HP-129 Subject: Re: hp49g+ Is there a flag to keep all angles positive in polar form? >.... Polar and Rectangular functions that were in the 41 > and used by surveyors for many years. Don't know why it was dropped > from the 48/49 though... >I don't get it, what was dropped in the 48/49? Hi Reth - I was refering to this function which I haven't found in any of the the literature of the 48. It was a very useful tool for us, but it didn't seem to make the cut in the 48. I have found where others have noticed it missing from the functions too. So unless I missed it somewhere ... :) . Scott H. === Subject: Re: hp49g+ Is there a flag to keep all angles positive in polar form? >.... Polar and Rectangular functions that were in the 41 > and used by surveyors for many years. Don't know why it was dropped > from the 48/49 though... >I don't get it, what was dropped in the 48/49? > Hi Reth - I was refering to this function which I haven't found in any of the > the literature of the 48. It was a very useful tool for us, but it didn't seem > to make the cut in the 48. I have found where others have noticed it missing > from the functions too. So unless I missed it somewhere ... :) . > Scott H. Now I see... they are there (P-R and R-P) but in SysRPL; in the 48/49 you wouldn't need them but if you do: RP :: CK2&Dispatch 2REAL %REC>%POL ; @ PR :: CK2&Dispatch 2REAL %>POL%REC ; @ The above will work as the 41 ones; you might modify them slightly: RP :: CK2&Dispatch 2REAL :: %REC>%POL %360 (to get angle ) %MOD (between 0-360) %>HMS (and even in DMS format ; ; @ PR :: CK2&Dispatch 2REAL :: %HMS> (input angle in DMS format) %POL>%REC ; ; @ === Subject: Open Fire remarks X-RFC2646: Format=Flowed; Original I'm testing OpenFire on my new HP49G+ Remark 1: in your example you don't call SET4 and by default = #0 then I don' see anything Remark 2: prefix you code example with !NO CODE !RPL Remark 3: I heard a sound but I don't see a 4colors picture Remark 4: DONEGS does not display the stack. Please call a sysRPL function in DONEGS to refresh the screen. Lilian --- Outgoing mail is certified Virus Free. Checked by AVG anti-virus system (http://www.grisoft.com). === Subject: Re: Open Fire remarks Hello OPEN FIRE fans, i just updated the OPEN FIRE with screen refresh on DONEGS as Lilian sugested and optimized the code a bit so now the whole OPENFIRE library is few bytes maller :-) (it's getting smaller as i'm adding features to it :-) manjo === Subject: Re: Open Fire remarks X-RFC2646: Format=Flowed; Original cn0aao$6mj$1@ls219.htnet.hr... > Hello OPEN FIRE fans, > i just updated the OPEN FIRE with screen refresh on DONEGS as Lilian > sugested > and optimized the code a bit so now the whole OPENFIRE library is few > bytes > maller :-) > (it's getting smaller as i'm adding features to it :-) New feature seems to be a bug :( --- Outgoing mail is certified Virus Free. Checked by AVG anti-virus system (http://www.grisoft.com). === Subject: Re: Open Fire remarks > Remark 1: in your example you don't call SET4 and by default = #0 then I > don' see anything > Remark 2: prefix you code example with !NO CODE !RPL > Remark 3: I heard a sound but I don't see a 4colors picture > Remark 4: DONEGS does not display the stack. Please call a sysRPL function > in DONEGS to refresh the screen. All true Lillian :-) the INITPF demo in fact has no picture -it initiates 4GS, and installs sound compatibility routine i didn't care enough to put in some nice image, so i did the basic initialisation stuff just so you and interested people can se the PRINCIPLE who it's done. I thought it would be enought to present the idea, however i might do another DEMO or TEST with all functionality shown :-) About DONEGS, true -when used with INITPF it may come to the situation where display wouyld not bee refreshed i will add display refresh when screen is back to normal manjo === Subject: Re: Open Fire remarks X-RFC2646: Format=Flowed; Response I've converted the Ice-Cube About (the screen with 2 faces) http://perso.atsat.com/pigallio/Hp48/icecube/ice.htm and... that works with OpenFire!!! But it's not perfet: to be compliant with the old HP48/49 screen you need to copy only 64 lines (and not 80) On HP49G+ maybe keep the header (like the HP49G+ done when we launch games) On HP48: who can test it with the original Ice-Cube game and with the new OpenFire version ? Manjo: send me a email to: pigallio (at) atsat (dot ) com to get this new version... Lilian. 41928ae6$0$10809$8fcfb975@news.wanadoo.fr... > I'm testing OpenFire on my new HP49G+ > Remark 1: in your example you don't call SET4 and by default = #0 then I > don' see anything > Remark 2: prefix you code example with !NO CODE !RPL > Remark 3: I heard a sound but I don't see a 4colors picture > Remark 4: DONEGS does not display the stack. Please call a sysRPL function > in DONEGS to refresh the screen. > Lilian > --- > Outgoing mail is certified Virus Free. > Checked by AVG anti-virus system (http://www.grisoft.com). --- Outgoing mail is certified Virus Free. Checked by AVG anti-virus system (http://www.grisoft.com). === Subject: Re: Open Fire remarks X-RFC2646: Format=Flowed; Response I am sorry : I don't understand how to download on my HP 49G+ Openfire (and any of the application of this web site). I tried but with no success. Could you please help me? jajacom 4192a075$0$10818$8fcfb975@news.wanadoo.fr... > I've converted the Ice-Cube About (the screen with 2 faces) > http://perso.atsat.com/pigallio/Hp48/icecube/ice.htm > and... that works with OpenFire!!! > But it's not perfet: to be compliant with the old HP48/49 screen you need > to copy only 64 lines (and not 80) > On HP49G+ maybe keep the header (like the HP49G+ done when we launch > games) > On HP48: who can test it with the original Ice-Cube game and with the new > OpenFire version ? > Manjo: send me a email to: pigallio (at) atsat (dot ) com to get this new > version... > Lilian. > 41928ae6$0$10809$8fcfb975@news.wanadoo.fr... >> I'm testing OpenFire on my new HP49G+ >> Remark 1: in your example you don't call SET4 and by default = #0 then I >> don' see anything >> Remark 2: prefix you code example with !NO CODE !RPL >> Remark 3: I heard a sound but I don't see a 4colors picture >> Remark 4: DONEGS does not display the stack. Please call a sysRPL >> function in DONEGS to refresh the screen. >> Lilian >> --- >> Outgoing mail is certified Virus Free. >> Checked by AVG anti-virus system (http://www.grisoft.com). > --- > Outgoing mail is certified Virus Free. > Checked by AVG anti-virus system (http://www.grisoft.com). === Subject: Re: Open Fire remarks > I am sorry : I don't understand how to download on my HP 49G+ Openfire (and > any of the application of this web site). I tried but with no success. > Could you please help me? > Best regard Hello jajacom, to sucessfuly install OPEN FIRE or any software distributed in form of library follow these 3 easy steps: they may sound like a special procedure at first but once you get familiar with it will be normal routine procedure Described procedure assumes you have your calculator in RPN mode, to tell you the truth i don't know how to do it in ALGEBARIC mode at all :-) 1. DOWNLOAD ------------------ download applications you're interested in, unzip them and transfer files (in most cases there will be only one file with .LIB extension) to your G+ (G+ only). My prefered method is to copy the files to SD card and load the card in to my G+. But you can use Conn4x as well. 2. INSTALL (attach is the HP term) ------------------------------------- put the file (object) on the stack type the port number (2 for example) and press (or type) STO Example: 2: LIBRARY bla bla (whatever library it is) 1: 2 (you may use other PORT numbers 0, 1 or 2 -port number will normaly be recommended by library autor) STO [enter] press and hold ON pres-and release C, release ON (restart) your G+ to allow library to install (attach) itself. (this is more-less standard procedure to install library to HP calculator -from series 48 up to date) 3.EXECUTE --------------- you can browse to the library using built in FILER, but to make it easier for you thee are commands: for sound check type: SNDCHK for grayscale passians game type: PASSGS in both cases you will have a menu with few options displayed like ABOUT, HELP and the actual program, just press A key or menu item 1 and the program will start. -in case of soundcheck you will hear short, nice music loop, -in case you're trying PASSGS the it will shuffle and deal the cards and game will start :-) since OPEN FIRE is ment to be as a dvelopment extension it put itself to APPS (key) menu it will be listed as OpenFire1.8 near the bottom of the choosebox. Select it, press ok and you will get the list of available commands as menu items. i hope you will find some of the presented software usefull :-) good luck, and... stay tuned :-) manjo === Subject: Re: USB News! X-RFC2646: Format=Flowed; Original > Where did you get the info about where the pins are connected?. Do you > have any calc with the cpu removed or a schematic? > Steve Sousa That info has nothing to do with where the pins go *outside* the CPU, and the statement is true regardless of what HP did with the signals once they get onto their schematic. The simple fact is that the signals *inside* the CPU go to the same I/O pins on the bottom of the CPU. I don't know how to be more clear about this. It has nothing to do with the 'pins' of the USB socket, or the 'pins' of any extra chips that may have to be inserted to get the signal voltages to the correct levels for USB transmission. The same 'pins' exiting the CPU perform the duties of both the USB host and device controllers. This info, along with many other goodies, is available in the product manual of the CPU, referenced elsewhere in this thread, but repeated here for your convenience: http://www.samsung.com/Products/Semiconductor/SystemLSI/MobileSolutions/Mobi leASSP/MobileComputing/S3C2410X/S3C2410X.htm Luke === Subject: Re: Hp 49 keyboard question Three pieces of advice I can give ... 1. Switch off the clock display 2. Turn ON keyclick (if practical) 3. If particular keys are troublesome, hold then down firmly and wiggle them for 5 seconds. > Hello all - I just started here so... I've heard that the 49 has a few > keyboard issues. I recently purchased one but it hasn't arrived yet. > My question is will the keystrokes be sufficiently registered so as to > give consistant results? I'm going to sit for an exam in 6 months and > needed a fast & reliable machine. I'm well informed on the 48 and even > dabble in programming it. But last year when I sat for the exam it > seemed I always had to wait for the 48 to catchup. Hence the need for === Subject: Re: Hp 49 keyboard question >3. If particular keys are troublesome, hold then down firmly and wiggle >them for 5 seconds. Hammer is very helpful. A.L. === Subject: Re: Hp 49 keyboard question > Three pieces of advice I can give ... > 1. Switch off the clock display > 2. Turn ON keyclick (if practical) > 3. If particular keys are troublesome, hold then down firmly and > wiggle them for 5 seconds. The last advice will probably break the plastic hinges off the key, but oh well... The problem is that the above really shouldn't be necessary. Granted, they are workarounds, but only very temporarily. Why include a clock in the product if you aren't able to see the time? Personally I can't accept having key click enabled on my calculator - I'm sure my colleagues wouldn't be impressed by it (and I wouldn't find it relaxing either). People on exams might feel the same way? The wiggle fix destroys the keyboard physically. I could accept points 1. and 2. above for a short period of time, if I knew HP was working on a fix. HP haven't said diddly about this (other than denying there is a problem), so I don't think they are working on it. They are way below TI in my book these days, and have been for years (privately and professionally). Steen === Subject: Re: HP 9G bug? Just have noted that on the HP site at http://www.hp.com/calculators/graphing/9g/ the HP 9G named as financial calculator (on the title). Sergei Frolov http://www.leninburg.com === Subject: Re: I had to try it.... X-Ref: news.freedom2surf.net ~XNS:00000150 >> After exiting the function, the original value of all labels used in the >> function are unaffected. >Yes, this is one of the paradigms of programming. It's called >functional programming. It's becoming popular these days. I wish HP would do this. Imagine if a function would always function the same way given specific inputs, regardless of the existence or setting of variables such as x, and not changing them during execution. Bliss. === Subject: Re: I had to try it.... >> After exiting the function, the original value of all labels used in the >> function are unaffected. >Yes, this is one of the paradigms of programming. It's called >functional programming. It's becoming popular these days. > I wish HP would do this. Imagine if a function would always function the same > way given > specific inputs, regardless of the existence or setting of variables such as > x, and not > changing them during execution. Bliss. The hp48/49 series already does this with local variables. Note the backslash. , precedes expressions for special characters. For example, save << -> x y << -> x y 'x^2+y^2' EVAL >> EVAL >> or, equivalently. << -> x y << x SQ y SQ + >>> as f, then f can be invoked as a function with x and y as local to the body of the function. In particular,'f(3,4)' evaluates to 25, and 'f(A,B) evaluates to 'A^2+B^2', provided neither A nor B is the name of anything. === Subject: hp49g+ weird low battery message, using brand new batts! Hi ppl, I new here and this thing on my hp49g+ is driving me crazy! here is the thing When I turn on my hp49g+ It shows this message Warning: LowBat(P1,P2) ; Alarm and Im using new batts! I changed the batterys 2 times and allways get the same message. any help would be much much appreciated! THANK YOU! === Subject: Re: hp49g+ weird low battery message, using brand new batts! > When I turn on my hp49g+ It shows this message > Warning: > LowBat(P1,P2) > ; Alarm > and Im using new batts! I changed the batterys 2 times and allways > get the same message. The above message don't make any sense on a HP49G+. Your calc may be defective. Which ROM version are you using? Steen === Subject: Re: Need algorithm to solve a question ....... Matteo Miglioranzi a .8ecrit : > About this problem, I'd like to know how to implement an algorithm to > solve this for 40, 50 ,345 ..... n objects. > In particular: > if you have a list of samples of weighs, like for example {1kg 5kg 9kg > 13kg}, > and you have a number like 6, that is 6kg, the algorithm that I need > must tell me if 6kg can be expressed like a linear combination of the > samples in the list. I have just read a demonstration about your problem two days ago. Let two positives integers : a and b. Let assume that gcd(a,b)=1. Then, if n>=(a-1)*(b-1) then you are sure you can write n as a linear combinaison of a and b. You can also prove that (a-1)*(b-1)-1 cannot be written as a linear combinaison of a and b. But I don't know more about the problem. === Subject: Re: Need algorithm to solve a question ....... Hello Matteo If one pan of the scale holds the measured weigh and the other holds the reference or sample weigh then you can use the SUM(of the first N integers) to measure up to (N^2+N)/2 kilograms. So, for a total of 40kg you would need the first 9 discrete units, that is, {1,2,3,4,5,6,7,8,9}. If you need something different then I misunderstood you. :-) !Demeter! === Subject: Re: Need algorithm to solve a question ....... > Hello Matteo > If one pan of the scale holds the measured weigh > and the other holds the reference or sample weigh > then you can use the SUM(of the first N integers) > to measure up to (N^2+N)/2 kilograms. So, for > a total of 40kg you would need the first 9 discrete > units, that is, {1,2,3,4,5,6,7,8,9}. > If you need something different then I misunderstood you. > :-) > !Demeter! Hello everyone. Yes Demeter, but consider that 9 samples of weigh (1kg, 2kg, 3kg,...9kg) are too expensive, and I want to save money buying less samples (consider that every sample has the same price, so 1kg is expensive like a sample of 20kg or 30kg or ..., figure it, even if it is not realistic): I must search the minimum number of samples to verify the weigh of every object. For 40 objects ( the first obj weighs 1kg, the second 2kg,.... the 40¡ weighs 40kg), you can use these samples of weighs (one sample for each weigh): - 1 kg - 3 kg - 6 kg - 10 kg - 20 kg so instead of 9 samples you have 5 samples... 1¡ obj = 1kg 2¡ obj = 3kg - 1kg 3¡ obj = 3kg ... ... 8¡ obj = 6kg + 3kg - 1kg ... ... 25¡ obj = 20kg + 6kg - 1kg ... ... 39¡ obj = 20kg + 10kg + 6kg + 3kg 40¡ .... so for the measure of the 25¡ obj, for example, the scale is balanced and it is the following --------------- | | 25¡ obj sample of 20kg + + sample of 1kg sample of 6kg 5 samples could be the solution but I think you can do all the measures using only 4 samples of weigh.... Matteo === Subject: Re: Need algorithm to solve a question ....... > If you need something different then I misunderstood you. He needs an algorithm that: 1) Given a reference weight AND a list of discrete weights... 2) ...outputs: A) The combination of discrete weights that equal the reference weight or... B) ...a statement that says the reference weight can't be expressed in terms of the discrete weights. Condition: Each discrete weight may only be used once (aka they are real ;-). I guess he looks for more than a brute force approach. It's a rather simple problem, but I can't help but wonder if this is a school assignment (which is why I haven't answered yet)? Steen === Subject: Re: Need algorithm to solve a question ....... >> If you need something different then I misunderstood you. > He needs an algorithm that: > 1) Given a reference weight AND a list of discrete weights... > 2) ...outputs: > A) The combination of discrete weights that equal the reference > weight or... > B) ...a statement that says the reference weight can't be expressed > in terms of the discrete weights. No this is *not* what he is looking for. What he is actually looking for is this: given a set of objects, what is the minimal set of weights that can be used to verify each object's weight. An object can not be used to verify another object's weight, only weights can do that. You are allowed to place weights on either side. This is why you can verify a 2 unit object using 1 and 3 unit weights (2+1 = 3). Quoting Matteo: ``Well, which is the minimum number of weighs that you can use? (you must verify the weigh of every object doing only a measurement).'' > Condition: Each discrete weight may only be used once (aka they are > real ;-). > I guess he looks for more than a brute force approach. It's a rather > simple problem, but I can't help but wonder if this is a school > assignment (which is why I haven't answered yet)? If you think it is so simple, why don't you produce an algorithm? Since he posted here, show it in a language that can be executed on a HP calculator. The program should ask for a set of objects, for example { 1 2 3 4 } and give the smallest set of weights that can be used to verify the weight of each object in the set (this is what you want, right Matteo?). Example: { 2 4 } -> { 1 3 } (just one possible answer). Maybe another condition should be to find the solution with the lightest total weight. It is a tricky thing, and I think you are better off asking this question in comp.programming. The people over there are more likely to come up with an algorithm. -- Daniel 2.6.3-7mdk-i686-up-4GB GNU/Linux 20:08:00 up 4 days, 3:19, 0 users, load average: 0.26, 0.35, 0.39 === Subject: Re: Need algorithm to solve a question ....... > No this is not what he is looking for. Yes, Daniel, it seems it is: In particular: if you have a list of samples of weighs, like for example {1kg 5kg 9kg 13kg}, and you have a number like 6, that is 6kg, the algorithm that I need must tell me if 6kg can be expressed like a linear combination of the samples in the list. > If you think it is so simple, why don't you produce an algorithm? A slight reading disorder? Because I suspect it's a school assignment. > Since he posted here, show it in a language that can be executed on a > HP calculator. If I posted an answer here, there is a more than slight chance that I would do just that. If I tried *really* hard, I might be able to bash a UserRPL program together. I'm almost positive I could succeed on that, but if I get stuck, I'll ask this group for help on syntax, deal? > It is a tricky thing, and I think you are better off asking this > question in comp.programming. The people over there are more likely > to come up with an algorithm. Because people here can't come up with an algorithm? LOL Steen === Subject: Re: Need algorithm to solve a question ....... > Hello. > Sometimes, I like to spend some time to solve a problem using only > machine algorithms, without to solve it first by hand, and the most > difficult thing is to write the algorithm really (but when you can > find the right procedure, it is very pleasant to see the answer given > by the machine). > The problem is the following: > You must verify that 40 objects weigh respectively 1 Kg, 2 Kg, 3Kg, > .... , 40 Kg (every object weighs really 1 kg 2kg etc, and you must > only verify it). > You can use a scale with two scale pans and some samples of weigh. > Well, > which is the minimum number of weighs that you can use? (you must > verify the weigh of every object doing only a measurement). If you have weights in powers of 3 up to a total as large as your heaviest weight, and sufficiently strong balance scales, you can weight any positive integer weight object up to and including the total of your weights. Powers of three: 1=3^0, 3=3^1, 9=3^2, 27 =3^3, 81 =3^4, etc., each being 3 times the previous one. === Subject: Re: Need algorithm to solve a question ....... X-GPG-FINGRPRINT: 6813 0372 183A 4E76 FB00 C858 1FFC A3A7 110E 3327 X-GPG-PUBLIC_KEY: http://n.ethz.ch/student/hasslerf/download/public_key.asc > If you have weights in powers of 3 up to a total as large as your > heaviest weight, and sufficiently strong balance scales, you can weight > any positive integer weight object up to and including the total of your > weights. > Powers of three: 1=3^0, 3=3^1, 9=3^2, 27 =3^3, 81 =3^4, etc., > each being 3 times the previous one. I agree. But are you shure (or can you proof) that your solution is the minimal one? Fabian === Subject: Re: Need algorithm to solve a question ....... > If you have weights in powers of 3 up to a total as large as your > heaviest weight, and sufficiently strong balance scales, you can weight > any positive integer weight object up to and including the total of your > weights. > Powers of three: 1=3^0, 3=3^1, 9=3^2, 27 =3^3, 81 =3^4, etc., > each being 3 times the previous one. > I agree. But are you shure (or can you proof) that your solution is the > minimal one? > Fabian I am sure [sic] that I can prove it to be A minimal one, in terms of numbers of weights. One can show that there is only one combination of weights on the scales for each integer weighing from 1 up to the maximum needed, which proves minimality. I am not sure that I can prove that there is no other, unless one requires tht the same rule must be applicable for all integer weight ranges 1..n. In that case, one need only show that the solution is unique for maximal weighings of 1, 1+3 = 4, 1+3+9 = 12, 1+3+9+27 = 39, etc. === Subject: sRPL question (=...) Hello! I've been trying to find some information on implementing a function that would allow to evaluate expressions depended on a variable (ex. A) in a way it's done with (sigma) E(=,,). Unfortunately I haven't found anything about it neither in manuals nor over the net :( What I am trying to do, is to make MUL(=,,) to be able to counts things like: _9_ | | | | (2+3*i) i=0 Alphajet PS. Sorry for my not always perfect English... === Subject: Re: sRPL question (=...) > What I am trying to do, is to make MUL(=,,) to > be able to counts things like: > _9_ > | | > | | (2+3*i) > i=0 You have it working in UserRPL? You just do the same in SysRPL. If your problem lies with not being able to enter 'MUL(I=0,9,2+3*I)' instead of 'I=0' 9 '2+3*I' MUL, then you'll have to develop your SysRPL program on a PC with Debug4x. That's the simplest way to configure your SysRPL program to behave like a function. Steen === Subject: Re: sRPL question (=...) >>What I am trying to do, is to make MUL(=,,) to >>be able to counts things like: >>_9_ >>| | >>| | (2+3*i) >>i=0 > You have it working in UserRPL? You just do the same in SysRPL. If your > problem lies with not being able to enter 'MUL(I=0,9,2+3*I)' instead of > 'I=0' 9 '2+3*I' MUL, then you'll have to develop your SysRPL program on > a PC with Debug4x. That's the simplest way to configure your SysRPL > program to behave like a function. > Steen Hmm, I don't know if I get it right. I know how to write it, I mean the loop etc. The problem is, that I don't know how to set up my procedure (written in sys-RPL) to take a function and a VAR as parameters, and then evaluate them in a loop. When you type HELPWITH RECURSE for instance, you get sth like: RECURSE(,,,) where: is the A..Z variable which is used for counting; is a function depended on ; is the initial value of and the 2nd is its final value; And this's exactly what I want to get when I type HELPWITH MUL: MUL(,,,); Alphajet === Subject: Re: Cube root -1 > Is there a way to get the hp 49g+ to return the real cube root of -1 > instead of the imaginary one? Use approx numbers - ie -1. instead of the integer -1 (notice the radix). Steen === Subject: Re: Cube root -1 > Use approx numbers - ie -1. instead of the integer -1 (notice the > radix). Just one more thing: The method Werner points out works for integers -> symbolic results too, in case you're interested in the exact answer. Example (to show the difference): XROOT(3,-5.) -> -1.70997594668 SIGN(-5)*XROOT(3,ABS(-5)) -> '-XROOT(3,5)' Steen === Subject: Re: Cube root -1 I was wondering if there was a built in function that would still calculate roots of any number, but return the real root of -1, not the imaginary one. Kind of frustrating to have a simple integral return an imaginary number because it is using the imaginary root of -1, not the real one. > Is there a way to get the hp 49g+ to return the real cube root of -1 > instead of the imaginary one? === Subject: Re: Cube root -1 Bonjour, > I was wondering if there was a built in function that would still > calculate roots of any number, but return the real root of -1, not the > imaginary one. Kind of frustrating to have a simple integral return > an imaginary number because it is using the imaginary root of -1, not > the real one. Someone asked you this question: what is the real cube root of -1? Someone else proposed to solve for X in the equation X^3=-1. That's not a valid answer. When I was in school, I learned that a^b was equal to e^(b.ln(a)) with e the base of the natural logarithm. It's easy from this expression to understand that -1^(1/3) has no real answer, since ln(-1) has no meaning in the real numbers. > Is there a way to get the hp 49g+ to return the real cube root of -1 > instead of the imaginary one? -- Erwann ABALEA - RSA PGP Key ID: 0x2D0EABD5 ----- Je pense qu'un lecteur assidu se reconnaitra. James, si tu veux que je r.8eponde .88 ton message, m.90me pour te dire que je n'ai pas envie de te r.8epondre, laisse-moi une adresse de retour ! -+-DM in : Guide du Neuneu d'Usenet : Je t'aime moi non plus -+-