.3481 === Subject: RCLF/STOF... [Re: values not showing when code is run (userrpl)] Hi Dieter, I think you know where the problem was;-) The sentence '..RCLF in order to save the default flags state..' is misleading, or at least not exact, and I think this is why Giancarlo jumped in. RCLF recalls the flags to the stack, as said in the user doc, it's up to the user to take care of _saving_ the stack contents elsewhere. The stack is not to be considered as a object save location, it's a temporary workspace. HP chose a specific perspective when creating the respective names, see HP-41CX Owner's Manual, section 19: Flags, pages 292-297, where RCLFLAG/STOFLAG are explained in deep detail, and HP-48G User Guide 'page' 24-7, where RCLF/STOF are mentioned 'on the go', but correctly. The HP-48 G Series Advanced User's Reference Manual, Command Reference 3-262 to 3-263 (RCLF) 3-327 to 3-328 (STOF) explains RCLF/STOF in deep detail. These manuals and the 50g equivalents are downloadable from the HP website and www.hpcalc.org , so these commands are not subject for philosophical discussions, their meaning can be found out by simply RTFM;-) The commands work the same way on the 50g, the only difference is that the 50g has more flags than the HP-48. and thus the list may have more elements... Raymond === Subject: Re: RCLF/STOF... [Re: values not showing when code is run (userrpl)] Raymond Del Tondo ha scritto: Yes, right! And I also would have been pleased to provide all the insights others did ;-) Giancarlo === Subject: Re: RCLF/STOF... [Re: values not showing when code is run (userrpl)] thank you very much everybody for clarifying everything i asked for, now its crystal clear, its even explained in different sorts of understanding it. thank you very much. :) cheers! elly === Subject: Re: HP50G HP docs - direct download URLs Direct download URLs for HP English HP50G documents (PDF), as of this posting date: User manual, User Guide, AUR (go here for descriptions and direct links) Complete Training Guides (10MB) http://h20331.www2.hp.com/Hpsub/downloads/hp50gPDFfiles.zip Training guides list (all dated 2006/07/21): SE00 Training aid introduction.pdf SE01 Using the EquationWriter.pdf SE02 Using the EquationWriter Part2.pdf SE03 Working with matricies.pdf SE04 Solving linear systems using matrices.pdf SE05 Base Conversions and arithmetic.pdf SE06 Operations on binary numbers.pdf SE07 Solving for roots of polynomials and quadratics.pdf SE08 Working with units.pdf SE09 Curve Fitting.pdf SE10 Using the Numeric Solver to solve a formula.pdf SE11 The basics of plotting functions.pdf SE12 Calculations involving plots.pdf SE13 Numeric Differentiation.pdf SE14 Numeric integration.pdf SE15 Symbolic integration of polynomials.pdf SE16 Symbolic integration of trig functions.pdf SE17 Symbolic Differentiation.pdf SE18 Changing date and time.pdf SE19 Setting an alarm.pdf SE20 Solving for zeroes of a function.pdf SE21 Calculator modes and customization.pdf SE22 RPN and Algebraic mode.pdf SE23 The USER keyboard.pdf SE24 Confidence Intervals.pdf SE25 Hypothesis tests.pdf SE26 Finding Limits.pdf SE27 Working with fractions.pdf SE28 Probability - Rearranging items.pdf SE29 Probability distributions.pdf SD00 Using an SD Card.pdf SC01 Complex numbers.pdf SC02 Working with Parametric Plots.pdf SC03 Working with Polar Plots.pdf SC04 Working with Fast 3D Plots.pdf SC06 Hyperbolic functions.pdf SC07 Solving differential equations.pdf SC08 Using Taylor Series.pdf BF01 Sinking Fund.pdf BF02 Bond Yield.pdf BF03 Bond Price.pdf BF04 Lease Payments.pdf BF05 Amortization Schedules.pdf BF06 Return on investment.pdf BF08 Trend Lines.pdf BF09 Cost Estimation Linear Regression.pdf RE01 House Payment.pdf RE02 Property Appreciation.pdf RE03 House Pmt Qualification.pdf RE04 House Amortization Schedules.pdf RE05 Remaining Loan Balance.pdf RE06 Loan Down Payments.pdf RE08 Average Sales Prices.pdf RE10 Confidence Intervals.pdf RE11 Hypothesis tests.pdf === Subject: Re: Printing: HP-50G to 82240B The problem was the DELAY. I had to set it to 2.2, after seeing your posts, to make it work. I should have mentioned in orig post that I was using AC power on the printer; distance between units about 0.5 in.. Also, as I said, the printer works fine with HP-42S. Altho calculator batts were about 50%, putting fresh batteries into calculator made no improvement. I did notice, after increasing DELAY, that the arrow stayed on until printing was complete - it had vanished earlier before. However, I was not sure that the printer was not accumulating data in a buffer. So the DELAY is the problem. === Subject: Re: Printing: HP-50G to 82240B Delay's the thing... said Hamlet :) For onlookers, printing to this IR printer is one-way transmission; the printer can't say I'm full, so the calculator must avoid trying to feed it too fast. Older calculators (e.g. HP18C/19B/17B/42S etc.) which correspondingly can only transmit IR have built-in knowledge of this printer's digestion rate (including its buffering capability for the first few lines), and can pace themselves accordingly. The 17B, for example, even has two IR printing modes, called AC Adapter and No AC Adapter, which slows delivery for use of the printer on batteries. === Subject: What happened with the mini challenges??? Hi People... since i`ve been reading this news group i have not seen a new mini challenge... Mini Challenges are very nice becouse it helps new User RPL programmers to learn and becouse it`s a motivation to programm in User... === Subject: Re: What happened with the mini challenges??? The formula below (in FORTRAN) works very well: What makes it interesting is it uses integer arithmetic to make decisions about leap years, and whatnot (decision made as the result of truncation). For the computer, there is also some benefit to using all integer arithmetic (just adds computations on the HP) - although I haven't done any benchmark testing against other more traditional formulas that use lots of if statements. The formula itself is from the Explanatory Supplement to the Astronomical Almanac, and I believe it won a contest ages ago for being the most concise method of converting between the two formats. It is also valid for all Gregorian dates (most formulas you run across are only valid for +/- 100 years or so). I use it on my 32SII for date math. It comes in handy on the 48G for - Kurt === Subject: Re: What happened with the mini challenges??? [Julian date] On that page is explained: The Julian date (JD) is a continuous count of days from 1 January 4713 BC (= -4712 January 1), Greenwich mean noon (= 12h UT) For example, AD 1978 January 1, 0h UT is JD 2443509.5 and AD 1978 July 21, 15h UT, is JD 2443711.125 How about: Flag -42 affects date format, but programs work either way (within the Gregorian calendar range of the calculator, which begins on Oct 15, 1582) You could simply add/subtract your local time offset in the first program, but in the second program, local time outside of 0 to 23.5959 should be normalized, and the date accordingly adjusted, e.g. === Subject: Re: What happened with the mini challenges??? [Julian date] (results are now slightly better :) For example, per http://aa.usno.navy.mil/faq/docs/JD_Formula.html === Subject: Re: What happened with the mini challenges??? [Julian date] (should work for dates after 11.111582) === Subject: Re: What happened with the mini challenges??? [Julian date] Cool. I ran it through a bunch of conversions, and they all worked well. I wish I could do something like this for the 32SII because it has such limited memory. It has no built in date math like DDAYS though. I experimented with using binary base just to make a simple date to JD converter for that calculator (binary so it forces integer arithmetic), but that took up more memory than calling IP after every divide. - Kurt === Subject: Re: values not showing when code is run (userrpl) Reminds me of a comic picture which showed a real programmer's keyboard, with no more keys than necessary: One for '1' and one for '0'... BTW: If you're interested in a real programmer's editor then please take a look at www.semware.com , the maker of TSE Pro ! Raymond === Subject: Re: This is Crazy...!!! re: HP48 Font for Macintosh that would be nice; i tried loading the font collection ( there are oodles of special characters, but curiously; there are several missing...??? 31 [ # b ] = ellipse 96 [ # 60b ] = single left quote 126 [ # 7Eb ] = tilde 159 [ # 9Fb ] = infinity 165 [ # A5b ] = yen 168 [ # A8b ] = naked diaeresis 169 [ # A9b ] = black-dot 3 172 [ # ACb ] = logical not 174 [ # AEb ] = black-dot 2 180 [ # B4b ] = single right quote / acute 184 [ # B8b ] = naked cedilla ??? === Subject: Re: This is Crazy...!!! re: HP48 Font for Macintosh I don't have an OSX capable Mac at the moment so can't help I'm afraid. -- Bruce Horrocks Surrey England (bruce at scorecrow dot com) === Subject: Re: This is Crazy...!!! re: HP48 Font for Macintosh And then you loaded them anyways...!!!??? Or did you think of some way to disentangle the bad ones from the good ones... it seemed to me no matter which one i hit, ( that is, any one of the many little sample windows ) if i hit the litle triangle on the left of the name, all the fonts would be listed, and some of them would appear to be OK, BUT-- if i went to that sample window, the one(s) that seemed to be OK, it would show that that it Wasn't OK and then, i think- if i hit the one that showed previously that it was OK, and but now showing that it wasn't OK, it just Wouldn't load anyways...!!! ??? !!! ??? [ !!! ( this is not really yelling, interpret this as frantic eye rolling and flipping ones hands up and down ) ( i have asperger's; ( which was undiagnosed throughout most of my life, so i never got any 'life style counseling' )( but that's not a very good excuse...!!!??? ) ] [ if you want to go to a mean group, with really Snotty people, try the comp.sys.mac.system room; they are just so mean to me...!!! i cry and cry whenever i have to go over there, they just a bunch of big mean bullies...!!! ] thanx! === Subject: IR to PC? Just wondering. Is there a way to link an HP49/50 to a PC with in IR link? Haven't had any luck. (I should actually try experimenting more before posting, but maybe one of you has already gone thru the drill...) === Subject: Re: IR to PC? Not sure if you're trying to connect the HP via IR to the IR on a laptop for example, but if you are, you can create a virtual comm port using the IR with this utility: http://www.ircomm2k.de/English/index.html I used it with my old laptop about 6 months ago and it worked great. === Subject: Re: IR to PC? Check that last comment. When I try to connect, the calculator flashes the usual stuff, including the little I/O icon. But the program on the PC eventually says it can't find the calculator. I've tried it at several speeds. === Subject: Re: IR to PC? Very strange. I can get it to work both ways with HyperTerminal, but not with the HP comm software. === Subject: Re: IR to PC? Yes, I think it was the Hyper Terminal I used also to connect the HP to the PC through IR, but I no longer have a computer with an IR, only a desktop now and running Vista which no longer has the Hyper Terminal, so I can't investigate. I believe ircomm2k sets up the IR as a virtual comm port, then you can use the Hyper Terminal to connect, which sounds like what you have done. Apologies for not being more clear, but it's been a while and don't remember the details. Someone, somewhere posted a set of step by step instructions including all the settings to successfully accomplish this but I just cannot remember where I found it at the time. Maybe someone else remembers where and by whom? === Subject: Re: IR to PC? This looks like the final word: I tried it with the older HP Connectivity Kit which uses the Kermit protocol. It worked fine - 2- way comms with the calculator. I can get it to work with HyperTerminal, Kermit, but not with the newer CONN4X HP software which uses XModem. The issue with XModem isn't worth further analysis at this end. === Subject: Re: IR to PC? Tell me more. I installed IrCOMM2k, and I can see COM5. What settings to I use on the calculator? I went into APPS, I/O, Transfer... then IrDA and XModem. Then I put the calculator into XModem server mode and opened the connectivity application, set to COM5 and tried to connect at different speeds. No response. === Subject: Re: IR to PC? It works fine. It exposes a serial device which can then be used in the usual communication applications. Samuel === Subject: Re: TED and 30 Columns Email. +++++++++++++++++++++ Check your email, many programs are attached. Enjoy it. - GaaK - === Subject: [User RPL] One-Minute Marvels Question... Hi people... I've been reading and practicing with the One-Minute Marvels book by Wlodek and Richar Nelson... that i downloaded from http://www.hpcalc.org/search.php?query=one+minute+marvels I've some questions about the first and second example... The first example is about Next Prime: 'np' << DUP 2 MOD { 3 WHILE DUP2 MOD OVER s < AND { DROP DUP } IFT } 2 And it's used as subroutine in the following two examples about prime factors... 'pfa' << { } SWAP DO np ROT OVER + ROT ROT / DUP UNTIL END 'pfb' << { } SWAP DO np ROT OVER + ROT ROT / 1 OVER MOD NOT UNTIL END I edited ROT ROT for UNROT... and then i tested the programs... For 1235 'pfb' returns { 5. 13. 19. } But 'pfa' gives an error for IFTE command... Could anybody gives me the reason? becouse i don't find where is the bug... I still dont handle IFT and IFTE commands very good... === Subject: Re: Wrong answer on HP50g I figured out the discrepancy for the value. I used 0.25 ft instead of 3 inches. However, I am still having a problem with the units not switching from metric to english. === Subject: Re: Wrong answer on HP50g The meaning of the menu settings is that newly created variables will be assigned units (SI vs. English vs. none) according to the flag(s); once created with attached units, existing stored variables are not converted (this allows you to even use mixed units, if you want). If you want to change units, you could delete the variables and start over, or else perform unit conversions on currently stored values. === Subject: Re: [not] Wrong answer on HP50g To add one recommendation re the Equation Library: You need to set the desired unit system *before* pressing SOLV; in fact, the easiest place to set it is immediately upon entry to the first EQNLIB selection list (topics); otherwise you'll later have to press VARS then NXT, to get back to the same menu. However, there is also a PURGE key in that later menu, after you have selected an equation, then VARS NXT, which you can use to purge all the variables first; in that case, you _can_ choose your unit system later, but this will create new (zeroed) values, rather than convert any previously stored values. This behavior has been the same ever since the original version of this library (for the HP48). === Subject: Re: Wrong answer on HP50g I do believe your posts are becoming more and more like blank verse. My prosody is a tad weak these days (along with other things :-), but it seems to be iambic pentameter. The English language drifted away from the iambic form post- Shakespeare. Poets have pointed out that this is a shame, since the iamb has the same meter as the beating of the human heart. Bill === Subject: Re: Wrong answer on HP50g There once was a man named McAmiter, Who sported both length and diameter. It wasn't his size That gave girls a surprise, It was his rhythm - iambic pentameter :-) -- Bruce Horrocks Surrey England (bruce at scorecrow dot com) === Subject: Re: Beats, the Poetry and Beauty of Existence Because they emerge directly from a blank mind :) And so nearly does this mathematical verse by Frederic Soddy (Nobel laureate, Chemistry, 1921), re four tangent circles [illustration at http://mathworld.wolfram.com/SoddyCircles.html ] Four circles to the kissing come. The smaller are the benter. The bend is just the inverse of The distance from the center. Though their intrigue left Euclid dumb There's now no need for rule of thumb. Since zero bend's a dead straight line And concave bends have minus sign, The sum of the squares of all four bends Is half the square of their sum. [published in Nature, June 20, 1936] I first read the above in Coxeter's Introduction to Geometry (a math book to put others to shame), but had never heard that Soddy played a part in the U.S. developing an atomic bomb: http://www.pballew.net/soddy.html (how history may turn on a train of accidental thoughts and incidents) The math (and art) can become beautiful in themselves: http://www.ams.org/featurecolumn/archive/kissing.html A mathematician, like a painter or a poet, is a maker of patterns. If his patterns are more permanent than theirs, it is because they are made with ideas. The mathematician's patterns, like the painter's or the poet's, must be beautiful; the ideas, like the colours or the words, must fit together in a harmonious way. Beauty is the first test; there is no permanent place in the world for ugly mathematics. - G. H. Hardy, A Mathematician's Apology Other quotes: Scientific men can hardly escape the charge of ignorance with regard to the precise effect of the impact of modern science upon the mode of living of the people and upon their civilisation. For them, such a charge is worse than that of crime. The catastrophe which has recently engulfed the world has, however, not been without its own intellectual renaissance. The effects are not yet apparent, but, at least, perhaps not all of us are now totally blind to the dangers ahead, or to the need of that impersonal but remorseless re-examination of the foundations of society, which Science has already applied to the mechanism of the physical universe. Possibly it may fail. Perhaps it may be too late. Even so, yet I cherish the fancy that, whatever may happen in the crowded and fevered countries that are still ranged in fratricidal animosity, here at least, here in the Northlands, truth will endure. [Frederick Soddy, at the Nobel Banquet in Stockholm, December 10, 1922] http://nobelprize.org/nobel_prizes/chemistry/laureates/1921/ http://nobelprize.org/nobel_prizes/chemistry/laureates/1921/soddy-speech.htm l Plus .8da change, plus c'est la m.90me chose. -[ ]- === Subject: SOLVER giving messages in French? Howto fix. My HP50g is giving messages in french when using the SOLVER, like Resolution de Vo or Erreur: Mauvaise Estimation. How do I reset these back to english? Is there any chance the EQW will ever handle units? Really puts a limit on the usefullness of a major feature of the CAS system. === Subject: Re: SOLVER giving messages in French? Howto fix. Steve a .8ecrit : Vous avez quelque chose contre la langue fran.8daise ? :-) === Subject: Re: SOLVER giving messages in French? Howto fix. Nope. TW === Subject: Re: SOLVER giving messages in French? Howto fix. But what needs fixing? French is perfect -- the HP49 series was programmed in French :) -[ ]- === Subject: Re: SOLVER giving messages in French? Howto fix. this calculator. Sorry for such a simple question but don't have the manual in the office, over 800 commands and a simple web search didn't find the answer. I seem to remember something similar happening years ago so one of my programs must be doing something strange.......... or it's subtle revenge with the HP over time reverting to it's native language. === Subject: Re: Battery Charge in hp48g Added! +++++++++++++++++++++++++++++++++++++ Source Code ChkLowBat for 48 series +++++++++++++++++++++++++++++++++++++ :: CK0 CODE GOSBVL =SAVPTR * ST=0 15 * disable Interrupts GOSBVL =ChkLowBat * carry S/C GOSBVL =AllowIntr * enable interrupts GOVLNG =GPPushT/FLp * ENDCODE ; ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ Source Code ChkLowBat for 49 series ( ROM 1.18 or higher ) ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ :: CK0 ( no arguments required ) ' ( no eval next_code ) CODE GOSBVL =SAVPTR * ST=0 15 * disable Interrupts GOSUB ChkLowB * carry S/C GOSBVL =AllowIntr * enable interrupts GOVLNG =GPPushT/FLp * A=DAT0 A * #Addr = FPTR 1 75 LC(5) #C7 * :-) A=A+C A * #Addr =ChkLowBat ENDCODE FLASHPTR 001 0000 ( EVAL from bank, imperative! ) ; ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ Source Code MASD syntax for 49 series ( ROM 1.18 or higher ) ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ !NO CODE !RPL :: CK0 ' CODE 33 8FB976084F7E008F767628D332531BB6404142347C000CA81B2 FPTR 1 0 COERCEFLAG ; @ +++++ Note: =ChkLowBat extracted from SetEcma94 for 48 and 49 series. +++++ - GaaK - PS: 49 series: 49G - 48gII, 49g+, 50G. === Subject: 50G - sporadic pauses - garbage collection ? My HP-50G frequently goes into sporadic delays when executing keyboard functions or recalling variables. Is this garbage collection? I've installed fresh batteries - makes no difference. Dick Chaffer Bozeman, MT === Subject: Re: 50G - sporadic pauses - garbage collection ? Other possible causes: o Viewing (or even listing) SD card contents. o Spontaneous re-organizing of user flash memory (Port2). o Inadvertently running a program or keyboard function. o Loose battery connection? o Internal intermittent connection? o Alien space/time warp fields (are you near Area 51?) === Subject: Re: 50G - sporadic pauses - garbage collection ? Ah! I think your first post answered the question, at least for me. === Subject: Re: 50G - sporadic pauses - garbage collection ? [FMEM] Back when I lived in New York, the Garbage Collectors sometimes went on strike, causing an even greater (and more uncomfortable) delay :) How often, and what value is returned by MEM? Are the batteries making firm contact with those springy things in the battery compartment? If it is due to GC, rather than to other causes, then a strategy can be employed to cause more frequent (but shorter) cleanups, similar to the strategy of doing some chore a little bit every day, rather than waiting until you have to drop everything for a whole day, just to catch up with a big load that has been left to accumulate. The very small program below is called FMEM, and it accepts as an argument how much memory (bytes) you want to leave free If you leave 256K free, then you wait a long time between GCs, but each GC takes a longer time, whereas if you leave 32K free, you still have plenty of free memory and will not run out of memory, but the system will do a much shorter GC at more frequent intervals. Smoothing out the workload in this fashion gives especially noticeable relief from long pauses in the slower machines, e.g. original HP48[S/G] and original HP49G (before 49G+/50G/etc.) which you had left free, you can re-run FMEM whenever you want, to set the amount of free memory back up again to a comfortable margin. To free the maximum amount of available memory, request 1 EEX 7 bytes, (the exact amount actually left free is always returned as a result). Here is an updated FMEM for all members of HP49/50 series: (do HOME 256 ATTACH before downloading or typing the program, because MAKESTR needs built-in library 256 to compile properly) @ HP49/50 series only << 5000. MAX :0:JUNK DUP PURGE MEM ROT - Here is an updated FMEM for original HP48[S/G][X/+] only: (try it if you've got 128K or merged 256K): @ HP48S[X]/HP48G[X/+] only << 5000 MAX :0:JUNK DUP PURGE MEM ROT - #18CEAh SYSEVAL #61C1Ch SYSEVAL The standard precaution applies (especially to the HP48 version, which employs internal system function addresses, errors in usage of which can cause a crash or memory clear): Back up your memory before trying any new program The last prior version of FMEM was posted five years ago (when all available models ran on slower, real Saturn CPUs) === Subject: Re: 50G - sporadic pauses - garbage collection ? [FMEM] suspect you're right on, again, again. Dick Chaffer === Subject: Re: 50G - sporadic pauses - garbage collection ? [FMEM] This meant how often does the delay occur (not how often changing batteries :) === Subject: Re: 50G - sporadic pauses - garbage collection ? I've noticed that too sometimes. I don't know what causes it. Especially when switching between soft menus. === Subject: =?gb2312?q?Antique_1912-1949_Porcelain_teapot_528_chinaware_400=A3=ACinvitati on_visit?= Chinese antique, wonderful treasure big collection £¬invitation visit Antique 1912-1949 Porcelain teapot 528 chinaware 400£¬invitation visit Jianxiong Xu City Shengzhou Zhejiang China Tel:£¨86-575£©83103731 Website:http://www.xjx588.com E-mail:xjx 588@tom.com === Subject: Re: Holland HP48SX Lib - REF? I sent an e-mail to the address listed in the book, and got no reply. This has been several years ago. By the way did you get the addendum for using the card in the GX? If you did could you send me a copy by e-mail or post it in the body of a message to this group? Harold A Climer Dept. of Physics,Geology, and Astronomy U.T. Chattanooga Harold-Climer@utc.edu === Subject: Re: Holland HP48SX Lib - REF? Hi. I haven't tried to contact the author yet, too busy. I bought a 50G and decided to use that because the functions are there already. After looking at the GAUS and AGAUS again, it seems that using both of them together will perform Gaussian elimination on ONE row of the matrix, and it only works for square matrices. There are a lot of other useful functions in there that would allow you to write your own program to do the whole Row-Echelon conversion, but it doesn't seem to be built in. Kind of suprising! Mark Smart http://www.marksmart.net === Subject: Batteries Reports About 18 months ago Joe Horn reported on experiments with Panasionic Oxyride batteries in the 49G+ with positive results. With it's additional current draw sounds like they would be good on the 50G too. Has anyone tried them in this application? Joe also noted that he intended to experiment with the 1.5V Energizer Lithium batteries in the 49G+ and report on his results. I have not found this report yet and wondered if it was posted? Rich Wood === Subject: HP30S firmware update During some discussion at hpmuseum.org, it became apparent that my old HP30S was behaving differently than newer ones. So I went out and bought another HP30S. My old unit has a serial no. of CN0019 and the new one has a serial of CN0143. The new one behaves differently. I get the same result as Karl and Jesse on the various tests we've been discussing. I notice that on the old one, SIN(3.14159265358) returns 9.793238461E-12, whereas the new unit returns exactly zero. I experimented around and found that they have done some more of that purification of certain results. The HP30S can only accept 13 digits as keyboard input, so to perform calculations on 24 digit inputs, one must do arithmetic in the input string. So, if I type: sin(3.1415926535+7391741495627E-23)*1E17-1587582 enter I get: .3506383 which is more or less consistent with internal 80 bit (24 decimal digit) arithmetic. But, if I type: sin(3.1415926535+7391741495628E-23) enter I get exactly zero. So, if you get close enough to PI on input, they return a result as if the input was exactly PI. The older unit doesn't do this for this particular calculation. The new unit seems to use a 77 bit internal representation for PI, whereas the older one only uses 70 bits. Also, I notice that the new unit returns a correct 24 digit square root. Joe Horn had noticed that the HP30S didn't return a result accurate to 24 digits when the calc was first released, but that has been fixed now. So, the firmware has definitely been changed since the initial introduction of the HP30S, but I don't know when. Whether it's an improvement or not depends on your point of view. === Subject: Re: HP30S firmware update -- PING RODGER Rodger, Would you send me a working email address please, to: editor at hpcc dot org -- Bruce Horrocks Surrey England === Subject: Re: HP30S firmware update The 35s returns 9.7932...E-12 as well so the firmware revision didn't make it into this model yet. Tom Lake === Subject: Re: HP30S firmware update The 35s works completely different from the 30s. It uses standard BCD arithmetics like most other pocket calculators. This helps to prevent those well-known strange results found in most computer languages, or even Excel (try 100 - 99.99 - 0.01 and see what you get). That's because for most decimal values there is no *exact* binary representation. The 30s uses the same kind of binary arithmetics, but there are enough bits acting as some kind of guard digits that can be rounded away so that the user sees a valid result. Of course this implies some special kind of rounding. If the results get very close to zero the machine has to conversion or it's indeed the correct result. That's what can be seen here in the 30s example. In the sin(x) case the result for arguments extremely close to pi should be quite exactly pi - x. An that's what you get with the (BCD) 35s: pi = 3.14159 26535 89793 23846 26433 ... Dieter === Subject: HP49G+ usb power without batery Somebody knows about or have a 49g+, specially the last models with usb power without capability? Humberto Estrada === Subject: Re: HP49G+ usb power without batery Here is a recent post regarding a 50G which will not work with batteries but only through the USB port power. Sounds to me like bad workmanship is allowing units with wiring problems to the battery compartment to be built. It almost sounds like functional testing of completed units is being done at the factory, if any, via powering the units through the USB port only. If so battery compartment wiring up problems would never show up. If a new unit under warranty I would demand a warranty replacement from HP. Rich W === Subject: Indeterminate Integration in 48g Hi people... A friend of a spanish forum is looking for the way to do the indeterminate integration of: in a 48g... We`ve readed the procedure of the User Guide but it doesn`t gives us the correct answer... Answer: 1/2008(x^2008Lnx-(x^2008)/2008) I`ve tryed in my 50g too but i`ve not finded the way to do indeterminate integration with it... i know it can do this but i dont find the way, and the user guide doesn`t explain that... I think about to do a taylor aproximation for the function but my 50g have been working about 10 minutes without an answer... Can anybody help us...??? === Subject: Re: Indeterminate Integration in 48g OK. In exact mode: INTVX(X^20) works fine... and fast INTVX(X^200) works fine... and quite fast INTVX(X^2000) works fine... and no so fast INTVX(LN(X)) works fine... and fast INTVX(X^20*LN(X)) works fine... and quite fast minutes on a HP49 emulator on an Intel Macintosh!) So be (very very) patient. FYI the answer takes less than 1 second on a ti nspire ;-) === Subject: HPgcc for 49G+ Hello I am new to this board so please bear with me. However , here is what's happening when I tried to run program written,complied and send to my Hp 49G+. Nothing never happens, meaning the program doesn't run as expected here are few step takens so far : 1) I install the Set,bin file on port2 2) Compile program, compiled correctly 3) used hp connectivity , move program to user Stack 4) User prRun function from ARMtool to run here is the program: void main(void) {printf(please enter 2); } . === Subject: Re: HPgcc for 49G+ Sorry for the double post. . . is it just going so fast you don't see anything? If you don't have a while(!keyb_isON()); or something at the end that might be happening. TW === Subject: Re: HPgcc for 49G+ Try installing it in port 1 and see what happens. Sometimes some other libraries in port 2 prevent it from being aligned properly in memory. If this is your problem, it will then work and run fine. TW === Subject: 48GX Manufacturing Differences I have two 48GX Units. The Singapore made unit is dated SG722xxxxx and the Indonesia made one is dated ID744xxxxx. Per serial numbers this makes them theoretically manufactured 22 weeks apart. I presume this makes the Singapore made unit manufactured near the end of production there and the Indonesia made unit from near the start of production. The Singapore made unit has double shot injection moulded keys while the Indonesia made unit uses silkscreened keys. Antone else noted this manufacturing difference? Also the Indonesia made unit seems to show greater temperature sensitivity affecting operating speed. On a 2 minute run time program the indonesia made unit seems to show a run time variation of up to 7 seconds depending on calculator temperature while the Singapore made unit only shows about a 2 second difference. Both run faster when cold. Rich W === Subject: Re: 48GX Manufacturing Differences The fact that you measured this shows you have too much time on your hands. Isn't there some local community organization you can volunteer this time to, like mentoring youth, taking hot meals to the elderly? -- Dominic Richens | knob@storm.ca If you're not *outraged*, you're not paying attention! === Subject: Re: 48GX Manufacturing Differences The fact that you posted this message shows you have too much time on your hands. Isn't there some local community organization you can volunteer this time to, like mentoring youth, taking hot meals to the elderly? Oh. Wait. Damn. Interesting find, richwood... any idea about possible hw differences that might account for this? === Subject: Re: 48GX Manufacturing Differences Reading some old posts here the clock is apparently a LC or RC circuit and is known for it's instability compared to a quartz crystal frequency control. I would suspect it is a matter of different component vendors or lots. Enough variation from unit to unit per posts so apparently some people get good clock accuracy and others do not. Rich W === Subject: Re: 48GX Manufacturing Differences Rich W === Subject: Connecting the 50G to the internet I can connect my 50g to my computer using Usinagaz and pppd, but I can't get a connection to my network or the internet. Do I need to set up some sort of proxy or port forwaring? If so, is there a guide anywhere? === Subject: Re: Connecting the 50G to the internet I don't think you can surf the internet using a calculator. === Subject: Re: GPS Orbital Path Question; Arent the GPS satellites low enough so that the user's position on Earth needs to be taken into consideration for any skyplot? I would think otherwise the parallax would be a killer. Maybe I have misunderstood but I have not seen this mentioned in the thread. Sounds like need GPS info fed into the program too for a reasonably accurate plot. Rich W === Subject: Re: GPS Orbital Path My url has basic orbit lessons with HP48 programs. http://www.grahamkendall.net/ === Subject: Re: GPS Orbital Path Yeah. And probably much more than what I'd need. Especially since it is for display on a TINY screen. If I remember right, the GPS sat positions don't change a whole lot, correct? So if I code something up will it still be mostly correct in a year or two? I know I'd need the ability to turn off broken sats, but that isn't a big issue really. TW === Subject: Re: GPS Orbital Path The orbital period of a GPS sat is ~ 11 hr 58 min, so from one day to the next there would be little difference on the 50g screen. But at 4 min. per day it wouldn't be long before positional discrepancy became apparent. Chuck === Subject: Re: GPS Orbital Path http://en.wikipedia.org/wiki/Orbital_elements ? Steve Sousa === Subject: Re: GPS Orbital Path I won't help you much but if you agree to some of this it may change your direction just enough to realize... When i was working on a GPS thing (WinCE based) i was about to display positions of satellites when i realized: Who would be interested in acquiring positions of the satellites in application which should actualy tell you where YOU are, especially knowing that these don't provide any other service ? So i decided : For most users it's important to display their position as acurate and as prominent as possible error parametters) I did display how many satellites are in sight and their signal strengths as a feature basicaly it's also nonsence, but one may get in to situatuion where it would say : 5-6 sats. in sight, without position fix This way one should be able to see that there is no signal and may realize that they're in the tunel :-) conclusion : To display GPS sat. trajectories is only for cosmetic purposes People who look at these probably have nothing better to do at that time OR may crash their car while doing that :-) (in car navigation applications) However i do admit/realize that application which does display sat. positions may LOOK more hight-tech to some which are not realizing that they are looking the wrong way (missing the point of the app.) as a result: i'm parsing all data but i choose not to bloat the display with it that was on WinCE 240x320, 50g is 131x80 much less space so... however you decide... have fun ! manjo === Subject: Re: GPS Orbital Path Well it already tells you where you are, it mainly has to do with planning WHEN during the day would be best to work at a site. Check. Yup. One of those impressive things that doesn't have a whole ton of use, but does have use for some. If it wasn't going to be too tricky I was thinking of implementing it. TW === Subject: Re: GPS Orbital Path Roger that :-) but then, GPS delivers azimuth and elevation (current position) of the satellites in sight to know the trajectory (for planing) you will have to do additional research on trajectory parametters on each of the GPS satellites which is bothering all by itself but then if you already have these -then it's just a matter of graphical representation if i had all that i would figure out a function of my own to aproximate azimuth and elevation in polar system (132x80) of course i would use built in ploter to plot trajectory without axes of course GOR a satellite number/symbol to current position (could look pretty impressive :-) manjo === Subject: Re: GPS Orbital Path Yup. My first one took me about 30 minutes and 100 commands in sysrpl, but I think it works fine. Redraw time is less than a second including parsing so I figure that is good enough since this isn't a speed critical thing, but I am sure there are ways to improve it. If anyone has any suggestions I'd be happy for them. http://pssllc.com/pub/pics/satview.png Yup. :-) TW === Subject: Re: GPS Orbital Path very nice, less than a second is decent speed too (for the task) now immagine all the same, but in grayscale :-) (maybe few circles less to make it look less like a target cross) but then if it's light gray it could look nice just the way it is also sattelites with lower signals would be brighter-stronger darker could look just a bit more impressive :-) (then again you're on the right track as is.. and if rest of the application is not GS it could stand out too much) manjo === Subject: HP 50g SOLVEVX I came across this recently (in C~) 'X^2 + X + 1' SOLVEVX returns {'X=(-.5, -.866...)' 'X=(0.,1.732...)-1)/ 2.' } Why this odd (not completely evaluated) representation for the second root? EVAL EVAL takes care of it, of course, and also 'X^2 + 1.*X +1' returns both fully evaluated roots. Any ideas?