HP-13 > Yes it can easily be done. You must use the IrDA, and you can obtain a > stand alone adapter or use the built in IrDA on your laptop. If you > use the IrDA on you laptop. Set your BIOS to IrDA standard not FIR. > You will need to obtain a copy of IrCOMM2K for free on the internet, > and it will let you set your laptops IrDA to be a com port on a > Windows computer. I use Hyperterminal for the transfer. There are two > types of stand alone IrDA adapters that you can buy. One type is > intelligent ie. you just plug it into a standard DB-9 serial port > and it works. This type can be used with radio modems and other non > Windows type devices. The second type requires Windows drivers, and > only functions with a PC. > Now the bad news... I think there is a bug in the IrDA software on > the 49G+. The following comments are only relevant to larger files, > but this issue is extremly important to me. On downloads everything > goes rather well, I only see a few retries on the 49G+ display, but on > the PC end I see a rather large number (18 - 25)resent packets show > up, but the transfer usually being succesful. Now sending back to the > 49G+ is much more of a problem. On small files you will be O.K., but > too many retries occour and failure results as the files become > larger. About 80% of the time on large file uploads, the 49G+ will > completly (paper clip time) lock up. > > John Evers If you have the option, try transferring in dark settings. Some sensors have trouble with interference from other sources radiating at frequencies that you may not realize. It's not likely, but not impossible. Sharp's IS1U60 receiver is noted on http://www.lirc.org/receivers.html is one example, although at more of an extreme. The faster data needs to go, the more sensitive the communication link becomes. ==== Pardon me for intruding, but Sony fixed my broken television IR remote control by having me change the light bulb in the lamp (about 12 ft. away from both the TV and the remote) from the energy saving fluorescent-type bulb to the old fashioned incandescent bulb. He was very amused at the number of TVs his employer was getting back because of broken remotes. I doubt that his employer was. In all IR transfer functions here at my home I try to make sure that I have only beautiful, yellow, incandescent bulbs in the lamps. Jim : If you have the option, try transferring in dark settings. Some : sensors have trouble with interference from other sources radiating at : frequencies that you may not realize. It's not likely, but not : impossible. Sharp's IS1U60 receiver is noted on : http://www.lirc.org/receivers.html is one example, although at more of : an extreme. The faster data needs to go, the more sensitive the : communication link becomes. ==== Warning: product endorsement follows. > I try to make sure that I have only beautiful, > yellow, incandescent bulbs in [my] lamps. Ever tried GE's Reveal light bulbs? The bulbs themselves look pale blue, but the light that they make is true white light; they're essentially incandescent bulbs with built-in white-balance filters. They make everything appear as colorful as they do in sunlight. I use them exclusively, and highly recommend them. They're available wherever light bulbs are sold. More info here: http://www.gelighting.com/na/reveal/media_product.html -Joe- Dislaimer: I don't work for GE, HP, or anybody else. ==== If you don't work for anyone it must be hard to make a living... ;) > -Joe- > Dislaimer: I don't work for GE, HP, or anybody else. ==== > Warning: product endorsement follows. > > > I try to make sure that I have only beautiful, > > yellow, incandescent bulbs in [my] lamps. > > Ever tried GE's Reveal light bulbs? The bulbs themselves look pale > blue, but the light that they make is true white light; they're > essentially incandescent bulbs with built-in white-balance filters. > They make everything appear as colorful as they do in sunlight. I use > them exclusively, and highly recommend them. They're available > wherever light bulbs are sold. More info here: > http://www.gelighting.com/na/reveal/media_product.html > > -Joe- > Dislaimer: I don't work for GE, HP, or anybody else. Let there be light! You work for JHWH (just say Adonai)! AND don't forget *that*! Brother-Peter ==== Greetings! I recently received my 49+, on the whole, a very nice machine. I have an existing 64mb MMC card made by Memorex. I used to use the card in my kept the card partitioned and formatted with ext2, however I partitioned and reformatted it on the Zaurus with mkfs.vfat /dev/... prior to using it in the 49+. After I upgraded through USB I inserted the card and nothing could be stored or recalled. I believe it is because vfat is fat16 with long filenames. I reformatted the card on the hp49+ and now I can store and recall objects without any problems. However, when I mount the card on the Zaurus all I get is garbage (many files/directories with incomprehensible names, symbols etc.) when reading the directory tree in the file manager and bash. I've tried mounting it by explicitly defining the file system type in fstab (vfat, dos, msdos) however each yields the same result. The card still works fine after I insert it back into the 49+ with all the objects I put there earlier. I don't have a MMC or SD card reader for my pc. Perhaps the solution lie in that. But I don't see why I can't read it on my PDA. Whether this is due to my lack of know. Does having a partition table affect how the 49+ reads the card? I heard that most memory cards come without partition tables from the factory, but thats just what I heard. For those interested, here is some information on the Zaurus: http://www.sharpelectronics.com/products/ModelLanding/0,1058,698,00.html Any help or experiences would be greatly appreciated. -Tony ==== demosthenes meinte >Greetings! > >I recently received my 49+, on the whole, a very nice machine. I have an >existing 64mb MMC card made by Memorex. I used to use the card in my >kept the card partitioned and formatted with ext2, however I partitioned >and reformatted it on the Zaurus with mkfs.vfat /dev/... prior to using >it in the 49+. After I upgraded through USB I inserted the card and >nothing could be stored or recalled. I believe it is because vfat is >fat16 with long filenames. I reformatted the card on the hp49+ and now I >can store and recall objects without any problems. However, when I mount >the card on the Zaurus all I get is garbage (many files/directories with >incomprehensible names, symbols etc.) when reading the directory tree in >the file manager and bash. I've tried mounting it by explicitly defining >the file system type in fstab (vfat, dos, msdos) however each yields the >same result. The card still works fine after I insert it back into the >49+ with all the objects I put there earlier. I don't have a MMC or SD >card reader for my pc. Perhaps the solution lie in that. But I don't see >why I can't read it on my PDA. Whether this is due to my lack of >know. Does having a partition table affect how the 49+ reads the card? I >heard that most memory cards come without partition tables from the >factory, but thats just what I heard. > >For those interested, here is some information on the Zaurus: >http://www.sharpelectronics.com/products/ModelLanding/0,1058,698,00.html > >Any help or experiences would be greatly appreciated. I think some time ago I had a similar problem with MOD disks. The difference was to take care of wether the disk was partitioned or not. That is wether it was treated like a hard disk - with partition information - or like a floppy without partition information. The result was very similar to your observation. If you apply fdisk -l and the card shows a wealth :-) of various unknown partitions, then you should be pretty close. With my MOD a partitioned disk would be mounted as e.g /dev/sdb1 while the unpartioned disk would be mounted as /dev/sdb thus mounting the entire disk rather than a non existing partition. Perhaps your situation is similar, just give it a try. Gru§ G.9fnter schink -- ==== This could be useful if it were possible. ==== Yes. I printed mine at two pages per side, duplex, then had it bound at the local Kinko's for less than $5.00. It is quite readable at this size, and fits nicely in a backback. I can lay it flat on a desk and see four pages of content at once. > version of the user manual? I know there is a PDF version in the CD > > > Rod ==== I often end up with a function on the stack that I would like to see plotted. The plot environment on the HP49 expects me to enter the plot application and then enter the desired function. An old RPN fart like myself, often want to operate on things that I already have entered, and get frustarated when I dive into plot, just to realize that I can't take the function with me. This is easy, I suppose: what commands do I need to take a function from stack level 1 and make the best of it as a plotted function? STEQ? What more? Or is there function within the the plot environment that pulls a function from the stack into the application? Per. HP21->HP25->HP29C->HP28->HP48SX->HP49 (One could come to the conclusion that I should know how to use these gizmos by now...) ==== If you just want to see it, you could make a little program: << DUP 'EQ' STO ERASE DRAX DRAW 0 WAIT DROP >> It'll just take the symbolic object from level 1 and store it into 'EQ', erase the old graph grob, plot the function, and then wait for you to press a key. You'll be left with the function on level 1 as before (unless you press ON, then you get the function on level 2 and a 0 on level 1). If you still want the GUI, just do << DUP 'EQ' STO >> and then invoke the GUI as usual. These are for the HP48, btw. I assume they'll work on the 49. If you wanted to use your 48SX, you could download PowerPlot: http://www.exzodia.net/hp48/software/index.shtml It does what you want by default and there's a good chance that it'd still be faster running on an SX than the 49 is. I've never fully tested PowerPlot on an SX, so I cant say how well it will work.. but I believe it should be okay. Aaron >I often end up with a function on the stack that I would like to see >plotted. The plot environment on the HP49 expects me to enter the plot >application and then enter the desired function. An old RPN fart like >myself, often want to operate on things that I already have entered, >and get frustarated when I dive into plot, just to realize that I >can't take the function with me. > >This is easy, I suppose: what commands do I need to take a function >from stack level 1 and make the best of it as a plotted function? >STEQ? What more? > >Or is there function within the the plot environment that pulls a >function from the stack into the application? > >Per. >HP21->HP25->HP29C->HP28->HP48SX->HP49 >(One could come to the conclusion that I should know how to use these >gizmos by now...) > > ==== You cold always press the 'HIST' when entering a function... This lets you pull things off the stack. cheers, Al http://alpage.ath.cx/hptute/ ==== > > > > Please, where could someone buy the 49G+ in London? > > I will be spending the next year here doing an MSc in Imperial College, > > and I think that it would be very helpful to have a new calc, > > although my 49G and 32SII are always there when I need them. > > > > > > Alex Markatis > > Civil Engineer > Probably in a month or two Tottenham Ct Rd will be swarming with them. I ll be waiting for them! > Doesnt the HPCC have meetings at Imperial (read up on comp.sys.hp48 > and visit the freqntuly mentioned web sites to learn more). I think > www.hpcc.org is the place you want. Yes, I know about HPCC, but I hadn t found the time to search this until now. But probably one of these days I will try to see what happens with HPCC, although I often wonder if a beginner like me can survive among those HP gurus! :-) > good luck! Alex ==== Bonjour, Perso, je commande pas de mat.8eriel ailleurs qu'en CEE. Trop risqu.8e avec les pb de douanes et de TVA... Sans parler de guarantie et service apr.8fs vente. En plus, je pr.8ef.8ere payer ma TVA en France ou en tout cas en CEE. Question de principe d'un habitant de la vieille Europe :-) Ce que j'aimerai : un bon vieux magasin pas du tout virtuel ou je peux toucher la b.90te vers Lille Gilles 1g37giu.m7uyd4xhvg7oN%thomas-ml@deniau.org... > > > > Pouvez pr.8eciser les URL de ces deux sites ? > > http://www.educalc.net et http://www.samsoncables.com > > Attention aux frais de douane... > > Beware, the customs may charge *a lot*... > > -- > Thomas Deniau > Unix is user friendly. It's just selective when choosing friends. ==== I too like the 49+, but disagree that the keyboard issue is minor. If it were just a matter of using greater force to get the response, ok. But the fact that the key clicks down, yet no action is taken is a serious error for a calculator. Whilst HP should be given credit for bothering to upgrade the HP49 to this faster model, they have added no real improvements as far as the algorithms go - for example an improved plot or CAS. The least they could do is make sure the input/output interface works ok. At least they got the output right :-) dave > Whether you should buy it is really up to you... but the few issues that > exist are minor. The keyboard problems are really just a matter of > those of us who are used to the softer 48G/GX keyboard getting used to > the 49G+ keyboard, which is stiffer. Sure, it's annoying, but it's > nothing to worry about, just something to get used to. The screen > glitches have no actual impact, just a little flicker every so often. > Other than that, the only glitches I know of are software errors, the > most serious of which (battery drain) was solved in the latest ROM. I > wouldn't worry about it from a quality perspective... > > > Should I buy the 49g+? > > I want to, but there seem to be a lot of bugs and a lot of weird > > things, and other not so good comments. I understand products have > > bugs from time to time, I mean, usually when you buy a software it was > > some bug here or there, and then a patch of update fixes it. > > But it might be different with a calculator (I don't know a lot about > > them, and specially about HP calcs), but I bought the TI's voyage 200 > > when it came out and it has done wonders for me, but I personaly wan > > to use a HP calculator, and well, I want this 49g+ to do the trick. > ==== I might be inclined to agree with you if I could get mine to click without registering, but when I try I cannot... the only keyboard issues I have other than not pressing hard enough are that I sometimes accidentally double-press the alpha key, locking it on (minor), and I often hit enter instead of + since the arithmetic keys are shifted up from where the 48 had them, which is quite annoying, but just a matter of not being used to the keyboard layout. The only way I have found of pressing a key and getting it to only go partially down is to press it slowly, at an angle from the top... If I press the key straight on or from the side/bottom, or even quickly from the top, it either clicks and registers or doesn't move. If I press slowly from the top, I can get it to slowly press but there's no click as it doesn't go down suddenly... it's possible to press a key only partially this way, but I have yet to duplicate a click with no keypress registering after several minutes of intentionally trying to do so. > I too like the 49+, but disagree that the keyboard issue is minor. > If it were just a matter of using greater force to get the response, ok. But > the fact that the key clicks down, yet no action is taken is a serious error > for a calculator. > Whilst HP should be given credit for bothering to upgrade the HP49 to this > faster model, they have added no real improvements as far as the algorithms > go - for example an improved plot or CAS. The least they could do is make > sure the input/output interface works ok. At least they got the output right > :-) > > dave > > > > > >>Whether you should buy it is really up to you... but the few issues that >>exist are minor. The keyboard problems are really just a matter of >>those of us who are used to the softer 48G/GX keyboard getting used to >>the 49G+ keyboard, which is stiffer. Sure, it's annoying, but it's >>nothing to worry about, just something to get used to. The screen >>glitches have no actual impact, just a little flicker every so often. >>Other than that, the only glitches I know of are software errors, the >>most serious of which (battery drain) was solved in the latest ROM. I >>wouldn't worry about it from a quality perspective... >> >> >>>Should I buy the 49g+? >>>I want to, but there seem to be a lot of bugs and a lot of weird >>>things, and other not so good comments. I understand products have >>>bugs from time to time, I mean, usually when you buy a software it was >>>some bug here or there, and then a patch of update fixes it. >>>But it might be different with a calculator (I don't know a lot about >>>them, and specially about HP calcs), but I bought the TI's voyage 200 >>>when it came out and it has done wonders for me, but I personaly wan >>>to use a HP calculator, and well, I want this 49g+ to do the trick. >> > > ==== I should add mine is a CN331 model that came with ROM 1.20. > I might be inclined to agree with you if I could get mine to click > without registering, but when I try I cannot... the only keyboard issues > I have other than not pressing hard enough are that I sometimes > accidentally double-press the alpha key, locking it on (minor), and I > often hit enter instead of + since the arithmetic keys are shifted up > from where the 48 had them, which is quite annoying, but just a matter > of not being used to the keyboard layout. > > The only way I have found of pressing a key and getting it to only go > partially down is to press it slowly, at an angle from the top... If I > press the key straight on or from the side/bottom, or even quickly from > the top, it either clicks and registers or doesn't move. If I press > slowly from the top, I can get it to slowly press but there's no click > as it doesn't go down suddenly... it's possible to press a key only > partially this way, but I have yet to duplicate a click with no keypress > registering after several minutes of intentionally trying to do so. ==== I wonder how widespread this problem really is? I have had my 49G+ for three days and have absolutely no problem with missed entries. The keys press a little harder than I am used to, but when they click, they register. My John > I too like the 49+, but disagree that the keyboard issue is minor. > If it were just a matter of using greater force to get the response, ok. But > the fact that the key clicks down, yet no action is taken is a serious error > for a calculator. > Whilst HP should be given credit for bothering to upgrade the HP49 to this > faster model, they have added no real improvements as far as the algorithms > go - for example an improved plot or CAS. The least they could do is make > sure the input/output interface works ok. At least they got the output right > :-) > > dave > > > > > > Whether you should buy it is really up to you... but the few issues that > > exist are minor. The keyboard problems are really just a matter of > > those of us who are used to the softer 48G/GX keyboard getting used to > > the 49G+ keyboard, which is stiffer. Sure, it's annoying, but it's > > nothing to worry about, just something to get used to. The screen > > glitches have no actual impact, just a little flicker every so often. > > Other than that, the only glitches I know of are software errors, the > > most serious of which (battery drain) was solved in the latest ROM. I > > wouldn't worry about it from a quality perspective... > > > > > Should I buy the 49g+? > > > I want to, but there seem to be a lot of bugs and a lot of weird > > > things, and other not so good comments. I understand products have > > > bugs from time to time, I mean, usually when you buy a software it was > > > some bug here or there, and then a patch of update fixes it. > > > But it might be different with a calculator (I don't know a lot about > > > them, and specially about HP calcs), but I bought the TI's voyage 200 > > > when it came out and it has done wonders for me, but I personaly wan > > > to use a HP calculator, and well, I want this 49g+ to do the trick. > > > > ==== > I wonder how widespread this problem really is? I have had my 49G+ for three > days and have absolutely no problem with missed entries. The keys press a > little harder than I am used to, but when they click, they register. My > Fantastic, John! Maybe you have a revised model? CN331...and CN334...are from a different week, so there is hope that the kb issue was just the first batch (and even not all of them) Any others with ROM 1.22 and better kb? ==== thanks for the comments >Whether you should buy it is really up to you... but the few issues that >exist are minor. The keyboard problems are really just a matter of >those of us who are used to the softer 48G/GX keyboard getting used to >the 49G+ keyboard, which is stiffer. Sure, it's annoying, but it's >nothing to worry about, just something to get used to. The screen >glitches have no actual impact, just a little flicker every so often. >Other than that, the only glitches I know of are software errors, the >most serious of which (battery drain) was solved in the latest ROM. I >wouldn't worry about it from a quality perspective... > >> Should I buy the 49g+? >> I want to, but there seem to be a lot of bugs and a lot of weird >> things, and other not so good comments. I understand products have >> bugs from time to time, I mean, usually when you buy a software it was >> some bug here or there, and then a patch of update fixes it. >> But it might be different with a calculator (I don't know a lot about >> them, and specially about HP calcs), but I bought the TI's voyage 200 >> when it came out and it has done wonders for me, but I personaly wan >> to use a HP calculator, and well, I want this 49g+ to do the trick. ==== I agree. I used a 49G for years, and dealt with its share of problems (most notably its slowness). The 49g+ addressed all of my major complaints with the 49G. The keyboard is not bad, and I love the rest of it features. I am very satisfied, and I think once the initial compaints slow down, you'll find a lot of other happy g+ owners. For what it's worth, I recommend it. I'm not going back to my 49G without a fight. I also own a TI-89, but basically never use it. Once you become comfortable with RPN and spend some time with the HP, I think you'll like it. Do spend some time, however, as it is very easy to get frustrated early. You'll miss out if you quit early. > thanks for the comments > > > > >Whether you should buy it is really up to you... but the few issues that > >exist are minor. The keyboard problems are really just a matter of > >those of us who are used to the softer 48G/GX keyboard getting used to > >the 49G+ keyboard, which is stiffer. Sure, it's annoying, but it's > >nothing to worry about, just something to get used to. The screen > >glitches have no actual impact, just a little flicker every so often. > >Other than that, the only glitches I know of are software errors, the > >most serious of which (battery drain) was solved in the latest ROM. I > >wouldn't worry about it from a quality perspective... > > > >> Should I buy the 49g+? > >> I want to, but there seem to be a lot of bugs and a lot of weird > >> things, and other not so good comments. I understand products have > >> bugs from time to time, I mean, usually when you buy a software it was > >> some bug here or there, and then a patch of update fixes it. > >> But it might be different with a calculator (I don't know a lot about > >> them, and specially about HP calcs), but I bought the TI's voyage 200 > >> when it came out and it has done wonders for me, but I personaly wan > >> to use a HP calculator, and well, I want this 49g+ to do the trick. > ==== > I also own a TI-89, but basically never use it. Once you become comfortable > with RPN and spend some time with the HP, I think you'll like it. I presume you know about the RPN interface for the TI-89? -- Bhuvanesh ==== > Similar to what you described > in another post, I'm getting some funny looking files when displayed > on the 49G+ filer but everything appears normal on my Mac. These files are normally invisible to the user (of the Mac). Yoe may safely upgrade the ROM from the SD-card, but you must not touch those Mac-files, anyway. > Should I go > ahead and attempt the ROM upgrade anyway or will this possible do > something serious to the calculator? Just go ahead! Michael -- -= Michael Hoppe , =------ ==== > Just go ahead! > Well I did and my 49G+ locked up each time displaying a message that says something like No Card, Error in reading card. I'm not sure what to try next. The card seems to read fine on my Mac and on a neighbor's PC (which said that the card was FAT formatted). I guess I'll keep playing around with this. Tom ==== I'm a neophyte at programming the 49g. I have done simple functions and some keyboard assignments so far. I'd like to be able to take a list of something. Lets say materials density. object1 number object2 number. object3 number etc. Have it stored in a variable (but it would be a function I think) press a softkey, my list pops up, scroll the list, press enter (or another softkey) and have the number appear on the top of the stack.. Anyone have a template for me to follow for doing this in user rpl? Or which help manual I need to read. Perhaps it's too advanced for an afternoon's fun? Or already been done in a way that I can easily use? Al... ==== You can create the next two (2) little programs in UserRPL, as follow: Ç -3 SF 2 FIX MODIFY DATA { { D1 Data D1 0 } { D2 Data D2 0 } { D3 Data D3 0 } { D4 Data D4 0 } { D5 Data D5 0 } { D6 Data D6 0 } { D7 Data D7 0 8 } { } } { 2 } mike DUP IF INFORM THEN OBJ-> DROP 'D7' STO 'D6' STO 'D5' STO 'D4' STO 'D3' STO 'D2' STO 'D1' STO 'D1' RCL 'D2' RCL 'D3' RCL 'D4' RCL 'D5' RCL 'D6' RCL 'D7' RCL 7 ->LIST 'mike' STO CLLCD STD END È Save this with a name, p.e. : MODIFYDATA Now create a other .p.e. : NEWDATA Ç -3 SF 2 FIX NEW DATA { { D1 Data D1 0 } { D2 Data D2 0 } { D3 Data D3 0 } { D4 Data D4 0 } { D5 Data D5 0 } { D6 Data D6 0 } { D7 Data D7 0 8 } { } } { 2 } { } DUP IF INFORM THEN OBJ-> DROP 'D7' STO 'D6' STO 'D5' STO 'D4' STO 'D3' STO 'D2' STO 'D1' STO 'D1' RCL 'D2' RCL 'D3' RCL 'D4' RCL 'D5' RCL 'D6' RCL 'D7' RCL 7 ->LIST 'mike' STO CLLCD STD END È Note: First charge the DATA with NEWDATA; next step, change the values with MODIFYDATA Miguel Angel CAPORALINI HERK (M.A.C.H.) **************************************************************************** **** > I'm a neophyte at programming the 49g. I have done simple functions and some > keyboard assignments so far. > > I'd like to be able to take a list of something. Lets say materials density. > > object1 number > object2 number. > object3 number > etc. > > Have it stored in a variable (but it would be a function I think) > > press a softkey, my list pops up, scroll the list, press enter (or another > softkey) and have the number appear on the top of the stack.. > > Anyone have a template for me to follow for doing this in user rpl? Or > which help manual I need to read. > > Perhaps it's too advanced for an afternoon's fun? Or already been done in a > way that I can easily use? > > Al... ==== Unfortunately HP does all the work here, so I doubt it'll fill an afternoon:-). dave {{object1 number}{object2 number}{object3 number}} @L1 Stack /<< title SWAP 1 @number that gets initial highlight - if =0 no object can be selected, view only CHOOSE DROP />> number @L1 Stack > I'm a neophyte at programming the 49g. I have done simple functions and some > keyboard assignments so far. > > I'd like to be able to take a list of something. Lets say materials density. > > object1 number > object2 number. > object3 number > etc. > > Have it stored in a variable (but it would be a function I think) > > press a softkey, my list pops up, scroll the list, press enter (or another > softkey) and have the number appear on the top of the stack.. > > Anyone have a template for me to follow for doing this in user rpl? Or > which help manual I need to read. > > Perhaps it's too advanced for an afternoon's fun? Or already been done in a > way that I can easily use? > > Al... > > ==== I've made a few changes made to the programs, directory structure etc on the Hp49G+ and wanted to (as painless as possible) transfer to the Hp49G. First problem is that I can't do a backup (via Conn4x) from the G+ - it works fine from the G though. Don't know if I'm making a mistake, or it's a general issue. Anyway then tried another approach - copied a HP49G+ archived backup from SD card to PC - its fast! The problem then was to get it to the G. Tried a manual download (XRecv) and drag-and-drop, via Conn4x. In both cases it was transfered to the calc as a directory structure - I suppose thats whats supposed to happen. Still it messed up my last idea for doing this - any others would be gratefully xreceived. dave ==== Apart from speed? No. > For those who have the 49G with it's skimpy manual, can the 49G Plus manual > be used as a substitute? Are there significant differences in the operation > of the 49G Plus and 49G? Thx. > > > > ==== > > > > > Some people have a slight problem with the keyboard. They can learn to > > > press a little harder. > > > > > > > > I don't agree. This keyboard causes a lot of mistakes and slows down > X > ??? > I tested some calculators (49G+) on users with various backgrounds > and NONE of them hit the keys so that there would be mistakes > Looks like only an old HP 48 user is going get some misses old HP 48 User yesyesyes HP 48 SX 1989 > I should try more people - you need a light touch. > Hmmm....maybe because the situation and the calc is new to them > people tend to press the keys carefully? > This needs more testing > PS: Only some calcs have these slight problems with kb > BUT almost every *old* hp user is going to have some trouble > so I agree that it has to be fixed.... I think it could be fixed, so it should. once I paid 300$ for the HP 48 SX. I don't regret one of them. Stefan ==== Dear Vali, It is easy to test the calculator. Just follow these steps in order. 1. Install batteries. Be careful to use the correct polarity as marked. 2. Turn power on by pressing the ON switch. It's in the lower left corner. 3. With the unit on, a. Press and hold the ON key with your left hand. b. Simultaneously press and hold the F1 and F6 keys with your right hand. These are the keys at both ends of the top row of keys. c. You now have all three keys pressed down at the same time. 4. Release both the F1 and F6 keys at the same time. 5. Now release the ON key. 6. You will see a menu of choices appear in the display. It looks like this: 1. LCD 2. Key 3. Flash 4. SRAM 5. IRDA 6. USB 7. Card 8. Power 9. Print 0. Buzzer +. IR -. EMI A. AUTOTEST ON. Reboot 7. In the list above, 1 means Press the 1 key to test the LCD Display. 2 means Press the 2 key to test all the keys in the keyboard. 3 means Press the 3 key to test the built-in flash ROM. 4 means Press the 4 key to test the built-in SRAM. 5 means Press the 5 key to test the Infrared communications. (This cannot be used with only one machine.) 6 means Press the 6 key to test the USB communications. (This only works if you have installed the USB cable between the 49g+ and a PC. You will also have to have installed the USB drivers.) 7 means Press the 7 key to test the SD card. (This only works with an SD card inserted into the slot at the bottom of the 49g+.) 8 means Press the 8 key to test power consumption. (I think this is what it means. My machine passed this test even with ROM image version 1.20.) 9 means Press the 9 key to test communications with the printer. (This will only work with a printer connected.) 0 means Press the 0 key to test the beep. (This also has a section to adjust the pitch of the beep. + means Press the + key to test the IR port. (I don't know the difference between this test and #5) - means Press the - key to test for electro-magnetic interference(?). (This is intriguing to a ham radio operator, but I don't know how this test works.) A means Press the A switch to run all these tests automatically. ON means Press the ON key to Reboot the calculator. (This only works to reboot after running the manual tests. It doesn't work after running the A autotests. I'm sorry to have to admit this, but I cannot make the calculator stop running the autotests once I have them started. The only thing that works for me is to open the battery compartment and pull out one of the batteries. Since I'm a mechanical engineer, this should surprise nobody. We always try to come up with a solution like this one. Here is one other thing you might want to know. Here in the USA the calculator and its book, manual, batteries, etc. are enclosed in a surround of very heavy clear plastic, not a box. You can see the items inside, but you cannot touch them. The only way to open the surround is to use either a strong knife or a very sharp and strong pair of scissors. The stores must do that here in the US because theft is such a big problem. If your friend wants to try out the calculator before he buys it for you, I am almost certain that he will not be permitted to do so because he would have to cut the heavy plastic open first to get the calculator out. The way some of us solve this problem here is first to buy the calculator, then, if the store will permit you to do it, open the calculator there in the store, put batteries into it, and test it there. Most of us, though, buy the calculator, take it home, open it and run the tests there. Only if we are then dissatisfied with the calculator do we take it back to the store and request either a replacement or a refund of the purchase price we have paid. If it is within two weeks after buying the calculator, we do not accept a request from the store to return it to the manufacturer. We insist that the store replace the defective unit with other units they have in the store. Best of luck to you, Vali. Jim Chumbley ____________________________________ to tell this guy to test the calculator : before I get it. I'm interested in the display problem, keyboard, : flickering and if there are other bugs, which are and how to test : against. ==== Is there anyway to test the CPU speed. It was one of the self tests on Dave ==== To get to the self test menu press the 'ON' key and the 'D' key, marked '2D/3D' above it in blue. The ON-A-F combination results in a memory clear. To exit the test menu press 'ON' & 'C'. rdb. > b. Simultaneously press and hold the F1 and F6 keys with your > right hand. > These are the keys at both ends of the top row of keys. > c. You now have all three keys pressed down at the same time. > 4. Release both the F1 and F6 keys at the same time. > 5. Now release the ON key. ==== NEVER use ON-D on the 49g+ unless you need to format the SD card! All of its tests are more easily accessed via ON-F. That's right; there are TWO self-test menus built into the hp49g+, not only ON+D but also ON+F. They are as follows: ON+D menu --------- 1. LCD 2. KEY 3. FLASH 4. SRAM 5. IRDA 6. USB 7. CARD 8. BUZZER 9. FORMAT (Formats the SD card w/o warning!!!) 0. AUTOTEST (same as pressing ON+E) The only way to exit this menu is ON+C. ON-F menu --------- 1-7: same as above 8. POWER 9. PRINT 0. BUZZER +. IR -. EMI A. AUTOTEST ON. REBOOT (no need to press ON-C to exit). A complete write-up about all these would be greatly appreciated. I cannot figure out the difference between the IRDA and IR tests, and the USB and EMI tests. -Joe- ==== > NEVER use ON-D on the 49g+ unless you need to format the SD card! All > of its tests are more easily accessed via ON-F. > > That's right; there are TWO self-test menus built into the hp49g+, not > only ON+D but also ON+F. They are as follows: > > ON+D menu > --------- > 1. LCD > 2. KEY > 3. FLASH > 4. SRAM > 5. IRDA > 6. USB > 7. CARD > 8. BUZZER > 9. FORMAT (Formats the SD card w/o warning!!!) > 0. AUTOTEST (same as pressing ON+E) > The only way to exit this menu is ON+C. > > ON-F menu > --------- > 1-7: same as above > 8. POWER > 9. PRINT > 0. BUZZER > +. IR > -. EMI > A. AUTOTEST > ON. REBOOT (no need to press ON-C to exit). be a repeating self-test like on the 49G, but now that I think about it, the ON- operations changed from the 48 series to the 49G, so we shouldn't be too surprised at more changes. > A complete write-up about all these would be greatly appreciated. Agreed! > I > cannot figure out the difference between the IRDA and IR tests, Maybe IR would be for the Serial IR as used for communications between the 48 series calculators, or for the Red Eye IR for printing to the 82240A/B? > and > the USB and EMI tests. Maybe EMI stands for Electro-Magnetic Interference? Could it be that it checks for sensitivity to EMI, or whether it causes too much EMI, or perhaps both at once? -- James ==== : To get to the self test menu press the 'ON' key and the 'D' key, : marked '2D/3D' above it in blue. The ON-A-F combination results in a : memory clear. To exit the test menu press 'ON' & 'C'. ____________________________ The ON A F combination results in a memory clear only when you release the ON key first followed by the A F combination. Interestingly the No thank you very much response to the question you are asked about making an attempt to recover memory should be a depression of the F6 key, but THIS DOES NOT WORK. F6 is ignored until something else, say F5 is pressed first. Apparently the F5 means nothing to the machine and serves only the function of releasing the lock-out of the F6 key. The ON A F works as described in my earlier postings if you first release the A F combination before releasing the ON key. the self-test and the effects of the ON F3(C) keys to kill the otherwise runaway self-test. Jim Chumbley ==== I have experimented with this and have found the following: ON-D calls up one self test menu with ten options, exit with ON-C. ON-F calls up another self test menu with 14 options, exit the menu with ON. Exit a test with ON-C. If you press ON-A-F, depending on the order in which they are pressed and the order in which they are then released, you may get the ON-F self test menu or the recover memory question, and depending on how the recover memory condition is 'invoked' F6 will clear it or it may require F5 then F6. Very strange. Perhaps someone else would care to confirm this behaviour? rdb. ==== > a. Press and hold the ON key with your left hand. > b. Simultaneously press and hold the F1 and F6 keys with your > right hand. > These are the keys at both ends of the top row of keys. > c. You now have all three keys pressed down at the same time. > 4. Release both the F1 and F6 keys at the same time. > 5. Now release the ON key. Hu? The autotest is accessed with On-A-F? Strange... > I'm sorry to have to admit this, but I cannot make the calculator stop > running the autotests once I have them started. The only thing that works > for me is to open the battery compartment and pull out one of the batteries. > Since I'm a mechanical engineer, this should surprise nobody. We always try > to come up with a solution like this one. Maybe you could try the On-C combination (try On-F3). -- ----- Dis tu nous l.89ches un peu ? C'est un sujet assez important pour qu'on en parle, peu importe le groupe. C'est .88 cause de gens comme toi si c'est la merde sur le net. Tes le.8dons de n.8etiqu.8equette, tu peux te les... OK? -+- Ghost in : Guide du Neuneu d'Usenet - Terrorisme neuneulectuel -+- ==== : : : > a. Press and hold the ON key with your left hand. : > b. Simultaneously press and hold the F1 and F6 keys with your : > right hand. : > These are the keys at both ends of the top row of keys. : > c. You now have all three keys pressed down at the same time. : > 4. Release both the F1 and F6 keys at the same time. : > 5. Now release the ON key. : : Hu? The autotest is accessed with On-A-F? Strange... *** As Wayne pointed out in a parallel posting with yours, ON F6 works in exactly the same way. No need to press the F1 and F6 keys together. : : > I'm sorry to have to admit this, but I cannot make the calculator stop : > running the autotests once I have them started. The only thing that works : > for me is to open the battery compartment and pull out one of the batteries. : > Since I'm a mechanical engineer, this should surprise nobody. We always try : > to come up with a solution like this one. : : Maybe you could try the On-C combination (try On-F3). ***Yes! You are exactly right. Both you and Wayne pointed this out for me. : : -- : ----- : Dis tu nous l.89ches un peu ? C'est un sujet assez important pour qu'on en : parle, peu importe le groupe. C'est .88 cause de gens comme toi si c'est la : merde sur le net. Tes le.8dons de n.8etiqu.8equette, tu peux te les... OK? : -+- Ghost in : Guide du Neuneu d'Usenet - Terrorisme neuneulectuel -+- : ==== > Dear Vali, > It is easy to test the calculator. Just follow these steps in order. > 1. Install batteries. Be careful to use the correct polarity as marked. > 2. Turn power on by pressing the ON switch. It's in the lower left > corner. > 3. With the unit on, > a. Press and hold the ON key with your left hand. > b. Simultaneously press and hold the F1 and F6 keys with your > right hand. > These are the keys at both ends of the top row of keys. > c. You now have all three keys pressed down at the same time. > 4. Release both the F1 and F6 keys at the same time. > 5. Now release the ON key. > 6. You will see a menu of choices appear in the display. It looks like > this: Are you sure about this keystroke sequence? On the HP48 those would be the clear memory keystrokes. Isn't it just ON F6 for the self-test? > I'm sorry to have to admit this, but I cannot make the calculator stop > running the autotests once I have them started. The only thing that works > for me is to open the battery compartment and pull out one of the batteries. > Since I'm a mechanical engineer, this should surprise nobody. We always try > to come up with a solution like this one. On the 48, ON C (I guess that would be ON F3 on the 49) stops the self-test. -- Wayne Brown (HPCC #1104) | When your tail's in a crack, you improvise fwbrown@bellsouth.net | if you're good enough. Otherwise you give | your pelt to the trapper. e^(i*pi) = -1 -- Euler | -- John Myers Myers, Silverlock ==== When I EXPAND the expression (A+B+C)^3, I obtain: C^3+3.C^2.B+3.C.B^2+B^3+(3.C^2+6.C.B+3.B^2).A+(3.C+3.B).A^2+A^3 Do you know how to: 1) expand the remaining parenthesis? 2) factorize for any given monom, e.g. A.B.C? Yves PS: I am running rom 1.19-6 on the HP49G. ==== > What about HP Solve algorithm?? I love it and wonder how it works Dec 79, Aug 80, May 83 and Aug 87. http://library.hp41.org/LibView.cfm?Command=List&CategoryID=7 ==== > Several of us worked on these, which are available free at the links > below. > > using the 49G+, ranging from solving differential equations to doing > base conversions. > > Students and educators > http://www.hp.com/calculators/graphing/49gplus/educators.html > > Scientists and Engineers > http://www.hp.com/calculators/graphing/49gplus/scientists.html > > Business & Finance > http://www.hp.com/calculators/graphing/49gplus/business.html > > Real Estate > http://www.hp.com/calculators/graphing/49gplus/realestate.html > > Hope these are useful. I view these as a good thing HP is > doing...since they didn't have to do these at all. If you find > > (Sorry if this ends up getting posted more than once...my news server > I usually post from has been really acting up). > > Gene Very nice! Martin Cohen ==== I understood that an IR transfer is possible between the 49G+ and the 48GX. I haven't been able to make it work. All I am trying to do is transfer a variable or program from one to the other. I have set both machines for Kermit (since GX won't do Xmodem) and IR. They are not able to connect. John ==== > I understood that an IR transfer is possible between the 49G+ and the 48GX. > I haven't been able to make it work. All I am trying to do is transfer a > variable or program from one to the other. I have set both machines for > Kermit (since GX won't do Xmodem) and IR. They are not able to connect. The 48G series can do Xmodem. See the XRECV and XSEND commands. -- James ==== Just an speculation: Could HP 49G+ use IRDA while 48G uses HP-IR (non IRDA compatible)? Can somebody confirm or deny it? James M. Prange escribi.97 en el mensaje > > I understood that an IR transfer is possible between the 49G+ and the 48GX. > > I haven't been able to make it work. All I am trying to do is transfer a > > variable or program from one to the other. I have set both machines for > > Kermit (since GX won't do Xmodem) and IR. They are not able to connect. > > The 48G series can do Xmodem. See the XRECV and XSEND commands. > > -- > James > ==== > Just an speculation: Could HP 49G+ use IRDA while 48G uses HP-IR (non IRDA > compatible)? > > Can somebody confirm or deny it? Well I know it doesn't use the now defunct and worthless HP Jetsend IR protocol my HP-1000 printer uses. ==== Yup. Figured that out right after I posted. However, it doesn't work with Xmodem either. Or at least, I can't make it work. John > > I understood that an IR transfer is possible between the 49G+ and the 48GX. > > I haven't been able to make it work. All I am trying to do is transfer a > > variable or program from one to the other. I have set both machines for > > Kermit (since GX won't do Xmodem) and IR. They are not able to connect. > > The 48G series can do Xmodem. See the XRECV and XSEND commands. > > -- > James > ==== The funny thing is if you buy through www.hp.com the shipping is sales tax in Delaware, buying through the compaq side of the site the 49g+ with shipping is cheaper than the calculator alone anywhere else I've found. Can't wait for it to arrive now! Mike > >> http://www.hp.com/calculators/ >> >> >I just (10/20/03) ordered the 49g+ from hp at the $149 price. > >Estimated ship date is 10/23. > >Cost: > >49g+: $149.99 >Tax (CA): 12.79 >Shipping: 5.15 >--------------- >Total: $167.93 > >Martin Cohen > > ==== > And now I have it!!!! > Give us a report on that keyboard please. ==== >>And now I have it!!!! >> > > > Give us a report on that keyboard please. So far, the keyboard seems fine. I turned on the click, though I usually (in the past) have it off. No presses missed yet (fingers of steel!!!). I played with the fast3d plotting using the instructions just added to HP's site. One funniness - when I tried sin(x+y), it did not look like a function - seemed to be 2 surfaces. I saw the quake once so far. All in all, I like it. However, I think HP made a mistake in using USB instead of rs-232. IMHO, the speed gain does not make up for the capability loss. Martin Cohen ==== > However, I think HP made a mistake in using USB instead of rs-232. > IMHO, the speed gain does not make up for the capability loss. I completely disagree with you there. Capability loss? I haven't used a computer with serial for YEARS! Next you'll want a floppy drive. ==== > > >>And now I have it!!!! > >> > > > > > > Give us a report on that keyboard please. > So far, the keyboard seems fine. I turned on the click, though I usually > (in the past) have it off. No presses missed yet (fingers of steel!!!). The keys have an easier click than on HP49G(less stiff). They definitely have the same feel as on the HP48 (clicks a little bit louder and sounds hollow). A funny detail about the keys is that the top 6 function keys have a steeper slant (due to the shorter width of the keys--where the top and bottom edges have the same heights as the other keys, the short width made the keys slant more). As a result, the top edge gap between the case and the key is wider, and you can see a part of the inside of the keyboard through the gap. Weird. Also, the arrow keys are somewhat loose: you can move the keys up/down/left/right a little bit when you press on the arrow keys, like the keys on a TI calculator, but less dramatic. Dave ==== > > >>And now I have it!!!! > >> > > > > > > Give us a report on that keyboard please. > So far, the keyboard seems fine. I turned on the click, though I usually > (in the past) have it off. No presses missed yet (fingers of steel!!!). Sometimes, my key press doesn't register. I don't know whether it's because I didn't press it hard enough, or because the system is busy with something (no busy aunciator though) and didn't catch my key press, no matter how I pressed the key for the few times that I pressed over and over. When you turn on the calc, the screen displays junk for half a second, then turns to normal, like on the old HP48, except it's much more apparent this time. You can adjust the contrast of the anunciators together with the rest of the screen when you press ON/+/- to adjust the contrast. And this time, it's an all black LCD. When you turn the clock on, there's screen flickering on the menu bar. After upgrading to ROM 1.22, the flickering becomes a little bit less noticable. And upgrading the ROM only takes a minute. The calc is much faster. However, it still takes a moment to load some menus with user input values in them like the plot setup or plot window setup. The wait for plotting wireframe graphs is much shorter, maybe more or less half a minute. There is also a few seconds wait time to graph regular 2-D graphs. Scrolling through a list in the browser and cursor movement are the same speed as before. Overall, the 49g+ feels more like a computer. It's fast and snappy inputing data and going through choose boxes. The speaker is quieter than in previous calcs. The error beep is more high pitched. Other beeps like the wrong key beep and alarm have the same tones. The inside of the pouch is cloth while the outside is leather with the HP logo on the lower right of the flap. Dave ==== > >I can't get it NOT to work. I tried deliberately touching ever so lightly >and increasing pressure slowly but then the key clicks down and registers. >For every click I get the proper character or function. No problems at all. I can easily and repeatedly press the backspace key lightly enough to cause a click, but no keypress and no speaker keyclick. Pete M. Wilson Gamewood, Inc. wilsonpm@gamewood.net ==== Update to my calc. I brought it to work where we have florescent lighting and the flicker is present. I din't notice it under my home incandescent lights. Also the range on my IR interface is about 1 not the 18 it's supposed to be. I get perfect results every time at 1 but garbage at any longer distances. Tom Lake ==== > Well, finally I can post to my own poll! Mitch and to all other posters to this thread, thanks, very useful. I for sure will upgrade after watching this thread for a few days. ==== How does the unit conversion compare to the one on the 48GX? Better/Worse, Quicker/Slower?? ==== If you switch to RPN + Softmenus, it's the same interface... Naturally since the calculator is faster the conversions will technically be so as well, but there isn't exactly a noticable delay doing unit conversions on the 48... > How does the unit conversion compare to the one on the 48GX? > Better/Worse, Quicker/Slower?? ==== > How does the unit conversion compare to the one on the 48GX? > Better/Worse, Quicker/Slower?? I don't quite understand the question... The 49G already had a few more units and the 49g+ is always the fastest around Just remember to switch to RPN ==== > I should add that in all cases, where you say the 15-digit result ends in > exactly ...500, > in fact the Sticky Bit is set - meaning the exact result is ...500xxxx, > meaning > 'nonzero digits trailing'. A subsequent PACKSB will then round *up*, of > course, > since the result is *not* exactly ...500, but 'larger than', and it doesn't > matter by how much. > That's how the system distinguishes between exactly ..500 and not: by the > SB. > > If you debug (using Jazz, on a 48GX) the following, you can see for > yourself: > > 262144. INV (result before PACKSB is called: 3.81469726562500 E-6, Sticky > Bit clear.) > That's exactly the midway case, so PACKSB will perform round to even and end > up with 3.81469726562E-6. > > 1998. INV, on the other hand, will show 5.00500500500500 E-4, Sticky Bit > set. > PACKSB will interpret that correctly as 'larger than' and subsequently round > *up*. I see your points about how even if the last 3 digits of the 15 digit mantissa are 500, the calculator may know that there must be a non-zero digit somewhere beyond that, and therefor quite correctly round the 12th digit up. Very good design. But what about my example of: %%1.000000000025E0 %%>% returns %1.00000000003 I was hoping that the calculator would round to even in that case, but it rounded up instead. Does that mean that the sticky bit was stuck set from a previous operation? I wonder when the calculator clears the sticky bit. -- James ==== > > I should add that in all cases, where you say the 15-digit result ends in > > exactly ...500, > > in fact the Sticky Bit is set - meaning the exact result is ...500xxxx, > > meaning > > 'nonzero digits trailing'. A subsequent PACKSB will then round *up*, of > > course, > > since the result is *not* exactly ...500, but 'larger than', and it doesn't > > matter by how much. > > That's how the system distinguishes between exactly ..500 and not: by the > > SB. > > > > If you debug (using Jazz, on a 48GX) the following, you can see for > > yourself: > > > > 262144. INV (result before PACKSB is called: 3.81469726562500 E-6, Sticky > > Bit clear.) > > That's exactly the midway case, so PACKSB will perform round to even and end > > up with 3.81469726562E-6. > > > > 1998. INV, on the other hand, will show 5.00500500500500 E-4, Sticky Bit > > set. > > PACKSB will interpret that correctly as 'larger than' and subsequently round > > *up*. > >I see your points about how even if the last 3 digits of the 15 digit >mantissa are 500, the calculator may know that there must be a >non-zero digit somewhere beyond that That's the purpose of the sticky bit; to know if any non-zero digits got shifted off to the right. , and therefor quite correctly round >the 12th digit up. Very good design. > >But what about my example of: > >%%1.000000000025E0 %%>% returns %1.00000000003 It looks like %%>% rounds up; why don't you also try %%1.00000000000035E0 > >I was hoping that the calculator would round to even in that case, but >it rounded up instead. Does that mean that the sticky bit was stuck >set from a previous operation? I wonder when the calculator clears the >sticky bit. The sticky bit is cleared before every arithmetic operation where it would subsequently be needed to do proper rounding (12 form result is where proper rounding to even is done). The Saturn arithmetic routines, starting with the HP71, never have done proper rounding to even when 15 form numbers are the result. The HP71 just truncates always when returning a 15 form number. Apparently the HP48 ALWAYS (that is, not paying attention to even or odd) rounds a trailing 50000000... (all the digits after the 5 which are still in the machine are zero, and the sticky bit is clear) up after an addition, but truncates after a multiplication when returning a 15 form number. ==== >I finally caught the dramatic 49g+ display earthquake on film! It >doesn't happen during most garbage collections, but once in a while it >does, and when it does it's quite startling! > I also noticed my calculator has changed it's initial clear of the display since I installed an SD card. Without the SD card, it shows black vertical stripes (thick, about 5) across the display that quickly fade to the regular display. With the SD card installed, they fade to a reverse black stripe pattern sprinkled with random on/off pixels, which then clears and becomes the regular display. Pete M. Wilson Gamewood, Inc. wilsonpm@gamewood.net ==== > I finally caught the dramatic 49g+ display earthquake on film! It > doesn't happen during most garbage collections, but once in a while it > does, and when it does it's quite startling! > > These two images are the same, but the first is smaller for those with > low bandwidth: > > 62K, 562 x 459: > http://holyjoe.net/images/earthquake_s.jpg > > 142K, 1125 x 918: > http://holyjoe.net/images/earthquake.jpg > > Notice the absolute rightmost three pixel columns: they are what's > usually the left-most three pixel columns, shifted up one pixel. If > you look carefully, you'll see (top to bottom) R (from the RAD > annunciator), { (the path delimiter), the status area line, the > numbers 7 down to 1, and a menu label. > > However, I can't make heads or tails of the garbage pixels. Can > you? > > -Joe- I just saw my first quake. Glad I was forwarned. Martin Cohen ==== > > Trying to view the result of the SERIAL command, I hit down arrow to > > edit the string. I then used SKIP-> to try to see the end. When the > > cursor was on the C... at the end of the display, pressing SKIP-> > > caused the cursor to appear at the end of the ..., and the string was > > not correctly scrolled! > > Same in 49G 1.19-6 > Have you ever tried to change the sign to negative on a number when the display has an 'algebraic' long enough to cause a scroll? ==== > > The 49G+ and 48Gii seem to have some similarities, but must be different in some > > aspects. What are the primary differences, and what would make one choose one > > over the other? > > http://www.hpcalc.org/images/graphcalccomp.pdf sums up graphing > calculator comparisons nicely. Differences that I found worth noting > are the following (in order of 48Gii followed by 49G+): > - processor: 48MHz vs. 75MHz (I haven't heard any benchmarks between > the calcs though, which would be more valuable) > - display size: 131x64 vs. 131x80 > - serial port: RS232 vs. USB > - expansion port: 49G+ only > - flash rom: 49G+ only > - memory: 512K vs. 512K+0.8M 128K vs 384K+768K (1280K~1.3M) port 1: 49G+ only corrections by PS: the net RAM is even lower. The 48gII has less than 1/10th of the memory (without 128MB SD) or less than 1/1000th when 128MB SD card is used in the 49g+ ==== I was wondering if any of you could help me out. I just purchased a HP-12C calculator from ebay. It came in the original package, but the calculator is made in China and there is no serial number on it. The manual appears genuine, but the calculator takes one lithium battery, not the three LR44's. Does HP really sell calculators made in China with no serial numbers? ==== > > I was wondering if any of you could help me out. I just purchased a HP-12C > calculator from ebay. It came in the original package, but the calculator > is made in China and there is no serial number on it. The manual appears > genuine, but the calculator takes one lithium battery, not the three LR44's. > > Does HP really sell calculators made in China with no serial numbers? > Yes and no! Yes, the HP calcs has been made in Malaysia, Indonesia, China, Singapore? No, you should not get a repair unit without the serial hand-engraved ==== > >> >>I was wondering if any of you could help me out. I just purchased a > > HP-12C > >>calculator from ebay. It came in the original package, but the calculator >>is made in China and there is no serial number on it. The manual appears >>genuine, but the calculator takes one lithium battery, not the three > > LR44's. > >>Does HP really sell calculators made in China with no serial numbers? >> > > Yes and no! > Yes, the HP calcs has been made in Malaysia, Indonesia, China, Singapore? And let's not forget the good ol' U.S. of A., and I'm almost positive Brazil as well. > No, you should not get a repair unit without the serial hand-engraved Some of the earlier models had the serial number on a sticker, but a recent production unit without a serial number does seem rather out of the ordinary. I was under the impression that even pre-production units were marked as such. I wonder whether perhaps this unit somehow accidentally got out of the factory in someone's pocket or lunch bucket. But a unit that somehow just didn't get a serial number in the factory due to a mix-up wouldn't be too terribly surprising. Sometimes a production process doesn't run quite as smoothly as we'd like. Maybe a collector's item? Is it a 12C or a 12C Platinum? As for the batteries, I'm under the impression that they switched from three LR44s to one lithium battery quite a while ago. -- James ==== This is a 12C non platinum model. It seems very cheaply made; the keys rattle when I lightly shake it. Not at all as well made as my 1 year old 12C made in Malaysia. If this is genuine, it is a real disappointment and HP really has gone way downhill. ==== > Joe's PDQ scout will be very interesting - just imagine if it > can casually deliver fractions in size order ... That's what it does, although the user always has the option of leaping over the intermediate convergents, which is useful when there are a huge number of them (half as many as each huge partial quotient). > ... or even just count the sum total of all possible fractions > - that would be AMAZING!! Hmmm... Hadn't thought of that. It'd be easy to add a Total Number of Best Fractions feature. What should I call it: The Hutchins Index? TONY(x)? What number has the LARGEST such value? It's NOT the golden ratio, which has the FEWEST possible intermediate convergents (namely, zero!). My guess it that it'd be whatever has the GREATEST possible number of intermediate convergents, something perhaps like sqrt(10^2n+1). Interesting... Oh, I have an idea! (Ouch! Ouch! I *hate* it when that happens!) Since the sequence of best fractions is deterministic, each has a fixed ordinal number. For example, the sequence for pi, with indices and CF expansion, is: 1: 3/1 {3} 2: 13/4 {3 4} 3: 16/5 {3 5} 4: 19/6 {3 6} 5: 22/7 {3 7} 6: 179/57 {3 7 8} 7: 201/64 {3 7 9} etc. So, when the Fraction Scout shows 201/64, it should also show 7 as its Hutchins Index, no? And maybe the CF expansion too, just for fun. -Joe- ==== -=[ Thu, 23.10.03 1:57 p.m. +1300 (NZDT) ]=- in message ID <87233f9e.0310221604.27f3e501@posting.google.com> : > > Joe's PDQ scout will be very interesting - just imagine > > if it can casually deliver fractions in size order ... > > That's what it does, although the user always has the > option of leaping over the intermediate convergents, Great! > which is useful when there are a huge number of them (half > as many as each huge partial quotient). Always a new challenge - hehe I'll have to think about the parenthetical statement above ... I get the idea - when we suddenly hit a huge partial quotient, rising up like Mt Sinai, we find many intermediate convergents before it. And yes they can stretch back to the midpoint. > > ... or even just count the sum total of all possible > > fractions - that would be AMAZING!! > > Hmmm... Hadn't thought of that. I'm not surprised as even my brain started bursting when I thought of how many there might be. I did try adding them up in the .500000000001 case but got rather confused. > It'd be easy to add a Total Number of Best Fractions feature. Wow, really? I would find that really really interesting. > What should I call it: The Hutchins Index? TONY(x)? LOL! You are welcome to call it the Hutchins Index :) Really I would take that as an honour to be included along with other H's like Horn, Hurwitz, and also Huysegoms - I can't wait to see Werner's new non-iterative method. > What number has the LARGEST such value? Oh no, another challenge :) I wonder if there is one such number. Root 3 I always found remarkable > It's NOT the golden ratio, which has the FEWEST possible > intermediate convergents (namely, zero!). LOL! My thoughts exactly, it is a sheer desert - not a large partial quotient in sight :) > My guess it that it'd be whatever has the GREATEST possible > number of intermediate convergents, something perhaps like > sqrt(10^2n+1). Interesting... Very. > Oh, I have an idea! (Ouch! Ouch! I *hate* it when that > happens!) Hehe. An interrupt! No, no, inspiration :) > Since the sequence of best fractions is deterministic, I was never quite sure about that - but yes it is. > each has a fixed ordinal number. For example, the sequence > for pi, with indices and CF expansion, is: > 1: 3/1 {3} > 2: 13/4 {3 4} > 3: 16/5 {3 5} > 4: 19/6 {3 6} > 5: 22/7 {3 7} > 6: 179/57 {3 7 8} > 7: 201/64 {3 7 9} > etc. Ahha! I never did the CF expansion for the intermediate ones. That almost gives me another clue as to how to find them. > So, when the Fraction Scout shows 201/64, it should also show > 7 as its Hutchins Index, no? Yes, I like that! > And maybe the CF expansion too, just for fun. Yes - the more interesting numbers, the better, IMHO :) It adds to the educational value. -- Tony Hutchins Wellington New Zealand #69 Are you part of the solution, or the precipitate? ==== OK all the examples work without bisecting. It has come to a point where calculating the simple continued fraction takes up the bulk of the work in some cases (like PI), so I'll combine it all in one program. The worst case should then be solved in less than one second as promised. (.500000000001 is solved in 0.7 secs but is not the worst case any more) Werner ==== > OK all the examples work without bisecting. I'm VEY impressed! I tried to do the same, but kept running into the problem of roundoff error sometimes giving false results. Bisection avoids that problem because it only uses integer linear combinations. Have you let both the old and new programs run on random numbers until they find different answers? I'm very much looking forward to seeing your code. If it always works, I'll modify the PDQ Algorithm accordingly and include you as co-inventor in the PDQ patent. Warning: If you calculate the backwards jump to the appropriate intermediate convergent using the same method as my DEC2FRAC program (on Goodies Disk #3), it will jump to the wrong answer in some pathological cases when zero error is specified. I very much hope that this is not the case! -Joe- ==== > > > OK all the examples work without bisecting. > > I'm VEY impressed! I tried to do the same, but kept running into the > problem of roundoff error sometimes giving false results. Bisection > avoids that problem because it only uses integer linear combinations. > Have you let both the old and new programs run on random numbers until > they find different answers? I'm very much looking forward to seeing > your code. If it always works, I'll modify the PDQ Algorithm > accordingly and include you as co-inventor in the PDQ patent. Naaah! Since Werner doesn't use bisecting at all he has a different and *better* algorithm. (-: You name will not be included in his patent. You have already been obsoleted, sorry! )-: (-; ==== -=[ Thu, 23.10.03 07:41 a.m. +1300 (NZDT) ]=- in message ID <44ec85ff.0310220520.21a6d18e@posting.google.com> : > OK all the examples work without bisecting. You cracked it! > It has come to a point where calculating the simple > continued fraction takes up the bulk of the work in some > cases (like PI), so I'll combine it all in one program. The > worst case should then be solved in less than one second > as promised. (.500000000001 is solved in 0.7 secs but is > not the worst case any more) Fascinating!! Yes the .500000000001 is really very short on partial quotients. CONGRATULATIONS WERNER! I really look forward to seeing your all-in-one program. -- Tony Hutchins Wellington New Zealand #185 To live is so startling it leaves little time for anything else. Emily Dickinson ==== > OK all the examples work without bisecting. > It has come to a point where calculating the simple continued fraction > takes up the bulk of the work in some cases (like PI), so I'll combine > it all in one program. The worst case should then be solved in less than one > second as promised. (.500000000001 is solved in 0.7 secs but is not the > worst case any more) > Werner Awesome That is much faster than my srpl version (197.5 bytes, #476Bh) with .87 sec. on the hp49- Gjermund ==== X > on the hp49- Do you mean the 48gII *without* the Flash memory ??? PS: I have a 49g- (aka 48gII), and a 49G, plus a hp 49g+ ==== > -=[ Sat, 18.10.03 9:37 p.m. +1300 (NZDT) ]=- > > > in message ID : > > > Nono -*you* have won it. > > All I did was copycat your work and write it the way I > > am used to, and I do have a fair bit of 'mini-challenge > > experience' ;-) > > There should be a library of mini-challenges - must be some > great coding tips in those answers. > > And you did much more than copycat - you gave me many puzzles > eg - I wondered about > > @ stack: R[i+1] R[i] a[i+1] 0. > > where you use a[i+1] as the maximum integral N - it took me a > while to confirm this. IMHO your bisection code is the BEST :) > I never quite realised how to use the stack this way. The challenge isn't over yet - even if you have won it already. I did a bit too much copying.. Bisection is unnecessary - I can determine the fraction directly. It'll take me awhile to get the algorithm sorted out for all cases and program it, I'm quite certain to get the timing of .500000000001 under 1 second on anything (48SX, to name the slowest). Wait and see, Werner ==== -=[ Thu, 23.10.03 07:29 a.m. +1300 (NZDT) ]=- in message ID <44ec85ff.0310220418.14dcc23f@posting.google.com> : > The challenge isn't over yet - even if you have won it already. > I did a bit too much copying.. > Bisection is unnecessary - Ahha! Yes, I looked at that before I did bisection but the algebra sort of put me off, and I was in a hurry. The error increases as K increases. I had an equation for K involving an explicit error like .5E-12 and it had a q^2 like in Hurwitz's formula. I didn't like having to code in an explicit error.... but maybe it is not necessary!! > I can determine the fraction directly. This will be fascinating to see!!! > It'll take me awhile to get the algorithm sorted out for all cases > and program it, I'm quite certain to get the timing of .500000000001 > under 1 second on anything (48SX, to name the slowest). > Wait and see, I will. A pleasure is in store :) -- Tony Hutchins New Zealand ==== > > > I like the 179/57 [approximation to pi]. > > I guess pi 2.9 PDQ will give that one :) > > Input: > > 2: 3.14159265359 > 1: 2.9 > > Output: > > 4: 179. > 3: 57. > 2: X: 2.90595659819 > 1: H: -.955278116548 Well, I've written a program that gives the same approximation, namely 179/57, based on the Stern-Brocot-tree, in 0.9609 seconds. But is very slow at converting 0.500000000001. How long does your version need to convert that beast? Michael -- -= Michael Hoppe , =------ ==== -=[ Wed, 22.10.03 11:21 p.m. +1300 (NZDT) ]=- in message ID <1g3897k.141xa5w1i0fv9cN%mh@michael-hoppe.de> : > Well, I've written a program that gives the same approximation, namely > 179/57, based on the Stern-Brocot-tree, in 0.9609 seconds. But is very > slow at converting 0.500000000001. How long does your version need to > convert that beast? Just a second on the 49g+, from memory. Also, from memory I remember trying a Farey series after my impression was it might take forever on this .500000000001. I must have a look at this Stern-Brocot though - it sort of looks a similar type of approach to the Farey? That .9609 seocnds is pretty quick. Have you read back in the thread? - timings were published I think. -- Tony Hutchins Wellington New Zealand #294 The unknown always passes for the marvelous. Tacitus ==== > That's often said, even by trustworthy mathematicians, but it's simply > not true. Proof: Using the continued fraction method, the next best > approximation to pi after 22/7 is 333/106. But that totally misses > 179/57, which lies between them both in size and in closeness to pi. > Go ahead, check it. See what I mean? Not quite. If rounding errors are not to heavy, I get: 22/7 = 3.1428571... 333/106 = 3.1415094... 179/57 = 3.1403508... 179/57 lies between 22/7 and 333/106 in size? Michael -- -= Michael Hoppe , =------ ==== -=[ Wed, 22.10.03 8:57 p.m. +1300 (NZDT) ]=- in message ID <1g379hi.1vd3y481ac0n2sN%mh@michael-hoppe.de> : > 179/57 lies between 22/7 and 333/106 in size? 179 lies between 22 and 333 57 lies between 7 and 106 -- Tony Hutchins Wellington New Zealand #56 A clean desk is a sign of a cluttered desk drawer. ==== > > > > > And maybe even more unfair if they've visited > > > http://www.smb.compaq.com/sp_main.asp?hp_url=searchresults.asp?store_id=8&sea rch_id=164&dept_id=1001&dsc=calculator > > > > > and read: 2.5 MB of total memory - 512K of RAM and 2MB of flash ROM. I > > don't doubt that their advertising is true, but they've neglected > > mentioning how much memory is used by the operating system. > > I notice that http://www.hp.com/calculators/graphing/49gplus/ says: > > 2.5 MB of total memory - 512K of RAM and 2MB of flash ROM (800KB > available to the user) > > I suppose that we might quibble about whether 800KB is accurate, and > whether the 800KB is flash memory only, or does it include port RAM and > user RAM, and no doubt it will vary with the ROM version, But at least > it makes it clear that nowhere near the total memory will be available > to the user. Hey! It changed! Now it says: 2.5 MB of total memory - 512KB of RAM and 2MB of flash ROM (330KB RAM and 800KB flash ROM available to the user) Could it be that someone at HP actually cares about what's posted on this newsgroup? Or is it just a coincidence? -- James ==== > > You can add up to 128MB SD into your system. > > So the 256KB, 512KB and 1GB SD cards won't work in the 49g+? > > Tom Lake I'd love to know myself. I have heard of the 256MB (assuming kb was meant to mean mb) cards at least report the correct size in filer, but apparently no one speaking up on here that I've heard has ran extensive tests in the short time that they have been out. I await a word on the other 2 sizes, but would love to point out the fact that my graphing calculator has more memory than the ram of all but the newest PCs I've been around lately. Ed Sutton P.S. I admit what an idiot I feel like when my 1GB card search failed as searches like pricewatch.com showed nothing for 1024kb sd memory card =/ ==== > > So the 256KB, 512KB and 1GB SD cards won't work in the 49g+? > > > > Tom Lake > > I'd love to know myself. I have heard of the 256MB (assuming kb was > meant to mean mb) Oops! KB, MB whatever! 8-) > P.S. I admit what an idiot I feel like when my 1GB card search failed > as searches like pricewatch.com showed nothing for 1024kb sd memory > card =/ seen any for sale either. Tom L ==== Responses follow: : : I'm glad to hear that. Was the card preformatted, or did you format it : in a card reader/writer, or did you format it with the 49g+? *** I tried the write/read cycle on the SD card right out of the box. Panasonic must have shipped the units correctly programmed because everything worked correctly with no intervention required by me. : : I take it the cluster size is 4KB? Store a small file and see how much : less the filer reports for port 3. : ***It will take me until Monday or Tuesday to do this correctly, James because I am going out of town to hear my daughter sing. : Please do report here if you run into any problems. : : > The real question, I think, for the larger cards is going to be: How : > many directory entries does the SD card allow to be made? Sometimes you : > can run out of available directory entries long before you run out of true : > space on the card/disk. : : Agreed, particularly as most files from a calculator could easily fit : into a single cluster with slack left over. : : Assuming that the card is formatted FAT16, there are of course 2^16 : (65,536) possible numbers, #0000h through #FFFFh. Of course each entry : in the FAT is two bytes. My reference book tells me that #0000h is used : for free clusters, #FFF0h-#FFF6h are reserved, #FFF7h for bad clusters, : and #FFF8h-FFFFh for the end of a cluster chain. The first two entries : in the FAT (which would otherwise be used for cluster #0000h and #0001h) : are reserved. The first entry is a copy of the media descriptor byte : followed by an FF byte, and the second entry is two FF bytes, so #0001h : isn't available either. So that tells me that 18 numbers (#0000h, : #0001h, and #FFF0-FFFFh) aren't available for indexing clusters, leaving : me with the possibility of up to 65,518 clusters, which places a maximum : on the number of files. : : I'm not sure how exactly an SD card follows the format of a disk, but on : a disk, the first sector begins the boot record (which, among other : things, contains information as to how the disk is formatted), which : ordinarily takes only a single sector, then comes two copies of the FAT, : and then the root directory. It seem to me that each FAT could take up : to 256 sectors, but the actual FAT size depends on how many clusters the : disk actually has. The root directory of a hard disk allows up to 512 : entries, and each entry is 32 bytes, so that's 32 sectors. That's : potentially up to 545 sectors used up before any data is stored. : Beginning with the first sector after the root directory, the data area : begins. Since the first two entries in the FAT are reserved, the first : cluster is numbered #0002h instead of #0000h (otherwise the first two : clusters could never be used). Unless the cluster size is 1 sector (512 : bytes) (a 32MB or smaller disk), I'll very likely have at least one : sector at the end left over which won't make up a full cluster, and so : can't be used. : : For a 128MB disk, I'm starting out with 262,144 sectors. If I assume : that 545 sectors are used up by the boot record, FATs, and root : directory, that leaves me 261,599 sectors for the data area. Since the : cluster size for a 128MB disk would ordinarily be 4 sectors (2KB), that : leaves me with 65,399 clusters, which indeed requires each FAT to be 256 : sectors, so at most 65,399 files. : : So it turns out that the possible numbers from 16 bits, less the : reserved numbers, wasn't the limiting factor after all. : : But wait, the root directory has only 512 entries, so for that many : files, I'm going to need at least one subdirectory, which takes at least : one cluster. Ok, suppose that I use all 512 root directory entries for : subdirectories, each using at least one cluster; that leaves me with at : most 64,887 clusters left for files. Each entry in a subdirectory take : 32 bytes, so accommodating that many files means that the subdirectories : would take up at least 2,076,384 bytes, or 1,014 clusters. That would : leave me at most 64,385 clusters for files, but 64,385 files requires : only 1,007 clusters for subdirectories, so would leave me 64,392 : clusters for files, which still requires 1007 clusters for : subdirectories. : : So assuming that I've figured out everything correctly (a dangerous : assumption), it would seem that not counting subdirectories, I could : have up to 64,392 relatively small (<=2KB) files on a 128MB hard disk. I : guess that the same would be true of a 128MB SD card, except that a 2KB : file from a calculator would be relatively large. : : As for how many files could fit on your 256MB card, I think that I'll : leave that as an exercise for the reader. : : Of course, if FAT32 were used, it would seem that, on extremely large : cards, roughly 4,000 times as many files would be possible. On small : cards, doubling the size of the FATs would be a disadvantage. The : question that I have is: : : How large does the card (or HD) have to be to make it advantageous to : use FAT32? : : -- : James : Sorry not to be able to finish this test and report back until Monday or Tuesday, James. Jim Chumbley ==== James, I cannot remember how to format an SD card. I've tried to find it in all the manuals for the 49g+, but I can't find it there. Could you please remind me how to format the little card? Jim Chumbley ==== >James, > I cannot remember how to format an SD card. I've tried to find it in all >the manuals for the 49g+, but I can't find it there. Could you please remind >me how to format the little card? >Jim Chumbley > Press On-D then pick Format from the menu. Requires ROM v1.22 Pete M. Wilson Gamewood, Inc. wilsonpm@gamewood.net ==== listened to the loose marbles rolling around inside my cranium. It truly astounds me what I am capable of forgetting. Again, though, many, many thanks for reminding both James and me of how to format the SD card. We really appreciate it. Jim Chumbley : : Press On-D then pick Format from the menu. Requires ROM v1.22 ==== > James, > I cannot remember how to format an SD card. I've tried to find it in all > the manuals for the 49g+, but I can't find it there. Could you please remind > me how to format the little card? Sorry Jim, my 49g+ isn't scheduled for delivery until tomorrow evening, so I don't have one to experiment with. And even if i did have the 49g+, I haven't even ordered an SD card yet. when I searched the downloaded PDF documentation, I didn't find much on SD cards. That the 49g+ can use an SD card, where it shows up in the filer, how to store things to it and recall them from it, but nothing on what sizes it can use or how to format it. Maybe someone else will be good enough to respond. Offhand, my advice would be that if it ain't broke, don't fix it. Is this your 256MB card? -- James ==== HA!! I thought I was the only one. Now you have a chance to join the group (It is comprised of all of us who got no sleep the night before the 49g+ was delivered out of anticipation, and then got no sleep the night it was delivered from finally getting to play with it.) Serves us both right! You are so right about the documentation. I couldn't find much in there about the SD cards at all. : Maybe someone else will be good enough to respond. Offhand, my advice : would be that if it ain't broke, don't fix it. ***Usually the very best advice, but I wanted to determine whether my 49g+ would format the big SD card successfully, and I wanted to run this test sometime when a failure would not cause any loss of real (maybe that should say valuable) data. It worked well. It's just my own brain that cannot be trusted.*** : : Is this your 256MB card? : ***Yes, it is. I found the Panasonic brand for sale at WalMart for just at $98. Its cost in dollars per megabyte of capacity was half that of smaller cards such as the 32MB card, and that's why I picked the largest card they had for sale. For convenience, I suspect that an 8MB or 16MB card may prove superior, if only in the speed of sorting, repairing a directory, and completing searches for an erasure. 256MB also makes little sense unless HP enables a directory/subdirectory structure in a future ROM release. Jim Chumbley ==== > HA!! > I thought I was the only one. Now you have a chance to join the group > (It is comprised of all of us who got no sleep the night before the 49g+ was > delivered out of anticipation, and then got no sleep the night it was > delivered from finally getting to play with it.) Serves us both right! > You are so right about the documentation. I couldn't find much in there > about the SD cards at all. Here's the Google entry: displayed. If you select it, it will format your installed SD card to FAT16. I tried it, and it works fine. Also the SD card recognition bug has been fixed, you don't have to take out and re-insert the card any more to use it. Thomas ==== Dear Tom, can speak for both James and me when I say that we might have had very bad evenings without your information. We really appreciate it. Jim Chumbley : Here's the Google entry: : [snips] : I was very pleased to see that they added the capability to format SD cards : with this new release. Just press ON-D, and a format option will be : displayed. If you select it, it will format your installed SD card to FAT16. : I tried it, and it works fine.