A111 You can easily make your own OBJFIX program on the HP49. Argument: 1: Name of a variable that contains a String with HPHP49-Cxxxxxx on the stack 256 ATTACH Then compile this program: << MEM DROP RCL ->H 26 + H-> NEWOB >> > > > > There may be a simple answer to this problem: > > When I download from hpcalc and transfer the program to my 49, upon > > execution, I get a string printed on the stack. Specifically, I have > > Similar problems can happen with the 48. What does the start of the > string look like? > > Has someone managed to port OBJFIX to the 49 yet? :) > Ray-O-Vac is also saying that its newest power stations > are ready to recharge anything: Renewal, NiMH, or NiCd > (did they update the electronics, or just the packaging?) I hope they updated the electronics :-) AIUI, NiCds need constant current charging with some form of end-of-charge-voltage detection, Renewals need current pulse charging (with rests to allow the open circuit voltage to drop), terminated by the average pulsed current dropping (nearly) to zero. ---------------------------------------------------------------------- I keep getting this message when I turn on the GX. I know there is a command to re-initialize all the ports, but just dont remember it. Any help would be appreciated. ---------------------------------------------------------------------- > I keep getting this message when I turn on the GX. I know there is a command > to re-initialize all the ports, but just dont remember it. Any help would > be appreciated. PINIT >Details should popup soon on various www-pages. I was at there also and got the free HP49G. I like the calculator but it doesn't give me the impression that it is made to last like my 48Gx. Was quite distressed for quite some time when on first power up, I couldn't get the menus to appear on function keys; instead the menu items were in choose boxes. Finally stumbled into flag -117 that controls this!! Setting this flag (via MODES / FLAGS) yields menu choices on function keys. Also, the Handbook is incorrect. According the the Handbook: Set: The default menu form is a choose box Clear(default setting): The default menu is a set mapped to the function keys This is incorrect; it should read (at least according to the operation of my machine): Set: The default menu is a set mapped to the function keys Clear(default setting): The default menu form is a choose box. ---------------------------------------------------------------------- On power up, my HP49 gave a 'Invalid Card Data' message. Then, it displayed some type of graphic for a split second: ------------------------------------- | Powered by: | | | |<> | |cw ver3.0 | ------------------------------------| The machine will repeat this sequence on a F4 keystroke. Any comments/other details? ---------------------------------------------------------------------- The line near the bottom of the previous post should read: The machine will repeat this sequence on a F4 Q sequence. ---------------------------------------------------------------------- Don't mean to be 'whining' here; just found an interesting difference here on the 49 vs the 48: A feature on my HP48 which I enjoyed and used quite often is the Fn to store the bottom stack item in the VAR indicated by the Fn key. On the HP49 (where the blue shift of the STO key is RCL), the Fn seems to do a STO. Further, if one is to tick the variable via the apostrophe ( ' ), the apostrophe on the 49 is a shifted stroke, so it is a bit more cumbersome to get at. ---------------------------------------------------------------------- Ok, so I can find the 256 Menu that JKH says contains Hacker's tools. Can I use one or more of them to compile SysRPL programs right on the 49? 256.01 [ ->H ][ H-> ][ ->A ][ A-> ][ A->H ][ H->A ] 256.02 [ ->CD ][ CD-> ][ S->H ][ H->S ][->LST ][->ALG ] 256.03 [->PRG ][COMP->][->RAM ][ SREV ][ POKE ][ PEEK ] 256.04 [APEEK ][ R~SB ][ SB~B ][ LR~R ][ S~N ][ LC~C ] 256.05 [ASM-> ][BetaT ][CRLIB ] ----- Poke, peek, CRLIB I can get, but ->H, ->A... I'm clueless... Any info or pointers to info would be greatly appreciated. Yes and no. You can hack together anything you like, but almost all the System RPL entry points have moved, so you have to be careful. As best as I can figure, here's what the commands are: [->H] To Hex. Same as ->ASC in the 48 HACK library. [->A] To Address. Same as ADDR in the 48 HACK library. [A->H] Address to Hex. Converts #12345h to 54321. [H->A] Hex to Address. Converts 54321 to #12345h. [->CD] To Code. Turns a string of hex digits into a Code object. [CD->] Code Out. Turns a Code object into a string of hex digits. [->LST] Make List. Composite or meta-ob --> list. [->ALG] Make Algebraic. Composite or meta-ob --> algebraic. [->PRG] Make Program. Composite or meta-ob --> program. [COMP->] Composite Out. Composite --> meta-object. [->RAM] To RAM. Dumps any ROM object into RAM. [SREV] String Reverse. Turns strings backwards. [POKE] Pokes raw hex into RAM. [PEEK] Extracts the raw hex at any address. [APEEK] ? Indirect PEEK? [R~SB] Real <-> System Binary. Converts reals to bints & back. [SB~B] System Binary <-> Binary. Bints to user binaries & back. [LR~R] Long Real <-> Real. [S~N] String <-> Name. Change VERSION to 'VERSION' & back. [LC~C] Long Complex <-> Complex. [ASM->] Assembly Out. Disassembles Code objects into source code. [BetaTesting] --> 184-byte string useful for testing string routines. [CRLIB] Create Library. Works similar to HP48 lib creators. > [ASM->] Assembly Out. Disassembles Code objects into source code. No [->ASM] ? Anybody want to poke around ASM-> and see where the entries table is? (ie, compile a list of entries, port to HP49, and decompile?) ---------------------------------------------------------------------- Well, according to Emeric Vernat, no [->ASM], that's right! Maybe another command does it? Oh yes, you can use them to compile SysRPL programs ! I think these commands come from the library Dev by Etienne de Foras. ( I With H-> you can find pointers for every command. We can for example find pointers for every old-UserRPL command and so, we can write a converter of HP48G to HP49 for UserRPL program. (We should assume that these commands have the same behaviours) . Take a look at: http://www.hpcalc.org/apps/mk/mk230pdf.zip Chapter 4: Miscellaneous Utilities. Most of the menu commands from menu 256 are there, but there still doesn't seem to be an assembler/compiler command. But I think Etienne de Foras and Dev 3.1 is off the hook on this one. SO, I ask again: How do I compile a SysRPL program on the 49G, given the program as a string on the stack (JYA? JKH?) - C. Fishburn > Oh yes, you can use them to compile SysRPL programs ! ---------------------------------------------------------------------- I have recently purchased a 48gx but I noticed that for $70 more I could get a 49g. I can return the calculator and get the 49, is it worth it? I am majoring in math and science at university and I wish to teach ---------------------------------------------------------------------- I just got off the phone with Wholesale Advantage (I know...not the same, but I figured the input would probably be appreciated)...they gave me a ship date of Thursday, August 26. That would put the Calc into my hands on Wednesday, September 1. I know I was about 120th on the list... ---------------------------------------------------------------------- Treat it like a textbook: at the end of the semester, if you like it and think it's valuable (or expect to need it for a future course), keep it. If not, sell it; the depreciation in value is the part of the cost of education. ---------------------------------------------------------------------- You can find a lot of info on the 48 at www.hpcalc.org. our Speed UP have 30-50% for every thing you made with the HP49G (alos HP48) and 100 % for connection to another HP49 with speed up so transferspeeds up to 30720 baud are possible !!! >Since the CPU/controller on the 49 uses the high voltage pin (previously >for the IR) for falshing the ROM, how does this work. we split the normal rs232-Port to 2 Ports - rs232 and IR, thats all > Is another chip >(ie controller) added ? no it is the same. M >screen: > even though the screen is higher constrast, > it is covered by super shiney plastic that glares > under most lights and if the light isn't positioned > right, shadows are cast under the lcd characters. The plastice lens (screen shield?) almost negates the benefits of the higher contrast screen. It get's annying to look at under certain lighting conditions--especially under polarized light when you get to see rainbow light interferance patterns. They should have use glass or an optical plastic. >I've only had the 49 one day and > there are already scratches > on the shiney plastic screen cover. I've had the 48 > for a few years and there are no scratches on it's screen Big Timne Ditto. It gets worse after w a week. I've even tried using car wax to hide the scratches but it only helped a little. > >fonts: > while there are 3 main fonts (6,7 & 8 pixels tall) on > the 49, they don't have one as large as the normal > 48 sized font (9 pixels tall) - I suppose you can > eventually download larger fonts into the flash? The small fonts don't delight me to much, especially if I have to strain my eyes to see the screen with the plastic lens. >>layout: > it's hard to get used to the enter key at the > bottom right. > in RPN mode some keys require > pressing a shift key WHILE simultaneously pressing > another button which isn't easy if you are holding > the calculator (entering the graphing screens > requires this) > >cover: > is there a trick to taking the cover on and off? > it takes so much force to pull it off I'm afraid > I'm gonna fling the calculator across the room.. I am going to fling the cover across the room. It will loosen up in a few days. I started using a old nylon 48 case which also holds the pocket guide nicely. At first I liked the idea of a cover because I broke a 48 screen (hence extra case), but I don't like the unit not seated snugly in the cover when it is being used (it's a sloppy fit and a bit hard to get off). ---------------------------------------------------------------------- When you take the HP 49G out of the blister pack in which it shipped, the first thing you do is ***REMOVE THE PLASTIC SHIPPING FILM*** which covers the real screen! I pointed this out to some of attendees at the 1999 Handheld Users Conference Vancouver, WA. And yes, as several fellow attendees to the conference have pointed out, Hewlett Packard did indeed give a brand new production HP 49G to each and every person who attended conference, a most gracious act on their part. > > even though the screen is higher constrast, it is covered by super > > shiney plastic that glares under most lights and if the light isn't > > positioned right, shadows are cast under the lcd characters. > The plastice lens (screen shield?) almost negates the benefits of the higher > contrast screen. It get's annying to look at under certain lighting > conditions--especially under polarized light when you get to see rainbow > light interferance patterns. They should have use glass or an optical > plastic. > >I've only had the 49 one day and > > there are already scratches > > on the shiney plastic screen cover. I've had the 48 > > for a few years and there are no scratches on it's screen > Big Timne Ditto. It gets worse after w a week. I've even tried using car > wax to hide the scratches but it only helped a little. ---------------------------------------------------------------------- > After about a 4-5 days worth of taking the case off / putting back on, > it will be much easier to do so. > I have found that depressing the case in the center while sliding the > cover off helps considerably. > > I agree about the screen, perhaps HP wanted to cut down on the number > of smashed screen complaints. Whatever the reason, HP should've known > better than to put a reflective surface over the screen (mine also has > a slight rainbow discoloration to it, enhanced by the position of the > light) > > I believe the graphing speed hype is from the FAST3D plot type. It > allows you to rotate a 3d plot very quickly. Although it is nice, I > don't know how useful it could be. > > One thing that has haunted me: is there some ASIM functionality of this > calculator? Aside from the Algebraic Simp that we found in Alg48, I > also detect a lack of numeric->rational conversion. One program I found > useful for that was Qpiracine (from hpcalc.org) it could take .707... > and turn it into sqrt(2)/2. > > Something strange (although it is correct) I have found is when you > want to take higher roots of numbers, sqrt of 120 displays as usual. > When you take the 3rd root, it is in terms of ln and exp. It is indeed > correct to write that, I just wasnt expecting that representation. > > Some other observations: > there are two algebraic tick symbols. One is a fancy ' and the other > is the standard '. To get the fancy one hold down the right shift and > press EQW, to get the standard tick, press (do not hold) the right > shift and press eqw. > to purge from the stack, press the tool button: then it is F5 > I found that left-shift mode was to hard to access the Custom menu, so > I re-assigned the CAT button to do the same: <<1 menu>> > > > > cover: > > > is there a trick to taking the cover on and off? > > > > One hand holds the 49 by the rubber ears at the top. Other hand > grips > > the sliding cover at the *bottom* between thumb on top and next 2 or 3 > > fingers between cover and case. Use these fingers to apply > > pulling pressure and you'll find the cover pops off nicely... with > > the 49 still in your hand :) ---------------------------------------------------------------------- Okay, I was depressed to find out all the work apparently went into fancy calc. and alg. stuff, not usability (defind as easy to use when you actually learn how to use it!). I still figured the screen and memory sounded like a good upgrade from the 48. Now, as I understand it from info. on this group, 1) The screen in no better, at best, than the HP48 2) The calc is just about no faster for the _work_ I do with it 3) Garbage collection is worse 4) The calc can be slower and harder to use with less soft keys 5) The keyboard is cumbersome because of layout and hard-to-press keys. ---------------------------------------------------------------------- > Okay, I was depressed to find out all the work apparently went into > fancy calc. and alg. stuff, not usability (defind as easy to use > when you actually learn how to use it!). Very unfair statements, I would say, spoken from extensive non-experience. Why not let the collective statements of all those people who do have the calc in their hands stand for themselves, and not dilute them by reinforcing negativity from people who did not learn all the helpful things that evidently *do* enhance usability, as discovered and related for others' benefit by people who have really taken the time to use the thing for a while. ---------------------------------------------------------------------- This morning, i got my flashy new Hp49G. However, there are some stuffs written in the manual that don't work with my calc : -The blue-shifted graph function (Y=, WIN, GRAPH, 2D/3D, TBLSET and TABLE), don't work at all in RPN mode (it is just like i pressed w/o the blue-shift). However, it wirks perfectly in Algebraic mode... -Where is the Coma mode ?? I tried ON-C but it didn't work. In RPM mode, hold the blue shift down while you press the function key. (Only needed for the special functions on the Fkeys, not needed for normal shift operation.) ---------------------------------------------------------------------- In RPN mode *hold down* left-shift while pressing F1. > -Where is the Coma mode ?? I tried ON-C but it didn't work. If you mean using comma as a fraction mark then press MODE and check the FM, box > In RPN mode *hold down* left-shift while pressing F1. > > -Where is the Coma mode ?? I tried ON-C but it didn't work. > > If you mean using comma as a fraction mark then press MODE and check > the FM, box No, i mean to put the calc. in the COMA status. With the 48, you just have to press ON+SPC. But with the 49, it doesn't seem to work... > Have you tried ON+SPC ? 'ON-C'...:o) ---------------------------------------------------------------------- Can anyone advise me on how to prompt with text for data input on HP48 Also is it possible to plot STD curves as a line curve and not ---------------------------------------------------------------------- There's a simple, scriptable RPN calculator, and some simple graphing software, but nothing (to my knowledge) at all near the abilities of even the 28. Keep in mind the focus of the Palm and its market -- it's a PDA, not a scientific calculator. A friend suggested a Palm 7, linked to a Mathematica engine. IMHO it'd be cool, except for the fact that Mathematica's rather overpowered for anything where you're not using an actual computer. However, it would be interesting if the screen was sized up to 8.5x11. The AUG is the manual that tells (hopefully) all that the manual coming with the HP 49G doesn't tell you. A simpler definition would be : the AUG is the RPN and programming manual. ---------------------------------------------------------------------- It will probably do justice to the flags, commands, errors, status messages, etc.. But as far as Assembly language goes, I didn't think that was part of the deal. I thought the understanding there was that the development software would be included, but not supported -- i.e. you're on your own. So while it might explain UserRPL and HPBasic (Algebraic version of UserRPL?), SysRPL and Asm would be very much at your own risk. ---------------------------------------------------------------------- > I'm not following you. Where, exactly, is it? The only semicircle I > see on top is to orient the serial cable. At the top of the lcd (where indicators such as the shift and alpha keys are shown). It looks like an oil ring you would see in a parking lot (or, more exactly, on the top of my newton's screen if lots of oil gets on it). (Did that help?) -- The pins are completely different. (10-pin vs 4-pin) I believe the answer is no. ---------------------------------------------------------------------- > DUP DUP AMUL DUP DUP AMUL DUP DUP AMUL DUP > DUP AMUL AMUL AMUL AMUL AMUL AMUL > >> > << DUP DUP AMUL DUP AMUL AMUL DUP DUP AMUL DUP AMUL AMUL DUP DUP AMUL DUP AMUL AMUL DUP AMUL DUP AMUL DUP AMUL << will do it in about 1.6s, then. ---------------------------------------------------------------------- I also encountered a bug like the one you're giving, and I also got a memory Clear. Here are some others : -> Comma as fraction mark option doesn't work at all. Just try it : set comma FM mode, enter a complex number (2 3) in the stack, it will appear like this : (2,,3,) Nice, isn't it ? If you edit this number, it becomes (2,;3,), and the editor is unable to validate it (doesn't understand the semicolon). Obviously this feature hasn't fairly not been tested and is unusable. This is very bad. -> If you hold down the shift key and press the NXT key, you get the last menu. Fine. But if you do it inside the editor running itself inside the Files application this causes a hangup. I've found some other situations where the calculators hangs up. In some cases, if you are lucky, you can stop it with a system halt (ON+C). If not, use the reset hole, and maybe get a Memory clear... -> In the HP48, CHOOSE command with option 0 inhibits the cursor. This is useful to display long lists. This doesn't work in the 49. -> While viewing some huge matrices, my screen AND MY VARS got garbled. During 10 years of intensive 48 use (A,E,M), all the bugs I know are those I've read about in errata or in FAQS. I never encountered one by myself. On the other side, I think HP49 has a very good design. I find the keyboard layout has been very cleverly defined, the algebraic capabilities of the calc are really excellent, and some speed improvements on several applications are welcome (very good and usable equation editor, filer, pretty stack display...). Apart from the rewritten applications, execution of an RPL program is not faster than with the 48. In some cases (just try a boolean graph and compare by yourselves) the 49 can be much slower (why ?). I think this is a very good product but there is an urge for serious debugging (fortunately we can upgrade the system). Please ACO guys be quick, your sales will tightly depend on it ! ---------------------------------------------------------------------- I transferred some stuff from my hp48gx to my brand new 49g, and noticed something weird... This is quite a long description, please read it, I'm quite desperate because of the problem... I have an algebraic expression, quite a complicated one, and when I try to graph it or just evaluate it by storing numbers to the variables involved, the hp49g simply freezes, while my 48gx evaluates the expression in a fraction of second. Here's how the expression looks like: 'LOG(ABS(c/f/*((1-COS(2**f/c*l))/SIN(2**f/c*l))/20))*20' (the expression depicts something related to a transmission line to a dipole antenna, don't remember what, I derived the formula for some time ago) the values for the variables are: f: 0 l: 20 c: 299792458 Actually, the equation produces an error on my hp-48gx, because there is a division by 0, but I could not get my 49g do anything else than just freeze. I tried everything I came to think about, i.e. tried to set 'numeric mode' on in CAS setup. I also tried to both evaluate the expression as it was after transferring from the 48gx (in ASCII mode) and after keying it in by hand. With values of 'f' that don't produce an error on the 48gx, the expression seems to evaluate reasonably on the 49g. Have I missed something crucial related how to evaluate things just numerically or what is going on? ---------------------------------------------------------------------- > You can change this by checking the approx setting in the CAS set up > under MODES. Right... actually, I did this, too, sorry for not mentioning it... and to be more accurate again, when I try to graph the function (with setting 'f' as the indep. var of course), the calc complains about / : undefined result. I tried with adjusting the flags so that both infinite and overflow produce an 9e499, not an error. I tried everything I could think about, but still didn't get the 49g to graph the function. So far I am not very impressed on the 49g, but I guess it is mainly that it takes time to get used to it, which was the case with the 48gx as well. I like the 48 very much of its speed in numerical computation. I tested some of my old 48gx UserRPL programs on the 49g, and those ones that had lots of loops run very very slowly. ---------------------------------------------------------------------- > I've got a very annoying bug with my HP49: > If I try to multiply a real matrix (type 3) with a real vector > (type 3), the calc just freezes, Oh, right. The funniest part is that, if you replace the second array by [5 6], i.e. use integers instead of reals, the calc is able to compute the solution. But your example does freeze the calc, indeed. Here's another one : Most HP49 predicates return integers : 1 for true, 0 for false. In some cases, the returned number is a real (type 0) : ex CHOOSE returns 1. or 0., in other cases, it's a an integer (type 28) : ex comparison functions (==, <, etc.). Why this inconsistency ? Could not booleans always be coded in the same type ? The fonction IFTE has a strange, strange bug : << 1 2 == Cool Oops IFTE >> works perfectly, but << 1 2 == 1 0 IFTE >> returns 'IFTE(0,1,0)'. No evaluation is done if the three arguments of IFTE are integers. You have to force it with ->NUM. I'm not trying to devaluate the HP 49, but I think there's an emergency to fix some big problems before the calc acquires a bad reputation. We all know this could be the worst thing to happen to it : not to sell because of being too buggy. And I repeat, I love the calc, but precisely, please ACO do something quick. ---------------------------------------------------------------------- let us have a look at the new possibility of doing hypothesis tests in our 49, e.g., ``z-test for 1 mu''. (Please forgive me not knowing all technical terms in english, but all who know about tests will surely understand what I mean.) I've written programs like these years ago but didn't published them for several reasons. The main reason still is that there is no agree upon which version of the rejection rule(?) is applicable. Suppose we have xbar = 9, sigma = 2, alpha = 0.05, n = 50 and we want to test the exspected value against 10 in a one-sided (?) test, i.e., mu > 10 or mu < 10. Then there are several possibilities to state the reject rule for H_0: mu > 10 against H_1: mu < 10: (0) ``Reject H_0 iff (xbar-mu_0)*sqrt(n)/sigma < -z_(1-alpha).'' (1) ``Reject H_0 iff (mu_0-xbar)*sqrt(n)/sigma > z_(1-alpha).'' (2) ``Reject H_0 iff xbar < mu_0-sigma*z_(1-alpha)/sqrt(n)'' (3) Here z_p denotes the p-quantile of the Standard Normal Distribution. Obviously the built-in programm uses the rejection rule (1) and achieves the correct result. But: *You must not have the choice between different alternative hypothesis* in our case, because if you choose (wrongly!!!) H_1: mu > 10, the rejection rule for H_0 reads: ``Reject H_0 iff (xbar-mu_0)*sqrt(n)/sigma > z_(1-alpha).'' (4) Well, here we have xbar < mu_0, so the left-hand side of (4) is always negative regardless the values of n and sigma an you will *never reject H_0.* In one-sides tests you have no choice choosing H_0 or H_1, the rule is: ``Choose H_1 such that the stated H_1 becomes true if you plug in xbar for mu.'' Ex.: Here we want to test H_0: mu (against) 10 vs. H_1: mu (against) 10. We have: xbar = 9 < 10. so we *must* choose H_1: mu < 10, otherwise we would always accept H_0 ... This choice can be done automatically by the program (as my programs do :). HP, we have Flash ROM! :-)) >please don't bash me if i'm mistaken.] numbers and other objects >can be mixed on the stack. you can pre-program a hot-key like >[left shift] [alpha] [.] that does a vector dot product, which Not user-defined keys, actually, but you can assign programs to the number keys; a program named kbdprgm1 will be executed on Diamond-1 HP beats TI here. >questions for the TI: hey, where is the list concatenation >function? can i pass a function or a function ref as an >argument into a program? [i could pass in say f(x) but that >is treated as an expression in x and i have to assume that >the expression is in x in the program. what if i want it >to be flexible and accept f(z), f(k) or f with other letter?] > You can perform indirection which has the highest priority of any operation. If you put # in front of a function or variable name, it uses the results of the function or contents of the variable, kind of like a pointer. Look up indirection in the manual for details. Don't know about the list concatenation. Overall, having read your post, I'm really starting to miss my 48. ---------------------------------------------------------------------- Well, there's not really a ton of stuff converted over to the 49G format yet. Because the internals were switched around, running any 48G program on a 49G could be disasterous, or at the very least displeasing. > What kind of cable/connector do I need to be looking for to plug it > up to the PC? Will an HP48g* cable work with the HP49g? NNNNNOOOOOO!!!!!! Don't use a 48G cable! Apparently the 49G's ten-pin configuration isn't compatible with the old cable (10-pin was used on the 200LX). It uses the same four pins as the 48G for regular data transfer, but the rest are for the video bus. If you can't find a 49G connectivity kit, don't be surprised. The 38G kit is supposed to work. Although I'm wondering if the 49G kit is going to be released with the Advanced Users Guide and the SDK on CD? ---------------------------------------------------------------------- The contest this year was to improve the following program by reducing the program bytes required. The technique of making the program a string and using OBJ-> to save bytes is *NOT* allowed. This adds so much to the run time - usually many minutes - that it is not practical. The program is a menu-prompting clock-adjust program to allow seconds, minutes, and hours adjustment. It displays the clock to allow the time to be visible and turns the clock off when the adjustment environment is exited. 'CADJ' << -40 SF { { S+ << 8192 CLKADJ >> } { M+ << 491520 CLKADJ >> } { H+ << 29491200 CLKADJ >> } { S- << -8192 CLKADJ >> } { M- << -491520 CLKADJ >> } { H- << -29491200 CLKADJ >> } { EXIT << -40 CF 0 MENU >> } } TMENU >> 282.5 Bytes, # DFFFh. Two approaches are possible. Improve the program using TMENU or use a different user interface that uses the same basic elements: 1. Turn on the clock when the program is run. 2. Provide an identified key for each of the six time adjustments. 3. Provide an EXIT function. 4. Turn off the clock when the program is EXITed or terminated. 3. User code only, no SYSEVALs, LIBEVALs, or machine code allowed. Here are the winning programs in order. 'JGS' << H- M- S- S+ M+ H+ 7 DISP 5 8 * NEG DUP SF WHILE 0 WAIT DUP PICT TYPE < REPEAT 13.6 - 5 3 %T OVER ABS 2 INV - ^ 8192 * SWAP SIGN * CLKADJ END DROP CF >> 147.0 Bytes, # C021h. John Gustaf Stebbins. 'JYA' << 8 -5 * SF S+ M+ H+ S- M- H- 7 DISP IFERR DO 1 5 3 %T DUP DUP * 3 ->LIST 8192 * DUP NEG + 0 WAIT IP 1 ALOG - GET CLKADJ UNTIL 0 END THEN 2 DROPN END 8 -5 * CF >> 'DTM' << -5 8 * SF{:H-:-29491200:H+:29491200:M+:491520: M-:-491520:S-:-8192:S+:8192} OBJ-> TMENU DO DUP IFERR INPUT THEN DROP2 1 ELSE IFERR OBJ-> CLKADJ THEN 1 ELSE 0 END END UNTIL END -5 8 * CF 0 MENU >> 177.0 Bytes, # F9#7h. Detlef Muller Every programmer looked at the others and saw ways their program could be improved. le Program and be less than 147 bytes. Judging these is always difficult. It must be done is a short time, everyone is waiting, and it should be fair and consistent. Did 'DTM' violate the no string, OBJ-> rule? > The Metakernel and CAS splash screens can be seen briefly during boot. If > you press ON at the right time, it will stay on screen for inspection. That's press AND HOLD the ON key at the right time during boot-up. ---------------------------------------------------------------------- I am thinking of buying a 4 MB ram card and also a 128k card from Klotz for my 48GX. I realise that the Klotz cards are not HP approved cards. I was wondering if anyone could help me with the following: Has anyone used the 4 MB Klotz card or other Klotz cards? Are there any special problems with fitting two non-approved cards in the same calculator? I just picked up a 49G two days ago, but it is already showing a considerable number of scratches on the screen (none viewable from head on though). This seems to be a common occurrence doesn't it? I have taken off the sticky plastic cover used for protection during shipping and display, and (fwiw) I am one who takes great care to not scratch screens (my newton, without a screen cover, is four years old and still doesn't have a single scratch on its screen).