arm-bands to keep you afloat for a while though just in case we want you back for some silly reason) TW ==== Yes it does: I can beta test but I cannot help for the other two problems :) Why not just release the source to the public? (just an innocent idea) -- Aaron SysRPL prog. This makes it a problem. And sorry for my ignorance, but what is the normal form of a square root? I have a hard time seeing (-((2z^2-1)*sqrt(z-1))+sqrt(z-1)*sqrt(4z^4-4z^2+z)*sqrt(1-2z^2+sqrt(4z^4-4z^ 2+z))/(z-1) as a simplification of sqrt(1-2z^2 + sqrt(4z^4 - 4z^2 + z)) ==== Try setting the -125 sturmflag, it will disable simplification of sqrt in the stack. MfG Joaquin Enriquez Beltran ==== For whatever reason (probably my own stupidity) after looking through the cas and AUG 1.3 I can't see anything on how to solve a equation in terms of it's own variables. A super simple example of what I'd like to do is: A = c * B original equation solver for c and get c = A/B Is there any way to get the HP49G to do this or will it only work with numerical solving? Todd Thomas -- Thomas Rast t.rast@iname.com ==== Could someone possibly port Exal (sp?) or Omnibus to the 49? I am looking to try it but I need a step-by-step How to decompile and recompile for dummies type of instruction. Basically yes. I haven't done much of this, but I bet if someone could write some good documentation (or bad, something would help) There would be quite an increase in good new 49 (aka: old 48) programs coming out soon after. . . I happen to know that a port of Exal is in the works (and it works well as far as I can see ;-) but it isn't complete yet and I am assuming when the person who is doing the porting feels that it is complete enough to release they will. If requests and info on release dates, etc. Just be a little more patient. . . TW ==== Not that i dislike their messages, but there are too many of them, and it's wasting both my time and my hard-drive space. noise. Maybe improve the in a LONG time!!! ==== Hmmm, I don't know what I am Wolfgang, but if you say so, who am I to dare insist the opposite? You are always right. If someone says no , then this someone is automatically rubbish to you? You name me (indirectly) a terrorist because I said that there are much more important things to be ashamed of? (Which you strategically cut out of your responce.) Because I have the opinion that I can feel respect for persons that *I* choose? Because to me a calc is more a hoby than profession? Who did I terrorize, Wolfgang? Whom did I ever tell, he/she should shut up? Did I ever call any person names? Could it be that you cannot accept any person that thinks different, Wolfgang? (This time , I don't mean Apple ;-) ) Many thanks for your deepest psychological/sociological analysis, with which you found out that I am the worst kind of rubbish, this world ==== think there is a little difference between real respect and respect on paper. I could sit down and post hymnes to the ,oh most clever programmers ever. Since none of them knows what I really think, this would be pretending and not respect. I respect everybody much better, trying to be honest and direct to them. Not speaking something and thinking something else. I show my respect expecting that we argue about things, not about the quality of the makers of things. I am not a diplomat, I know, but at least you know *exactly* what my opinion is, when I say this opinion. go NK go. i am with you %100. it is all in the attitude (perception). i just want to let you know that you never bother me. on the contrary, i wish more people were like you. our cultural lense dictates our expectations. keep sending good vibrations ;) Karagiaouroglou says... works great so far. Ray equation |1/(i*w)| - 1. Solving for w leaves me with w = EXP(i*x1). I'm not that into complex math and such, so I assume this is correct in a way. But what is the purpose of introducing the variable x1?? Solving this by hand leaves |1/(i*w)| - 1 = 0 1/|w| = 1 w = (+-) 1 Or is there something I am missing here?? ==== Yeah, great improvement, I must say! I can hardly wait to see what the next update will bring us. ==== Your solution is correct, but it is only *one* of all possible correct solutions. In complex mode the variable w is assumed to be complex, except if you have added it in your REALASSUME list. Now, because w is complex, the equation 1/|w| = 1 that you found, tells us that the absolute value of the complex variable w must be 1. Remember that a complex number can be denoted as x+i*y where x and y are the real and the imaginary part, or as r*EXP(i*theta) where r is sqrt(x^2*y^2) and theta is atan(y/x). The quantity sqrt(x^2*y^2), the magnitude or absolute value of the complex number, must be 1 according to your equation. The complex numbers that have an absolute value of 1, sit all on a circle which has its center at the point (0,0) and radius of 1. The equation w=EXP(i*x1) tells exactly this thing, x1 being an angle, which is arbitrary. There is an infinite number of complex numbers that satisfy your equation, each one having a particular angle x1 with the x-axis. Assuming you use a monospaced font to read this: | w <- This is where an arbitrary solution sits. | /_ |/ <- This is the arbitrary angle x1 --------------- | | | to get the complex number and then press ABS to find that all of them have an absolute value of 1. Your solution w=1 belongs to the set of complex numbers that satisfy the equation: | | | ------------w-- | This is where your solution sits. | The angle in this case is 0 because you found | one real solution. Hope it helped. ==== The 'x1' means that you can plug in any real number for x1 and the result will still be valid. That is, it is an arbitrary real constant ==== Nagging? I wouldn't consider this nagging. Let's look at this. When I solve for this absolute value equation I get w= -i/EXP(i*x1) on the calculator which is correct as you have mentioned. I'm not very good at complex math either, but I'll show you why the above answer has the mysterious x1 parameter in it. Notice if we substitute our answer into the original equation, we get: (1) |1/(i*(-i/EXP(i*x1))| = 1 and we can symplify this to: (2) |1/(1/EXP(i*x1))| = 1 and this boils down to (3) |EXP(i*x1)| = 1. Now here is where the complex math comes in. EXP(i*x) can be written as COS(x) + i*SIN(x) (this is called the Euler identity). And from our above equation the absolute value of this new complex form is equivalent to the magnitude of this form. And recall, |x + i*y| = SQRT(x^2 + y^2). Now let us apply this to our actual problem: We can now write (4) |COS(x1) + i*SIN(x1)| = 1 using the Euler identity and calculate the magnitude as (5) SQRT(COS(x1)^2 + SIN(x1)^2) = 1. But we know cos(x)^2 + sin(x)^2 = 1. So 1=1 and the above solution is correct. You see, the x1 parameter was introduced because we ultimately end up with the Pythagorean identity, so any value of x1 is valid. hope this helps, --Rahul Hore program works under rom 16-19.5. You may need to update your hp49g. Then there is the issue of editing maually the program files. These are intypes, types,accnts, and the account name. I recomment that you call your account : checking . I reccomend that you erase all the files created by the program and start over. Also note that you should try to work in approximate mode. It will work in exact mode, buyt then you will have to add a decimal point to all the numericals. -- http://home.earthlink.net/~fguzman1 work FYI, my famous, fabulous www.jps.net/joehorn website got nuked by Earthlink. Now only pweb.jps.net/~joehorn works. For some reason this also works: home.jps.net/~joehorn It'll take me a few days to reconstruct my old site at the new URL. It'll be the same but without the wallpaper archive. Warning: The HP section will shrink as I replace it with a new section about digital photography. -- Great minds have purposes; little minds have wishes. -- Washington Irving Sad news. ==== Why did you not include this in the text !!! Does feeding it into megabloat JAZZ qualify as entering ML from 48 K.B. ?!! ================================================ (A) I did. Try reading it: ... And you can do it too! HP has made a System RPL Development Tools package available. Get TOOLS.EXE from your favorite ftp site, or order Goodies Disk $4 for $4.95 from EduCALC. Everything you need is there, including well-written documentation for it all. But the really exciting news is that you don't even need a computer to write System RPL any more (and you don't have to be a genius hacker with a prodigious memory and infinite patience, like Alonzo Gariepy or Rick Grevelle, either!). Detlef Mueller and Brian Maguire have independently developed libraries that effectively replace the built-in User RPL command names with HP's in-house System RPL names! ... 1993. I could not explain the string input method of Jazz or MASD was correct at the time and I left out nothing available at the time. (1) Yes. It *is* entered from the HP48 keyboard, even if it requires more steps than you expected. (2) Jazz is NOT bloated software. To the contrary, it is a marvel of compact coding. I'll retract that claim as soon as you show me anything you've written that beats it in functionality and size. (3) You'll live longer if you become a little patient with your own learning curve, and stop expecting worthwhile efforts to be easy. (4) Try decaf. ==== I haven't been able to get the HP49 to provide a solution to the following equation. Probably because I'm only superficially familiar with the operation of CAS. ==== There is no point for adding DROP3, DROP4 and DROP5, as it would not save space neither time. All new commands added in the HP49 must be done using XLIB (through a ROMPTR call). A ROMPTR call is using 11 nibbles (5 (prologue) + 3 (lib id) + 3 (command number). Each time you would call DROPx memory would have to be allocated etc... Doing 3. DROPN is using only 10 nibbles (5 for a pointer to %3, and 5 for a pointer to xDROPN) Adding commands in the library 2 is first not possible (you would shift many entry points) but even if feasible, it would be risky. Any call to commands in library 2 is done using pointers. So if you saved a program calling DROP3 and sent it to another HP49, not running the same ROM version you would surely get a Memory Clear. While using a ROMPTR at worse you would get the error: Undefined XLIB Name ==== Salut, Monsieur Avenard ! OK, it's not easy to squeeze them into Lib2 anymore BUT the point is: F(X,Y,U,V)='v/(SQ(X+U)+SQ(Y+W))' DEFINE Then Modify the last commands to be And the LASTARG will return your input parameters like any build-in function. would return 4. This is not what we want. I think you now see why DROP3, DROP4, DROP5 are asked for. Werner Huysegoms seems to agree, too. Any others?? So: if you can squeeze them in 1.19-6 (or -7) ==== On Sat, 29 Sep 2001 16:46:33 +0300, (snip) As this problem would exist just the same if he added DROP503201509, it seems to me that the problem is not a reasonable justification for adding the commands you suggested. Since there are infinitely many other cases where you might want to use LASTARG in the manner described above, it seems to me that this problem is not a reasonable justification for adding ANY command such as DROP3 to get around it. Not as a sole reason, anyway. A better request would be for a general solution that would allow you to use LASTARG and not get a certain argument, if this is feasable. A better idea might be to not use LASTARG at all, and accept that the trade-offs are in favor of a change in your program's design. Of course, if you frequently need DROP3 or DROP4 or DROP5, you can code these yourself quite simply in ML. (snip) (snip) decent condition and the thought of buying it to run emu48 immediately crossed my mind. It`s key specifications are: CPU: some MIPS chip running at 70MHz Ram: 8MB Screen: 256 color TFT, 320x240 OS: Windows CE 2.11 Will it support a HP emulator? I already have a real 48GX so I`m not in dire need of it. Admitely, I see it more as a game or a geek challenge. I am not so keen on the WinCE/PocketPC platform so probably it won`t see any other applications running except the emulator. According to the words out there the CE 2.11 was a horrible OS, CE3 is considered way better. So, will it work or it would be a complete and utter waste of money? -- ==== congratulations on your future new toy (the one who dies with the most toys wins). usually the GUI is not good due to the low resolution of the display. it is like going to a retaurant and eat with a pocket knife (a little unconfortable). also, i have tried emus before and they froze my PDA from time to time (CE2.1). the emu will take about 5 Mgs of your memory. i think it should work, but the performance will be slow. if i were to buy a PDA and i had the money, i would consider a linux PDA instead ==== That will be tough, especially the 8 MB of RAM. On my HP jornada 548 with 32MB of memory the emulator runs at approximately 1.5 times the speed of a real HP4x. The jornada is using a 133Mhz CPU. Aero`s price jumped to $160 (forgot to mention that it was an online auction). So there goes my future toy... I must find somenthing else to play with :-) -- Zorz, http://users.auth.gr/iilias ==== I'm actualy using eql49 to store all the ecuations that i need, but i can't tidy the ecuations like i want. I want to do a eql48.lib but not whit the ecuations that come in the original one, i want to do one whit my equations, equations, eql48.lib is compiled so the order of groups and equations can't be rearranged. But you can build your own list of equations, stored in a variable USEREQ (if I remember well). There's a format to respect to permit to eql49 to correctly read the file, and you can also add images and comments. The format is described in the documentation which cames with the library, the eql49r27.doc file where you find all the informations you may need. Also try to use the library etools.lib which may help you creating the USEREQ file. Hope this can help! (anyway try to contact the author!) ==== example: DERIV(X,X) or DERIV(Z,Z) or DERIV(Y^2,Y) press enter and i get: bad argument type message DERIVX works fine. I guess that is due problems i had during upgrading bios. (had to do it 2 times. 1st time i had message that(if i recall correctly) bank(or port) 0 not empty - how to delete it before upgrade. What can I do to make sure upgrading is to be succesful - deleting what, reseting, purging ......? ==== DERIV is OK in RPN mode Not working in Algebraic mode. ????? -0400) what, Read and follow directions at http://www.epita.fr/~avenar_j/hp/49.html? ==== For most libraries you store them in port 2. If you've transfered L902 to your home directory simply move it to port 2 with the FILES menu. If L902 is on the stack then store it directly to port 2 with 2 STO. Some libraries prefer to be placed in port 0 or 1 for speed or other reasons if thats the case the procedure is still the same just stor it in port 0 or 1. You then need to reboot the calculator with a simultaneous press of ON F3 to attach and make the library available. After rebooting you should be able to display and access the the library functions with RShift LIB. Some of the better libraries may also add entries in to the APPS menu. All the loaded library functions and commands will also be listed in the CAT menu in italics. If the library is corrupted or causes problems rebooting then read the manual on how to reboot without loading libraries and then delete the offending library from port memory. It's also amazingly simple to create your own UserRpl libraries to make your own commands allways available and protected from accidental deletion or modification. ==== You can execute programs by simply presing their soft menu key in the VARS menu. But what you have is not a program but a library. Press its soft menu key to put the library on the stack. Then do 2 STO. This stores the library in port 2. Turn the calculator off, then back on. Press now red-shift, 2 to go to the libraries menu. The library has now a new soft menu key, which you must press to see the commands of the library. These commands are also accessible from the catalog and you can also type their names in the command line. Did that help? Greetings, Nick. P.S. Turning the calc off, then back on, should be enough to activate the library. But some libraries have special installation procedures, so ypo should also read the readme files. -0400) ==== Oh yes, that's right. The procedures for the 48 and the 49 got mixed up in my mind. Sorry for that. ==== purposes, the visible rompointers may not yet be usable or even visible. ON hold C has to be done anyway to be shure. PS. Sorry, if Nick, as usual, answered to your reply because my system kills his messages. I see I did the ==== I have tryed all the af the tricks in the answers - but it will not execcute. I can see the lib. in the file menu but I can't get it executed. -- MVH nyhedsmeddelelse:3BB4EB66.E33179A6@iname.com... config object, which in most of the case is only attaching the library. So, unless the config object is creating some internal used objects, ATTACH is enough. And if it wasn't it means the library is not well built anyway. At least Nick is always nice. It would be nice if this newsgroup went back to its original good athmosphere. or 2). In most cases you'll want to put the library into port 2, i.e. the flash ROM. menu will show up on the bottom of the screen. Now look for the library you just installed, e.g. in the case of elepo the softkey might be labeled Elepo . If it's not on the first page, press NXT until it shows up. Then press the corresponding menu key; e.g. if the label Elepo is above F2, press F2. Now the menu belonging to elepo shows up. Each command in the library has its own menu key in this menu. Since I don't have elepo, I can't help you any further, but there are usually three ways to proceed: 1. Trial and error , i.e. try all commands, starting with the first. 2. Look out for a command named ABOUT , HELP or similiar. 3. Read the documentation that came with the library. I hope that helped; if not, feel free to ask again =) Thomas -- Thomas Rast ==== We probably wish all just this! We all should be happy that such wonderful calculators like the 48 and 49 exist and that we users have at least a a sense a guarant for the former good athmosphere (from the moment on when he moderated his famous auto-reply). I also will take out Nick from my filter provided he starts really thinking twice to one a day in the average. If, however, he does really *not* care about being on the kill-list of Detlef and you, for instance, and feels even happy with this, then he'd better create his own newsgroup. ==== To give a look to this program, that should turn a matrix of 0 into a matrix of 9 with the pretense put , because it doesn't work? << [ [0;0] ;[0;0] ] sto A; << FOR( I ; 1 ; 2 ) FOR( J ; 1 ; 2 ) ==== f ' is 1st dir' , F' is antidir' . f'( F ) = f(x) =================================== You enter System RPL source code from the calculator keyboard between double quotes. The entire source code thus goes on the stack as a single string object. The next step is to convert it from the source code string into the actual desired object. This is done by feeding the string into a compiler such as Jazz or MASD or whatever your favorite compiler is. Since each compiler has its own personality and quirks, be sure to follow the syntax and requirements of your compiler when writing the source code string. Otherwise you'll get errors or worse. Of course, this requires you to read the compiler's documentation. This can be a daunting task, but be patient and persistent. It's worth it. This is only a tiny part of a huge topic. Please feel free to ask specific questions for more specific answers. Hope this helps! -- MSGBOX D/dx= DERV(s1^2,s1): Hey you wanted simple. . . ;-) Just remember that to do CAS cacluctions from the home or program area, ==== Very trivial, easier to do in other ways and perhaps not what you were looking for but... ERASE: DISP 1; First 50 primes are... : 2/STO/X: FOR I=1 TO 50 STEP 1; DISP 2;X; WAIT 0.25; NEXTPRIME(X)/STO/X: END: -- Colin Croft Old mathematicians never die; they just lose some of their functions. ====================================== ==== I've posting this message to all the addresses I've found in text files concerning upgrade of hp48 calculator. All those files have been downloaded from site www.hpcalc.org. However, only Stefano Garavaglia replied till now, so I decided to post this message to this newasgroup. The thing is, I would like to upgrade my hp48. Since I'm not very familiar with electronics, soldering and stuff, my mate who is an electronic professional (don't know how it's said in english) is going to solder the ram chips instead of me. We decided the best option would be to solder some smd chips instead of making a pcb and using the dip chips. My task is to get the plans, texts about upgrade, buy all the neccessary items and bring them to him; his only task is to solder everything together. I'm from Slovenia, I have some questions regarding SMD ram chips : -First of all, I'd like to know whether it's possible to upgrade to more than 256kb mem using ONLY 2 smd chips, one of them being a replacement for the original 32kb chip (can I use 2 512kb chips instead of 2 128kb chips - if not, why so??) -Which access time is the best? I saw on one picture that one of the original chips being used in hp48g is sony's cxk58257am-70L (working at 70ns) and read somewhere that the other one is mitsubishi's m5m256bfp-15L (however, there isn't any 15ns version of mitsubishi chip existing, so it might be a mistake and the 45L or 55L chip was meant). In those upgrade-texts, there are several 128kb chips mentioned : - hitachi's hm628128 (version LFP-7,LFP-7SL,LFP-8,BLFP-8SL) (working at either 70 or 85ns - what does SL mean?) - samsung's km681000BLG-7L (working at 70ns) - nec's pd431000a-70L (working at 70ns) the only hint I got were two sentences saying ram shouldn't have more than 120ns access time and that the author isn't sure about exact access time needed but that the 100ns chip works in his G. so, which speed is the optimum one? some are using 100ns chips, others 70ns, third ones 85ns - I really don't know which one should I choose/buy? -If I'd like to upgrade my HP48 to 256kb RAM, I need 2 low-power static ram chips organised 128K * 8 (SOP32 version)! Am I right? The only chips I've found here in Slovenia that almost match these criteria are Hyundai's (now Hynix's) HY628100B-LLG55 chips (128k x 8bit 5.0V Low Power CMOS Slow SRAM - low low power, 55ns). I'm saying almost because I'm not sure about the speed - I haven't noticed any chips being used for the upgrade with 55ns access time. Can these chips be used or are they too fast ? Furthermore, can you please explain me why this chip is entitled slow sram (does this mean it is slow-working and should therefore be avoided?), although in the pdf manual there is written that it is designed for high speed low power circuit tehnology? What's more, can you please explain me the difference between low low power chip and low power chip (concerning power consumption)? Is LL chip suitable for upgrade or is there anything wrong with it? Last but not least : have those Hyundai chips the same shape as the others (so, can they be attached to the existing 32pins on the hp48's pcb without any problems?)? PLEASE, PLEASE check out the www.hynix.com/datasheet/eng/sram/detail09.jsp?PartNo=HY628100B&product_kind= 1M or even better, please download the pdf manual from site http://www.hynix.com/datasheet/pdf/sram/HY628100B.PDF and PLEASE tell me whether I can use these chips and if there's something about this chip that is not good for upgrading (voltage, ...). I really don't know anything about ==== Does someone what's happen to hpcalc.org? It's been more than one month without processing updates! Looks so much. Does Eric have any problem? I Hope not! Luis. -- --------------------------------------- Luis Morales Boisset ==== using ROM 1.19-5 beta What is it???? What caused it? Any solution? I also had some problems upgrading to 1.19-5 bios - message : bank ) not empty! so i had to do it twice! HELP! ==== Use PINIT command. Piotr Kowalewski For example? Did you read the rest of the document? It explains the differences between System RPL and User RPL, which you seem to be confused about. Actually, it's quite the other way around. I never worked for HP and in fact have pretty much given up on 'em, but my writing has always kicked ass. So there. ;-) -- ==== But most links as, e.g., http://pweb.jps.net/~joehorn/contents.htm lead to File Not Found (Error 404). BTW: I unfortenutely don't know why he doesn't post in c.s.hp48 anymore, do you? Michael -- See Joseph K. Horn's post in the thread Re: Any news about HHC 2001:A handheld odyssey . anymore, I count four posts from Joseph yesterday evening. As for why he doesn't post more often, I don't know either; perhaps other commitments, or the low signal to noise ratio and rudeness in this newsgroup lately, or maybe he just hasn't felt much need to comment. James Check now. The pweb site is being constructed out of the pieces of the old www site. It should be finished within 24 hours from now. Yes. -- Time's fun when you're having flies! -- Kermit the Frog If Netscape 6 is like Netscape 4, just go to the desired newsgroup, click on Edit, then Message Filters. There are a ton of options; it's very powerful. -- ==== especially Matt Bryant ==== The filter I have set doesn't seem to work! I don't know what fogie means - could you explain this? If you change your name and change the way you post with it, we'll have no problem! But you should know that I wouldn't put you into my killfile because of one posting only - I read your answers and some other posts and it wasn't as bad as the one you think of, but somewhat aggressive enough. . . Christoph ==== I would like to write a ML program waiting for a key to be pressed. Anyone could help me ? (I tried with OUT and IN, but it just crashed my HP :)) I have a little program that tells me the in and out codes of all keys it waits for a keypress (doesn't work for ON) It is barely commented but you should be able to get enough information from this to get underway Once compiled it should be 50.5 bytes and have a checksum of #C535h CODE SAVE GOSBVL Flush % flush keybuffer *Start C=0 A A=0 A C+1 A *Main OUT=C % set mask A=IN2 C+C A % shift mask one bit GOTO Main *Push R0=A A R1=A A GOSBVL Flush % flush pressed key LOAD GOSBVL PUSH2# % push IN and OUT RPL ENDCODE @ -- This message was written with 100% recycled electrons ==== Because of a design problem of the internal Yorke chip the r=IN command must be executed on an even address. The solution is to call an ASM routine in ROM where the IN Command is on an even address. So replace your A=IN with GOSBVL =AINRTN or C=IN with GOSBVL =CINRTN . ==== must ==== I've just recently been answering support questions for people at calc.org. I am disgusted at some of the questions that come there (and surprised at the difference between the questions) There were 10 the last time. 2 HP questions. 1 of the TI questions was Where does the cord plug in on my calculator? 2 were where does it plug in on the computer? 1 was Do I need a cord to send things to my calculator from the computer? 4 were I've sent a game to my calculator and it won't run. What do I do? (the whole question, nothing else, no title of game, nothing, just that one sentance, and all four were different people) Now the HP questions. Something has happened to my PORT2 memory (any guesses on what is going to be my answer. . . ;-) and it displays a bunch of crap. i've tried deleting it manually using the FILER and on the home screen using PURGE. I read in my book and can't find what is wrong. Do I need to reinstall my ROM or send it back? what happened to my calculator? (notice the difference between the questions) The second was I've tried using several different programs for note saving but none have all the things I'd like. I'd like one that (list of things). Do you have any recomendations for a program that does that? UUGH! First I'd like to ask any TI person who regularly visits the DG's or anything (where are those by the way? I'd like to check out the sorry affair myself) how they manage to stand it? I'm basically CRAZY with the stupid questions I've gotten. I consider myself a pretty easy-going person, but the way I've received these questions is ridiculous. I've receive about three intelligent TI questions of the many I've answered there. Almost all of the HP questions qive details and things they have tried already(and many say i read in my book. . . Think I heard that once in a TI related question). Why can TI people not do this? I know that there are many more that are much younger and many that only care about games, but why do ==== Well, My guess is, and may I add I ask a lot of HP questions and some of them could be more detailed :), is that you have a much larger I hate to say stupid but shall we say non serious calculator users using TI calcs because every HS, or college I've been to or heard about uses/pushes/requires the Ti's. So you probably have a 20...30....maybe 100 Ti to one HP users in the younger ages. I can say here in ND, that after taking all of HS, pre- calc, 2 calc classes, stats, univ physics, discrete, linear algebra, that of all those classes mine is the only HP graphic I have seen. Most my instructors don't even know that HP makes a graphic calc. They all require, teach students how to use the Ti's. Anyway my point is I guess that you probably have a lot of Ti users that unless their instructor hands them a step by step instruction sheet with their lesson for the day they couldn't use their Ti if they had to aside to add/sub/mul/div etc. So if any of these few decide to go a bit further they are pretty much at ground zero. I can stumble my way around a Ti86 better than most of the people I know in class can use one, and they own one. I think with HP users most of us have been using HP for awhile, and went that way because we knew exactly what we wanted, not just cause the class list said I have to have one. Many of the people using HP are professionals and serious students, because if your going to use a HP in HS/college/etc. your going to have to be fairly knowlegeable cause your going to have to learn how to use it, not just press the buttons the sheet your instructor gave you has on it. Kinda like linux -vs- windows, if you want to use unix/linux your going to have to learn something, not just point and click stuff. Course that's a HUGE generalization and probably sounds eliteist, and I'm sure there are many very skilled Ti users out there, but I think there are just so many that aren't that they show up more than the HP users. Other than that I think the only other possible reason would be of course ==== (snip) My theory is this: These are young people (lots more teenagers with TIs than HPs?) who have yet to grasp how you ask questions. They use dialects in their speech, using different vocabulary and pronounciation to different people. They *ought* to learn some time through gradeschool that English classes are there to teach a particular formalized English, and that a difference exists between the language taught there and what idiom they pick up elsewhere. They've spent much of their life so far asking questions in a particular way, want to ask questions in that way, and are annoyed at people like you and me who don't answer questions in the way they expect. Some of those people who ask you questions want what I call Conversational Debugging, and it looks like this: B: Well, that depends on your problem. What's wrong? A: This program doesn't work. B: How doesn't it work? A: Well, *duh*, I try to run it and nothing happens. B: What program is this? B: How did you install it? (and so on) This makes sense in a classroom where long and involved questions will only be cut off or interrupted and you'll be forced into a Conversational Debugging session *anyway*, but no sense at all over questions were asked, you might want to write short and clear instructions are how good questions should be given to you, with lots of RIGHT and WRONG -type examples. After giving this to people who ask such questions as in your examples, you can start to acquire a hard line about bad questions ( I've already sent you the instructions. This question fits BAD#4. If you send it to me *again* I'm just going it ignore it, OK? ) and focus on answering the good ones. Your only danger is having Mr. BAD#4 deciding to send you ten thousand copies of his question. Don't be sorry. Rants are healthy and good -- and rants on Usenet are *especially* good, because they lack all problems that a rant otherwise has, like boring your audience or ==== Here's a link to the TI-89/92+ discussion group: http://www-s.ti.com/cgi-bin/discuss/sdbmessage.cgi?databasetoopen=calculator s&topicarea=TI-89/92+Plus&do_2=1&viewmethod=Thread This Calculator Comparisons discussion group attracts many HP'ers: http://www-s.ti.com/cgi-bin/discuss/sdbmessage.cgi?databasetoopen=calculator s&topicarea=Graphing+Calculator+Comparisons&do_2=1 ==== Using the HP48G+ for statistics. I am taking a statistics class with a second one next semester. I have both a TI83+ and a HP48G+. I am more comfortable with the HP48G+ but it is imperative that it has at a minimum, the statistical functions of the TI83+. The professor uses a TI83+ and I need to keep up with her. I have loaded STAT48PRO into my HP48G+ and learned how to use the data manager. Part of the problem is I don’t know statistics well enough to know what capabilities I will need until it is too late. Is there a simpler program out there that will calculate median & mode for example? OK. I lost my first attempted reply, so I decided to give that instead. The idea is that you set flag -85 and then immediately realize that, by gum, How utterly amazing! Then you're supposed to try and compile a few short programs, and make the connection that UserRPL is a special case of SysRPL, that then it is obvious that any given UserRPL has analogous SysRPL code, that therefore anything you can do in UserRPL, you can do in SysRPL -- and in very much the same way. For more information, see hpcalc.org and its SysRPL documentation. I realize that this is not necessarily what you're looking for, but it's helpful anyway and will do what you want to do in SysRPL. (snip) entries on the computer? I've tried using the dissasembler, but I can't find a way to use it to search through the ROM to find what I need. Anyone have a suggestions or a different program to use (besides Jazz or any other on calc programs)? TW ~The enemy's gate is down. You can check out SAD (sadhp105.tgz), which is available on hpcalc.org . There's an option in SAD which shows you the unsupported entries. Raymond entries on to use suggestions or Good point, and one I did not address in my original post. My students were instructed, in this case, to simply write Obviously wrong, out of time on their test paper. At which point they get the normal partial credit for everything they did except the error. The key here is that I ==== recently I have experienced some really sad TTRM events during the use of the Xmodem connectivity kit. Thus I checked my WSLOG (mostly out of curiosity) and found some entries which are not explained in the '49 AUG. F-0000... @ ??? 4-0000... @ system time is corrupt (really? =) E-01.01.91 00:00:00 @ ??? 3-01.01.91 00:05:53 @ run through address 0. Nick and Richard have put together a complete list in a recent thread: http://groups.google.com/groups?threadm=cd9ca36b.0108210306.a5b8c07%40postin