HP-12 > I recently started using the 49G+ (I used to use the 12c) and I'm having > difficulty working in the Statistics mode. I can get the data entered in > the Matrix editor, but I can't figure out how to store it in the statistics > (?DAT) variable. Can some one help me understand the right keystrokes? > > > Mark Gannon > mark at truenorth dot nu Another thing you might try, esp. if you are going to re-use the matrix -- just give it any variable name which makes sense to you. Then when you go into stat mode, select choose from the menu and that will give you a list of the arrays and matrices in the directory you're in. If you want to choose a matrix in another directory, press tree which will give you a list of the directories on your calculator, and you can choose from there. John ==== > and what about rom upgrades? > would i be able to upgrade roms with my linux box? Well, I don't know whether the tool for hp49g might be adapted to send hp49g+ roms. But anyway, it's hardwarly possible, only the software has to be written. Samuel Thibault ==== I would like to write some software for my brand new hp 49g+. The last hp much has changed. Can some kind person please point me in the direction of tools, books and/or tutorials that can help me to master System RPL? Tanya ==== > I would like to write some software for my brand new hp 49g+. The last hp > much has changed. > Can some kind person please point me in the direction of tools, books and/or > tutorials that can help me to master System RPL? I think the best SysRPL tutorial is Eduardo Kalinowski's one : http://www.hpcalc.org/details.php?id=1758 Yoann. ==== > > I would like to write some software for my brand new hp 49g+. The last hp and > > much has changed. > > Can some kind person please point me in the direction of tools, books and/or > > tutorials that can help me to master System RPL? > > I think the best SysRPL tutorial is Eduardo Kalinowski's one : > http://www.hpcalc.org/details.php?id=1758 > I found the 2nd edition at http://www.hpcalc.org/hp49/docs/programming/ Its huge! Tanya ==== >> I would like to write some software for my brand new hp 49g+. The last hp >> much has changed. >> Can some kind person please point me in the direction of tools, books and/or >> tutorials that can help me to master System RPL? > I think the best SysRPL tutorial is Eduardo Kalinowski's one : > http://www.hpcalc.org/details.php?id=1758 there is a 2nd edition from eduardo with carsten dominik, somewhere on hp homepage mike ==== > > This means that all programs written for HP48 or HP49G and changing > > the contrast WILL NOT work on HP49g+ ! > > Is this kind of application common for the HP48/49G? On the 68k, it's > quite uncommon for a program to change the contrast. On the 68k doesn't mean anything to me. On a TI calc, (which uses 68k), it might be useful to control contrast, to make a nice fade in / out in a game, for instance. Samuel ==== In message <7d896fcd.0310281216.181d00c3@posting.google.com>, Bob >I work for a company that uses the HP 48G+ (G & GX) as a programmable [Snip] The short answer is that the 48GII is not really a direct upgrade of the 48. It is actually an alternate upgrade of the 49G. Consequently, all of the software incompatibilities that you note you would have encountered had you tried to switch from the 48 to the 49G previously. -- Bruce Horrocks Surrey England ==== > In message <7d896fcd.0310281216.181d00c3@posting.google.com>, Bob > >I work for a company that uses the HP 48G+ (G & GX) as a programmable > [Snip] > > The short answer is that the 48GII is not really a direct upgrade of the > 48. It is actually an alternate upgrade of the 49G. > > Consequently, all of the software incompatibilities that you note you > would have encountered had you tried to switch from the 48 to the 49G > previously. > Any chance that the 48GII then can transfer data with the HP49G/HP49G+ without much trouble, or is it far too different? It seems that port wise, it at least connects nicely to both since it has both ir and serial available. Does anyone know if the 48GII or 49G+ can be used to control ir receiving devices? In my math class, there's an overhead projector with the remote always locked away in a desk which my teacher doesn't know the combination to since its too much trouble to get the remote to usually work anyways with interference to the device in the room messing up remote reception. I've got my HP49G+ on the way, and thought it would be nice to show her that she can turn it's ir port into a remote for things like the projector if that's an option. If so, any ideas of an average of how far it can be from the device it talks to? I hope this port can turn out to be quite fun yet after all. Ed Sutton ==== The specs indicate that the 48GII should be able to transfer data with the 49s using the IR port. You would not be able to do it using the serial ports since the 48GII uses an RS-232 port while the 49s use USB. The IR signal strength of the 48GII has been greately reduced. I can't get it to work with HP's own IR printer (82240B) outside of a few inches. Because of this, I doubt that it would work very well in this application. > Any chance that the 48GII then can transfer data with the > HP49G/HP49G+ without much trouble, or is it far too different? It > seems that port wise, it at least connects nicely to both since it has > both ir and serial available. > Does anyone know if the 48GII or 49G+ can be used to control ir > receiving devices? In my math class, there's an overhead projector > with the remote always locked away in a desk which my teacher doesn't > know the combination to since its too much trouble to get the remote > to usually work anyways with interference to the device in the room > messing up remote reception. I've got my HP49G+ on the way, and > thought it would be nice to show her that she can turn it's ir port > into a remote for things like the projector if that's an option. If > so, any ideas of an average of how far it can be from the device it > talks to? > I hope this port can turn out to be quite fun yet after all. > Ed Sutton ==== > i thought this was comp.sys.hp48! shouldn't we be promoting the > purchase of hp calculators, not telling me to use my head and pencil > and paper!?! > > (just kidding). Well, also buy an HP calculator and learn how to use it well (that means in RPN mode). For that matter, buy a whole bunch of them! Personally, I took to the 28S like a duckling takes to water. What a wonderful revelation after my previous calculators! A calculator that actually works like I think! But then too, I've heard lots of hints that my mind might be just a tad out of the ordinary. But I fear that if you've been indoctrinated into using one of those arcane algebraic input calculators, the 49g+ may seem a bit backwards at first, and may take a little getting used to. And if do you have previous experience with RPN style calculators or you just naturally take to it right off the bat, you may be having so much fun with the new calculator that you'll forget that you have to study (and eat and sleep, for that matter). > anyway, i've decided to tough it out, and wait for the next version of > Power48 (http://power48.mobilevoodoo.com), which does not currently > work on my particular pda model, but there is an upgrade on the way. Oh, so you do have some experience with RPL calculators, or at least emulators? > i found a place in phoenix, called Holman's, which has one 49g in > stock, but they want $199 for it. i told them about the new 49g+, > which happend to be in their computer for $139, so the manager was > nice and said i could buy the 49g for $149, if i wanted it. i couldn't > do it, knowing the 49g+ is 10-12x faster. i couldn't do it even though > i could use it on my test tonight. > > the manager also said he was going to transfer (2) 49g+'s (from his > albuquerque store), in case i wanted one on wednesday. i'm going to > get one, because they're good looking (except for that crazy alpha > button. not sure what color that is), and might be nice to have when > my teachers figure out how much info i have on my pda (notes, pdfs, > entire books, etc). > > i just talked to the manager at Holman's (http://www.holmans.com) > again. he said there are (3) 49g+'s in albuquerque, and (2) will be in > phoenix tomorrow (i have one reserved). But still, I don't know that I'd care to take any test with a new calculator that I wasn't yet familiar with, unless it were for very basic arithmetic. > wish me luck on my test, Good luck on your test. -- James ==== > (2) will be in > phoenix tomorrow (i have one reserved). > > wish me luck on my test, > greenchile505 Good luck. ==== I've been lurking here for quite a while. This is definitely a nice place. Anyway, it's nice to know that the 49G+ is coming to town, even if you two did get one that's rightfully mine :) I'll probably just wait until Fry's gets them anyway. They got the 12c Platinum in just a little ago. I'm sure the 49G+ will be on its way too. > > >> (2) will be in >> phoenix tomorrow (i have one reserved). >> >> wish me luck on my test, >> greenchile505 > > Good luck. > > ==== > > See: > http://groups.google.com/groups?threadm=L8zmb.1692%24Px2.595%40newsread4.news .pas.earthlink.net > > HP is aware of the problem, and I'm sure that they're looking for a > solution. For now, I hope that you have an SD card and reader to do any > transfers. The HP people seems to be aware of the problem, at least they say they contacted the developers to make them know it doesn't work. Yoann. ==== > And how do you find this bug & solution ? > On Hp49+ or on an emulator ? I tested the first lines of Navigator, being connected. And it appeared that the first subprogram called made the HP crash. It's the 'Connected?' program, and I immediately saw that I used an unsupported entry point, that began with a 1xxxx... while supported entry points often have upper values. And of course, I found it on a real HP49g+ (manipulated with rough finger-pressing, though) ! Who do you take me for ? :-) Yoann. P.S. Would you be kind enough to test Navigator 0.07f 49g+ version on your 49g+ if you have one ? ==== > And of course, I found it on a real HP49g+ (manipulated with rough > finger-pressing, though) ! Who do you take me for ? :-) Had it been on an emulator, many many people would have been really much interested... Samuel Thibault ==== > Had it been on an emulator, many many people would have been really much > interested... Anyway, if it works on the calculator, it should work on an emulator, souldn't it ? Yoann. ==== > > Had it been on an emulator, many many people would have been really much > > interested... > > Anyway, if it works on the calculator, it should work on an emulator, souldn't it ? Of course, but the eventual existence of an emulator would have been really interesting. ==== As funny as it sounds, its true. My puppy is going through his adoloscent period, and he decided that the HP49g+ USB cable would be a good thing to chew on. Its now ruined and I want to know where I can buy a new one. ANy ideas I've looked at the HP site and nothing and ==== > Its now ruined and I want to know where I can buy a new one. You're in luck: it's a standard USB cable with a mini plug! Just take your dead cable to any electronics store, visually match the plug with any of the USB cables they have, and buy it. Get the cheapest one they have, because the 49g+ is USB1, not USB2. -Joe- ==== > As funny as it sounds, its true. My puppy is going through his > adoloscent period, and he decided that the HP49g+ USB cable would be a > good thing to chew on. Its now ruined and I want to know where I can > buy a new one. ANy ideas I've looked at the HP site and nothing and Any USB to microUSB cable would do, quick google revealed -> http://www.bhphotovideo.com/bnh/controller/home?A=ShowProduct&Q=&O=&sku=2851 92 ps to whom it may be interesting, yes me got meself one of those too :) -- Mario Mikocevic (Mozgy) mozgy at hinet dot hr It's never too late to have a good childhood! The older you are, the better the toys! My favourite FUBAR ... ==== If you can salvage the specialized 4 pin HP49G+ end of the cable, you can try to rewire it. But first I'd make sure the the pinouts, etc. are obtainable from HP or someone knowledgeable here about the signal and ground connections in such a new machine. You don't want to fry it. Once you have the wiring paths, then it's safe to at least temporarily have a home-kluged cable that works and then get a new one when they become available. EGL > As funny as it sounds, its true. My puppy is going through his > adoloscent period, and he decided that the HP49g+ USB cable would be a > good thing to chew on. Its now ruined and I want to know where I can > buy a new one. ANy ideas I've looked at the HP site and nothing and > ==== > If you can salvage the specialized 4 pin HP49G+ end of the cable, you > can try to rewire it. But first I'd make sure the the pinouts, etc. are > obtainable from HP or someone knowledgeable here about the signal and > ground connections in such a new machine. You don't want to fry it. Actually, the HP49G+ cable is an absolutly STANDARD Mini USB cable. The connector on the calcaultor is defined in the USB norm as mini USB and was devised especially for small devices such as PDA and other... ==== LOL! I AM getting old! I was so used to the HP-48 4-pin connector being different- I heard that the closest thing and it wasn't exactly the same was the 4-pin connector you find at the end of something like a floppy to a motherboard! I am very happy to hear that it is standard item that stores carry! > > >>If you can salvage the specialized 4 pin HP49G+ end of the cable, you >>can try to rewire it. But first I'd make sure the the pinouts, etc. are >>obtainable from HP or someone knowledgeable here about the signal and >>ground connections in such a new machine. You don't want to fry it. >> > > Actually, the HP49G+ cable is an absolutly STANDARD Mini USB cable. > The connector on the calcaultor is defined in the USB norm as mini USB and > was devised especially for small devices such as PDA and other... > > > > ==== Any digital camera USB cable does the job - doesn't have to be HP. I haven't used mine yet, just been using the camera one. Works perfectly. > If you can salvage the specialized 4 pin HP49G+ end of the cable, you > can try to rewire it. But first I'd make sure the the pinouts, etc. are > obtainable from HP or someone knowledgeable here about the signal and > ground connections in such a new machine. You don't want to fry it. > > Once you have the wiring paths, then it's safe to at least temporarily > have a home-kluged cable that works and then get a new one when they > become available. > > EGL > > > > As funny as it sounds, its true. My puppy is going through his > > adoloscent period, and he decided that the HP49g+ USB cable would be a > > good thing to chew on. Its now ruined and I want to know where I can > > buy a new one. ANy ideas I've looked at the HP site and nothing and > > > ==== Ditto here. I've been using the same USB Cable from my Cannon Digital camera for the hp49g+, the camera, the SD Imagemate, my wife's digital camera, (different brand), a Roland SC D70 sound module, blood pressure monitor, hmmm ... I think that may be all. I haven't got any drivers for the dog .... but if he ate my cable, I probably would install my own driver in him! I just plug/unplug the devices and they look up their respective drivers all installed in WinXP. ALL using the SAME single USB cable. USB cables are as generic as toilet paper. >Any digital camera USB cable does the job - doesn't have to be HP. I haven't >used mine yet, just been using the camera one. Works perfectly. > >> If you can salvage the specialized 4 pin HP49G+ end of the cable, you >> can try to rewire it. But first I'd make sure the the pinouts, etc. are >> obtainable from HP or someone knowledgeable here about the signal and >> ground connections in such a new machine. You don't want to fry it. >> >> Once you have the wiring paths, then it's safe to at least temporarily >> have a home-kluged cable that works and then get a new one when they >> become available. >> >> EGL >> >> >> > As funny as it sounds, its true. My puppy is going through his >> > adoloscent period, and he decided that the HP49g+ USB cable would be a >> > good thing to chew on. Its now ruined and I want to know where I can >> > buy a new one. ANy ideas I've looked at the HP site and nothing and >> > >> > ==== > If you can salvage the specialized 4 pin HP49G+ end of the cable, you > can try to rewire it. But first I'd make sure the the pinouts, etc. are > obtainable from HP or someone knowledgeable here about the signal and > ground connections in such a new machine. You don't want to fry it. > > Once you have the wiring paths, then it's safe to at least temporarily > have a home-kluged cable that works and then get a new one when they > become available. Why would you bother? It's an industry standard cable. You should be able to get one from most electronics shops. Al ==== I had a puppy chew on my HP48G once - it was lying on the floor in its soft case and the puppy found it. Fortunately the only damage there was cosmetic - a few tooth marks beside the screen and a tear on the case... screen didn't break. Unfortunately, I can't help you with a place to buy a 49G+ USB cable, but your thread just reminded me of that instance... when people ask what happened to my 48, I tell them my dog ate it :) > As funny as it sounds, its true. My puppy is going through his > adoloscent period, and he decided that the HP49g+ USB cable would be a > good thing to chew on. Its now ruined and I want to know where I can > buy a new one. ANy ideas I've looked at the HP site and nothing and ==== Anyone know of a TI-86 newsgroup? ==== > Anyone know of a TI-86 newsgroup? bit.listserv.calc-ti has topics about all TI calculators -- Tom Lake ==== > Anyone know of a TI-86 newsgroup? I think the closest is bit.listserv.calc-ti , but thats pretty much dead. There are more forums on ticalc.org. Keep in mind that ticalc is mostly posts like 'how do I play games' etc - maybe thats what you are after, I don't know. cheers, Al ==== > entirely undeserved > KW > > > >This post is spam itself. Don't bother to reply. > > I apologize, I made a mistake. Even much worse off-topics that this don't deserve being called spam. I am so fed up of spam that tend to see spammers everywhere. I solutions to fight spam, or (b) methods for becoming yourself a spammer. But I judged too quickly in this case. I'm sorry, RaM. ==== that lead me to subscribe to the grc newsgroup to begin with. Of coarse after I got there and also read all of the advice on Steve Gibson's website (grc.com) that I figured out I was REALLY wide open to problems. I have since made major modifications to how I browse, what software I run, and how my computer is configured.

I have listened here for several years even though I personally am not a HP calculator operator. My daughter has a HP49G, but since she switched to an english major .... :-)  I am also a friend of Tim Wessman, and have kept him posted on some of the high points while he has been gone. In general this is just a great group and I just enjoy the conversations even if they are over my head.

I resisted posting a warning here for quite a while, but a survey about a week ago showed some people are still getting 200 to 300 swen obvious that they were not, and the swen problem was not improving, I figured I would post warning and wait to get flamed.

Since you are a fellow spam hater, I should point out the reason for infected machines as spam relays to hide their identity, talk about your slime balls!! What ends up happening is the poor sucker who doesn't know he is infected suddenly gets canceled by his ISP for being a spam generator. This virus was meant to make somebody else money at your expense.

             kw


entirely undeserved KW 
This post is spam itself. Don't bother to reply.
 
      

I apologize, I made a mistake. Even much worse off-topics that this
don't deserve being called spam.
I am so fed up of spam that tend to see spammers everywhere. I
solutions to fight spam, or (b) methods for becoming yourself a
spammer.
But I judged too quickly in this case. I'm sorry,
RaM.


==== > Pi can be easily found to 24 decimal places on HP calculators, by making > use of sin(x) ~= x for small x (in radians). I am using an HP-32SII, > which provides 12 significant digits: > > Enter pi. See 3.14159265359. > 3.14159265358 has to be slightly less than pi. > > sin(3.14159265358) = 9.79323846264e-12 > pi ~= 3.14159265358 9 79323846264 (24 digits) > > Trying the same technique on a TI-83+SE (10 significant digits, so I > would hope of easily getting 20) shows a problem with sin(x): > > Enter pi. See 3.141592654. So far so good. > 3.141592653 has to be slightly less than pi. > > sin(3.141592653) = 5.898e-10. PROBLEM. The answer has dropped to 4 > significant digits instead of 10, so the TI misses 6 important > significant digits for arguments close to pi. The HP calculator > correctly gives: > > sin(3.141592653) ~= 5.89793238463e-10, or > pi ~= 3.141592653 5 89793238463 > > I understand why the TI does this, and it isn't optimum. > > Rick > > > can seem to check the results, but they like to round too much, which makes the truth hard to see. =( Something seems flakey somewhere in this to me though. if you need sin(x)~=x to have a value very close to x, (taking note that 3.14159265358 is less than pi), is there a way to use 3.14159265359 to find the extra pi digits since it is closer to pi (and also less than pi), while not being equal to pi? By the way, HP49G returned the same results as your HP-32SII. Ed Sutton ==== > > can seem to check the results, but they like to round too much, which > makes the truth hard to see. =( > Something seems flakey somewhere in this to me though. if you need > sin(x)~=x to have a value very close to x, (taking note that > 3.14159265358 is less than pi), is there a way to use 3.14159265359 to > find the extra pi digits since it is closer to pi (and also less than > pi), while not being equal to pi? > By the way, HP49G returned the same results as your HP-32SII. > Ed Sutton Of course, it works but 3.14159265359 is very stlightly more than pi so you get sin(3.14159265359)=-2.0676154E-13 Which is by how much it is slightly bigger. 1+(-.2067154)=.7932846 Which are the numbers you are after. Arnaud ==== > > > > can seem to check the results, but they like to round too much, which > > makes the truth hard to see. =( > > Something seems flakey somewhere in this to me though. if you need > > sin(x)~=x to have a value very close to x, (taking note that > > 3.14159265358 is less than pi), is there a way to use 3.14159265359 to > > find the extra pi digits since it is closer to pi (and also less than > > pi), while not being equal to pi? > > By the way, HP49G returned the same results as your HP-32SII. > > Ed Sutton > > > Of course, it works but 3.14159265359 is very stlightly more than pi so you get > > sin(3.14159265359)=-2.0676154E-13 Which is by how much it is slightly bigger. > > 1+(-.2067154)=.7932846 > > Which are the numbers you are after. > > Arnaud Hmm, seems I managed to screw up my decimal place when checking my last digits quite a few times yesterday then while looking at it. That did answer my next question of if it was still usable though too so thank you very much. Now I wonder what other fun things there are to do with my calc. =) Ed Sutton ==== > can seem to check the results, but they like to round too much, which > makes the truth hard to see. =( > Something seems flakey somewhere in this to me though. if you need > sin(x)~=x to have a value very close to x, (taking note that > 3.14159265358 is less than pi), is there a way to use 3.14159265359 to > find the extra pi digits since it is closer to pi (and also less than > pi), while not being equal to pi? The Pi key (2nd ^) gives the value of Pi to as many places as the calculator can use anyway so what's point in computing it? -- Tom Lake Experience keeps a dear school but fools will learn in no other - Poor Richard's Almanack ==== it's a standard mini-usb cable...your local computer supplier should have one. Failing that, try a stockist of digital cameras or PDAs. On a related note, I used to keep fancy rats and had a fair bit of trouble with them chewing cables...a simple deterrent is to wipe at-risk cables (especially mains electrical cables!) with tabasco sauce...they only do it once after that! JasonG ==== Sorry, this was supposed to go under 'My Dog Ate It!' > > it's a standard mini-usb cable...your local computer supplier should > have one. Failing that, try a stockist of digital cameras or PDAs. > > On a related note, I used to keep fancy rats and had a fair bit of > trouble with them chewing cables...a simple deterrent is to wipe at-risk > cables (especially mains electrical cables!) with tabasco sauce...they > only do it once after that! > > > JasonG > ==== I just received some brand new HP32SII service replacements. They come with slipcase alone. Please remove NO SPAM ==== send info and pictures to: wilbertindall@berkeley.k12.sc.us > I just received some brand new HP32SII service replacements. They come > with slipcase alone. > Please remove NO SPAM ==== Given the documentation for the new HP--49G+ I am still lacking the AUR that was optional to the 48 G series. Anybody know if there will be one? JYA? Otherwise it«s a fine machine, but my has a (defect of course, but serious?) that if I dr.97p the top end ie display end a few cm it kind of warmstarts«, the display goes blank but the stored items remain. SD card intact as well. BR Matti ==== > > Given the documentation for the new HP--49G+ I am still lacking the > AUR that was optional to the 48 G series. > Anybody know if there will be one? JYA? Well, check out page C-13 of the 49g+ user's guide. It seems that you're expected to use the 48G series documentation if you want good references for the non-CAS commands. -- James ==== > Given the documentation for the new HP--49G+ I am still lacking the > AUR that was optional to the 48 G series. > Anybody know if there will be one? JYA? > Otherwise it«s a fine machine, but my has a (defect of course, but > serious?) that if I dr.97p the top end ie display end a few cm it kind > of warmstarts«, the display goes blank but the stored items remain. SD > card intact as well. If you'll settle for the AUG for the 49G in PDF format, it's available at http://www.hpcalc.org/hp49/docs/misc/ most of the commands of the 49 g+ are in there. Tom Lake ==== I wonder if and ifso how one might get the 48 style solver menu by storing the equation using STEQ and then typing ..... MENU to get to the 48 style solver menu, if there is one for the 49G+. It«s too deeply rooted in my mind having used it that way the last 13 years..., but perhaps I must change to the odd new solvesys type interface. Matti ==== example: Y=SIN(X) 'EQ' STO 30 MENU and there you are! RSHIFT followed by menu-key recalls the value stored while LSHIFT followed by menu-key solves for the corresponding variable. The menu-key itself stores the stack value under the corresponding variable name. Happy computing with the new test-machine :-) !Demeter! ==== > > example: > > Y=SIN(X) > 'EQ' > STO > 30 > MENU > /matti ==== Dear fellow user of the new DUT (device under test). The 49g+ has indeed a couple of bugs (those that came under my attention that is) that hopefully will be fixed with the new ROM revision. First, the machine will some times freeze especially during turn-on. You have to insert a paperclip underneath to make it recover. Second, sometimes during turn-on again, it will execute the alarms (while it should not) and the rescheduling of at least the first one will be screwed up. !Demeter! ==== > > Dear fellow user of the new DUT (device under test). > The 49g+ has indeed a couple of bugs (those that came > under my attention that is) that hopefully will be > fixed with the new ROM revision. > First, the machine will some times freeze especially > during turn-on. You have to insert a paperclip underneath > to make it recover. > Second, sometimes during turn-on again, it will execute > the alarms (while it should not) and the rescheduling > of at least the first one will be screwed up. > > !Demeter! calculator -- some place in Oregon. They should have a central reporting site for bugs, don't you think? Do you know when the next ROM revision will come out? And do they specifically list the bugs they have fixed? Lastly, I didn't know I had signed up as a beta tester! How do I get my beta tester doscount? ;} -- John ==== Aloha! I«ve been thinking about how to use this new gadget, the 49g+. As I see it the USERRPL programs works, but what other programs work with it? (Yep, never touched a 49g only good old 48«s 41«s 28«s 15«s and so on.) This is a seriously asked question without any intention to hurt anybody nor hp. Adios, Matti ==== > I«ve been thinking about how to use this new gadget, the 49g+. > As I see it the USERRPL programs works, but what other programs work with it? > (Yep, never touched a 49g only good old 48«s 41«s 28«s 15«s and so on.) The 49g+ emulates the Saturn assembly, so at the condition to change the adresses of entry points, 48G(X) programs can be converted to that calculator. Compatibility is greater for 49G programs, of course, since the 49G has a very close rom to the 49g+. Good bye, Yoann. ==== that's my question. I think that 80k is not too much. ==== I work for a company that uses the HP 48G+ (G & GX) as a programmable controller for part of a small system. We have been using the HP calculators for almost 10 years now by preloading our software into the units before sending them to our customers. Needless to say, we were quite concerned when HP announced that they would be discontinuing the 48G+ and replacing it with the 48GII. We received our first batch of 48GII last week and have started evaluating the units. Of course, our worst fears were realized when the existing software would neither load nor run on the new platform. I decided to start this thread in an attempt to document some of the compatibility issues we have found with the new units. Please feel free to add to the list to give other developers and programmers a head start when porting their software to the 48GII. The Good: 1. New memory backup design. Lithium cell is more reliable and longer lasting than capacitor design of the 48G+. 2. New serial connector. More rugged design should be less prone to bent pins. 3. Serial cable included. Although the 48GII is slightly more expensive than the 48G+, they are now throwing in a $20 cable. 4. Screen appears to have slightly better contrast than the 48G+. 5. Fast. Appears to be approximately 2x the speed of the 48G+. The Bad: 1. Speaker volume is softer in the 48GII. Although fine in a quiet environment, it is very hard to hear when using the unit in real world applications. Should have been made louder than the 48G+ or (even better) variable volume. 2. IR range has been reduced. IR communications are now limited to only a few inches (from around 18 in the 48G+). 3. Analog clock ö this has to be a joke? If you want a good laugh, give it a try. 4. Intermittent Menu flashes are distracting. 5. Manual errors/omissions. a. Chapter 26 ö This appears to be the documentation for a different product (probably the 49?). b. No documentation on flag functionality ö Approximately 60 system flags have been added without any documentation on functionality. The Ugly: 1. The new fonts. Very hard to read in small sizes and in some instances do not reflect the original character at all. Try turning the # character into a graphic object using GROB with size 1. You will get something that looks like a not-equal sign. Some of the numeric characters fonts are also very difficult to read. 2. Keys. Very bad. Although I cannot remember a single missed key stroke on our baseline 48G in almost 10 years of use, missed key strokes occur on the 48GII on such a regular basis it would be hard to use as a calculator (we use it as a programmable controller so it is not much of an issue for us). I have tried 5 different units ö all exhibit the problem. 3. Compatibility Issues. Why, why, why? This unit should run 48G code straight out of the box - without modification ö PERIOD. Think -- very large installed software base. I'm all for progress and added functionality but the baseline configuration (from the factory) should be compatible with the 48G code. All non-compatible added functionality should be enabled via system flags. Here are some of the issues I have run into so far: a. IR compatibility. Basically it doesn't appear as though you will be communicating with any existing 48G, 48G+ or 48GX. This is odd since they did decide to keep the existing IR transmission protocol which allows it to continue to communicate with the 82240B printer. The IRDA is fine, but the legacy IR receiving protocol should have been included also for compatibility sake. b. Algebraic mode is factory default (not PRN). This, of course, is just plain backwards for compatibility. c. Printer errors ö probably due to an increase in IR output speed due to the faster processor? Could be fixed by adding printer delay using the DELAY command? c. Key mapping ö if you used many KEY command in your code get ready for some work. Most of the key locations have changed. d. Real numbers are now displayed with decimal points when operating in standard numeric mode. Although still logically correct, this affects precisely formatted field output to the screen or printer. e. The default font size for Choose boxes has been changed to MINI (from normal size on the 48G+). f. Symbolic representation rules have changed. For example, the factory result for the program <<1 TAN >> on the 48G (factory default) was 1.7455E-2. Running the same program with factory default settings on the 48GII returns TAN(1). g. RS-232 communication problems ö Cannot communicate with several RS-232 devices that worked fine with the 48G+. h. Cannot seem to load library routines written for the 48G+ onto the 48GII from a PC. ==== I just remembered seing a bunch of empty SD Card packages hanging from a shelf at Sam's Club. So, go to Sam's: you do not have to fight the plastic there. Toby > It was a subcontract, but still... > > The last time I tried to open one of those I cut myself with the plastic. I > still have part of it, inside my HP 48GX soft case, for extra protection. > > Toby > > > Wal-Mart. > > > > The SD Cards they sell are exactly the same as the ones sold by > everyone > > else. They are all encased by that extremely hard, see-through plastic > which > > it takes either new scissors or a very sharp knife to open. Wal-Mart is > your > > best bet. That's where I bought mine. > > > > Unless, of course, you might be boycotting Wal-Mart because they > > place. > > > > > > : for the calculator? any good ideas/tips? > > > > > > > ==== > I just remembered seing a bunch of empty SD Card packages hanging from a > shelf at Sam's Club. So, go to Sam's: you do not have to fight the plastic > there. > > > Toby > > > It was a subcontract, but still... > > > > The last time I tried to open one of those I cut myself with the plastic. > I > > still have part of it, inside my HP 48GX soft case, for extra protection. > > > > Toby > > > > > Wal-Mart. > > > > > > The SD Cards they sell are exactly the same as the ones sold by > everyone > > > else. They are all encased by that extremely hard, see-through plastic > which > > > it takes either new scissors or a very sharp knife to open. Wal-Mart is > your > > > best bet. That's where I bought mine. > > > > > > Unless, of course, you might be boycotting Wal-Mart because they > > > place. > > > > > > > > > : for the calculator? any good ideas/tips? > > > > > > > > > > > > I like hearing the places, but what about the prices? Ed Sutton ==== On page 549 of the HP48SX Owner's Manual, Volume II there is a nice little recursive program, called FIB1, for calculating the nth Fibonacci number: << ->n 'IFTE(n<=1,n,FIB1(n-1)+FIN1(n-2)' >> 13 FIB1 gives 233. This takes 34 seconds on the HP48SX. In approx mode: On the HP49G it takes 309 seconds!!! On the hp49g+ it takes 71 seconds. Does anyone know why this is so much faster on the HP48SX? Is there a way to speed it up on the 49G/g+ ? - Tony ==== > On page 549 of the HP48SX Owner's Manual, Volume II there is > a nice little recursive program, called FIB1, for calculating > the nth Fibonacci number: > > << ->n 'IFTE(n<=1,n,FIB1(n-1)+FIN1(n-2)' >> > Does anyone know why this is so much faster on the HP48SX? > Is there a way to speed it up on the 49G/g+ ? This may not be precisely what you're looking for, but a non-recursive way to calculate the nth Fibonacci number is: g = the golden ratio = (1+sqrt(5))/2 fib[k] = round(g^k/sqrt(5)), where round means round to the nearest integer (It can easily be shown that this gives the correct result for all k >= 0). Given the 12-digit precision of the HP48 series floating point, a straight-forward implementation in RPL gives the correct result as long as it is fewer than 12 digits. There are single-digit errors in the 12 digit results. Three out of the twelve digits have errors when the answers are as large as can be represented (approx. 1e500). This suggests that a SYSRPL implementation in 15 digit floating point would produce accurate 12-digit results. Scott -- Scott Hemphill hemphill@alumni.caltech.edu This isn't flying. This is falling, with style. -- Buzz Lightyear ==== -=[ Thu, 30.10.03 3:21 p.m. +1300 (NZDT) ]=- in message ID : > This may not be precisely what you're looking for, but a non-recursive > way to calculate the nth Fibonacci number is: > > g = the golden ratio = (1+sqrt(5))/2 > > fib[k] = round(g^k/sqrt(5)), > where round means round to the nearest integer 1961 book by N.N. Vorob'ev (called Fibonacci Numbers) and sure enough, he has that result. Yup, he calls g^k /sqrt(5) the kth term of the geometric progression whose first tern is g/sqrt(5) and whose common ratio is g. -- Tony Hutchins Wellington New Zealand #66 Alimony is like buying oats for a dead horse. ==== > On page 549 of the HP48SX Owner's Manual, Volume II there is > a nice little recursive program, called FIB1, for calculating > the nth Fibonacci number: > > << ->n 'IFTE(n<=1,n,FIB1(n-1)+FIN1(n-2)' >> > > 13 FIB1 gives 233. > This takes 34 seconds on the HP48SX. That FIN1 is supposed to be FIB1, right? Tom Lake ==== -=[ Thu, 30.10.03 3:30 p.m. +1300 (NZDT) ]=- in message ID : [...] > > << ->n 'IFTE(n<=1,n,FIB1(n-1)+FIN1(n-2)' >> [...] > That FIN1 is supposed to be FIB1, right? Right. Typo, and no excuse!! Must be a pure typo :( -- Tony Hutchins Wellington New Zealand #333 If you wish to learn the highest truth, you must begin with the alphabet. Jap. Proverb ==== > On page 549 of the HP48SX Owner's Manual, Volume II there is > a nice little recursive program, called FIB1, for calculating > the nth Fibonacci number: > > << ->n 'IFTE(n<=1,n,FIB1(n-1)+FIN1(n-2)' >> > > 13 FIB1 gives 233. > This takes 34 seconds on the HP48SX. > > In approx mode: > On the HP49G it takes 309 seconds!!! > On the hp49g+ it takes 71 seconds. > > Does anyone know why this is so much faster on the HP48SX? > Is there a way to speed it up on the 49G/g+ ? Due to the introduction of the CAS on the HP49G, evaluation of expression is now handelde by the CAS. This allows the powerfull CAS to work, on the other hand, is slows down some processing. Althroug this example is extremly elegant, you could rewrite it simply in the following form: << -> N << N 1. <= N << N 1 - FIB N 2 - FIB + >> IFTE >> >> in which case it would only take 6 s on a HP49G+. If you are after speed, you could even rewrite it in non recursive: << IF DUP 1 > THEN IF DUP 2 == THEN 1 ELSE 1 1 3 4 ROLL START DUP ROT + NEXT SWAP DROP END END >> it does 13 FIB in 0.04s ==== For ultimate speed, try instead the direct formula Fibonacci(n) = ((1+sqrt(5))/2)^n/sqrt(5), rounded to the nearest integer where you should pre-compute sqrt(5) and (1+sqrt(5))/2 [the golden ration] to further speed things up. This way you'll need just one raising to power, one division and one rounding operation in all, for any n. This is non-iterative, non-recursive and thus will be much faster, specially for large n. > > > > On page 549 of the HP48SX Owner's Manual, Volume II there is > > a nice little recursive program, called FIB1, for calculating > > the nth Fibonacci number: > > > > << ->n 'IFTE(n<=1,n,FIB1(n-1)+FIN1(n-2)' >> > > > > 13 FIB1 gives 233. > > This takes 34 seconds on the HP48SX. > > > > In approx mode: > > On the HP49G it takes 309 seconds!!! > > On the hp49g+ it takes 71 seconds. > > > > Does anyone know why this is so much faster on the HP48SX? > > Is there a way to speed it up on the 49G/g+ ? > > Due to the introduction of the CAS on the HP49G, evaluation of expression is > now handelde by the CAS. > This allows the powerfull CAS to work, on the other hand, is slows down some > processing. > > Althroug this example is extremly elegant, you could rewrite it simply in > the following form: > << -> N > << N 1. <= > N > << N 1 - FIB N 2 - FIB + >> > IFTE > >> > >> > > in which case it would only take 6 s on a HP49G+. > > If you are after speed, you could even rewrite it in non recursive: > << IF DUP 1 > THEN IF DUP 2 == THEN 1 ELSE > 1 1 3 4 ROLL START DUP ROT + NEXT SWAP DROP END END > >> > it does 13 FIB in 0.04s > ==== > ... > Althroug this example is extremly elegant, you could rewrite it simply in > the following form: > << -> N > << N 1. <= > N > << N 1 - FIB N 2 - FIB + >> > IFTE > >> > > > in which case it would only take 6 s on a HP49G+. > > If you are after speed, you could even rewrite it in non recursive: > << IF DUP 1 > THEN IF DUP 2 == THEN 1 ELSE > 1 1 3 4 ROLL START DUP ROT + NEXT SWAP DROP END END > > it does 13 FIB in 0.04s For Fibonacci elegance, how about << [[ 0 1 ][ 1 1 ]] SWAP ^ 2. GET >> (from programs posted by Tom Gutman and Joe Horn in May 2000) Works for Exact Integers < 10000 on the 49G and presumably the 49G+. (Faster than the looping method above for N > ~170, at least on later ROMs. Does not work on the 48.) CdB Question: Why the arbitrary 9999 limit for raising an integer or matrix to an integer power???? I know it can be slow, but the power algorithm was speeded up in ROM 1.16. Tom ==== > > ... > > Althroug this example is extremly elegant, you could rewrite it simply in > > the following form: > > << -> N > > << N 1. <= > > N > > << N 1 - FIB N 2 - FIB + >> > > IFTE > > >> > > > > > > in which case it would only take 6 s on a HP49G+. > > > > If you are after speed, you could even rewrite it in non recursive: > > << IF DUP 1 > THEN IF DUP 2 == THEN 1 ELSE > > 1 1 3 4 ROLL START DUP ROT + NEXT SWAP DROP END END > > > > it does 13 FIB in 0.04s > > > For Fibonacci elegance, how about > > << [[ 0 1 ][ 1 1 ]] SWAP ^ 2. GET >> > > (from programs posted by Tom Gutman and Joe Horn in May 2000) > > Works for Exact Integers < 10000 on the 49G and presumably the 49G+. > (Faster than the looping method above for N > ~170, at least on later > ROMs. Does not work on the 48.) > > CdB Question: Why the arbitrary 9999 limit for raising an integer or > matrix to an integer power???? I know it can be slow, but the power > algorithm was speeded up in ROM 1.16. > > Tom Could someone post 49G and 49G+ timings for this (or whatever the fastest method for different ranges is)? I'd like to compare them with those of my 68k FibNum function. n Time (s) Result ---------------------------- 13 0.26 Exact 113 1 Exact 1113 12.2 Exact 1113. 0.74 Approximate 4444. 0.21 Approximate Bhuvanesh. ==== > >Could someone post 49G and 49G+ timings for this (or whatever the >fastest method for different ranges is)? I'd like to compare them with >those of my 68k FibNum function. > >n Time (s) Result >---------------------------- >13 0.26 Exact >113 1 Exact >1113 12.2 Exact >1113. 0.74 Approximate >4444. 0.21 Approximate > >Bhuvanesh. Using matrix power method for exact and sqrt(5) for approximate, on 49g+ n Time (s) Result ---------------------------- 13 0.2129 Exact 113 0.4075 Exact 1113 1.9919 Exact 1113. 0.0898 Approximate 4444. 0.0903 Approximate Same on 49G n Time (s) Result ---------------------------- 13 0.6907 Exact 113 1.2448 Exact 1113 4.4794 Exact 1113. 0.3137 Approximate 4444. 0.3130 Approximate Using non-recursive loop for both on 49g+ n Time (s) Result ---------------------------- 13 0.1216 Exact 113 0.5227 Exact 1113 5.4762 Exact 1113. 5.4861 Approximate 4444. 26.3077 Approximate Pete M. Wilson Gamewood, Inc. wilsonpm@gamewood.net ==== [...] >> For Fibonacci elegance, how about >> >> << [[ 0 1 ][ 1 1 ]] SWAP ^ 2. GET >> >> >> (from programs posted by Tom Gutman and Joe Horn in May 2000) >> >> Works for Exact Integers < 10000 on the 49G and presumably the 49G+. >> (Faster than the looping method above for N > ~170, at least on later >> ROMs. Does not work on the 48.) [...] >Could someone post 49G and 49G+ timings for this (or whatever the >fastest method for different ranges is)? I'd like to compare them with >those of my 68k FibNum function. > >n Time (s) Result >---------------------------- >13 0.26 Exact >113 1 Exact >1113 12.2 Exact >1113. 0.74 Approximate >4444. 0.21 Approximate With Joe Horn Program on an HP49G+ according to TEVAL: n Time (s) Result ---------------------------- 13 0.21 in exact mode 113 0.41 in exact mode 1113 2 in exact mode 1113. - in approximate and exact mode: Error: Bad Argument Type (raise to power doesn«t seem to work with a matrix) 4444 19.8 in exact mode Seems to be fast....;) For interest: FIB(4444) ist about 929 numbers long. Mathias -- Mathias Habel mathias.habel_no-spam_@t-online.de Remove _no-spam_ before replying ==== > For Fibonacci elegance, how about > > << [[ 0 1 ][ 1 1 ]] SWAP ^ 2. GET >> > > (from programs posted by Tom Gutman and Joe Horn in May 2000) Wow! Instantaneous! All these alternatives are great but the question remains. Why is the original program so slow on the 49 g+? -- Tom Lake Experience keeps a dear school but fools will learn in no other - Poor Richard's Almanack ==== -=[ Thu, 30.10.03 8:04 p.m. +1300 (NZDT) ]=- in message ID : > > For Fibonacci elegance, how about > > > > << [[ 0 1 ][ 1 1 ]] SWAP ^ 2. GET >> > > > > (from programs posted by Tom Gutman and Joe Horn in May 2000) > > Wow! Instantaneous! All these alternatives are great but the question > remains. Why is the original program so slow on the 49 g+? Cyrille answered that one almost before it was asked. -- Tony Hutchins ==== -=[ Thu, 30.10.03 10:42 a.m. +1300 (NZDT) ]=- in message ID <3fa02b16$1@usenet01.boi.hp.com> : [...] > Althroug this example is extremly elegant, you could rewrite it simply in > the following form: > << -> N > << N 1. <= > N > << N 1 - FIB N 2 - FIB + >> > IFTE > >> > >> > > in which case it would only take 6 s on a HP49G+. -- Tony Hutchins New Zealand ==== Wouldn't have to be - it could be written into a future version of Conn4x, and parallel versions could be created for Mac and *nix platforms if desired... not that HP will do that, but it is possible :) > > >>and (2) the PC should have a >>utility that re-synchs the pseudo-FAT with what's actually in the >>card. > > > No, please! I presume that would be some Windows thing -- i.e., > OS-dependent. > > Michael > ==== How can you tell? > > Is your model one of the earlier ones? > > dave > > > Just got it and was surprised at how cheap the thing feels. Much worse > than > > the TI's. Think $11.99 Walmart special. Really quite lame. Worst of all my > 7 > > key only works about 1/3 of the time. > > > > Construction issues aside, it's a powerful tool, but I doubt it will last > me > > 14 years like my 48GX has. > > > > > > > > Does anyone know of a program to produce Jacobian matrices? > > > > > > > > > > ==== This must have been a topic of discussion here before, but I wasn't able to find anything too informative, so excuse me if my question has been answered before. I currently own a 49G among some other HP calculators and I was wondering if there is any information available on how the Saturn emulation is done. What kind of slowdown can I expect? The ARM runs at almost 20x the clockspeed of the original Saturn, so there must be quite a bit of power going to waste running an emulator. One project I'd like to work on for example would be a native ARM operating system. With 75mhz multitasking would not be a problem, so you could even run the emulator with the original ROM as a seperate process if needed. Anyway, I don't really know what I'm getting at here yet. I just want some information for starters, so if anyone could point me to some manuals or source code, I would greatly appreciate that. ==== > This must have been a topic of discussion here before, but I wasn't > able to find anything too informative, so excuse me if my question has > been answered before. > I currently own a 49G among some other HP calculators and I was > wondering if there is any information available on how the Saturn > emulation is done. What kind of slowdown can I expect? The ARM runs > at almost 20x the clockspeed of the original Saturn, so there must be > quite a bit of power going to waste running an emulator. One project > I'd like to work on for example would be a native ARM operating > system. With 75mhz multitasking would not be a problem, so you could > even run the emulator with the original ROM as a seperate process if > needed. > Anyway, I don't really know what I'm getting at here yet. I just want > some information for starters, so if anyone could point me to some > manuals or source code, I would greatly appreciate that. I haven't heard much on the details myself either. Mine doesn't arrive until tomorrow, but even on the HP49G, I never got around to finding out if there is more than 2 voice polyphony available on the HP49G, but would like to hope that it could be available on the HP49G+ at least. I've seen music applications listed on hpcalc.org, then saw that they usually mention a 1 to 2 voice limit. If intended for music, I think 3 sould make a good minimum, 4 would be a but if a decent workspace, and more than that would make for some fun in terms of composing when chords are involved. I hope that building the new HP in an emulated mode means that it can now be seen as a bit more portable to different machines now and/or be able to be expanded beyond the limits of the origional saturn machines further. If I remember, the speed would take a lot of work to increase, and the screen used to be trapped to a max resolution based on the previous hardware HP calcs used to run on. I'd personally like to watch the new HP calcs expand into the new machine in which their code resides, assuming it builds off of the old HP trend of good calc design, which seems to not show so strong in some aspects of the more recent calcs (for whatever reason that may be). Ed Sutton ==== I don't have any info, but if you find any, please share it with me - I am very interested. Also, please note that some ROM routines are already native. Al ==== I just wanted to mention that the British HPCC group authorized the addition of another year of their Datafile newsletters (volume 21 from 2002) to PPC CD #3. This disk now goes to version 2.0 and is now titled Twenty-One Years of Datafile (1982-2002). As before, for each copy sold, U.S. $5.00 is going back to the HPCC group to help their continued efforts. Already, $300.00 has been generated for them at present. For more information on this disk, check http://www.magpage.com/~jakes/ppccdrom.htm on the web. Jake Schwartz ==== I just bought a new 256 MB SD card after reading posts where I thought people said it worked for them. I formatted it on my PC as a FAT file system. But when I insert the card into my 49G+ port 3 does not come up in the Filer. I know that the SD card works since I can copy and view files on it on my PC. I guess I'll just have to go out and get a 128 MB card. Hopefully a future ROM will support the 256 MB cards. I updated my calculator to 1.22 using the USB connection.before I got the SD card. -- Michael C. Polinske Milwaukee, WI ==================== Come to the WELS for new life in the living Word http://www.wels.net ==== > I just bought a new 256 MB SD card after reading posts where I thought > people said it worked for them. > > I formatted it on my PC as a FAT file system. > Did you format it as FAT16? How about trying to format it on the calculator itself? Al ==== I also tried to format it on the calculator, but it only allowed up to 128 MB, no option for 256 MB, and all of the options I tried ended with the message Format failed.. > > I just bought a new 256 MB SD card after reading posts where I thought > > people said it worked for them. > > > > I formatted it on my PC as a FAT file system. > > > > Did you format it as FAT16? How about trying to format it on the > calculator itself? > > Al > ==== Mike, what ROM version does your 49G+ have? I have updated to ROM v. 1.22 (immediately, so I don't know what the previous version did nor am I even sure of the previous version number) and as I type this, I turned mine on just to check- on the Filer, it says, 3:SD 246816KB (I haven't put anything on it yet.) And when I remove the card, it doesn't show. Oh, and make sure the SD card clicks in! You have to push it in fairly deeply. For a second a couple of days ago, I didn't think my finger could do it! You have to make sure it clicks in. Ed Look > I just bought a new 256 MB SD card after reading posts where I thought > people said it worked for them. > > I formatted it on my PC as a FAT file system. > > But when I insert the card into my 49G+ port 3 does not come up in the > Filer. > > I know that the SD card works since I can copy and view files on it on my > PC. > > I guess I'll just have to go out and get a 128 MB card. > > Hopefully a future ROM will support the 256 MB cards. I updated my > calculator to 1.22 using the USB connection.before I got the SD card. > > -- > Michael C. Polinske > Milwaukee, WI > ==================== > Come to the WELS for > new life in the living Word > http://www.wels.net > > > ==== OK, this is just weird...Today when I tried to read the card in my calculator it worked fine. I'll play with it more after work tonight. Last night when I put it in and it clicked in, the Filer would clock for a while but the SD card would not show up. I tried it several times. I have ROM version 1.22. > Mike, what ROM version does your 49G+ have? I have updated to ROM v. > 1.22 (immediately, so I don't know what the previous version did nor am > I even sure of the previous version number) and as I type this, I turned > mine on just to check- on the Filer, it says, 3:SD 246816KB (I > haven't put anything on it yet.) And when I remove the card, it doesn't > show. > > Oh, and make sure the SD card clicks in! You have to push it in fairly > deeply. For a second a couple of days ago, I didn't think my finger > could do it! You have to make sure it clicks in. > > Ed Look > > > > I just bought a new 256 MB SD card after reading posts where I thought > > people said it worked for them. > > > > I formatted it on my PC as a FAT file system. > > > > But when I insert the card into my 49G+ port 3 does not come up in the > > Filer. > > > > I know that the SD card works since I can copy and view files on it on my > > PC. > > > > I guess I'll just have to go out and get a 128 MB card. > > > > Hopefully a future ROM will support the 256 MB cards. I updated my > > calculator to 1.22 using the USB connection.before I got the SD card. > > > > -- > > Michael C. Polinske > > Milwaukee, WI > > ==================== > > Come to the WELS for > > new life in the living Word > > http://www.wels.net > > > > > > > ==== When I first got my 128MB SD card, I was a bit surprised to see that it showed up in my card reader as being only 120MB. Ok, I know that some of it's used just for keeping track of what's on the card. The system area that is, the boot record, FATs, and root directory, takes away from what's actually available for storing files. But 8MB out of 128MB? Off hand, that seems awfully high. I'd expect a 128MB card to have a total of 262,144 sectors of 512 bytes each, but every tool that I've used shows me a total of 245,919 sectors. And yes, that's including the system area. So what happened to the other 16,225 sectors (a bit under 8MB)? Maybe those sectors on an SD card are read only? Perhaps used for the Secure code designed to protect copyrighted materials and such by encryption or whatever? To be sure, 120MB is probably more than I'll ever actually need. But SD cards are currently marketed at least as low as 16MB. Do the smaller cards give up as many sectors before the user gets to store anything on them? Would we actually get more available storage space by using a MultiMediaCard (MMC)? -- James ==== Actually, you need to remember that NAND flash is not a perfect technology, so ever if the chips are 128MB, some sectors are kept on the side for future use, in order to replace defective sector and the like... for example, you know that the FAT is written often, and flash has a limited number of erase/write cycles, well if you dig in the controler, you will see that the FAT is relocated regullary in the physical memory (but not the logical)... it's a little bit like hard drive that have banks of sectors on the side that they use to replace defective sectors (or sector that they can predicts are getting in bad shape)... a SD card (or CF) is MUCH more complex that just some linear memory :-) > When I first got my 128MB SD card, I was a bit surprised to see that it > showed up in my card reader as being only 120MB. Ok, I know that some of > it's used just for keeping track of what's on the card. The system > area that is, the boot record, FATs, and root directory, takes away > from what's actually available for storing files. But 8MB out of 128MB? > Off hand, that seems awfully high. > > I'd expect a 128MB card to have a total of 262,144 sectors of 512 bytes > each, but every tool that I've used shows me a total of 245,919 sectors. > And yes, that's including the system area. > > So what happened to the other 16,225 sectors (a bit under 8MB)? Maybe > those sectors on an SD card are read only? Perhaps used for the Secure > code designed to protect copyrighted materials and such by encryption or > whatever? > > To be sure, 120MB is probably more than I'll ever actually need. But SD > cards are currently marketed at least as low as 16MB. Do the smaller > cards give up as many sectors before the user gets to store anything on > them? > > Would we actually get more available storage space by using a > MultiMediaCard (MMC)? > > -- > James > ==== On my new 256 MB card, Windows reports 241 MB free, so 15 MB must be used by the system some how. > When I first got my 128MB SD card, I was a bit surprised to see that it > showed up in my card reader as being only 120MB. Ok, I know that some of > it's used just for keeping track of what's on the card. The system > area that is, the boot record, FATs, and root directory, takes away > from what's actually available for storing files. But 8MB out of 128MB? > Off hand, that seems awfully high. > > -- > James > -- Michael C. Polinske Milwaukee, WI ==================== Come to the WELS for new life in the living Word http://www.wels.net ==== When I format my 128MB SD card using the card reader, Norton Disk Doctor tells me that everything's ok (though fewer sectors than I expected). But when I format the SD card using the 49g+ (ROM 1.22), NDD notifies me of an Error on drive F: -- Invalid Disk Table in Boot Record, and gives me a chance to correct it if I really want to. Ok, the card seems to work just fine whether I have Norton correct the problem or not, so no big deal. Looking into this a bit deeper, it seems the the first sector of the FAT is sector 3, instead of the expected sector 1 (sector numbering starts with sector 0, the boot sector). So it seems that the boot record is reserving 3 sectors? Looking at the SD card with a disk/sector editor, I find that the boot record really does reserve 3 sectors. NDD's correction doesn't mess with this. The physical description of the disk, 63 sectors per track and 255 heads, is a bit unusual, but that's not surprising, considering that the disk is really flash memory. It turns out that allowing NDD to correct the error changes this to 32 sectors per track and 64 heads. My guess is that these are merely dummy values, never actually used for anything although NDD notices that they're unusual. Sector 0 looks like an otherwise typical boot record. Sectors 1 and 2 look like the beginning of a FAT; whatever was there before I used the 49g+ to format it is still there. The first FAT begins with sector 3. As the 49g+'s format doesn't change anything in the reserved sectors 1 and 2, my guess is that the 49g+ doesn't actually use them for anything. By the way, the 49g+'s format doesn't write anything to the data area of the card, just the system area. Whatever was in the data area is still there. Why would the boot record ever reserve 3 sectors? I know that it can reserve more sectors, I suppose as many as a 2 byte hex number would allow, up to 255 sectors, but I don't believe that I've ever seen a boot record that reserves more than the single sector for itself. For my card, it makes no practical difference. Formatted in the card reader from MS-Windows, it has 2 sectors left over at the end that don't make up a full cluster, so are unusable. Formatted on the 49g+, it has 2 reserved sectors that I can't use, but no unusable sectors left over at the end. But I expect that on other cards, reserving 2 sectors may make a difference of 1 or 2 clusters unavailable for use, so it may be worthwhile to change the 49g+'s format so that it reserves only the 1 sector for the boot record. But for all I know, perhaps the development team really does have something in mind for those two reserved sectors, to be added later. If so, then when it takes effect, the card will have to be formatted on the 49g+, not with other ordinary methods of formatting it. -- James ==== I'm expecting my HP49G+ to arrive with v1.20 and (if not for this time, for future reference) was wondering how I could go about backing up my current ROM before upgrading and do problems with upgrading get resolved by the old HP49G steps? Also have any linux/unix people tested the HP49G updater for the G+, or are there differences that it doesn't handle? I hope I don't need to wait until I have access to a windows USB machine just to upgrade since I have no sd equipment (yet) and rebooting to what little of a windows setup I have on this machine just casuses more trouble than it is worth; if windows wasn't a problem on this machine, I wouldn't have gone against my computer's warranty and fixed my computer's OS problems for the last time by using something other than windows. Ed Sutton ==== > I'm expecting my HP49G+ to arrive with v1.20 and (if not for this > time, for future reference) was wondering how I could go about backing > up my current ROM before upgrading and do problems with upgrading get > resolved by the old HP49G steps? > Also have any linux/unix people tested the HP49G updater for the G+, > or are there differences that it doesn't handle? I hope I don't need > to wait until I have access to a windows USB machine just to upgrade > since I have no sd equipment (yet) and rebooting to what little of a > windows setup I have on this machine just casuses more trouble than it > is worth; if windows wasn't a problem on this machine, I wouldn't have > gone against my computer's warranty and fixed my computer's OS > problems for the last time by using something other than windows. Well, I think the options for updating the ROM on a 49g+ are an SD card and reader, a USB connection, or *maybe* IrDA. -- James ==== Given that my freshly formatted 128MB SD card has 61351 free clusters of 2KB each, and what's more Norton Disk Doctor says that it has 125646848 bytes available, I expect 122702KB to be free on it. Why does the filer show me only 122698KB on the card? What happened to the other 4KB? -- James ==== For some reason my 49g+ doesn't like solving DE's with Y(x), but U(x), V(x) or anything else is fine. My 49g accepts any user function, but the g+ says 'Invalid User Function' - any pointers? I've compared the flags etc and they're the same, but I don't think this is the problem. ==== You probably have a variable 'Y' stored in the current directory. If you purge it, it should be fine. > For some reason my 49g+ doesn't like solving DE's with Y(x), but U(x), > V(x) or anything else is fine. My 49g accepts any user function, but > the g+ says > 'Invalid User Function' - any pointers? I've compared the flags etc > and they're the same, but I don't think this is the problem. ==== I used a simple example on the stack with Y purged from the directory and in exact mode I got the correct answer thru DESOLVE. I use ROM 1.22 (when it is not crushing or screwing the first alarm :-) !Demeter! ==== > > Typing #33107 SYSEVAL on the calc returns BINT0 ok, but when I try on > Debug4x (#33107 xSYSEVAL) get an error. > I'm clearly doing something wrong and wanted to know how to use xSYSEVAL on > Debug4x, or if there's a better way to access system functions? > There is no point in using SYSEVAL in Debug4, why would you - you can directly write BINT0 or PTR 33107. What you did wrong is this: In UserRPL (or on the command line), #33107 is HEX string. In SystemRPL, #33107 is a BINT, a different object type. The user command SYSRPL expects a HEX string as an argument, not a BINT. In order to do what you wanted correctly, you would have to write in Debug4 HXS 5 70133 (note that the digit sequence is inverted) xSYSEVAL but as I said, this is not useful since SystemRPL gives you direct access to all internal commands. Hope this helps - Carsten ==== The BINTO was a (bad) example. I actually want to use the command RCLSYSF2 (recall sys. flags 65-128) - there doesn't seem to be any named command that will do this. I can see from your post that its pointless to use SYSEVAL in Debug4x, if all system commands can be otherwise accessed. The problem still is I can only access named commands, for example PTR #33107 is not accepted by Debug4x, and therefore doesn't return BINT0. Also tried some other commands with PTR in the name (eg SysPtr@) but with no success. If you could let me know the exact Debug4x syntax for accessing a system command by address it would be great. I apologise if I'm making a stupid mistake, but cannot find any examples of exactly how to do this in the SysRpl manual (probably expected to know it!). Dave > > > > Typing #33107 SYSEVAL on the calc returns BINT0 ok, but when I try on > > Debug4x (#33107 xSYSEVAL) get an error. > > I'm clearly doing something wrong and wanted to know how to use xSYSEVAL on > > Debug4x, or if there's a better way to access system functions? > > > > There is no point in using SYSEVAL in Debug4, why would you - you can > directly write BINT0 or PTR 33107. > > What you did wrong is this: > > In UserRPL (or on the command line), #33107 is HEX string. In > SystemRPL, > #33107 is a BINT, a different object type. The user command SYSRPL > expects a HEX string as an argument, not a BINT. In order to do what > you wanted correctly, you would have to write in Debug4 > > HXS 5 70133 (note that the digit sequence is inverted) > xSYSEVAL > > but as I said, this is not useful since SystemRPL gives you direct > access > to all internal commands. > > Hope this helps > > - Carsten ==== Jean-Fran.8dois : An excellent SYSEVALs list can be found in Eduardo Kalinowski's online SysRPL programmation tutorial, which you can download at : http://www.hp-sources.com/docs/download/stut_pdf.zip Yoann. ==== > Greetings! > > Sounds like a great idea!!! I'd love to crack this puppy open and regain > the header area for normal use (9 or 10 level stack all on the same > screen!!!). I can definitely vouch for 68000's powerful instruction set > (CISC), this usually makes programming for 68k machines a > simpler/shorter process. Doesn't the ARM family use a RISC (reduced) > instruction set? I have no experience in ARM assembly. Anyhow, that > would be a great endeavor! Long Live HARP! > > -demosthenes > I find RISC easier to program for actually. There are fewer instructions to keep in mind and they are more homogenous (at least with PowerPC). ==== > > Sounds like a great idea!!! I'd love to crack this puppy open and regain > > the header area for normal use (9 or 10 level stack all on the same > > screen!!!). I can definitely vouch for 68000's powerful instruction set > > (CISC), this usually makes programming for 68k machines a > > simpler/shorter process. Doesn't the ARM family use a RISC (reduced) > > instruction set? I have no experience in ARM assembly. Anyhow, that > > would be a great endeavor! Long Live HARP! > I find RISC easier to program for actually. There are fewer instructions to keep in mind and they are more homogenous (at least with PowerPC). Same with me, I could not agree more. I am not a fan of SH series, but I love the ARM, the ASM is so powerfull! you can have one instruction like: CMP r0, 15 LRDLT r15, [r15, +r0 shl 2] B default address case 0 address case 1 (for the one of you who do not understand, this is a switch (r0) { case 0: ... case 1: ... .... case 15: dafault:.. } all in 1 instruction you can also do stuff like: add(condition) r0, r1, r2, shl r3 (r0=r1+r2< > > Sounds like a great idea!!! I'd love to crack this puppy open and regain > > > the header area for normal use (9 or 10 level stack all on the same > > > screen!!!). I can definitely vouch for 68000's powerful instruction set > > > (CISC), this usually makes programming for 68k machines a > > > simpler/shorter process. Doesn't the ARM family use a RISC (reduced) > > > instruction set? I have no experience in ARM assembly. Anyhow, that > > > would be a great endeavor! Long Live HARP! > > > I find RISC easier to program for actually. There are fewer instructions > to keep in mind and they are more homogenous (at least with PowerPC). > > Same with me, I could not agree more. > I am not a fan of SH series, but I love the ARM, the ASM is so powerfull! > you can have one instruction like: > > CMP r0, 15 > LRDLT r15, [r15, +r0 shl 2] > B default > address case 0 > address case 1 > > (for the one of you who do not understand, this is a switch (r0) { case 0: > ... case 1: ... .... case 15: dafault:.. } all in 1 instruction > > you can also do stuff like: > add(condition) r0, r1, r2, shl r3 (r0=r1+r2< also with binary and other instruction)... > > this is so good for stuff like graphic manipulation! 1¡ But how much memory space does such an instruction take ? 2¡ Can you reveal to us the mystery of the new OpCodes ? How can one use them ? Why do we often find 81B1 instead of 142164808C for return-to-RPL ? Yoann. ==== Does anybody know if there is a rammap.a file for the 39G? Or is it identical to the 49G file? - Carsten ==== Manel. The expected (a rumor) date for publish the FLASH ROM 1.19-7 was HP Could be gathering errors from ROM 1.22 of HP49+ to fix them on new flash for HP49(old)? Is there any official date to publish the expected ROM? Goood-bye! ==== I'll be in London this friday and saturday. I wonder if those of you from the UK could give me some tips as to what not to miss, and especially what's a waste of time, because i only have saturday morning for sightseeing. I'll be staying near Hammersmith or Kensington, also if you know of a place i should stay clear from, please let me know! ;-) -- Steve Sousa ==== > > I'll be in London this friday and saturday. I wonder if those of you from > the UK could give me some tips as to what not to miss, Remember the wise words of Gerald Hoffnung: - all london brothels display a blue light - try out the famous echo in the British museum reading room - on entering a bus or tube train, it is customary to shake hands with all the other passengers. And if you follow his advice, I've got a bridge to sell you ;) Serious places to visit: Buckingham Palace & the royal parks (but I suspect the palace will be closed as it is not summer). Parliament square (Houses of Parliament, Big Ben) and then up whitehall. Visit the cabinet war rooms (used in WWII by churchill), and finish up in Trafalgar Square. Or visit the London Eye (large millenium ferris wheel) and if time, you can walk along the river to the millenium footbridge (and not marvel at how much it shakes ;)). Cross to go to the City Of London. The Bank Of England has a nice museum. HMS Belfast is moored on the river and open (but a full visit takes about 3 hours). The science Museum in kensington is probably worth a visit, as is Greenwich Observatory/Museum (but is probably too far for the time). ==== Steve Sousa a .8ecrit dans le message de > > I'll be in London this friday and saturday. I wonder if those of you from > the UK could give me some tips as to what not to miss, and especially what's > a waste of time, because i only have saturday morning for sightseeing. I'll > be staying near Hammersmith or Kensington, also if you know of a place i > should stay clear from, please let me know! ;-) Tower Bridge is a nice place, Hyde Park. I don't know about Buckingham Palace, I didn't get to see it when I went to London. I suppose Big Ben is worth seeing too. As for places to avoid... I stayed in the King's Cross area, and it was under works at that time, so apart from the big Church, it wasn't great, plus, it's crowded. > > > -- > Steve Sousa > > ==== > > I'll be in London this friday and saturday. I wonder if those of you from > the UK could give me some tips as to what not to miss, and especially what's > a waste of time, because i only have saturday morning for sightseeing. I'll > be staying near Hammersmith or Kensington, also if you know of a place i > should stay clear from, please let me know! ;-) > The London Eye is an absolute must, unless you're scared of hights! Dave ==== --------------------------------------------------------------------- > > > seem less possible. Also since the GII has no FLASH capability, the > > problems you outlined are not fixable by the software upgrade route. > > The specs for the GII say it DOES have FLASH. > > Product Features > Electronic Specification: > CPU: 48Mhz ARM9 48Mhz...??? is that fortyeight, or four point eight...??? fortyeight seems awfully fast for a 48g(x)...??? sproogles... ( '±' ) / `'*:-.,_,.-:*'``'*:-.,_,.-:*'``'*:-.,_,.-:*'``'*:-.,_,.-:*'` .---. .-. .---..-..-..-..-..---..-..-. .--. | | < / | | < > / | .` || O || '' |( -< `---'`--^--'`---' `-' `-'`-'`---'`----'`---) ,.-:*'``'*:-.,_,.-:*'``'*:-.,_,.-:*'``'*:-.,_,.-:*'``'*:-.,_ The Babynous Cult A SemiHedonistic Cult for The Whole Family! ^v^v^v^v^v^v^v^v^v^v^v^v^v^v^v^v^v^ --------------------------------------------------------------------- name=babynous.vcf filename=babynous.vcf begin:vcard n:;The Babynous Cult x-mozilla-html:FALSE adr:;;;;;; version:2.1 x-mozilla-cpt:;3 fn:The Babynous Cult end:vcard --------------E75A6DC944AC74684D7E3ADD-- ==== > > > > > > > seem less possible. Also since the GII has no FLASH capability, the > > > problems you outlined are not fixable by the software upgrade route. > > > > The specs for the GII say it DOES have FLASH. > > > > Product Features > > Electronic Specification: > > CPU: 48Mhz ARM9 > > 48Mhz...??? > is that fortyeight, > or four point eight...??? > > fortyeight seems awfully fast for a 48g(x)...??? That's forty-eight, OK. Remember, the ARM is emulating a Saturn so you won't see such a dramatic difference as you might expect from just the difference in clock speed. -- Tom Lake Experience keeps a dear school but fools will learn in no other - Poor Richard's Almanack ==== For these main reasons it would have been better to have named this model the 49GII and the 49G+ should have been called the 49GII+. Leave it to the marketing boneheads. My understanding is that the entire calculator division is comprised of only about 11 people so it has got to be a miracle that such a machine was ever produced in such shot time using contracted R&D and outsourced production. Greg S [snip] > 3. Compatibility Issues. Why, why, why? This unit should run > 48G code straight out of the box - without modification - PERIOD. > Think -- very large installed software base. I'm all for progress and > added functionality but the baseline configuration (from the factory) > should be compatible with the 48G code. All non-compatible added > functionality should be enabled via system flags. Here are some of > the issues I have run into so far: [snip] > c. Key mapping - if you used many KEY command in your code > get ready for some work. Most of the key locations have changed. > d. Real numbers are now displayed with decimal points when > operating in standard numeric mode. Although still logically correct, > this affects precisely formatted field output to the screen or > printer. > e. The default font size for Choose boxes has been changed > to MINI (from normal size on the 48G+). > f. Symbolic representation rules have changed. For example, > the factory result for the program <<1 TAN >> on the 48G (factory > default) was 1.7455E-2. Running the same program with factory default > settings on the 48GII returns TAN(1). > g. RS-232 communication problems - Cannot communicate with > several RS-232 devices that worked fine with the 48G+. > h. Cannot seem to load library routines written for the 48G+ > onto the 48GII from a PC. ==== > > > Tom Lake replied: > >>seem less possible. Also since the GII has no FLASH capability, the > >>problems you outlined are not fixable by the software upgrade route. > > > > > > The specs for the GII say it DOES have FLASH. > > > > Product Features > > Electronic Specification: > > CPU: 48Mhz ARM9 > > LCD: 131 x 64 pixels > > Display Type: pixel > > IR Port: IrDA > > Serial Port: RS232 > > Memory: 128KB > > Permanent Storage: Flash ROM > > Power: Three AAA and one CR-2032 (same battery used in most PC motherboards) > > Auto-power OFF: approximately 5 minutes > > Material of enclosure: Plastic > > Material of key top: Plastic > > Dimensions: 184.0 x 87.0 x 27.5 mm (7.2 x 3.4 x .09 in) > > Weight: 220 g (7.7 oz) > > Where did you get this? I got those specs from HPCalculators.com (Samson Cables) Tom L ==== There is something I as a simple user of 49G+ can not understand! Keying in an simple equation as 2X^3 - X^2+1 Then Solvevx in exact mode leaves an empty list. If I switch to approx mode it works just fine and I get the expected answer. { X = -.6666666667 X=0 } Why is not the calc suggesting switch to Approx mode if it can not solve in exact mode? And why can it not solve it? Anyway, this is the kind of things I do not like in a calculator for students. Torstein ==== > Keying in an simple equation as 2X^3 - X^2+1 Then Solvevx in exact > mode leaves an empty list. If I switch to approx mode it works just > fine and I get the expected answer. { X = -.6666666667 X=0 } The equation above has one real root and two complex (a conjugate pair that is). To get all solutions switch to complex approximate mode and press SOLVEVX. The solutions you posted are incorrect. My machine gives me -0.657298106138 (0.578649053069,0.652575763252) (0.578649053069,-0.652575763252) In exact mode the polynomial factoring seems to be too complex to handle. Nevermind. Get to know your machine to use it effectively. If you want something simple for a student then throw it away and get yourself a TI junk! !@*#^%$@$* !Demeter! ==== > > > Keying in an simple equation as 2X^3 - X^2+1 Then Solvevx in exact > > mode leaves an empty list. If I switch to approx mode it works just > > fine and I get the expected answer. { X = -.6666666667 X=0 } > > The equation above has one real root and two complex > (a conjugate pair that is). To get all solutions switch > to complex approximate mode and press SOLVEVX. The solutions > you posted are incorrect. My machine gives me > -0.657298106138 > (0.578649053069,0.652575763252) > (0.578649053069,-0.652575763252) > In exact mode the polynomial factoring seems to be too > complex to handle. Nevermind. Get to know your machine > to use it effectively. If you want something simple for > a student then throw it away and get yourself a TI junk! > > !@*#^%$@$* > > !Demeter! Waiting for my reply, aren't you? ;-) With my Cubic() function from MathTools, one can get exact solutions on the 68k. Exact solution of general cubics and quartics is not built-in on the 68k. -- Bhuvanesh ==== > Keying in an simple equation as 2X^3 - X^2+1 Then Solvevx in exact > mode leaves an empty list. If I switch to approx mode it works just > fine and I get the expected answer. { X = -.6666666667 X=0 } Then your calc is defective, as neither X = 0 nor X = -.6666666667 is a solution of 2X^3 - X^2 + 1. Michael -- -= Michael Hoppe , =------ ==== If I use Numeric Sover-->Solve Poly, I get 3 complex roots on my 49G. **Leigh > > > Keying in an simple equation as 2X^3 - X^2+1 Then Solvevx in exact > > mode leaves an empty list. If I switch to approx mode it works just > > fine and I get the expected answer. { X = -.6666666667 X=0 } > > Then your calc is defective, as neither X = 0 nor X = -.6666666667 is a > solution of 2X^3 - X^2 + 1. > > Michael ==== I don't get why the title of this thread has changed but looks like changing the links on the PCB and fitting a suitable 256K device is all that's needed to upgrade the RAM size. (Well, maybe). Did I miss something? ö yes ö surfacemount rework expertise!! It is a bit of a pity that the MR27T802F ROM is effectively a one-time programmable device, (the code is actually loaded at OKI's factory before shipment). I do like the HP49 idea of being able to reload the OS as new features are added. Possibly a much bigger RAM device would allow some proper scope for upgrading the OS. Actually, I'm a bit intrigued by the two links on the PCB. I can't see any real need to route address lines with them. (I would have thought the ARM software would do a SIZE routing during the cold boot-up sequence, to determine what memory device is installed - a bit like the old HP48 code did for 32K/128K devices.) Wild guess - if the links are provided as purely software readable configuration, then installing both *might* allow one to fit a 1M part, the biggest memory size for the TSOP2-44 footprint! One way to establish if it were at least feasible to replace the SRAM with something bigger than 256K would be to see if either or maybe both pins 23 and 28 of the SRAM have PCB tracks connected to them. (I'm assuming that pin 22 definitely has a track connected to it, to give the 256K address space). Frank > After opening the 48gII I saw the following: > Sideways the CPU chip: > SAMSUNG @ the manufacturer > S3C2410X01 @ type of ARM9 by Samsung > KP750 @ serial? 75.0MHz? what??? > ****************** > BSI chip CMOS memory: > BS616LV1010EC-70 > BS = Brilliance Semiconductor Inc. > 616 = probably the chip series it belongs to(?) > LV = low Voltage > 1010 = 1Mbit (128KB) > E = TSOP II (A=BGA) > [TSOP2-44, Vcc=3.0_V => 20_mA operating, 0.5_uA standby,] > C = Commercial +0..+70 (I=Industrial -40..+85) > -70= 70 ns > serial?=S31640FY46745.1 > date?=E0319 TAIWAN > ********************* > > The mainboard says: SG48-10-3 (Series G48 ? october 3rd? > > 256KB: S1-S4 SHORT > 128KB: S5-S8 SHORT > > Since the darn thing has only MEM 82737. > I surely would like to see a 256KB model > which would give MEM 213809. > Well, 82K is surely more than 32K and pretty enough > I'm just used to have more in my old 49G > This 49G- is not up to the same memory level > as the new 49G+ > for wich you have ~10* more memory 384K+768K=1280K > with SD 128MB it is ~1000* more.... > PS: There is also an OKI chip... > Looks like R27I802F > R = ROM > 27 = Series > I = Packaging (?) or CMOS or ??? > 80 = 1MB or rather 512K x 16 > 2 = 2*8 eg 16-bit data organization > F = Prodcution Programmed ROM - P2ROM ==== It seems the HP49G has a problem with inverting matrices. I want to invert a 3x3 matrix which consists of the following entries: inv([[11/8, 7/8, 3/8][7/8, 11/8, 7/8][3/8, 7/8, 11/8]]) this results in a matrix which is not really the inverted of the above one: [[1.2857, -1, .2857][-1, 2, -1][.2857, -1, 1.2857]] the correct result should be (as matlab or octave determines it): [[1.19048,-0.66667,-0.14286][-0.80952,1.33333,-0.14286][0.19048,-0.66667,0.8 5714]] when I pull out the common factor of the original matrix (1/8) and calculate the inversion the results match. do I have to make a trick to get the correct result without making further simplifications or is this a bug in the software which could not be fixed from me. thanks for your help christoph ==== > > It seems the HP49G has a problem with inverting matrices. I want to > invert a 3x3 matrix which consists of the following entries: > > inv([[11/8, 7/8, 3/8][7/8, 11/8, 7/8][3/8, 7/8, 11/8]]) > > this results in a matrix which is not really the inverted of the above one: > > [[1.2857, -1, .2857][-1, 2, -1][.2857, -1, 1.2857]] > > the correct result should be (as matlab or octave determines it): > > > [[1.19048,-0.66667,-0.14286][-0.80952,1.33333,-0.14286][0.19048,-0.66667,0.85 7 > 14]] > > when I pull out the common factor of the original matrix (1/8) and > calculate the inversion the results match. > > do I have to make a trick to get the correct result without making > further simplifications or is this a bug in the software which could not > be fixed from me. > > thanks for your help > > christoph > You apparently entered the wrong matrix into mathlab and octave. Instead of the desired matrix, correctly entered in the hp49: [[11/8, 7/8, 3/8][7/8, 11/8, 7/8][3/8, 7/8, 11/8]] you seem to have entered this matrix into mathlab and octave: [[11/8, 7/8, 3/8][7/8, 11/8, 3/8][3/8, 7/8, 11/8]] ^ PROBABLE ERROR In exact mode, on my 49, I get the correct exact inverse: [[ 9/7 -1 2/7 ] [ -1 2 -1 ] [ 2/7 -1 9/7 ]] In approximate mode, on my 49, I get what you got, except for more decimal places, and the product of original and this approximate inverse, rounded to 10 dcimal places, is the identity matrix. ==== I do not see any problem. In exact mode I get [9/7 -1 2/7] [ ] [-1 2 -1 ] [ ] [2/7 -1 9/7] which is indeed the inverse of the original (I multiplied the two to get the identity matrix). ??? !Demeter! ==== > It seems the HP49G has a problem with inverting matrices. I want to > invert a 3x3 matrix which consists of the following entries: > > inv([[11/8, 7/8, 3/8][7/8, 11/8, 7/8][3/8, 7/8, 11/8]]) > > this results in a matrix which is not really the inverted of the above one: > > [[1.2857, -1, .2857][-1, 2, -1][.2857, -1, 1.2857]] > > the correct result should be (as matlab or octave determines it): > > > [[1.19048,-0.66667,-0.14286][-0.80952,1.33333,-0.14286][0.19048,-0.66667,0.85 714]] If you had typed the entries of your matrix correctly in Matlab, you would have gotten the correct result, instead of the one above, too... there, instead of 7/8. -- Helen. ==== > this results in a matrix which is not really the inverted of the above one: > > [[1.2857, -1, .2857][-1, 2, -1][.2857, -1, 1.2857]] > That's correct. > the correct result should be (as matlab or octave determines it): > > > [[1.19048,-0.66667,-0.14286][-0.80952,1.33333,-0.14286][0.19048,-0.66667,0 > .85714]] Then matlab semms to be defective ... Here's what Maple gives: > B:=array(1..3,1..3,[[11/8, 7/8, 3/8],[7/8, 11/8, 7/8],[3/8, 7/8, 11/8]]); [11/8 7/8 3/8 ] [ ] B := [7/8 11/8 7/8 ] [ ] [3/8 7/8 11/8] > inverse(B); [9/7 -1 2/7] [ ] [-1 2 -1 ] [ ] [2/7 -1 9/7] > evalf(%); [1.285714286 -1. .2857142857] [ ] [ -1. 2. -1. ] [ ] [.2857142857 -1. 1.285714286] Michael -- -= Michael Hoppe , =------ ==== > Then matlab semms to be defective ... no, I was wrong. I got a little bit confused because I tried matlab, octave and the HP49. and two different results appeared. Maybe I had an error when typing the matrix. thanks for your fast help! christoph ==== > > It seems the HP49G has a problem with inverting matrices. I want to > invert a 3x3 matrix which consists of the following entries: > > inv([[11/8, 7/8, 3/8][7/8, 11/8, 7/8][3/8, 7/8, 11/8]]) > > this results in a matrix which is not really the inverted of the above one: > > [[1.2857, -1, .2857][-1, 2, -1][.2857, -1, 1.2857]] > > the correct result should be (as matlab or octave determines it): > > > [[1.19048,-0.66667,-0.14286][-0.80952,1.33333,-0.14286][0.19048,-0.66667,0.8 5714]] > The TI-89, TrueBASIC and Wang 2200 BASIC all get the same answer as the 49G. What do you get when you multiply the Matlab inverse with the original matrix? The problem might be the definition of inverse. The 49G, TI-89, TrueBASIC and Wang BASIC all define the inverse of a matrix to be the matrix which, when multiplied by the original, gives the identity matrix. When you multiply the 49G answer by the original matrix, you get [[1 0 0][0 1 0][0 0 1]] which is correct. -- Tom Lake Experience keeps a dear school but fools will learn in no other - Poor Richard's Almanack ==== We are also having problems communicating with RS-232 devices (which worked fine with the G+). Given the single connection cable issue you outlined below, it does appear that HP might have decided to design the GII with a non-standard (RS-232C) serial port. Has anyone been able to communicate with a RS-232 device (other than the one on their PC)? Bob > X > Now on to the wierd RS232 serial thing. > I used the provided serial cable to load my libraries from the PC to the > 48gii, > and that all worked OK. > I then tried to connect the 48gii to my survey instruments' radio modem > (hayes compatible). > well it did not work at first as expected, > considering that I assumed that the TX and RXD would need to be reversed, > as compared to connecting to a PC. > Well I swaped wires #2 and #3, and it still did not work. > I got out my ohm meter, and discovered that only pin #5 (signal ground) > has a connection from from the HP end to the DB9 end of the provided serial > cable. > Now my question is this. > How in the heck is the 48gii able to connect to the PC using only a signal > ground?? > And does the 48gii even have a real serial port at all, > or is it some form of a USB port in reality. > I find this rather stunning!!! > Would you please verify that my above statements are correct. > X > John Evers > *************************** ==== Took one for the team. My suspicions were realized as I cut open the molded plastic cover on the D-sub end of my brand new (now non-returnable) 48GII cable and dug through the potting material to see active components hidden underneath. Looks like (in order to save cost/power???), HP decided not to include a functioning RS-232 port on the unit itself. Instead, they have included the RS-232 transceivers in the custom cable they supply with each unit. They simply drive the logic levels of the uP UART out of the connector on the back of the unit. The active components within the cable then convert the logic levels to RS-232 signal levels. Problem is they did not supply the power necessary to drive these active components. Instead, they are stealing power from the modem control lines of the RS-232. This is fine as long as you 1) use the supplied cable, 2) don't use an additional cable extender that does not support the modem control lines and 3) connect your 48GII to a PC using HP's driver software which will toggle either the DTR or RTS lines of the PC to the proper levels. If you are attempting to connect the 48GII to any other RS-232 device you will need to make sure that it can supply the signals necessary to power the cable circuitry. This will probably include making sure it supports the modem control lines (many current RS-232 devices do not) and making sure the device's firmware/software drivers keep the modem control signals in a state that keep the cable powered properly. This may require a rewrite of the device's serial drivers. In some cases (such as a modem that uses these lines for it's operational functionality), you may be out of luck. Bob > X > Now on to the wierd RS232 serial thing. > I used the provided serial cable to load my libraries from the PC to the > 48gii, > and that all worked OK. > I then tried to connect the 48gii to my survey instruments' radio modem > (hayes compatible). > well it did not work at first as expected, > considering that I assumed that the TX and RXD would need to be reversed, > as compared to connecting to a PC. > Well I swaped wires #2 and #3, and it still did not work. > I got out my ohm meter, and discovered that only pin #5 (signal ground) > has a connection from from the HP end to the DB9 end of the provided serial > cable. > Now my question is this. > How in the heck is the 48gii able to connect to the PC using only a signal > ground?? > And does the 48gii even have a real serial port at all, > or is it some form of a USB port in reality. > I find this rather stunning!!! > Would you please verify that my above statements are correct. > X > John Evers > *************************** ==== >active components within the cable then convert the logic > levels to RS-232 signal levels. Problem is they did not supply the > power necessary to drive these active components. Instead, they are > stealing power from the modem control lines of the RS-232. This is > fine as long as you 1) use the supplied cable, 2) don't use an > additional cable extender that does not support the modem control > lines and 3) connect your 48GII to a PC using HP's driver software > which will toggle either the DTR or RTS lines of the PC to the proper > levels. I have to admit that is a kludgey way to do it - fine for PC's, but quite unreliable for anything else. to mind. If you give it an external 5V supply. it converts CMOS to RS232 and back again. Is there any chance you can post pictures of the 48gii's PCB? I'd like to compare it to the 49G, and see exactly where this serial line comes from. thanks, Al ==== > Took one for the team. My suspicions were realized as I cut open the > molded plastic cover on the D-sub end of my brand new (now > non-returnable) 48GII cable and dug through the potting material to > see active components hidden underneath. Looks like (in order to save > cost/power???), HP decided not to include a functioning RS-232 port on > the unit itself. Instead, they have included the RS-232 transceivers > in the custom cable they supply with each unit. They simply drive the > logic levels of the uP UART out of the connector on the back of the > unit. The active components within the cable then convert the logic > levels to RS-232 signal levels. Problem is they did not supply the > power necessary to drive these active components. Instead, they are > stealing power from the modem control lines of the RS-232. This is > fine as long as you 1) use the supplied cable, 2) don't use an > additional cable extender that does not support the modem control > lines and 3) connect your 48GII to a PC using HP's driver software > which will toggle either the DTR or RTS lines of the PC to the proper > levels. > > If you are attempting to connect the 48GII to any other RS-232 device > you will need to make sure that it can supply the signals necessary to > power the cable circuitry. This will probably include making sure it > supports the modem control lines (many current RS-232 devices do not) > and making sure the device's firmware/software drivers keep the modem > control signals in a state that keep the cable powered properly. This > may require a rewrite of the device's serial drivers. In some cases > (such as a modem that uses these lines for it's operational > functionality), you may be out of luck. > > Bob > > communicating with serial devices! What could they have been thinking?! The new calculators can't print and can't communicate. As I said before, I think that the USB is a mistake - genuine RS-232 is MUCH more useful. Another mistake is have made a 33 instead of a 43. Martin Cohen ==== what kind of level of complaining do we need to obtain in order for HP to address the lcd flickering problem some of us are having with a patch? i like the calculator for the most part, but the flickering LCD really makes the calculator seem cheap. (can't do much to remedy the hollow feel of the calculator itself, except maybe take the case apart and add balast). does HP normally jump on these kinds of issues, or is it going to be a long time before we see some kind of patch? if they are slow to correct this problem, i'm seriously thinking about taking the 49G+ back and replacing it with a 49G. does the 49G flicker?!? thanks. gc ==== I have just gotten my 49+ and before that owned the 48GX. What a disappointment the manuals are. I started with the HP45 then 55, then 65, then 34C, then 97, then 42S, then 41CX, then 48GX. The manuals included with these machines had to be the envy of the entire industry. They were humorous, succinct, fabulous explanations, spelling and grammar syntax-free, and just plain pleasurable to use. I sent a missive to HP about the quality of the user's manual (in print) and the User's Guide (in pdf) in the package. Is it just me? I see errors on every page! Either illustrations don't match keystroke examples or the English is just not very good. I don't even see a section on I/O procedures. I am getting that info from the 49 users guide as nothing whatsoever is in the 49+ docs. What else is missing? Did somebody in the China write this thing? I asked HP to do a re-write in the style and penache of yore and send a FREE copy to every registered user! What a dreamer! Will they do it? I don't think so either. I guess we have to wait for some enterprising manual-writer to do the job HP should have done. And pay for something that we should not have had to do in the first place! Corvallis section where are you? Comments? Mike Bloch ==== This morning I found a corrupted entry in my 49g+ HOME directory. It used to be a directory with quite a few programs in, but is now claased as an external of size 3. The name used to be 6 letters long but many characters now follow this, including one ot more null characters. I cannot delete this entry using any tools I know of - PURGE from command line or from filer (in fact Filer goes to sleep trying to purge this). Also using the USB, I cannot copy /move/delete it from a PC. Renaming is not possible. I don't know how to selectively format the HOME directory probably a silly question. Is there a way to rest the whole calculator to tabula rasa with all user storage blank? No, don't tell me to use a hammer reset instead of a paper-clip reset. - Tony #79 You have dialed an imaginary number. Rotate 90 degrees and try again. ==== The 2 manuals for the HP 49G+ are missing 2 rather important subjects from its contents: the I/O functions of the calculator and a complete commands reference. I guess they are explained in the 49G user guide and in the downloadable (www.hpcalc.org) 49G commands reference, respectively. But it's just that the manuals that came with the 49G+ doesn't explain anything about transferring files or printing the display, and it's also missing the commands reference section which would explain all of the 49G+'s functions in its entirety. Dave ==== I'm thinking that with a suitable SYSRPL or Saturn program (driver) the hp49g+ might be able to use the FTDI USB to RS232 hardware adapter to implement the needed (missing) RS232 feature. Maybe by using their D2XX v2.0 driver techology adapted to the hp49g+? Would this work? Anyone with experience in this willing to comment? ==== I ran my bogomips program on the hp49g+ and found ~1.2 Bogomips :) (as opposed to ~0.3 Bogomips on hp48g/49. For more details about bogomips, just google bogomips mini-howto Samuel Thibault ==== > I ran my bogomips program on the hp49g+ and found ~1.2 Bogomips :) (as > opposed to ~0.3 Bogomips on hp48g/49. Oops, I forgot to turn the screen off. This gives ~1.335 Bogomips :) Samuel Thibault ==== The USB driver on the HP CD I got with my 49G+ doesn't run under NT. This was confirmed by HP tech support. Anyone know where I can get a driver, or has anyone used a USB/serial converter to communicate COM1 or COM2?