A2 At least in RPN mode, hitting the menu key for a function/command immediately executes it. If you want to get it onto the command line instead, just hit ENTRY; you get an empty command line in PRG mode. Now you can hit the menu key, and the function name is echoed to the command line. E.g. when you want to get the stack diagram (if you have SDIAG of course) of a command in a menu. You'd else have to type it in. Does the HP49 have a function to get a function's inverse function? For example if I have y=x^2-9 I would like to get y=(+/-)sqrt(x+9). I'm also wonder if it's possible to get the inverse of ie. y=x^2-2x+4? If it's possible, how? Think about what it is you want to do. You want to isolate one variable in an equation, hence you SOLVE for it. Undrestand what you do before you apply the SOLVE command. ==== | But could you explain to me how to do it by hand? | If I have y=x^2-2x+4 I should get 1+sqrt(x-3) and 1-sqrt(x-3) according to | the calc. I can't find a way to isolate x. first rewrite it: (full square) y=x^2-2x+4 = x^2-2x+1+3 ^^^^^^^^ = (x-1)^2+3 next: y-3=(x-1)^2 then sqrt(y-3)=x-1 then it is what it should be Well, this is very simple! Write your equation in the form x^2-2*x+(4-y)=0, then solve for x (you know, delta=2^2-4*(4-y), ...quadratic equation) Remember that this is not the place where you would ask people to solve your school maths problems :-) Your teacher can help you much better! __ PS. You get 1+sqrt(y-3) & 1-sqrt(y-3), *not* 1+sqrt(x-3) & 1-sqrt(y-3) (in terms ==== I am running RH7.2 and have managed to get syslog to output on the HP49 using terminaltor, but how do you actully logon? I have nothing against a wishlist on the newsgroup as a place to discuss the desirability and feasibility of additional features, but I would mention that it seems appropriate to submit feature requests at http://bugs.hpcalc.org/ by choosing Enhancement for the Severity. Personally, my wish is that all known bugs be fixed, a new offical Commercial ROM be released, and the documentation updated and corrected, and then a new series of Beta ROMs with additional features be started. This might be acceptable for an unofficial ROM, but I'd like to see these Beta ROMs eventually released as Commercial, and I don't think it ==== Yes, this version was only a few hundred bytes. If I would isolate the current RPLCPL from Emacs, it would be a lot bigger since is does so much more now (variables, completion from user supplied lists, merging catalog and extable where appropriate, decompilation and help from the choose box...). That uses a lot of different parts of Emacs, so isolating it would not make sense. I could make a small stand-alone version with fewer features for the ROM, if that was what HP wanted. First) Commercial ROM with updated documents from the current HP with current sales of the 49G ???!! Keep on dreaming, James !! Second) A separate Library is a good idea !! I vote for a separate Downloadable Pack, which installes itself to the first User Bank and integrates itself to the system bankswitching schema!!! Is this possible, JYA ??? So - at least on the old 48-series - if you during the setting of a new alarm creation looked at your [VAR] you could then view or copy the ALRMDAT BUT generally you just use the User Interface and don't mess with it directly. Well, I wondered how long I had to wait for a word-joke (we say in this way in English?) like yours on my name! :-) Well, thanx for the info. Unfortunately I can't look at my VAR during an alarm creation/edit, or at least I don't know how to do it (VAR key is disabled). Also I wonder where the calculator store ALARMs information for not yet delivered alarms. Maybe in the hidden dir as Bruce Horrocks said me in a private reply to my post? Any hints about reaching this lost variable? looking for a converter that will convert text to an ascii code for computers. ie type in a string hit a softkey and return a list of ascii codes for each letter I recieved an answer from Calcpro. They are in business and are working to fix their broken web site. I placed my order this morning. Of course I mean compressing data *inside* a library. The easiest way to do this is to just pick an object in the library, compress it, and then make it self-compressible with Emacs (RPLED, build a SysRPL :: ; -thingy around the NIBHEX with your decompressor before the ; Simple, except for some strings that Emacs doesn't want to resolve to a NIBHEX (why not?); for these put UserRPL quotes and your decompressor in, select everything and paste it outside of Emacs onto the stack). With this method I nearly halved the size of Mendel v1.5, changing it from 30K to 17K. The overhead of decompressing is light with BZ, and with most programs you aren't going to incurr it at all very much -- just maybe once when data objects are pulled onto the stack at startup, or whenever you use a certain feature of the program. I apply essentially this method from the very day when we ported BZ to the 49. E.g., I made a Mig3BZ from the famous and huge Mig library which is only 4 KB (on my site only, for copyright reason :-) To see whether it pays to compress inside a lib, first compress the lib itself. If the compression rate is more than 25% (which never happens with my libraries :-) it probably pays. E.g., version 1.07 of Emacs (25.6 KB) has a compression rate of 25.45%. That means it can probably be brought down below 20 KB (I did it without loosing speed :-). I do not understand what Emacs has in common with the whole compression story. Of course, you may use its decompressor (which is just taken from OT49) but that's all. One must be familiar with some properties of BZ. For instance, just comressing a rompointer in a splitted library and storing it back may result in a crash if running the library lateron. To compress a rompointer's content, it *must* be recalled directly. E.g., easily done with DType from OT49. If first edited by Emacs and cast on the stack it may not recompile to its original. Also, you have to know whether certain rompointers are specially written for speed (usually in asm, as this is the case for several Emacs rompointers). Compressing such rompointers would slow down their execution and is contraproductive. Compression is an art and needs a lot of experience to be applied properly. I've got a pretty esoteric question: why does HP-16C index register have 68 bits when all normal registers can be at most 64 bit long? What those 4 extra bits could be used for, bar writing (nearly) endless ISZ loops, I mean. ==== Samson Cables... just what I was looking for ! They may be a little more expensive there... but they ship international with 2-day Priority UPS which is what I wanted... because I am now living in Barbados and my credit card statement address does not match my shipping address which LIMITS my options online I used to buy stuff for my 48SX from calcpro.com... but something gone wrong whatsoever. ==== Oh-no! There has been some development since 1.16 ROM like EXP2HYP ... > Try with EXP2HYP in CAT to do it. 55447677.0303050811.c5833c7@posting.google.com... > (e^x)-(e^-x)=3 <=> ((e^x)-(e^-x))/2 = 3/2 <=> sinh(x) = 3/2 <=> x = > asinh(3/2) > 3 2 / ASINH ->NUM gives the answer. > Is there any way to get the HP49 to do this conversion? I can > get it to go from sinh(x) to (exp(x)-exp(-x))/2, but I can't > get it to go the other way. --CS ==== > How would I go about solving for an equation with a variable in the > exponent, such as: (e^x)-(e^-x)=3? On my ROM v1.19.6 it works: 'EXP(X)-EXP(-X)=3' 'X' SOLVE -> { 'X=-(LN(2)+(-(2*i*n2*pi)-LN(3+SQRT(13))))' 'X=2*i*n1*pi+LN(-((-3+SQRT(13))/2))' } ==== > The number of Inquisition deaths throughout history obviously isn't > something people kept accurate records about, but I've seen the > estimate of nine million, which is close enough to JYA's statement. > This number might be utterly wrong, but I've seen it more than once, > so presumably a lot of people have read it too. Apparently there has been some scholoarly work done. See > http://www.bringyou.to/apologetics/num25.htm > which seems to place the numbers somewhere around 30,000 with a lot of criticism of one author who claims 300,000. The site is > interesting because the end of the page lists a great number of authorities. I'm glad we finally got away from that useless namecalling and onto something useful like whether the Inquisition or the Holocaust killed more people. The only thing important to take from any of this is that we humans have a very cruel and stupid side to our nature. Right now our civilization is probably in more danger than ever before. Some say Bush is the reason. Some say Sadaam is the reason. I think we're all the reason. Each of us has one and only one thing we can do about the world situation. We can control ourself. That's the only influence we have. I suggest using it. Barry ==== TomCee Or set flag -60 and press alpha once to lock, again to unlock. > Please remind me--what input methods allows me to temporarily set > alpha mode for data entry? > -- James ==== I have a list of reals and I want to work with them. The problem is this: For example, 5. in the stack is 33920 0000 0000 0000 0050 but inside a list is 0A9F2. Why is that? How can I recognize when it is encoded as a five nibble structure (I guess it's an address) or the usual 21 nibble structure of a real? I need to know this in order to handle the list of reals in a proper way. Any hints? -- Beto ==== > I have a list of reals and I want to work with them. > The problem is this: > For example, 5. in the stack is 33920 0000 0000 0000 0050 > but inside a list is 0A9F2. > > Why is that? > Even on the stack %5 will be a pointer to the built-in constant. > Any hints? > Use an array of reals instead. Then you can be sure to always have the data part inside your your structure, instead of pointers. Apart from that, an array saves ome memory because the prologue is stored only once. Oh, I see you're referring to a 49G w/o noting it in the header. Then it could be different. Raymond ==== >Use an array of reals instead. >Then you can be sure to always have the data part >inside your your structure, instead of pointers. >Apart from that, an array saves ome memory >because the prologue is stored only once. I'll check that and see how better is it. Right now I solved the *problem* by checking every prologue inside the list and processing the object accordingly. -- Beto ==== I have a list of reals and I want to work with them. >The problem is this: >For example, 5. in the stack is 33920 0000 0000 0000 0050 >but inside a list is 0A9F2. Why is that? How can I recognize when it is encoded as a five nibble structure (I >guess it's an address) or the usual 21 nibble structure of a real? I need to know this in order to handle the list of reals in a proper >way. Any hints? > First, read up on the internals of objects and how direct and indirect execution and the RPL inner loop work. Some sources were pointed out here : http://groups.google.com/groups?selm=riigqu4h17698m2i8122oqkeuhqh8viepp%404a x.com When you've done that, you can understand how the following code fragment works : C=DAT0 A CD0EX RSTK=C A=DAT0 A LC(5) =PRLG A=A-C A C=RSTK ?A#0 A GOYES + D0=C + D0=D0+ 5 A=DAT0 W D0=C Assuming the address of the real number object or object pointer is contained in D0, the above fragment returns the real number body in A.W . The only registers used are A.W, C.A and one RSTK level. All others remain unchanged. ---------------------------------------------------------------------------- --- Jonathan Busby - before replying. ==== >First, read up on the internals of objects and how direct and indirect >execution and the RPL inner loop work. Some sources were pointed out >here : >http://groups.google.com/groups?selm=riigqu4h17698m2i8122oqkeuhqh8viepp%404 ax.com Done, hard to digest for an ML beginner like me. >When you've done that, you can understand how the following code >fragment works : > C=DAT0 A > CD0EX > RSTK=C > A=DAT0 A > LC(5) =PRLG > A=A-C A > C=RSTK > ?A#0 A > GOYES + > D0=C >+ D0=D0+ 5 > A=DAT0 W > D0=C Assuming the address of the real number object or object pointer is >contained in D0, the above fragment returns the real number body in >A.W . The only registers used are A.W, C.A and one RSTK level. All >others remain unchanged. I've already done something similar, reading the prologue and process accordingly. -- Beto ==== I went on their website. But it was not their normal website, it was... HP/Compaq's Small Business Site. www.smb.compaq.com When I just checked, they were marked While Supplies Last http://www.smb.compaq.com/optionCategories.asp?oi=E9CED&BEID=19701&SBLID=&Su perCategoryId=163&SuperCatDesc=hp+calculators&ParentID=30&ParentDesc=softwar e +and++peripherals Goodluck >>I bought a new 49G from HP last week. > > > directly. > > I went on their website and couldn't find 49G listed anymore ==== X > http://www.smb.compaq.com/optionCategories.asp?oi=E9CED&BEID=19701&SBLID=&Su perCategoryId=163&SuperCatDesc=hp+calculators&ParentID=30&ParentDesc=softwar e+and++peripherals X Ah-haa! The F1633A is stil sold by Compaq part of the HPQ ==== The hp9g calculator show there is different from the one show in HP Calculators site, at least this one has a nice silver buttom instead the golden one I've seen: http://www.smb.compaq.com/images/hp--g.jpg Compare it with this (the official hp9g): http://hpshopping.speedera.net/www.shopping.hp.com/shopping/images/products/ f2222a_400.jpg J.Manrique #1077 HPCC Member Users Club from Gijon > I went on their website. But it was not their normal website, it was... > HP/Compaq's Small Business Site. > > www.smb.compaq.com > > When I just checked, they were marked While Supplies Last > > http://www.smb.compaq.com/optionCategories.asp?oi=E9CED&BEID=19701&SBLID=&Su p erCategoryId=163&SuperCatDesc=hp+calculators&ParentID=30&ParentDesc=software + and++peripherals > > Goodluck > >>I bought a new 49G from HP last week. > > > directly. > > I went on their website and couldn't find 49G listed anymore ==== I went to small business link and they are one of those that have to match up stament address to the credit card + they dont ship outside USA I am now living in BARBADOS and enjoying island life... so I bought it from Samson Cables who ship international 2-day delivery... so I payed the piper and called the tune After I get it... I may be able to answer your other question... who knows :) Have a great day have a great day ! ==== I have a series of x,y points which I believe are part of an ellipitical or circular curve. Does anyone have a pointer to a program which will allow me to deduce/discover the formula of the curve? I'm presuming some sort of least-squares calculation, but I have been unable to find a formula for fitting to curves (lines yes, curves no). I *think* there was one in one of the HP formula books for the HP-65/HP67 but my copies seem to have disappeared long since.. TIA Geoff -- We sleep soundly in our beds because rough men stand ready in the night to visit violence on those who would do us harm. George Orwell ==== > I'm trying to enter a complex number in polar form - you know, magnitude, > then the angle buttons, then the angle. I can enter a numeric quantity of > degrees fine, but when I try to enter something like pi/2 it stops me right ... I have tried to make the same using different ways. I think symbolics are not allowed as a part of a complex number. Bad Argument Regrettable that the greak pi is not handeled in the edit line for this paritcular case My recomandation, use DEG-mode and choose 90 Á instead of pi/2, than convert back to RAD-mode :-) The mode toggling you can wonderfull handle by keyman (www.hpcalc.org) ..Heiko ==== Use Eulers relation and enter it as an exponential 'r*e^(pi/2)*i' where r is the magnitude and (pi/2) is whatever angle formula you want in terms of pi then use the rshift+EVAL key to return a complex number whenever you want ==== > Folks: > > I'm trying to enter a complex number in polar form - you know, magnitude, Here's a tiny utility for the HP49 that will create vectors in the form (a,V2 -16 CF V-> R->C -16 SF >> Example: 5 2 --> (5.,<2.) Best Wishes, Bill ==== Sorry.... made a typo in my first reply The Parenthesis has to be around the entire Exponent say for example 25*e^(pi/2*i) which would give (25 <90) with calculator in degree mode if you did rshift+EVAL ==== > Sorry.... made a typo in my first reply Ok, you got me there :-) > The Parenthesis has to be around the entire Exponent Too late. Just want to ask a related problem, that always irritates me whenever this problem occurs, using EQW. Now IÇve entered: 'r*e^(pi/2)*i' on stack level 1, as you suggested originally. IÇve found that that rshift+EVAL does nothing, because of the missing parenthesis. So I press [down-arrow] and EQW-windows opens: I want to highlight i*e^(pi/2) but instead I get highlighted either: 1) 5 2) i 3) e^(pi/2) So then I press [F2/cursor] and get either: 1) 5 2) 5*i*e^(pi/2) 3) or any of the previous 1), 2), 3)....ARGH! :-) I want to use [cut] + [LS] + [()/-] and [paste], if not anyone else has a better suggestion... I donÇt know if IÇve seen the answer before, but it irritates me a lot, because I think it shouldnÇt be necessary to type in the whole thing from the beginning. Now this is a relative simple expression, but IÇve tried to manipulate much longer expressions, say i.e. I need an extra parenthesis somewhere and I found out too late... Any help on that? > 25*e^(pi/2*i) which would give (25 <90) with calculator in degree mode if you did rshift+EVAL Mine doesnÇt give that in degree mode. DoesnÇt help to switch from rectangular to polar mode. The closest thing mine comes to your result is in radian mode, where it gives me '25*i' - I guess that was not what you meant mode if you did rshift+EVAL. Mine rshift + EVAL does not change the expression - any comments on that? Martin J. ==== Now IÇve entered: 'r*e^(pi/2)*i' on stack level 1, as you suggested >originally. >IÇve found that that rshift+EVAL does nothing, because of the missing >parenthesis. that's NOT what I get.... I simply get a WRONG answer >Mine rshift + EVAL does not change the expression - any comments on that? The default flags should allow evaluation to a NUMBER. Have you changed the flags ? I am using an HP48SX with ROM version E what are you using ? ==== > Also look at MODES flag -114. You're talking about hp49? (Hp48 has 64 flags). miz. ==== > You're talking about hp49? (Hp48 has 64 flags). Yes, of course - sorry. On the '48 run this program: << DO DUP EXPAN UNTIL DUP ROT SAME END DO DUP COLCT UNTIL DUP ROT SAME END >> It's a very slow approach though, so if you need to do this alot, you'd want to install ALG48 and use its RSIM command. ==== Yes, I had already thought of that (since I couldn't see how many EXPANs I needed, do until the expression didn't change), but I don't think you need the second loop (just one COLCT needed). I just wanted to know if I was missing some hp built-in command or if I could do the same with a FOR loop instead of a DO loop. About Alg48, I'd like something that I can include in a small colection of programs, so the slow but small RPL program might be better. > You're talking about hp49? (Hp48 has 64 flags). > > Yes, of course - sorry. On the '48 run this program: > > << DO DUP EXPAN UNTIL DUP ROT SAME END DO DUP COLCT UNTIL DUP ROT SAME END > > It's a very slow approach though, so if you need to do this alot, you'd want > to install ALG48 and use its RSIM command. > ==== yeoldcoffeeshop.com and olde spellings yeoldecoffeeshoppe.com URL: http://cgi.ebay.com.au/ws/eBayISAPI.dll?ViewItem&item=3405639845 Open a cafe, put in some computers, and make the biggest frachise in history!! -- _____ ___ / ___/ / __ __ __ /__ / / /__/ _ / -_)(_- (_- _/_ / _ _/_//_/__/ /__)/__)/___/ www.chess3.com ==== > Sorry I forgot to say an INPUTFORM in SYSRPL for the HP40G... > in an aplet or a library... There are two kinds of input forms and choose boxes in the Elsie family (HP38/39/40): modal and modeless. The former being the same as the input forms and choose boxes in the HP48. Modeless environments allow the user to task switch any time. The differences are described in detail in the June 1998 issue of the HP Journal, on page 50: http://www.hp.com/hpinfo/abouthp/histnfacts/museum/personalsystems/0026/othe r/0026hpjournal01.pdf In fact, modal versions have been rewritten so that they meet the topic outer loop requirements. Therefore, while the descriptions of these tools for the HP48 and HP49 are still valid, some details in the message handlers are now different. For example, to retrieve the list of picked indices of a choose box, you have to use TOLVar41@ instead of 15GETLAM. You'll find the simulation aplet: http://www.hpcalc.org/hp39/math/simul.zip Here's a list of the input forms and choose boxes I've used in my aplets, hopefully they'll help you, although I don't have time to comment on them now. http://www.hpcalc.org/hp39/games/launch12.zip (modeless choose box) http://www.hpcalc.org/hp39/games/mh39-32.zip (modeless input form) http://www.hpcalc.org/hp39/apps/nf39-03.zip (modal choose box and input form) HPCC #1046 ==== To communicate between them I use a freeware software in iPAQ (3760) called Filegram (http://www.neutrio.com/wince/Filegram/filegram.html), which allows to communicate with the XModem protocol, and I change the communication settings in HP to be able to communicate: iPAQ (3760) Filegram: Protocol: XModem Flow Control: None Port: Com2 Speed: 2400 Slow Down Transmission: yes And in HP48GX we have (In I/O Settings): IR/Wire: IR ASCII/binary: binary baud:2400 parity: none 0 cksum: 3 translate: 1 Then you can choose to send a file (XSEND) or receive a file (XRECV), just put the name on the stack and run the command. There is another program more 'powerfull' that even supports KERMIT, but is not freeware, called vxHpc (http://www.cam.com/vxhpc.html), which I don't use because it corrupted the iPAQ memory a few times, but besides that is very good... I've made a library to HP48, that automatically changes the I/O parameters to allow communication with Filegram in iPAQ, and restores them after transferring in a transparent way, If you are interested it in HPCalc since there's not a lot of information available. Paulo Pinheiro > Any idea how I can get these to talk to each other with IR? > Anybody tried it? > I want to know if I can exchange information this way rather than get > the HP48G comms cable for my desktop PC. > > I have used PeaceMaker pro for my Ipaq in the past which has pretty > good for communicating with other devices by IR, for instance palm > pilots and handheld scanners. Not sure if it works for the 48G tho. > ==== Can anybody answer these questions: 1. Are the HP48GX and HP49G still being manufactured ? Who knows? > 2. If so, what is the latest ROM version on each of the these models ? VERSION command gives: HP48-R, so it's R-ROM 49: 1.18, this is the latest HO Commercial version 49beta: 1.19-6 (*) this is the latest beta (*) Jean.Yves Avenard has the 1.19-7 in his sleeves...not yet ready > 3. Does the HP49G surpass the 48GX in every way ? What about all the many A) no plug-in card slots @ who cares! It has 1MB of FlashRAM and extra 256KB of IRAM, system free is ~256KB B) no IR @ the serial works C) no EQLIB build-in @ get one from the www.hpcalc.org D) no tactile feel on the keyboard @ yeah, but the keybounde is adjustable AND you can use the cursor keys in the alpha mode AND you have additional levels of User Mode key assignments with build-in hold-shift AND you have more keys with a cleaner look AND the keyboard mechanism will last longer AND... BUT the ENTER key is in the wrong place and too small (compared to what the old-timers are used to) The key labels may wear off more easily the old injection molding is the best! education.... surely not being able to use these with the 49G is a disadvantage > ? Just load the needed libraries into the FlashRAM, where they will stay put! (until YOU erase them, no more backup batteries for the 1MB RAM modele) => The total overall value for money is more twice that of the HP 48G whether it is software (*) OR amount of RAM. The keyboard is crap. (*) The software of the HP 49G: B) ALG48 (or rather ALG49) build-in C) Erable (for 49) build-in D) MASD Assembler/SysRPL compiler buil-in E) Loads of other neat stuff, all nicely integrated into the system BUT The CAS sets your DEG/RAD, Exact/Approx., Real/Complex, etc. flags at will. The only remedy is have something like << { -17 -103 -105 } CF DROP >> stored in bENTER (b=Greek beta) and bith User Mode and Vectored ENTER on It's good to have in aENTER (a = Greek alpha) << STR->> to make everything work alright AND I could just go on, but - naah! Buy a CASIO Xpander instead... Veli-Pekka Nousiainen > Hoping for some advice before I plunk down more money (I already own a 48SX > flexibility.... it seemed to be the worlds only open-system calculator.... > right down to being able to program new function in assembly language !) Now it's time to move on... and I would like to hear from others who have been > there and done that > Stephen Mendes ==== I get the picture I ordered a 49G just got to decide if to cancel the 48GX or let both come I am a bit of a calculator freak... and I would suppose you might have both of them as well ?? ==== I get the picture I ordered a 49G just got to decide if to cancel the 48GX or let both come I am a bit of a calculator freak... and I would suppose you might have both of >them as well ?? Both are great calculators, If I had the money I'd buy a 48GX myself to use it specially for number crunching, because it's keys are far more better. In the meantime, I'm happy with my 49G, even though the ENTER, ON and Right-Shift are totally erased. -- Beto ==== I get the picture I ordered a 49G just got to decide if to cancel the 48GX or let both come I am a bit of a calculator freak... and I would suppose you might have both of > them as well ?? Try to buy them both used to save your money. ==== X > Both are great calculators, If I had the money I'd buy a 48GX myself I have them both, but use the 48GX seldom now. ONLY when I need to use units & EQW > to use it specially for number crunching, because it's keys are far > more better. In the meantime, I'm happy with my 49G, even though the > ENTER, ON and Right-Shift are totally erased. Naturally this means the key labels, which in my calculator are still intact after more than three years of (ab)use helping HP 49G users by the phone in Finland. -- > Beto Veli-Pekka Nousiainen ==== >X >> to use it specially for number crunching, because it's keys are far >> more better. In the meantime, I'm happy with my 49G, even though the >> ENTER, ON and Right-Shift are totally erased. >Naturally this means the key labels, which in my calculator are still intact >after more than three years of (ab)use helping HP 49G users by the phone in Finland. Three years and intact key labels? I thought yours should be more *erased* than mine... Do you really use it? :) -- Beto ==== ==== X >> ENTER, ON and Right-Shift are totally erased. >Naturally this means the key labels, which in my calculator are still intact >after more than three years of (ab)use helping HP 49G users by the phone in Finland. Three years and intact key labels? I thought yours should be more > *erased* than mine... Do you really use it? :) Not much on weekends, but you probably A) have a calculator from a different batch and B) have used the translucent slide-in cover C) have fingernails (: I bite mine) ==== > I bought a Dell laptop which has only USB ports and also a Targus port > replicator. Unfortunately I could not establish connection to the hp49 or > the 48GX. using the standard hp connectivity kit software and cable. After reading some the contributions to this group dealing with USB - Serial > conversion I wonder why the Targus failed. > Two years ago James Prange said a USB Serial Converter-UC 310 worked well > for him. Somebody else suggested the Belkin USB Serial Adapter for $74 and > others recommended the Keyspan high speed USB Serial Adapter. Are there any new insights into this? Has anybody had good or bad > experiences with USB-Serial adapters? Does the connectivity kit work with > these devices or does one go back to a (cumbersome) kermit protocoll? It seems that I owe you an apology; my converter doesn't seem to work quite as transparently as I had thought at that time. Using HPComm v3.0r4, ASCII and binary file transfers, Backup..., and Restore (all using the Kermit protocol) work fine, but Capture Image... doesn't work with the converter. If I use the built-in serial port instead of the converter, then Capture Image... does work. The HP Connectivity Kit XModem kit version 1.0.9 can't find the COM port when I use my converter. If I use the built-in serial port instead of the converter, then the XModem kit does work. http://www.hilgraeve.com/htpe/index.html) works with either the converter or the built-in serial port. I guess that the HyperTerminal that comes with MS Windows would also work, but I don't know that for certain. This works as a terminal emulator, allows you to Capture Text... (from XMIT or the various print commands), Send Text File... (using SRECV on the calculator), transfer using the Kermit protocol, and transfer using the XModem protocol. But note that it doesn't recognize the 49G's XModem Server mode, so you have to use XSEND and XRECV on the calculator and always provide a name on the receiving device. Also, for XModem, it initially expects CRC error checking, and it takes a considerable time to fall back to checksum error checking. All of this is using MS Window 98SE. With the various possible operating systems and software and hardware configurations, Your Mileage May Vary. > My Targus cost $86 and I am shipping it back. I'm not familiar with the Targus port replicator, but if it's just a USB to RS-232 serial converter, $86 seems expensive. But it seems that if you want to connect your calculator to the laptop, then you're going to need some sort of serial port. Before you send it back, try using HyperTerminal with it. Also make sure that you have everything set up for the right COM port, speed, and parity. -- James ==== I have been away from this forum for some time and was surprised to see the news that HP has for all intents and purposes given up on the calculator business. Does anyone know if this is due to the new managment and the merger with Compaq, or lack of interest by the public/HP management for the HP calculators? Also, are there any remnants of the US or Australian HP calculator development teams left which will be allowed to develop new products in the future? On a more technical note I checked the HP site for HP 49G support and nothing, what is the latest usable version of the HP 49G continues. ==== I'm very tired of people not using Google Search on this newsgroup! HP HAS NOT GIVEN UP ON THE CALCULATORS NOR RPN period Veli-Pekka Nousiainen > I have been away from this forum for some time and was surprised to see the > news that HP has for all intents and purposes given up on the calculator > business. Does anyone know if this is due to the new managment and the > merger with Compaq, or lack of interest by the public/HP management for the > HP calculators? Also, are there any remnants of the US or Australian HP > calculator development teams left which will be allowed to develop new > products in the future? On a more technical note I checked the HP site for > HP 49G support and nothing, what is the latest usable version of the HP 49G > continues. ==== Anyway... you know your calculators good.... so don't let the postings frustrate you we all need you in here :) ==== NO, but I voluntarely (as the current president of STak) help the local distributor and the users in Finland and here at comp.sys.hp48 > Anyway... you know your calculators good.... so don't let the postings > frustrate you Perhaps good, but not excellent - I have been often corrected by JYA WR & others > we all need you in here :) all? (-; ==== HP may not have give up, but Staples, Wal-Mart, Office-Depot, and Office-Max don't have any in various states where I've looked: NM, TX, FL, CA, and NV. ==== Has anyone tried to make this interface? http://home.nordnet.fr/~phpamart/pdf/fiche_cable_hard.pdf And perhaps made a useful program on the pc/calculator? I know that itÇs possibly to use the connectivity kit also, but thought that making your own program/interface perhaps could make more interesting programs for the pc like data-loggers etc. I saw a person had made a interface that could control win-amp... Sounds like a fun project.... Any comments? I think it could be interesting to make, but right now I donÇt have time enough..... Just wanted to hear if anyone has any experience, doing so... Martin Jorgensen. begin 666 Martin Jorgensen.vcf ==== I've heard so much about the HP48GX but I didn't get a chance to see it before they got obsoleted. In my current life, I don't think I'll ever use the full power of a HP48GX or a HP41CX. The 200LX built in calculator and Lotus123 is more than enough for I what I need. However, I was wondering if a HP48GX is worth a HP41CX? Thx weemeng ==== They are NOT obsolete..... they are still selling brand new everyday... from many stores all over the world. HP48GX is a superior machine to the HP41CX.... but then I am scientist as well as a computer programmer.... and if you are not, then its value to you... may be negligible ==== --------------------------------------------------------------------- Hallo ! I just analyzed the video signals on the 10 pin connector of the HP38G calculator. I think they are more or less identical to the HP48GX signals found on the internal connector. But I found a problem with the vertical synchronization. This bit sometimes vanishes, for instance in the plot screen without any menue bar. Does anybody have any further informations about the display video signals? There had been some overhad displays for the HP38G and HP48GX. How they done it ?? Greetings, Christian Daniel --------------------------------------------------------------------- name=christian.daniel.vcf filename=christian.daniel.vcf begin:vcard n:Daniel;Christian tel;work:++49/421/539-4694 x-mozilla-html:FALSE org:Astrium Space Infrastructure;IO 65, Data Processing Equipment adr:;;Huenefeldstrasse 1-5;Bremen;;D-28199;Germany version:2.1 x-mozilla-cpt:;29752 fn:Christian Daniel end:vcard --------------F8CDFB1CBD56B1F1C6F5C967-- ==== I can't comment on the electronics of it but I can tell you that the overhead display unit is exceedingly annoying because if you plug the calculator into it while it is switched on then the LCD in the panel almost always displays 'rolled' by about half a screen and you have to repeatedly hit a button to roll it down the screen until it displays properly. On the other hand if you plug it in with the calculator switched off and then turn the calculator on then there is usually no problem. This is the case for the 38G and the 39/40G. > Hallo ! > > I just analyzed the video signals on the 10 pin connector of the HP38G > calculator. I think they are more or less identical to the HP48GX > signals found on the internal connector. > > But I found a problem with the vertical synchronization. This bit > sometimes vanishes, for instance in the plot screen without any menue > bar. > > Does anybody have any further informations about the display video > signals? > There had been some overhad displays for the HP38G and HP48GX. How they > done it ?? > > Greetings, Christian Daniel ==== use this kind of calc, but the big prob is I dont know how to use that, so if you know where can i get a manual for this calc + card,ill be very happy. Fred ==== Frederic Durand schrieb im Newsbeitrag > use this kind of calc, but the big prob is I dont know how to use > that, so if you know where can i get a manual for this calc + card,ill > be very happy. on www.hpcalc.org you will find a lot of usefull stuff. And there you find also a manual for the HP48G/GX (works for the HP48SX too) in pdf-format. But be warned, it's a huge file! Roman Fred ==== > Frederic Durand schrieb im Newsbeitrag > use this kind of calc, but the big prob is I dont know how to use > that, so if you know where can i get a manual for this calc + card,ill > be very happy. on www.hpcalc.org you will find a lot of usefull stuff. And there you find > also a manual for the HP48G/GX (works for the HP48SX too) in pdf-format. But > be warned, it's a huge file! Well, except that some things (list processing, new commands, etc.) on the 48G series aren't the same as on the 48S series. At http://www.hpmuseum.org/cd/cddesc.htm you can purchase a CD-ROM set or DVD-ROM that has scanned-in PDF copies of the 48SX Owner's Manual Volumes 1 and 2 plus the 48SX Programmer's Reference manual. It doesn't have the Quick Reference (pocket) Guide though. It also doesn't have the 82211A Owner's Manual, but since most of what was on the card is built-in to the 48G series, I expect that the 48G Series Guide should be helpful for it. -- James ==== Usually the numeric solver in HP49 (the solve lin sys in this case) seems to work ok, but today it gave me really strange results. Either this is a bug or (more likely) I've managed to set some very esoteric flag. But I can't find what is wrong and a on+F3 doesn't helt. ROM version is 1.19-6 I have the following matrixes: A = [[-0.120273333333 -1] [0.64102 0]] B = [[-0.64102] [0.120273333333]] I enter num.solv with rshift 7 4. I choose (F2) the matrix A for A and B for B, scroll to X and press solve. Solution: [[-2.9095203426E-26] [-6.4164881584E72]] this is obviously wrong. However cond(A) = 1.7476 and if I remember my numerical linear algebra correct this indicates that the problem is NOT numerically instable. If I try X = A^(-1)*B I get the more resonable solution [[0.187628]] [0.618453]] (remaining digits omitted) which is about what MATLAB R12.1 also says. This is probably not how num.solv is implemented but since the condition number is so low we should expect the difference to be small iirc. Of course the problem is trivial to solve by hand. But there must be some reason the calc behaves in this way? What do you think? ==== Is anyone familiar with this calculator? It looks like they gave the 92 more memory, and a slightly different interface (still menus, no soft keys),and a better appearance. It seems to have the same (dis)abilities as the TI-89. I've seen one TI-200 around campus, and I was wondering if anyone had intimate knowledge of it. Are the commands different? Is it any better than the TI-92? I know that this is way off subject, but I also know that there are a lot of calculator fanatics on this NG. I'm a die hard HP user, but I work as a college tutor in the US. As far as I know, I'm the only one at my school using the HP49. There are three Engineering professors that use the HP48. Everyone else I've seen uses the TI's. Someone came in the other day proudly displaying his TI Voyage 200 and asked me if I knew anything about it. I played around with it but couldn't find anything new. --CS ==== I hope somebody, kindly soul reads this :-) I have this 2.order diff. equation, which I enter, this way in my hp49: d1d1Y(X) + 4*d1Y(X) + 4*Y(X) = e^(-3*x) 'Y(X)' Then I get a solution, which I cannot see whether is correct or not, because in my math classes we donÇt get solutions like that... My hp49 gives this result: { } By hand I get this solution: y(x) = c1*e^(-2*x) + c2*t*e^(-2*x) + e^(-3*x) ; c1, c2 is real numbers... My hand-solution looks much better than my hp49g-solution. IÇve tried to simplify my hp49g-result, but it wonÇt do any better. Anyone knows the secret here? I canÇt see if the two results are the same............ Another derived question is: When I get the result, { } and I press down-arrow, the editor opens in text-edit-mode (or whatever itÇs called). So I have to manually erase the { in the beginning and the } in the end, to be able to get the equation into the equation writer-editor... IsnÇt there a command for that, or is it really necessary to use the arrow-keys and press backspace and press down-arrow key + rightshift + right-arrow + backspace to get rid of these (stupid) {}-characters??? I hope my question is clear enough, if not - be sure to give me your comments, and thanks in advance for any posts that helps me (and others as well, I guess)... br, Martin Jorgensen. begin 666 Martin Jorgensen.vcf ==== Is it possible to simulate saturn cpu architecture in a fpga(field programmable gate array) Wherefrom i can get detailed architecture of saturn cpu? ==== > Is it possible to simulate saturn cpu architecture in a fpga(field > programmable gate array) Yes. > Wherefrom i can get detailed architecture of saturn cpu? ==== > Is it possible to simulate saturn cpu architecture in a fpga(field > programmable gate array) Yes > Wherefrom i can get detailed architecture of saturn cpu? Well, I do not know who has them, but if I was to redesign a SATURN, I would not reuse the same architecture (internal), but redesign it as a modern CPU, with a 5 level pipe and a cache. I would use a 32 bit or 64Bit UAL. This would allow most instruction to run in 1 or 2 cycles. The original Saturn is build around the Saturn 4 bit serial bus which is extremely ineffective. this would not be something I would copy. I do believe that someone started working on such a project, but never finished it. ==== Is there a way to solve equation with 3 unknow without using the Matrix Writer? Maybe put then on Stack and using cmd solve? Can you help me? E.M ==== What happened to Hrastprogrammer !!?? He seems very upset indeed, and has deleted from the web any info or details on his wonderful HP-41X, HP-42X and HP-71X emulators for the HP-48GX/49G !!!, just see this link: http://hrastprogrammer.tripod.com/ Can someone (hrastprogrammer ?) shed some light on just what happened ? Perhaps it would be useful to know, just in case ... ==== I went ahead and ordered a hp48gx from amazon. They have them for the cheapest plus they have free shipping. I dont have the converter piece anymore to use my old hp49 cable. Should I buy a hp48 cable or a hp49 cable? ==== > I dont have the converter piece anymore to use my old hp49 cable. > Should I buy a hp48 cable or a hp49 cable? > Use a HP-48 cable. Raymond ==== >Should I buy a hp48 cable or a hp49 cable? All you need for any of the 48 series calcs is a 48 cable.... it is a 4 pin to 9 pin cable and it sells for $24.95 at Samson Cables here's the link.. you can even see a picture of it... http://www.samsoncables.com/catalog/prodDetail.cfm?Prod_ID=195&Sku=SCHP48 ==== I noticed calcpro sells the cable for $15. Is calcpro a good company to deal with? ==== I called their phone and got an answering machine. If they are that hard to reach... how long do you think you will have to wait to get your order ?????? a Saturday night.... now that is what I call proper service. ==== Maybe ALG48 + Erable ?? > When watching the ebay auctions I noticed that the 48g+'s are going > for realitivly low prices? You said I could fit the CAS on an 128kb > RAMcard, so could I fit the CAS on the 48g+ which has 128kb? I'm a bit confused here - which CAS is being discussed here? Is it > Alg48, or some other CAS? Mike ==== > Maybe ALG48 + Erable ?? Yes ==== > Is there any way to change the tone and duration for the key click > option in the Calculator Modes form? What is the use of the Print form (to print an object)? The only port > setting is Wire. I think that the port setting is a carry-over from the 48G series, which gives you the choices of Infrared or Wire. Even though the only choice on the 49G's input form is Wire, you can still choose to print by infrared by clearing system flag -34. If you do -34 CF and then try to print something (without using the PRINT input form), then the I/O annunciator comes on for a while, but since the IR LED (and, I suppose, supporting circuitry) is missing, the printer doesn't respond. Using the PRINT input form forces flag -34 to set. There's a similar situation with the TRANSFER input form. In this case, you can choose to transfer by infrared by setting system flag -33, with similar results. -- James ==== nine hours or something for each message to be posted. This seems to be very limiting. (your questions take a long time to be answered. And why answer a simple question? It has probaly already been answered (or not)) It would be nice if this group was abandoned and one of those good ones was formed to replace it. I dont know what they are called but I can describe them. People register and can post messages like here. Moderaters control and there are many nice features such as the ability to post polls. And messages appear almost instantly. So you read through all of the threads and by the time you are done people have replied to what you have written. I am not sure what these are called, but I participate in several. If you want to see one look here http://www.bombermanboard.com It is for bomberman which is one of my favorite games ==== I am not sure what these are called, Trolls, they're called. richard ==== > nine hours or something for each message to be posted. My experience is that a post shows up on various usenet servers within a few minutes. Perhaps you're reading this newsgroup by using the Google newsgroup archives. It does indeed take a long time for posts to show up there, but the archives are meant mostly for archiving messages, not so much for real-time interaction. Try using a newsreader and a usenet server. > This seems to > be very limiting. (your questions take a long time to be answered. > And why answer a simple question? It has probaly already been > answered (or not)) Of course, I don't suppose that many of us are on the newsgroup 24/7, and even when I do see a post that I might respond to, I rarely respond immediately; I like to consider what (if anything) I might say, check that what I'm about to say is correct, and sometimes I simply have more important things to do. > It would be nice if this group was abandoned and one of those good > ones was formed to replace it. No! That would not be nice! This *is* a good group, and I see no reason to replace it. > I dont know what they are called but I > can describe them. People register and can post messages like here. > Moderaters control And who would the moderators be? To be sure, there are sometimes posts that *I* think should never have been posted, but who gets to decide? > and there are many nice features such as the > ability to post polls. And messages appear almost instantly. So you > read through all of the threads and by the time you are done people > have replied to what you have written. I am not sure what these are called, but I participate in several. If > you want to see one look here http://www.bombermanboard.com It is for > bomberman which is one of my favorite games -- James ==== input... > nine hours or something for each message to be posted. This seems to > be very limiting. (your questions take a long time to be answered. > And why answer a simple question? It has probaly already been > answered (or not)) It would be nice if this group was abandoned and one of those good > ones was formed to replace it. I dont know what they are called but I > can describe them. People register and can post messages like here. > Moderaters control and there are many nice features such as the > ability to post polls. And messages appear almost instantly. So you > read through all of the threads and by the time you are done people > have replied to what you have written. I am not sure what these are called, but I participate in several. If > you want to see one look here http://www.bombermanboard.com It is for > bomberman which is one of my favorite games ==== YOu are all right. I did not know how to use this correctly. I feel dumb for suggesting, but at least I now know how to do it. ==== > I'm very tired of people not using Google Search on this newsgroup! > HP HAS NOT GIVEN UP ON THE CALCULATORS NOR RPN > period I read this newsgroup nearly every day and all I've gotten from it is the rumors of new calculators. Barry ==== > HP may not have give up, but Staples, Wal-Mart, Office-Depot, and > Office-Max don't have any in various states where I've looked: NM, TX, > FL, CA, and NV. Agreed. I think actions speak louder than words. ACO is terminated. Kinpo calcs are being rebadged as HP, and no one sells HPs anymore. But there is an HP marketing memo saying they are still in the business. Let's see, which do I believe? Steve ==== Me > HP may not have give up, but Staples, Wal-Mart, Office-Depot, and > Office-Max don't have any in various states where I've looked: NM, TX, > FL, CA, and NV. Agreed. I think actions speak louder than words. ACO is terminated. > Kinpo calcs are being rebadged as HP, and no one sells HPs anymore. > But there is an HP marketing memo saying they are still in the > business. Let's see, which do I believe? Steve ==== Exactly. Rumors that they are coming, rumors that they are not. I wish HP clears the fog once and for all by announcing officially SOMETHING or states that it's over. -- Thierry Morissette thm47@msn.com > I'm very tired of people not using Google Search on this > newsgroup! > HP HAS NOT GIVEN UP ON THE CALCULATORS NOR RPN > period I read this newsgroup nearly every day and all I've gotten from it > is the rumors of new calculators. Barry > ==== ... >> Agreed. I think actions speak louder than words. ACO is terminated. >> Kinpo calcs are being rebadged as HP, and no one sells HPs anymore. BTW, has anyone out there actually _seen_ a 9S or 9G in a store? Craig <02ac30e9a098f22a308d27bd620a93ff@news.teranews.com> <898ba.818$mE2.56953@news20.bellglobal.com> ==== In message <898ba.818$mE2.56953@news20.bellglobal.com>, Thierry >Exactly. Rumors that they are coming, rumors that they are not. I wish >HP clears the fog once and for all by announcing officially SOMETHING >or states >that it's over. There is a press release announcing the 9s and 9g at: The following quote comes from it: The HP 9s and HP 9g are just the beginning of what we have in store for our customers throughout the coming year, said Fred Valdez, general manager, Calculator Division, HP Personal Systems Group. We've accelerated our product development plans and begun working with our new aggressive sales and marketing partners around the world. -- Bruce Horrocks Hampshire England bh@granby.demon.co.uk ==== > ... >> Agreed. I think actions speak louder than words. ACO is terminated. >> Kinpo calcs are being rebadged as HP, and no one sells HPs anymore. BTW, has anyone out there actually _seen_ a 9S or 9G in a store? So far I haven't seen it in any stores... and I live in Canada. ==== Maybe. But: 1) I do not consider the 9s and 9g as canon HPs. 2) We've accelerated our product development plans and begun working with our new aggressive sales and marketing partners around the world. can mean ABSOLUTELY anything or nothing. Plans are plans, and he only talks about SALES AND MARKETING, not actual R&D. 3) I'll believe only what I see. Right now, HP has gone the way of the super-crap TI-like too expensive for what it's worth TOY. -- Thierry Morissette thm47@msn.com > In message <898ba.818$mE2.56953@news20.bellglobal.com>, Thierry >Exactly. Rumors that they are coming, rumors that they are not. I wish >HP clears the fog once and for all by announcing officially SOMETHING >or states >that it's over. > There is a press release announcing the 9s and 9g at: The following quote comes from it: The HP 9s and HP 9g are just the beginning of what we have in store for > our customers throughout the coming year, said Fred Valdez, general > manager, Calculator Division, HP Personal Systems Group. We've > accelerated our product development plans and begun working with our new > aggressive sales and marketing partners around the world. -- > Bruce Horrocks > Hampshire > England > bh@granby.demon.co.uk ==== How about this for rumors: http://www.hpmuseum.org/cgi-sys/cgiwrap/hpmuseum/forum.cgi?read=29158 X > 2) We've accelerated our product development plans and begun working with > our new aggressive sales and marketing partners around the world. can mean > ABSOLUTELY anything or nothing. Plans are plans, and he only talks about > SALES AND MARKETING, not actual R&D. X ==== In ten years of extensive use, my HP48GX is starting to fail. I get a reset almost every day, and the memory is lost. It is not the battery and I'm not using any card, sometimes it just hungs up and performs a 'soft reset' as pressing ON-C, but sometimes I get the annoying hard reset, and I loose all memory. I guess it is time to upgrade to a new HP49, but you know that the future of HP calculator is uncertain. I am considering maybe going for a TI, but have no experience with such calculators. Can anyone with experience in both HP and TI tell me what are the advantages and drawbacks of each? I want to go for an HP, but I guess the future of TI is better. Can anyone give me a good reason why I should not go for a TI? Mariano ==== wee-meng lee schrieb im Newsbeitrag I've heard so much about the HP48GX but I didn't get a chance to see it > before they got obsoleted. In my current life, I don't think I'll ever use the full power of a HP48GX > or a HP41CX. The 200LX built in calculator and Lotus123 is more than enough > for I what I need. However, I was wondering if a HP48GX is worth a HP41CX? > As another poster said in other words, the HP-48GX is much more powerful than the HP-41CX. However, many people consider the HP-41CX to be more practicable, and easier to use. Nowadays a new 41CX costs much more than a brand new 48GX. Raymond ==== X > However, many people consider the HP-41CX > to be more practicable, and easier to use. > Nowadays a new 41CX costs much more than a brand new 48GX. I consider the HP 42S being even better than the HP 41CV ==== >HP48GX is a superior machine to the HP41CX.... No. In my opinion the just are not compareable. for instance the HP 41CX has a 4-Port Interface for locking Solid State Software Modules, Printers an a megnetic card reader as a mass stoage system, and, IIRC, they sold a digital cassette drive as a mass storage for some time. In addition to this, there was an interfacemodule for the HP 41 series to the HP-IB and it's devices. The HP 48 GX has none of the above features. But it has much more memory, it has graphic capabilities (the HP 41 don't) and it can solve task symbolically (the HP 41 can't). The only thing i would wish for my HP 48 is a portable mass storage system and something like the HP-IB for using the HP 48GX as a mobile measuring computer. Volker ==== >HP48GX is a superior machine to the HP41CX.... No. In my opinion the just are not compareable. X > The HP 48 GX has none of the above features. But it has much more > memory, it has graphic capabilities (the HP 41 don't) and it can solve > task symbolically (the HP 41 can't). So, basically: 1) The best number cruncher=HP 41 [42] 2) The best symbolic graphics calc= HP 48 [49] ? PS: I use the 41CX as a clock to wake me up and remainder for TV-shows ==== Was trying to decide whether to find someone to do a trade or to buy a 48GX to see what this 48GX is all about. Reading about it is one thing, having it in your hands to touch and feel is another. Thx weemeng ==== I owned a '41 from when they first came out - they really were waaaay ahead of their time. I've heard so much about the HP48GX but I didn't get a chance to see it > before they got obsoleted. In my current life, I don't think I'll ever use the full power of a HP48GX > or a HP41CX. The 200LX built in calculator and Lotus123 is more than enough > for I what I need. However, I was wondering if a HP48GX is worth a HP41CX? Thx > weemeng ==== > > Usually the numeric solver in HP49 (the solve lin sys in this case) > seems to work ok, but today it gave me really strange results. Either > this is a bug or (more likely) I've managed to set some very esoteric > flag. But I can't find what is wrong and a on+F3 doesn't helt. ROM > version is 1.19-6 > > I have the following matrixes: > > A = [[-0.120273333333 -1] > [0.64102 0]] > > B = [[-0.64102] > [0.120273333333]] > > > I enter num.solv with rshift 7 4. I choose (F2) the matrix A for A > and B for B, scroll to X and press solve. > > Solution: [[-2.9095203426E-26] > [-6.4164881584E72]] I got the same results. I tried a manual solve by dividing the constants by the coefficients, and that worked fine. Then I put the two matrices on the stack, started the solver, and used CALC-OK to insert them into the fields. That worked fine too. But if I open the solver and use the matrix writer to enter the matrices, I get those strange values. Can anyone shed any light on this? Bill Maybe it Expects Mattresses Markwick ==== > > > I have the following matrixes: > > A = [[-0.120273333333 -1] > [0.64102 0]] > > B = [[-0.64102] > [0.120273333333]] > > What do you think? > I think you should enter -1. and 0. in the matrix A (note the decimal points), it will work then. Apparently the numeric solver doesn't check the type of the input matrix (in this case, A is type 29. because it contains an integer - if you change the -1 and 0 to -1. and 0., the matrix type becomes 3. - real matrix.) Definitely a bug. Werner ==== Usually the numeric solver in HP49 (the solve lin sys in this case) > seems to work ok, but today it gave me really strange results. Either > this is a bug or (more likely) I've managed to set some very esoteric > flag. But I can't find what is wrong and a on+F3 doesn't helt. ROM > version is 1.19-6 I have the following matrixes: A = [[-0.120273333333 -1] > [0.64102 0]] B = [[-0.64102] > [0.120273333333]] > I enter num.solv with rshift 7 4. I choose (F2) the matrix A for A > and B for B, scroll to X and press solve. Solution: [[-2.9095203426E-26] > [-6.4164881584E72]] But if I open the solver and use the matrix writer to enter the > matrices, I get those strange values. Can anyone shed any light on this? Bill Maybe it Expects Mattresses Markwick matrices. Enter numeric ( e.g. by usig real values), or you convert them first. Then the results seems ok Try: 'A' RCL redshift and ENTER 'AA' STO 'B' RCL redshift and ENTER 'BB' STO Using the numeric solver with numeric matrices AA and BB gives [[ .187628 ][ .618453 ]] Using symbolic mattrices One result I got was [[[ -8. E-26 ][ -6.E72]] Later/next time I got Solutions: Symbolic Matrix Which is an object with size 7789. bytes, and at a closer look perhaps is a program/ list of pointers and other stuff. Tried to evaluate this object. Can't see anything happen. HOWEVER, trying a third time, result in a MEMORY CLEAR You should report this on the bug list as a SERIOUS bug. There is something strange going on here. Gjermund Skailand ==== Although it is almost the same as the TI 92 I would say, after reading the specs and holding it in my hands in a store (together with the TI 92) that the V200 is a much better product, BUT since I'm more towards a vertical model I would choose a TI 89 anyway (even if it is not as good) > Is anyone familiar with this calculator? It looks like they gave > the 92 more memory, and a slightly different interface (still menus, > no soft keys),and a better appearance. It seems to have the same > (dis)abilities as the TI-89. I've seen one TI-200 around campus, and I was wondering if anyone > had intimate knowledge of it. Are the commands different? Is it > any better than the TI-92? I know that this is way off subject, but I also know that there are > a lot of calculator fanatics on this NG. I'm a die hard HP user, but > I work as a college tutor in the US. As far as I know, I'm the only > one at my school using the HP49. There are three Engineering > professors that use the HP48. Everyone else I've seen uses the TI's. Someone came in the other day proudly displaying his TI Voyage 200 > and asked me if I knew anything about it. I played around with it but > couldn't find anything new. --CS ==== > Is anyone familiar with this calculator? It looks like they gave > the 92 more memory, and a slightly different interface (still menus, > no soft keys),and a better appearance. It seems to have the same > (dis)abilities as the TI-89. I've seen one TI-200 around campus, and I was wondering if anyone > had intimate knowledge of it. Are the commands different? Is it > any better than the TI-92? > > I know that this is way off subject, but I also know that there are > a lot of calculator fanatics on this NG. I'm a die hard HP user, but > I work as a college tutor in the US. As far as I know, I'm the only > one at my school using the HP49. There are three Engineering > professors that use the HP48. Everyone else I've seen uses the TI's. > > Someone came in the other day proudly displaying his TI Voyage 200 > and asked me if I knew anything about it. I played around with it but > couldn't find anything new. > > --CS It is basically a TI-92+ with more FLASH memory and a new operating system, AMS 2.07. I am not sure what this has that 2.05 doesnt but the newest 2.08, which came out near it and should be very similar, has an application desktop ripped off from Casio that supports only FLASH apps. ASM programs and BASIC programs cant be accessed from here. AMS 2.08 also uses the quartz cyrstal in hardware version 2 calcs for a time and date function. Time and date wont work on HW1 calcs since they dont have the quartz crystal but thats irrelevant since all V200s should be HW2. Mathematically you get 3 or 4 trig functions: sinh, tanh, cosh and one other thing I think. In AMS 2.07/2.08 you basically get a desktop, time, date, and a few trig functions for 140k of your archive. That may not be a problem on the V200 with all its memory but for TI89 users like me with only 768k FLASH, taking 140k for the above is not worth it. I heard something about V200 users getting some FLASH apps free from TI but Im not sure about that. There are TI message boards of course but if you dont want to get the forum OT you can ==== Elvis Mera schrieb > Is there a way to solve equation with 3 unknow without using the Matrix > Writer? > Maybe put then on Stack and using cmd solve? > Can you help me? a few days ago I read in the User Manual how to do that for 2 unknowns Chapter 8... Solving linear Equations with the command RREF Row-Reduced Echelon Form You just need to type in *one matrix and type RREF ENTER ..Heiko ==== > Is there a way to solve equation with 3 unknow without using the Matrix > Writer? Whether you are using an HP49 or HP48 the following method is the same. For a linear system described by A*x=b where A is a square matrix nxn and x and b are vectors nx1 then the solution is given by A^-1*A*x=A^-1*b or in words the product of the inverse of A times the vector b producing an nx1 vector. On the formidable stack place the matrix A in form [[row1][row2][row3]], then press the inverse key 1/x, then enter the b vector in the form [[b1][b2][b3]] and finally press the product key *. There you have it(?) !Demeter! ==== Depending on the model you can solve a system of three variables with / SOLVE SOLVEVX LINSOLVE RREF and perhaps ISOL or even GREDUCE Nope: with some of the commands you need to use SUBST Also sometimes the SYST2MAT is useful I would use / [vector] [[matrix]] / => [solution] On the 48SX I may refine the solution with a residual correction., RSD PS: you do have the HP 48G User's Manual, don't you? http://www.hpcalc.org/details.php?id=3937 Is there a way to solve equation with 3 unknow without using the Matrix > Writer? Whether you are using an HP49 or HP48 the > following method is the same. For a linear > system described by A*x=b where A is a square > matrix nxn and x and b are vectors nx1 then > the solution is given by A^-1*A*x=A^-1*b or in > words the product of the inverse of A times the > vector b producing an nx1 vector. On the > formidable stack place the matrix A in > form [[row1][row2][row3]], then press the > inverse key 1/x, then enter the b vector in > the form [[b1][b2][b3]] and finally press the > product key *. There you have it(?) !Demeter! ==== > Is there a way to solve equation with 3 unknow without using the Matrix > Writer? > Maybe put then on Stack and using cmd solve? On the HP49 in RPN mode, you put the coefficients and the solution term from the first equation on the stack then a 4 (number of soft key. This will create a 4X1 matrix on level 1 of the stack. You repeat this process for the remaining two equations. Now you should have three 1X4 matrices. Now type 3 [ROW->]. This will give you a 3X4 matrix. Now go back to the [MATRIX] menu and go to the [FACTR] menu. Press [RREF] and the calculator will put the matrix in reduced row-echelon form. If you have Step-by-Step mode active, it will show you how it's getting its answers. The other way to do it is to type the three equations (using equation writer or the stack) then typing 3 [ROW->] then type the variables you want to solve for, the 3 [ROW->] then press [SOLVE] from the [ALG] menu. You don't get to see how the calculator comes up with it's answers this way though, and it takes longer to type the equations in. I only very recently learned about this method (from this NG!), but I will probably only use it if I already have the equations on the stack. You can also do it in the numeric solver, but I'm also not real familiar with it; I don't know how to make it give me exact answers. I'm not sure how to do it on the 48. I think that it has a numeric solver similar to the 49. ==== arrange eq's like so: ax+by+cz=d ex+fy+gz=h ix+jy+kz=l enter your constants array [d h l] enter a 3x3 matrix [[a b c][e f g][i j k]] press divide key thats it! be sure to understand http://www.anti-matrix.net ==== This takes a while to do. I just pull out my ti-89 that does it when it is typed straight in. Someone should write a little program that counts the number of equations and then uses the method below to solve the system. I would but I am lazy and I am not the best programmer. I am asumming that you are using a hp49. Here is my method 1. type in equation. 2. type in any of the variables then use the SOLVE comand 3. now use the SUBST comand 4. solve for one of the remaining variables using SOLVE 5. use SUBST command 6. solve for the remaining variable (SOLVE) now you have the numeric solution to one of the unknowns 7. use SUBST command 8. solve for an unknown (SOLVE) 9. use SUBST 10. SOLVE now you have two unknowns SUBST them in and SOLVE and you are done This method is much easier if you press enter before you SOLVE becouse then you dont have to type it in again. ==== I looked at that link. I had never heard of a caclulator emulator for a calculator. I dont want to be one of the disrespectful individuals and wannabe programmers that he is talking about, but I have a question. Why? I thought that the only advantage of the old calculators was the hardware(pocket size with nice keys). There must be something that I have not thougth of. > What happened to Hrastprogrammer !!?? > > He seems very upset indeed, and has deleted from the web any info > or details on his wonderful HP-41X, HP-42X and HP-71X emulators for the > HP-48GX/49G !!!, just see this link: > > http://hrastprogrammer.tripod.com/ > > Can someone (hrastprogrammer ?) shed some light on just what happened ? > Perhaps it would be useful to know, just in case ... ==== > I looked at that link. I had never heard of a caclulator emulator for > a calculator. I dont want to be one of the disrespectful individuals and wannabe > programmers that he is talking about, but I have a question. Why? I thought that the only advantage of the old calculators was the > hardware(pocket size with nice keys). There must be something that I > have not thougth of. Visit: http://www.hpmuseum.org/ Lurk on the HP Forum a bit. -- James ==== > I would just like to say thank you for your wonderful > contributions to the HP48,49 community. I currently have 5 of > your libs installed in my HP49 (Keyman, Appsman, Tetri, OT49, > AdjClk) plus Filer2 and a modified ChKM program. Your programs > are efficient both in speed and size and I have learnt allot > by just poking around the code. I really got excited the other > month with your release of ChKM ( which you now incorporated > in Keyman). I stored both a full and basic (28 assignments) > set of keymaps and used the 0 DELKEYS STOKEYS commands to > switch between them. This method was so slow... TETRI game and AdjCl that makes the HP49 clock surprisingly precise) are essentially interface enhancements for a slightly more advanced user, to make the life on the HP49 as easy and convenient as possible. If Urroz would have known these tools and used in his book, the reader could advance twice as fast and Urroz could have saved a lot of paper. There is a library called Unitman which I consider as my best contribution to the 48/49. It may be less important for people permanently living in the United States, but for engineers and scientists outside US it is really useful. It allows a complete customization of the builtin units system, to add or cancel or rename the builtin unit categories into your language. Clearly, you can delete also individual units or rename or add such units like dg (d=deco, a common unit in some contries) into the builtin menus. The most important (on which I worked hard) is that a user unit operates *exactly* as a built-in unit, i.e., you can convert such a self-made unit with Leftshift *directly* into another unit if these units are convertible at all. To realize this I had to dig rather deeply in the OS (like a mole :-) ftp://ftp.math.fu-berlin.de/pub/usr/raut/HP49/science/ ==== > > What font size are you using on your HP49G for: > stack? > editor? > menues? > EQW? > > I am using the 4*5 minifont for all. > > I have no idea wether this applies to the other HPs as well, so answer at will. > > Michael I used to just have mini-font on the stack. After I got used to mini-font, I decided to have everything in mini-font. That way I can see more information at a time. --CS ==== >>What font size are you using on your HP49G for: >>stack? >>editor? >>menues? >>EQW? >>I am using the 4*5 minifont for all. >>I have no idea wether this applies to the other HPs as well, so answer at will. >>Michael > > > I used to just have mini-font on the stack. After I got used to > mini-font, I decided to have everything in mini-font. That way I can > see more information at a time. I'm using font size 8 to everything, because I think the display is pretty poor and in bad ligthning it's not easy to read small numbers and formulas on a bad display. I've also tried smaller fonts, but didn't like it. Martin J. ==== > I'm using font size 8 to everything, because I think the display is > pretty poor and in bad ligthning That's true, but there are 2 really good readable fonts of size 7 on hpcalc.org, Raymond's Rayfont and my LFont. Letters are similar. The advantage of LFONT is that you've got a lot of symbols in it used in logic and math. The first 32 ASCII symbols of LFONT are mainly upper and lower index number, extremely useful. I applied LFONT essentially to study the general font structure and to generate my Fontman library. I'm working nearly always with LFONT, in program development, but also for everyday tasks. A font of size 7 has the great advantage of displaying six stack levels :-) - Wolfgang ==== > I'm using font size 8 to everything, because I think the display is > pretty poor and in bad ligthning it's not easy to read small numbers and > formulas on a bad display. I've also tried smaller fonts, but didn't > like it. i really wonder why in '99 they didn't use a better display than the one HP49 is shipped with. if it had more pixels the readability would improve considerably (and minifont is really PITA). hell, psion in '95 had display with 640x240 pixels :/ i know it's not that easy - the cpu must be powerful enough to work with bigger displays but still ... -- fuf ==== > [BTW, Tim, still thinking of your data manager, it's just that you don't > need it till November, right? ;-] I was thinking more of the others that might need it. I donÇt really have much desire to use the calc other than to play around, but it would be usefull to some. TW (La Ceiba, Honduras, C.A.) ==== > Where do I find info on the bits in this nibble? http://www.hpcalc.org/hp48/programming/entries/ent_srt.zip (Appendix E: Library Property Fields) http://www.hpcalc.org/hp39/math/simul.zip (Read lib.s in source.zip) http://groups.google.com/groups?selm=3d7bf79e%40duster.adelaide.on.net Also search google for property field > If STRING(Text) could be used then: > STRING(ABC)+2*5+STRING(DEF) > would be ok - I dont think this is ok: > STRING ABC + 2*5 + STRING DEF > or ist it? It's ok. Try this: DISP 0; ABC 2*5 DEF Isn't it in the manual? Remember that : is used to separate commands, and that Ans is the way they have to share things. > I would prefer string variables. > Then the program is independent of the note(s). Why do you want that? Notes already have their own environment to be created and edited. If you fear that the notepad will be filled up with garbage notes, use my NoteFiler then. :) > Is is difficult to make T1..T5 or String1..String5? No, but they would't behave as the other built-in home vars. Anyway, notes are *the* string variables on the HP38G. > This note function you made - can it create new notes or must the note > exist already? That was just an example: it recalls existing notes only. Martin Lang and I are working on a string handling library for the HP39/40G. If there's enough interest, I'll try to release an HP38G version too with source code, then you'll have another resource to look at. > Has been testing with SYSEVAL on the emulator. > Also made a simple decompiler today to find Ans. What I use is a modification of J.F. Garnier's mon48: http://www.geocities.com/ResearchTriangle/2000/mon48.zip Indeed you don't need to recompile anything: just replace the 48 ROM file and the entry lists. To upload the ROM of your 38G: http://www.hpcalc.org/hp38/utils/romupl.zip I did some changes to allow bigger entry lists. > When I start making libraries is there som way to get a library into > the emulator directly? Google works quite well for this kind of questions: http://groups.google.com/groups?selm=3D733DF3.7090701%40iinet.net.au http://groups.google.com/groups?selm=al0cg5%24eo9%2407%241%40news.t-online.c om Don't worry, Google won't expect any thanks either. :) HPCC #1046 ==== If anyone needs to tell people what they have just acquired, why not do it in a way that all of us can enjoy, for example posting or publishing a program for it, or describing a trick, not necessarily new ? For example, something like this: Did you know you can speed up your HP-42S? It can be done manually or with a program. Unfortunately, the program listed at Gene Wright's website does not work on my recently purchased calc (a bargain indeed!). Can anyone help? instead of: Hey guys, you won't believe it! I've just bought a 42S for $5!!!! Oh Jesus, today I believe in God!!! ;-) Just a thought. HPCC #1046 ==== I have plot a function. I can use X/Y softkey from men.9f under the plot to lock the position of the cursor. 1) If I press the arrows the cursor moves in the direction of arrows, it's clear. 2) If I press the arrows with red shift the cursor moves to the end of the plot window, it's clear. 3) If I press the arrows with blue shift the cursor moves to infinite position, the cursor psition is not shown and I can not see a x or y position. I cannot move back the cursor, is that a bug by HP49G, or make I a mistake? Hans Joachim (.de) ==== > 3) If I press the arrows with blue shift the cursor moves to infinite > position, the cursor psition is not shown and I can not see a x or y > position. > I cannot move back the cursor, is that a bug by HP49G, or make I a > mistake? No Hans, you do no mistake. This activates scroll mode for the plot. You see, if the PICT is bigger than 131x64 pixels, then in this mode you can scroll around the picture. The screen is then sort of a window that lets you look at a 131x64 portion of the PICT. You can understand better what this mode does if you: Enter #262d #128d PDIM Setup some function to plot and plot parameters. Draw the plot and while in the plotting environment, press [blue-shift] and then scroll around the plot using the arrow keys. Happy scrolling, Nick. ==== > Does anyone know how to get the cube (or odd) root of a negative > number without going into complex mode in the 49g? > As I'm sure you know, there are in general n, nth roots of a number. i.e 2 square roots, 3 cube roots etc. Sometimes these roots all have the same value. The 3 cube roots of -1 include -1 and two complex answers of the form a+jb, a - jb (engineers use j instead of i). -- Jim Backus bona fide replies to jimb(at)jita(dot)demon(dot)co(dot)uk http://www.jita.demon.co.uk ==== gotten it. I just did not realize until now that it had to be in approximate mode. By the way, you wouldn't happen to know how to speed up the graphing, would you? Or maybe there's a program out there that graphs faster. I love my HP 49g and I would really hate pulling out the TI 89 during an exam. Nekkyo ==== Our lovely HP49G will give you all roots (including complex ones) when it is in complex mode. In real mode it will give you only real ones. IT HAS NOTHING TO DO WITH IT BEING IN EXACT OR APPROXIMATE MODE! !Demeter! ==== HI I AM MANEL. WAY 1 DIRECTLY YOU CAN ENTER A MATRIX WITH MATRIX WRITER (SHIFT-LEFT+O) ANOTHER WAY STEP 1: ENTER THE ARROWS OF MATRIX AS LISTS STEP 2: ENTER THE NUMBER OF ARROWS WILL HAVE THE MATRIX (IN THIS CASE 2) STEP 3: RUN THE PROGRAM 4: { 'Observations' 19.3456 23.4566 45.67 } 3: { 'Refraction ' 0.0039 0.0039 0.0039} 2: 2 1: L2MATRIX RESULT [ [ 'Observations' 19.3456 23.4566 45.67 ] [ 'Refraction ' 0.0039 0.0039 0.0039] ] SURE THERE ARE ANY OTHER WAY BUT IT IS ONLY AN IDEA... THE PROGRAM IN USR-RPL (RPN MODE) 3: << -> N << N ->LIST AXL >>> 2: L2MATRIX (or another name) 1: STO The N variable is the number of arrows NOTE: I think arrows means filas in Spanish. I am not sure It works with a HP49G. I do not know with a HP48. I wish it will be usefull for you. Bye. ==== >I/O port of the calc, but the only thing iÇve found is about >the rx,tx,gnd,and shield pins. whatÇs the purpose for the >other 6? I think some of then have something to do with the screen >since i accidentally connected two of them and the contrast >went almost black, but, is there any information about this? thank you in advance > I don't recall seeing any public definition of the pin out, but there have been some discussions of these being used for an external LCD projector. And, there have been many posts about improper cables corrupting, and even damaging the calculator or display. (list at www.holyjoe.org?) Be careful. If there is a public pin out list, you should be able to find it in the hardware section of the HP49 section of www.hpcalc.org. If not, look in the FAQ, also at hpcalc.org. Sorry for the lack of specifics. Your English is better than some of the native english speakers who post here. Bill alternate E-dress wtstorey@ieee.org.no.spam.please (Use the obvious) ==== knowing the pins are for managing the LCD iÇve found the following (this is estracted from a doc about the EPSON SED 1353, for the MC68K) **Voltage maximun absolute ratings: This applies for both input and output: -0.3V to Vcc + 0.3V, in all the devices iÇve read about, and all of them work with TTL logic (Vcc = 5). And also all of them advice you that any overshoot on an input could damage the pin (i imagine they refer to the logic behind) or even the LCD Pin functions: XSLC: output: Shift clock of display data. At the falling edge of this signal data is shifted to X driver on the LCD. LP: output: Latch clock of display data. At the falling edge of this signal line data on the LCD is latched and used for turning on de LCD Y driver. LD0-LD1: outputs: Display data in the dual panel mode (???) GND: ground :) The VCO pin is a little confusing, in the 1353 and a few disp iÇve also read about (they were quite advanced devices), there are 2 oscilator pins, one input and one output, depending on whether you want to use an external osc or not; so i donÇt know if the calcÇs one is an input or an output. Anyone knows the part number/model of the calcÇs LCD? thank you for your previous responses ==== minutes after my last post i found something even better :) of course the knowledge was at hpcalc.org as in one of the photos showing an opened calculator you can see the part name of the LCD controllers (it uses 2 so i suppose one is for half the screen and the other one for the other half). They are SED 1181Fla from EPSON so i wasn't too far from the truth but the voltage ratings are completely different. I found the datasheet in google without problems thank you ==== The FAQ at http://www.hpcalc.org/hp49/docs/faq/#ss1.11 has at least names for the remaining pins. But these don't seem to be known abbreviations, since google doesn't find much about them. Markus > IÇve searcheh hpcalc with the following topics pinout pin cable and > even readed the documents about radio and IR modules for the 49 but they > all use only the rx,tx,gnd,shield pins for their purpose, so nobody talks > about the remaining pins > > thank you <80004322.0302030538.1eb9d081@posting.google.com> <3e3eb4e9$1@pfaff.ethz.ch> ==== In message <3e3eb4e9$1@pfaff.ethz.ch>, Markus Keller >The FAQ at http://www.hpcalc.org/hp49/docs/faq/#ss1.11 >has at least names for the remaining pins. But these don't seem to be >known abbreviations, since google doesn't find much about them. I've long wondered what voltage level and switching speed the remaining pins work at? Are they essentially bus outs? I wonder if it would be possible to connect to a parallel port on a PC say and then read the data fast enough to be able to re-create the display on the screen. Why? Well, intellectual exercise mostly since, for display purposes, the various emulators do a pretty good job but I'm sure someone here might be able to come up with a good reason. -- Bruce Horrocks Hampshire England bh@granby.demon.co.uk ==== > I prefer using alpha for trigs, (alpha is the Nick's greek a) > so the ASSUME is always on (no need to 'X' UNASSUME) > 'alpha>=2*pi' ASSUME > 'alpha<=0' ASSUME years I'm sure I will put it to good use. Dave ==== schrieb im Newsbeitrag You must be a real newbee here... Ummm...do you know who Dave Arnett is? No, I didn't think so, you > must be a real newbie around here.... I have no will to take this 'discussion' any further. :) Great. Now be a good little net.cop and go police elsewhere. > As I said: You're acting like a child. If you were frequently participating or even reading this ng, you would know more about me. Since you don't, it would be wise to check your archives before posting such unqualified stuff. Raymond ==== > As I said: You're acting like a child. :) Oh, I'm hurt. Such cutting comments. Please don't upset me with your barbarous barbs! > If you were frequently participating or even reading this ng, > you would know more about me. > Since you don't, it would be wise to check your archives before posting such > unqualified stuff. I've haven't really noticed you, to be honest. Hmmm, maybe it's because your posts are boring? You sound a bit up tight. People I do notice are JKH, JYH, posted before, but, not that I've noticed. :) ==== schrieb im Newsbeitrag As I said: You're acting like a child. :) Oh, I'm hurt. Such cutting comments. Please don't upset me with > your barbarous barbs! If you were frequently participating or even reading this ng, > you would know more about me. > Since you don't, it would be wise to check your archives before posting such > unqualified stuff. I've haven't really noticed you, to be honest. Hmmm, maybe it's because your > posts are boring? You sound a bit up tight. People I do notice are JKH, JYH, > posted before, but, not that I've noticed. :) I don't care about such comments too much. There were always bigmouths like you trying to catch attention, but they usually don't have the guts to stand to their words. It's easy to shout senseless things while staying anonymous, like you. ==== > But this kind of unqualified comments like yours is one of > various reasons why most of the old HP fans and contributors > left this newsgroup. Yes, I confirm this, I know it from at least two of them. The level of this NG declined gradually. It is not because people who have really contributed to the 48/49 are called gurus, but check out how this bigmouth Nick interprets it. He is exactly of the same quality than the Nobody who attacked Raymond. If it is true that he didn't read Urroz's books then the worse! Working on the same field (in this case simply commenting the basic properties of the 49 to newbees) without at least quoting the predecessors, is considered as the most dirty buisness in software-engineering, even if payed only by chocolate. There is no excuse in this case that one hasn't read it. Whether commercial or not, it exist, and this is known to everybody in this NG. - Wolfgang ==== >> But this kind of unqualified comments like yours is one of >> various reasons why most of the old HP fans and contributors >> left this newsgroup. > Yes, I confirm this, I know it from at least two of them. So, a sample of two is supposed to be statistically relevant? > The level of this NG declined gradually. It is not because > people who have really contributed to the 48/49 are called > gurus, but check out how this bigmouth Nick interprets it. > He is exactly of the same quality than the Nobody who > attacked Raymond. If it is true that he didn't read > Urroz's books then the worse! Yes, everything is going to hell-in-a-hand-basket. This is the same cry you'll hear from generation to generation, generally by those people, who don't like change. As they age, they'll tell you all about how good it _used_ to be, and _bad_ it is now. Oh, and Raymond was never *attacked*. He was simply disagreed with, after expressing his opinions on the posting etiquette of another contributor. Don't make such silly emotive statements. > Working on the same field (in this case simply commenting > the basic properties of the 49 to newbees) without at least quoting the > predecessors, is considered as the most dirty buisness in > software-engineering, even if payed only by > chocolate. There is no excuse in this case that one hasn't read it. > Whether commercial or not, it exist, and this is known to everybody in > this NG. I have no idea what you're going on about here...but I guess you don't like something someone else posted somewhere. ==== > I don't care about such comments too much. That's good. :) > There were always bigmouths like you trying to > catch attention, but they usually don't have the guts > to stand to their words. > It's easy to shout senseless things while staying anonymous, > like you. I wasn't shouting - really Raymond. Like SAVERPN, I just couldn't let your net.cop hand waving pass. If you don't like pseudonyms, fine - you made that clear. I'm also feeling free to make it clear that I see no problem with it. I tend to notice the contents and quality of the post, I really don't care who's actually posting. You know, on the net, nobody knows you're a dog. :) > have any interest contributing meaningful things here. > You just want to try to upset other people for fun. Really Raymond! I believe it was you who started with the insults. I assume you wanted to upset me? I'm wouldn't be on Usenet if I was so easily upset. And I don't wish to upset people, but if you don't like my opinion, and it upsets you, so be it. I don't like your opinion (on pseudonyms) but it doesn't upset me. > But this kind of unqualified comments like yours is one of various > reasons why most of the old HP fans and contributors left this newsgroup. Oh rubbish! It's more to do with HP pulling the plug on their decent calculators, and the fact the people do move on. Wish Dave A was still here though. Hmmm, hope it wasn't my unqualified comments (whatever that's supposed to mean) that drove him off. ;) > So please don't waste my time with your noise. Feel free to ignore me then. You could just killfile me, if this offends you so much, and you can't help wasting your time by replying. HTH, HAND. ==== A small hint: Query hpcalc.org and hpmuseum.org, c.s.48 or Google by 'Raymond Hellstern' ==== > A small hint: Query hpcalc.org and hpmuseum.org, c.s.48 or Google by > 'Raymond Hellstern' This was pour moi? You should learn to attribute. Really, I do believe you. Your self importance precedes you. :) ==== schrieb im Newsbeitrag > A small hint: Query hpcalc.org and hpmuseum.org, c.s.48 or Google by > 'Raymond Hellstern' This was pour moi? You should learn to attribute. > Really, I do believe you. Your self importance precedes you. :) > So now we come to *your* valuable contributions... ==== > schrieb im Newsbeitrag >> A small hint: Query hpcalc.org and hpmuseum.org, c.s.48 or Google by >> 'Raymond Hellstern' >> This was pour moi? You should learn to attribute. >> Really, I do believe you. Your self importance precedes you. :) > So now we come to *your* valuable contributions... *sigh*. I'm sure you've contributed a whole bunch of stuff, it's just odd that you feel the need to have to point it out.... ==== >> You must be a real newbee here... > Ummm...do you know who Dave Arnett is? No, I didn't think so, you > must be a real newbie around here.... Actually, anyone know what Dave is doing these days? I miss his posts, I think his last was in 2001 or so. Now those were the days... ==== little. I just wanted to clarify that we're talking about the same thing (terms like UserRPL don't mean anything to me (yet!)) ... ... I'm talking pretty simple stuff here - prompting for a handfull of variables - doing some basic calculations - and presenting some answers. Biggest problem I'm facing is that once I've input the program it loses all formatting - and once it gets more than a few dozen lines long it becomes quite a jumble to sort through. Is there an easy answer to this issue? - if I could maintain the original formatting (or even have a few gaps to visually seperate sections of code) it would make things a lot easier. Whilst I've 'got you' - just a couple more quick questions if I may ... Is there a way to supress the variables created by a program from being displayed in the lables used by the function keys - it's my personal preference to have just programs there - it's difficult to eyeball all the data just to find the programs. Additionally, is there a way to step through program execution to help with debugging? Many thanks all. KJJ Now that my 49G and I have 'got to know each other a little' I'm at the > point of writing a few simple programs. I've quickly realised that it's pretty hopeless to do this on the 49G > itself - and I understand that it's possible to write the programs on a PC > (notepad?) and then transfer them to the 49G ... ... could someone please give me the basic steps to follow to accomplish > this? - I have the serial cable and a spare comms port - but no idea as to > what comms software to use, or what buttons to puch on the 49G. Any help truely appreciated. Many thanks, KJJ ==== > Biggest problem I'm facing is that once I've input the program it loses all > formatting - and once it gets more than a few dozen lines long it becomes > quite a jumble to sort through. Is there an easy answer to this issue? - if > I could maintain the original formatting (or even have a few gaps to > visually seperate sections of code) it would make things a lot easier. A way to keep the formatting of your programs is to start each new program within a string . Once you finish typing, save it as a string. If you want to execute it, you must get rid of the quotes, so just go to the [LSHIFT] [PRG] menu, [TYPE] and hit OBJ->. Even better to save your self from typing assign the OBJ-> operation to a key. The program will be left on the stack for you to execute. For example, to assign the key O to OBJ-> type: << OBJ->> 43.1 ASN and that will do the assignment. Next to illustrate the string method, type for example: << 1 4 FOR X X NEXT >> with spaces and new lines inside the << >> as you see fit. Save it as a string under, say, 'Atest' STO and then bring it on the stack by pressing the menu key that corresponds to that name. Press now [LSHIFT] [USER] O and there you have the program on the stack. Type [RSHIFT] [EVAL] and it runs producing the trivial count. As far as the local variables question look it up in the manual (PDF) :-) !Demeter! ==== > I just wanted to clarify that we're talking about the same thing (terms like > UserRPL don't mean anything to me (yet!)) ... The guys who post on this newsgroup are often gurus but they mean well of course. > > ... I'm talking pretty simple stuff here - prompting for a handfull of > variables - doing some basic calculations - and presenting some answers. > Biggest problem I'm facing is that once I've input the program it loses all > formatting - and once it gets more than a few dozen lines long it becomes > quite a jumble to sort through. Is there an easy answer to this issue? - if > I could maintain the original formatting (or even have a few gaps to > visually seperate sections of code) it would make things a lot easier. I've made up some tutorials, sample programs and programming guide for simple tasks with surveyors in mind. The principles would be directly the same for any other task though. Check; http://www.quickclose.com.au/tut.htm tutorials http://www.quickclose.com.au/workshop2.doc manual (you need charting.ttf font at http://www.quickclose.com.au/CHARTING.TTF http://www.quickclose.com.au/programs.htm sample programs > > Whilst I've 'got you' - just a couple more quick questions if I may ... > > Is there a way to supress the variables created by a program from being > displayed in the lables used by the function keys - it's my personal > preference to have just programs there - it's difficult to eyeball all the > data just to find the programs. Use what are called local variables. These variables exist only while the program is executed. For example you want to store variables 56 in variable name x [45 67] in variable name y First put 56 on level two of the stack (from command line or other program) and [45 67] on level one of the stack then use the following code; << -> x y << @ remark: x and y can be used only inside these delimeters > > Additionally, is there a way to step through program execution to help with > debugging? > 1. Enter the name of the program to debug (e.g. 'MYPROGRAM') onto level one of the stack 2. Type 41 MENU (using alpha keys) 3. Press DBUG to start the debug 4. Press SST to step through the program If you put a HALT (as many as you like) in the program and run the program normally, using SST directly will start debugging from the HALT. Very useful tip ! CONT skips the HALT when it comes up, good for viewing stack contents, and the like. Hope that helps ! Richard S. ==== > I've quickly realised that it's pretty hopeless to do this on the 49G > itself - and I understand that it's possible to write the programs on a PC > (notepad?) and then transfer them to the 49G ... Many people develop on their HP49, I prefer the PC. There is a new and excellent tool for what you want. It is called Debug4x and consists of an upgrade to HP Software Development Kit, the HPTools set of PC tools, the EMU48 emulator, built in debuggers for SysRPL and ASM, a communications package for XModem, good help files, documentation and many more features than I can mention in one note. The product can be obtained from HPCalc.org, search for Debug4x -- - - - - - - - - - - - - - - - - Bill Graves RKBA! bgraves@ix.netcom.com ==== >I don't understand Derive. It gives me 2.197224577 for log(9) when >the answer is obviously 0.9542425094. Why is log the same as ln? >Do base 10 logs no longer exist in Derive, only natural logs? Try LOG(x,10) to get base 10 logs. The default, if you omit the base, is e. -- Bruce Horrocks Hampshire England bh@granby.demon.co.uk ==== Elect49, downloadable at hpcalc, which calculates the atomic number of a substance, given the cuantic numbers of the laste e-. For some reason i cant get it to work.. I get some type of weird chars... Pleas help, its important. ==== >Here's a stock photo of Michael P. Anderson: >http://www.jsc.nasa.gov/Bios/portraits/anderson.jpg >He was 42 years old and had a master's degree in physics from Creighton >University, so it's safe to surmise that was familiar with HP calculators. I think most NASA scientist and astronauts are familiar with HP calculators. At least if the following add reflects the truth: The calculator that operates a space ship... ... was bought by NASA at an Authorized HP Dealer. Do the same in Brazil! The success of the space missions of shuttles Columbia and Challenger is associated to varied and complex calculations that ensure their advanced technology to work perfectly. Some of these calculations were entrusted by NASA to Hewlett-Packard model HP-41. The choice of Hewlett-Packard calculators to be used in the space shuttles Columbia and Challenger was the result of a study carried out by NASA to determine which calculator, amongst all available in the market, would fit the job. The calculators underwent tests as rigourous as those endured by the shuttles' equipments before they are declared apt to flight. One of programs stored in the calculator is related to the ship's equilibrium and the volume of fuel to be burned to establish the center of gravity necessary to the re-entry in the atmosphere. This program was qualified by NASA as of critical importance to flight. In Brazil, when you have important calculations to solve, do as NASA does: Go to an Authorized Hewlett-Packard Dealer and buy your HP-41. (*) HEWLETT-PACKARD - When performance must be measured by results. (*) The HP 41 calculators used by NASA were acquired normally in an HP Dealer without prior knowledge that they were to be used in a space mission. This was a famous HP add that appeared in various science magazines around here 20 years ago... It showed a nice picture of Columbia landing (or taking off) on a desert runway and a smaller picture of a 41CV. This specific add was translated from Ci.90ncia Ilustrada, AnoII, N.bc 10, Julho/83, Editora Abril, page 5. (Sorry if I haven't been able to translate it properly, I am not good at this kind of task). We are all sorry for this tragedy, but the space program must continue. If they had given up at the first shipwrecks 500 years ago, the New World would still be there waiting to be discovered. Gerson ==== > We are all sorry for this tragedy, but the space program must continue. Agree, though the responsibles should add a little more weight to technology issues and be more careful in future. Many of the participating engineers have repeatedly warned that there are problems, but the bean-counters had a different opinion. (Again...) And the astronauts and their families pay the bill. > If they > had given up at the first shipwrecks 500 years ago, the New World would still > be there waiting to be discovered. Yes, and if we don't manage to somehow go out to the stars, then... it doesn't look very well for our fate. Let's remember those astronauts for ever, are they with Gus now? Nick. ==== >Here's a stock photo of Michael P. Anderson: >http://www.jsc.nasa.gov/Bios/portraits/anderson.jpg >He was 42 years old and had a master's degree in physics from Creighton >University, so it's safe to surmise that was familiar with HP calculators. I think most NASA scientist and astronauts are familiar with HP calculators. At least if the following add reflects the truth: The calculator that operates a space ship... ... was bought by NASA at an Authorized HP Dealer. Do the same in Brazil! The success of the space missions of shuttles Columbia and Challenger is associated to varied and complex calculations that ensure their advanced technology to work perfectly. Some of these calculations were entrusted by NASA to Hewlett-Packard model HP-41. The choice of Hewlett-Packard calculators to be used in the space shuttles Columbia and Challenger was the result of a study carried over by NASA to determine what calculator, amongst all available in the marked, would fit the job. The calculators underwent tests as rigourous as those endured by the shuttles' equipments before they are declared apt to flight. One of programs stored in the calculator is related to the ship's equilibrium and the volume of fuel to be burned to establish the center of gravity necessary to the re-entry in the atmosphere. This program was qualified by NASA as of critical importance to flight. In Brazil, when you have important calculations to solve, do as NASA (*) : Visit an Authorized Hewlett-Packard Dealer and buy your HP-41. HEWLETT PACKARD - When performance must be measured by results (*) The HP 41 calculators used by NASA were acquired normally in an HP Dealer without prior knowledge that they were to be used in a space mission. This was a famous HP add that appeared in various science magazines around here 20 years ago... It showed a nice picture of Columbia landing (or taking off) on a desert runway and a smaller picture of a 41CV. This specific add was translated (at least I tried to...) from that shown in Ci.90ncia Ilustrada, Ano II, N.bc 10, Julho/83, Editora Abril, page 5. We are all sorry for this tragedy, but the space program must continue. If they had stopped at the first ship-wrecks 500 years ago, the New World would still be waiting for discovery. Gerson ==== Please, download from Eric Rechlin Web site (www.hpcalc.org) the pinouts connection for HP, by Enrico Carta. See the next explanations: **************************************************************************** *** Building the HP49G Serial PC-Link Cable 1.2 (details) 14KB Explains multiple ways to make a serial link cable for the 49G. Includes both HTML and ASCII text versions. By Enrico Carta (H). 2000/03/03 **************************************************************************** *** HP to Modem (details) 15KB Explains how to connect the HP49 to a modem. In HTML and text format. By Enrico Carta (H). 2000/07/01 **************************************************************************** *** Miguel Angel CAPORALINI HERK (M.A.C.H.) **************************************************************************** *** > Manel. > > Today I transfered a program to the hp49g with conectivity kit version > XMODEM 1.0.9. (x49kit2) and appeared a message Connection error. > Please check your connections. > > Whith version 1 SubVer 2 Build 17(x49kit) I have no problems to > transfer files but when I wanna rename a file or I wanna delete more > than one item at once the error appears. With v1.0.9 ther error > appears too. > > With HP connectivity kit 3.00 r04 I have no problems (well it's so > slow..) > > > I use the cable suplied by HP Espa.96a. Cable, RS adaptor, and two > adaptor for connecting two hp48's or connect a hp49 to hp48. > I use a digital multimeter to check cables are good and the result is > the follow: > > > HP-49 wire connector > ________*________ _______*_______ > | 1 2 3 4 5 |--0 | A B C D E |--0 > | 6 7 8 9 10 | | F G H I J | > ----------------- ----------------- > TO PC ADAPTOR TO HP-49 > > HP49-PC ADAPTOR > __________________ ______*_________ > e d c b a / | 5 4 3 2 1 | > i h g f /--0 |10 9 8 7 6 | > ------------- ----------------- > SIDE to connect to PC here you connect HP49 wire > > > (*) means a mark to plug it on the correct way > I have found only this connections. > > On cable: > 1 --0--A--0 ; 7 -- G ; 8 -- I ; 9 -- H ; 10 -- J > (2,3,4,5,6 y B,C,D,E,F ARE NOT CONNECTED) > > ON adaptor: > > 0 -- 10 ; b -- 8 ; c -- 9 ; e -- 7 > > Have I my cable ok? Are they the normal connections? Notice the > adaptor has less connection than cable. If it is ok. What is the > function of rest of the pins? > > Sorry but I am not English. > -------------------------- > CABLE DE CONEXI.94N DE HP-49 > ________*________ _______*_______ > | 1 2 3 4 5 |--0 | A B C D E |--0 > | 6 7 8 9 10 | | F G H I J | > ----------------- ----------------- > al adaptador del PC A la calculadora HP-49 > > EL ADAPTADOR DE HP49 AL PC > __________________ ______*_________ > e d c b a / | 5 4 3 2 1 | > i h g f /--0 |10 9 8 7 6 | > ------------- ----------------- > parte que se Aqu.92 se conecta > conecta al PC el cable de la HP-49 > > > Con un mult.92metro (la opci.97n que suena un pitido si dos cables tiene > conexi.97n) he comprobado las conexiones de KIT de conexi.97n. > > En el cable: > 1 --0--A--0 ; 7 -- G ; 8 -- I ; 9 -- H ; 10 -- J > NO hay ninguna conexi.97n m.87s (el 2,3,4,5,6 y B,C,D,E,F NO EST.8dN > CONECTADOS) > > En el adaptador de HP49 a PC > > 0 -- 10 ; b --- 8 ; c -- 9 ; e -- 7 > > ËEs normal que el adaptador tenga menos conexiones que el cable? > ËPara qu.8e sirven las dem.87s conexiones? ==== I have 4 NiMH batteries I use for my 48GX. I numbered them so I know which one I need to keep out of the next rotation. I haven't seen the quick warning-to-shutdown time reported by Peter. I do have to keep a spare set of alkalines around as suggested. That's usually only a problem when traveling (batteries alway choose a bad time to die). -Dave > > >>I was wondering if anyone uses niMH batteries in their calculators. I own a >>digital camera so I already have the charger and I'm contemplating >>purchasing a set of 4 niMH AAA batteries to use in my 48GX so I basically >>will never have to replace them again... Has anyone already tried this? > I did try and am back with batteries. The low-batt warning was very > late. If you want to use NiMHs, carry a (charged) spare pack (or > better spare alkalines, they don't discharge as much when not in use) > got less than 5 minutes ! > > Pete > ==== where do you get it? how much? ==== Years ago I ran across the hp42s and found it to be the greatest tool in my tool box(mold maker), after learning to program it. This tool took the entire shop and a few later to a new leavel of accuracy and speed. But now we can't get any more of the 42s's so I'd like to introduce the 48gx to the group and take the programs to the next level. Hence the problem. I don't have the time or the pacience to do this, this by all indications is a very complicated task in the 48gx. Is there any one who would be able to convert the old programs to the new program language. Eather for fun or will this cost and how much. ==== I am writing my first SYSRPL program. It calls the grob SS from memory and then copies another grob (MRK) onto that one. But it seems that when I call it to the stack with ID SS that the original gets replaced by the new formed grob. So I have to RCL it from memory somehow. Here is my problem. If I do ID SS @ then I get external results in the stack, and I have no idea whether I can work with that further on. Like displaying the grob. I'd like to see the grob, not 'External'. If I put mulitple @'s in the program then my calc crashes and clears my memory. What is the syntax to use for the @ command? Or SAFE@? ==== you need to copy the GROB to TEMPOB first. HTH Andreas > I am writing my first SYSRPL program. It calls the grob SS from memory > and then copies another grob (MRK) onto that one. > > But it seems that when I call it to the stack with ID SS that the > original gets replaced by the new formed grob. > > So I have to RCL it from memory somehow. Here is my problem. If I do > ID SS > @ > then I get external results in the stack, and I have no idea whether I > can work with that further on. Like displaying the grob. I'd like to > see the grob, not 'External'. > > If I put mulitple @'s in the program then my calc crashes and clears > my memory. > > What is the syntax to use for the @ command? Or SAFE@? ==== it will be available again May or June this year, is this just for Europe? in USA the price for 49G is going crazy....and no one said it will be produced again soon...... anthony thx....! : ) ==== I tried out the long string viewer Wolfgang Rautenberg made called VLS. It is a neat little program; it accepts a string as an argument and then displays it in the text viewer, but 'word wraps' the text. This is good for proggie. I found that I had to convert my numbers to strings before I could view them. The easiest way I found was to press the key, then +. This converts to a string. Is there an easier way to do this? Also, how do you convert from a string back to a number easily? ==== > I found that I had to convert my numbers to strings before I could view > them. The easiest way I found was to press the key, then +. This > converts to a string. Is there an easier way to do this? Well, not really easier, but you can also use ->STR, which converts anything to a string. >Also, how do you convert from a string back to a number easily? You can use OBJ-> to do that. Greetings, Nick. ==== Here is a picture where you can see how Lt. Michael P. Anderson use your HP48 in the Columbia for science activities. http://www.time.com/time/covers/1101030210/sctone.html R.I.P to this heroes. Gregorio Mu.96oz ==== > I'll repeat here a small challenge... We start with > f(0)=f(1)=1 and > the recursion formula > (1) f(n+2)= f(n) + i*f(n+1) (i = sqrt(-1)). > This defines a complex-valued sequence but (1) is very > similar to Fibonacci's formula for his sequence F from his > Liber Abacci (Pisa 1202) which starts with F(1)=1, F(2)=2, > hence the setting f(0)=f(1)=1. If the limes of the sequence > [f(n+1)/f(n)] exist in the complex plane then it is a root > of the equation > (2) x^2 - i*x -1 = 0. > As is easily shown, the roots of (2) are the complex > numbers (sqrt(3)-i)/2 and (sqrt(3)+i)/2. My question is: > Does lim [f(n+1)/f(n)] really exist and which of the two > roots of (2) is then the limes? > > The eigenvalues of [[i 1][1 0]], w1 and w2, are primitive > 12th roots of 1, so w1^n=w2^n = 1 iff n ==0 mod 12. > For any values of f(0) and f(1), there are suitable values > of a and b such that f(n) = a*w1^n + b*w2^n. > Then f(n + 12*k) = f(n) for all non-negative integers n and > k, and the values of f(n) are periodic with period 12. > Since f(n) is not constant unless a = b = 0, there can be no > limiting value to f(n+1)/f(n), as the ratio will be periodic > too. The limit of the ratio does not exist. > > That's altogether correct, hence, you are the winner. The result > could have been obtained even with elementary math, I mention this only > for the rest of the people :-) > > The sequence f when drawn in a clock face, f(1)=1 at 12 o'clock, > f(1)= 1 at 1 o'lock etc looks as follows: > > * > * 1 * > 1-i 1 > * * > -i 1+i > > * -i * i > > > * -i-i * i > > * -1 * -1+i > * -1 > > And this is the legend: When God had created Heaven and Earth and > Mankind he took a little rest and looked at the sequence defined > by f(n+2) = f(n) + i*f(n+1). Notwithstanding, God saw at once the > above circle. And he said > > Ok, this may be the clock face for my creatures > > There is a very small missing detail in Virgil's argument. It's > correct that the sequence [f(n+1)/f(n)] is periodic as well, but > the period may perhaps be 1. This is not so as is seen by the > circle. Actually, the period is 3. AND THIS HOLDS TRUE FOR ANY > sequence f starting with any two reals a,b distinct from 0 and > obeying (1) above. Hence, my next challenge: > > Write a shortest possible UsrRPL program which yields the > 3 different terms of the sequence [f(n+1/f(n)] where f is > defined above with initial values a,b <> 0. > > - Wolfgang I can do it with f(0) and f(1) as input on the stack, in levels 2 and 1 respectively, and the 3 ratios as output in a UsrRL program of 35.0 bytes, with checksum of # 39588d, though the last 2 ratios may appear in unfamiliar forms. ==== >> I'll repeat here a small challenge... We start with > f(0)=f(1)=1 and the recursion formula >> (1) f(n+2)= f(n) + i*f(n+1) (i = sqrt(-1)). >> This defines a complex-valued sequence but (1) is very >> similar to Fibonacci's formula for his sequence F from his >> Liber Abacci (Pisa 1202) which starts with F(1)=1, F(2)=2, >> hence the setting f(0)=f(1)=1. If the limes of the sequence >> [f(n+1)/f(n)] exist in the complex plane then it is a root >> of the equation >> (2) x^2 - i*x -1 = 0. >> As is easily shown, the roots of (2) are the complex >> numbers (sqrt(3)-i)/2 and (sqrt(3)+i)/2. My question is: >> Does lim [f(n+1)/f(n)] really exist and which of the two >> roots of (2) is then the limes? > The eigenvalues of [[i 1][1 0]], w1 and w2, are primitive >> 12th roots of 1, so w1^n=w2^n = 1 iff n ==0 mod 12. >> For any values of f(0) and f(1), there are suitable values >> of a and b such that f(n) = a*w1^n + b*w2^n. >> Then f(n + 12*k) = f(n) for all non-negative integers n and >> k, and the values of f(n) are periodic with period 12. >> Since f(n) is not constant unless a = b = 0, there can be no >> limiting value to f(n+1)/f(n), as the ratio will be periodic >> too. The limit of the ratio does not exist. > >That's altogether correct, hence, you are the winner. The result >could have been obtained even with elementary math, I mention this only >for the rest of the people :-) The following is a straightforward and elementary proof that doesn't rely on any linear algebra : First we have of course the equation f(n+1) = i*f(n) + f(n-1) . Based on this, we can rewrite it in terms of the components of the complex values, where f(n+1) = a(n+1) + i*b(n+1) , a and b being reals : (1) a(n+1) = a(n-1) - b(n) (2) b(n+1) = a(n) + b(n-1) Based on the above we can derive further identities: (3) a(n) = b(n+1) - b(n-1) , rewriting (2) b(n+2) = b(n) - b(n-2) , substituting (3) into (1) and rewriting (4) b(n) = a(n-1) - a(n+1) , rewriting (1) a(n+2) = a(n) - a(n-2) , substituting (4) into (2) and rewriting Thus we arrive at : a(n+2) = a(n) - a(n-2) b(n+2) = b(n) - b(n-2) which is just what we need since these sequences are only expressed in terms of themselves. Now, based on the above we arrive at our final relation : a(n+4) = a(n+2) - a(n) = a(n) - a(n-2) - a(n) = - a(n-2) This can be rewritten as : a(n+6) = - a(n) The analogous relation for b(n) holds as well of course. Looking at this, it is obvious that both sequences have at most a period of 12, and consequently so does the original sequence. Finally, it follows from the above that the ratio of consecutive terms of the original sequence is periodic with a maximum period of 6, and therefore cannot approach a limiting value. correct that the sequence [f(n+1)/f(n)] is periodic as well, but >the period may perhaps be 1. This is not so as is seen by the >circle. Actually, the period is 3. AND THIS HOLDS TRUE FOR ANY >sequence f starting with any two reals a,b distinct from 0 and >obeying (1) above. Hence, my next challenge: Write a shortest possible UsrRPL program which yields the >3 different terms of the sequence [f(n+1/f(n)] where f is >defined above with initial values a,b <> 0. > >- Wolfgang << / DUP DO OVER INV i + SWAP UNTIL DUP2 == END DROP2 >> 42.5 bytes , checksum # 6C49h (48) . ---------------------------------------------------------------------------- --- Jonathan Busby - before replying. ==== << / DUP DO OVER INV i + SWAP > UNTIL DUP2 == END DROP2 > >42.5 bytes , checksum # 6C49h (48) . > Boy is that unnecessarily complicated. ;) Here's a simplified version : << / 1 2 START DUP INV i + NEXT >> 32.5 bytes, checksum # AF23h (48) . I guess I broke the custom of not sharing one's code until the end, but it's too late to stop now. ;) Sorry if I spoiled the fun for anybody. ;) ---------------------------------------------------------------------------- --- Jonathan Busby - before replying. ==== > << / DUP DO OVER INV i + SWAP > UNTIL DUP2 == END DROP2 > Boy is that unnecessarily complicated. ;) > Here's a simplified version : > << / 1 2 START DUP INV i + NEXT > I guess I broke the custom of not sharing one's code until the end, > but it's too late to stop now. ;) Sorry if I spoiled the fun for > anybody. ;) You didn't, fun will be continue :-) that he may win the first price (time counts!), the disadvantage that his published solution may be used by others for improving his own. One rule should be observed, however: For a program like your's also the correctness should at least be explained. I don't comment on the correctness of your program at this moment, that's your task! The good thing is that you get a second chance by this :-) his elementary solution of the first question. He transformed the whole problem in real arithmetic. But usually, a problem in complex arithmetic is more easily solvable therein, and indeed, I have an elementary proof of cyclicity of the sequence f of a few lines. Also other people should still try to beat Jonathan (whenever his solution is correct), in particular Nick. - Wolfgang PS. I feel a bit sorry for Nick. Since his disaster in trying to solve the Preference Ordering challenge he never again participated in any competition, even not the relatively easy one by JKH. Why it is so that even in UsrRPL competitions usually the gurus win and not a UsrRPL programmer? Guess why ... ==== X > Fibonacci invented also the first iterative numeric procedure, taking > the example of computing with it the 3rd root of 47 > (which is irrational). Remember that he lived nearly 500 years > before Newton and 400 years before Decartes! I may present Fibonacci's > one person is interested in and if time allows. 2=? ==== > I often get very big results or very small results, > like 8438284321.54 or .000000000038, but how can I > make that into scientific notation without having to > [manually change display modes back and forth all the time]? > I'd like some kind of utility that you could personalise > that would automaticaly set the display type depending on > the size of the number. Say <10^5 use 4 FIX else 4 ENG etc. > Also like to have option for 1000's separator in STD mode. See this example: http://groups.google.com/groups?selm=3D7495E9.848BF8E8%40miu.edu Related example (to show leading zeros in binary values): http://groups.google.com/groups?selm=8rhhqe%24mri%241%40nnrp1.deja.com Thousands separators, showing all digits, but no trailing zeros: http://groups.google.com/groups?selm=6l338g%24q78%241%40news.iastate.edu For displaying exact integers with separators on HP49, see: http://groups.google.com/groups?selm=3AF38F47.4C05013D%40miu.edu Here's the program from the first link above, augmented to display, in the status area at the top of the screen, both SCI mode and FIX mode (showing commas and all significant digits) for any level-1 number, regardless of the current (unaffected) stack display mode; note that only significant digits are shown (no trailing zeros) in the generated status display: %%HP: T(3)F(.); @ Include for Kermit download to calculator. @ Comments are ignored when downloaded or pasted. << 2. ->HEADER @ set status area height, omit for HP48 DROP DEPTH { RCLF OVER DUP TYPE 28. == { I->R } IFT @ optional for HP49 exact integers DUP TYPE 10. == { B->R } IFT @ optional for user binary integers DUP TYPE NOT { DUP MANT STD ->STR SIZE 2. - SCI DUP 1. DISP 2. DISP 1. FREEZE DUP STD ->STR E POS NOT { DUP FP ABS DUP { ->STR } IFT @ handle 0 [48] or 0. [49] SIZE 1. - DUP FIX 11. > { STD } IFT @ simulate 12 FIX DUP a%>$, SYSEVAL 2. DISP @ see important *NOTE* below } IFT } IFT DROP STOF } IFT >> 'GbENTER' STO @ 'BetaENTER' [alpha right-shift B] -62. SF -63. SF @ set User mode and Custom Enter mode @ End of download Important *NOTE* Replace the SYSEVAL string argument a%>$, (entry point name) with the *hexadecimal* address for your calculator model: For 49 only: #25F0Fh, For 48 only: #162ACh A package is available to do this automatically, provided that you have the extable library in your 49, or the Jazz library [full version] in your 48: http://www.hpcalc.org/hp49/programming/misc/systrans.zip http://groups.google.com/groups?ic=1&selm=8618et%24p7v%241%40nnrp1.deja.com CAUTION: Back up your memory *before*installing* the above program, because if the program is entered incorrectly, a memory wipe-out could occur! [r->] [OFF] . ==== I'm in the process of reviving my HP48, which I haven't used other than to change the batteries since 2000 sometime. Amazingly it hasn't forgotten anything - the battery in the merged card in slot 1 had expired, meaning I lost libs and stuff, but the slot 2 card was OK, and I had backups of everything there, and (I must have been prescient!) a little program to recover everything from the copies in the high ports. So I have it all working as it was in 2000, and now, after documenting things so I don't get so stuck again, I'd quite like to gently upgrade some of the libs I have to faintly current versions. One problem is that a lot of the URLs seem to have gone away - for instance alg48 doesn't live where it did, and the Java URL I have doesn't seem to have it any more (it has Jazz still..). Searching for `Java' on google isn't going to find me the current page, either... Is hpcalc.org pretty much a definitive location for hp48 things? It has a copy of most stuff (as ever), but I'm not sure if it's up to date? ==== >I have both of Urroz's books, in electronic format only, and have printed >>them myself. Over the two volumes there is something like 700+ pages. You should be supporting Urroz for doing a good job! I'm a poor student, but I've just bought the book online, because I've > now heard too many good things about his books and I want to support the > man for doing a good job. If you're interested in the book, then you should go buy it. That's my opinion. ps: I can give you a feedback, when I'm finished with volume 1, if you > want it. > It was a joke (just in case). Send me the feedback on the book asap > please, gonna get the hard copy if its good. > [AC] I'll accept it as a joke.. As for feedback:I bought the hardcopy of the two books back when I was taking College Algebra. I also received them in PDF format. I used the books extensively through Calc 2. Calc 3 is next term!!! and I plan on using them also. There is an errata file for the books out there, unfortunatly I can't remember where. You might do a google search. I figure (Hehehe) that between the Urroz books and Nick K (and the rest of this NG). I made it through the 5 maths classes with flying colors. I would highly recomend you getting the two books AND the trig marathon and the series marathon by Nick K. ==== I write to point your attention to a behaviour of the SUBST command which IMHO could be improved. Executing this command on an differentiated argument automatically call the EVAL command - I think so - and this can be a bit fastidious in some cases. I'll make an example to explain better my POV. I've the equation: 'ddx(A+B+C)+T=0' With A, B, C that are constants for the HP49 but in my mind are functions and I've not explicitly declared their dependency from X to have a clearer notation. If I execute 'A=0' SUBST I have the ddx(A+B+C) automatically evaluated and so the result is 'T=0' which in fact is correct but it is not what I have in mind! :o) So there's a way to operate SUBST without calling EVAL? Maybe it could be modified in future ROMs? ==== > Is there any way to easily convert a 48 XLIB's rom addresses > to the 49's rom addresses? i.e. a conversion program... There are several ways, depending on what you really want to do, e.g. o Convert UserRPL (and some SysRPL) binary files [hpconvert]. o Convert decompiled (or decompilable) source [just decompile and recompile, UserRPL or SysRPL] What do you mean by a 48 XLIB -- do you mean an entire library? [sometimes you can split and convert each part]; do you have some specific, published program or library in mind? Generally you need to look on a case-by-case basis, and more detail is necessary to begin. . ==== And I forgot to list this other possibility: o Convert UserRPL containing SYSEVALs and LIBEVALs? A reference to an automatic converter for supported SYSEVAL addresses has just been posted in another thread. [r->] [OFF] ==== How can I evaluate a LOG in base 2 (something like 2 LOG X), if possible? ==== > How can I evaluate a LOG in base 2 (something like 2 LOG X), if possible? > the formula is LN(x) / LN(base) ==== This is the point where in real life, I would take you to the side and ask you about the obvious problem you seem to have. which is impossible in your case, and doesn't seem to even make sense. > How do we know you're who your headers suggest? > Perhaps you should attach a passport picture to prove it? > You must be a real newbee here... I have no will to take this 'discussion' any further. Maybe some day, you'll get the point, and let be this childish nonsense. ==== > You must be a real newbee here... Ummm...do you know who Dave Arnett is? No, I didn't think so, you must be a real newbie around here.... > I have no will to take this 'discussion' any further. :) Great. Now be a good little net.cop and go police elsewhere.