370 === Subject: Debug4x Not Displaying certain chars in Vista Other than Don't use Vista has anyone found a solution to certain charaters not showing up in Debug4x or HPUserEdit under Vista? Some chars don's show up at all, others as the wrong glyph. For example, the square root sign displays as a cursive f on my display. Am I the only one having this problem? Is there a correct font out there? === Subject: Re: Debug4x Not Displaying certain chars in Vista > Other than Don't use Vista has anyone found a solution to certain > charaters not > showing up in Debug4x or HPUserEdit under Vista? Some chars don's show up at all, others as the wrong glyph. For > example, the square root sign > displays as a cursive f on my display. Am I the only one having this problem? It is just the range of control characters. It does not seem to be all of the control characters either. If I run Charmap.exe to check on Vista, characters 0x007F to 0x009F do not appear. When I run Charmap.exe on XP, I see the expected characters. I have been looking for a solution for a long time and even asked in the Delphi groups (Debug4x is developed in Delphi) and there has been nothing that works. Windows Vista does not handle the control character range the same as all previous Windows. If you have any ideas or fonts or anything, let me know directly - I am out of ideas!! bill (who publishes Debug4x) -- - - - - - - - - - - - - - - - - Bill Graves RKBA! bgraves@ix.netcom.com === Subject: Re: Debug4x Not Displaying certain chars in Vista > Other than Don't use Vista has anyone found a solution to certain > charaters not > showing up in Debug4x or HPUserEdit under Vista? Some chars don's show up at all, others as the wrong glyph. For > example, the square root sign > displays as a cursive f on my display. Am I the only one having this problem? It is just the range of control characters. It does not seem to be all of > the control characters either. > If I run Charmap.exe to check on Vista, characters 0x007F to 0x009F do not > appear. When I run Charmap.exe on XP, I see the expected characters. I have been looking for a solution for a long time and even asked in the > Delphi groups (Debug4x is developed in Delphi) and there has been nothing > that works. Windows Vista does not handle the control character range the > same as all previous Windows. If you have any ideas or fonts or anything, let me know directly - I am out > of ideas!! bill (who publishes Debug4x) > -- - - - - - - - - - - - - - - - - > Bill Graves RKBA! > bgra...@ix.netcom.com Bill, that the font did not have some sort of tagging of Unicode names for the chars that were not showing up. Sorry I can't be more specific than that. I added what seemed right by analogy to other characters, but it had no effect. I'll try some more and report back if I have any success. BTW, great work on debug4x. Dan Doherty === Subject: Re: Debug4x Not Displaying certain chars in Vista posting-account=ky6NnQoAAAAl8HjjF10EUMKbzXiIKhTR Gecko/2008120122 Firefox/3.0.5,gzip(gfe),gzip(gfe) > Other than Don't use Vista has anyone found a solution to certain > charaters not > showing up in Debug4x or HPUserEdit under Vista? Some chars don's show up at all, others as the wrong glyph. For > example, the square root sign > displays as a cursive f on my display. Am I the only one having this problem? Is there a correct font out there? > went as far as creating my own font (from my handwriting), and interestingly enough when I set this font in HPUserEdit as the font in Options/Editor Options, all the characters display just fine. However with Debug4x I am completely at a loss. I looked at the registry to see if there was a way to change the system font used in debug4x (to the one I created) to see if it would solve the issue, but I never found a way to do that. This was a few months ago and I gave up because I ran out of ideas. What have you tried so far? There must be a way to solve this issue, I wouldn't mind spending some more time on this if I had a plausible idea to go on. Jacob === Subject: Re: Debug4x Not Displaying certain chars in Vista posting-account=kSyaOAkAAABaqKHaQ8qUzyG2kFubDVsp .NET CLR 2.0.50727; Media Center PC 5.0; .NET CLR 3.0.04506),gzip(gfe),gzip(gfe) Other than Don't use Vista has anyone found a solution to certain > charaters not > showing up in Debug4x or HPUserEdit under Vista? Some chars don's show up at all, others as the wrong glyph. For > example, the square root sign > displays as a cursive f on my display. Am I the only one having this problem? Is there a correct font out there? > went as far as creating my own font (from my handwriting), and > interestingly enough when I set this font in HPUserEdit as the font in > Options/Editor Options, all the characters display just fine. However > with Debug4x I am completely at a loss. I looked at the registry to > see if there was a way to change the system font used in debug4x (to > the one I created) to see if it would solve the issue, but I never > found a way to do that. This was a few months ago and I gave up > because I ran out of ideas. What have you tried so far? There must > be a way to solve this issue, I wouldn't mind spending some more time > on this if I had a plausible idea to go on. Jacob Well, I tried messing around with FontForge on Linux, because many of the characters are just wrong. I could not get any traction with FontForge, besides, I noticed that even many of the glyphs that were correct did not show up. Is the invisible glyph thing a Vista-specific problem? Anyway, having the wrong glyphs in the font must affect everyone. Dan === Subject: HHC2008 Year-2009 Calandars Are Now Available posting-account=K31tcwgAAABVnT_oG3A7BQonvJCmOPn4 Gecko/2008120122 Firefox/3.0.5,gzip(gfe),gzip(gfe) The second annual HP Calculator commemorative calendar is now available for those interested. Check http://www.pahhc.org/2009/Calendar.htm for details. The prices have dropped slightly from last year. Jake Schwartz === Subject: Syntax of all HP48 HP50g commands posting-account=ZnpnxQoAAAAXzWo26i-7z19miWsTIM3P Gecko/2008070208 Firefox/3.0.1,gzip(gfe),gzip(gfe) Hey Guys I had a HP48g and now a 50g. Sometimes I missed a Command Syntax Help to the calculator. Something that shows all commands AND the syntax of each command. A kind of HELP. This is because several times I just can«t remember the syntax of each command. And during my classes, I don«t have access to the Manual. This should be great even if it was just a TXT file saved in memory. In new 50g we have the CATALOG menu, but there are just some commands inside. Do you know something about this? === Subject: Re: Syntax of all HP48 HP50g commands >Hey Guys I had a HP48g and now a 50g. >Sometimes I missed a Command Syntax Help to the calculator. >Something that shows all commands AND the syntax of each command. A >kind of HELP. >This is because several times I just can«t remember the syntax of each >command. And during my classes, I don«t have access to the Manual. This should be great even if it was just a TXT file saved in memory. >In new 50g we have the CATALOG menu, but there are just some commands >inside. Do you know something about this? > http://www.hpcalc.org/details.php?id=5973 Damir === Subject: Re: Memory card reader You could also abuse an HP-48SX or GX as card reader. You could write a small program to copy every data bit of the card, and then send the data to a PC via the serial connection. I have the Needam's programmer with the EPSON card adapter, the only thing is there are no OTP and only a few EPROM cards left, and no, I won't let the adapter go:-) HTH jfphp schrieb im Newsbeitrag >I am looking for a standalone reader able to read the HP/EPSON (SMI, > ultimate goal is to transfer the data to a flash card. I have heard of > an Epson adapter for a Needham's programmer but the company has === Subject: Re: Memory card reader > You could also use an HP48SX or GX as card reader. > You could write a small program to copy every data bit of the card, > and then send the data to a PC via the serial connection. Doesn't one generally need only to recall each object that's stored on the card, then store in main memory, then simply transfer from main memory to a computer? The reverse can then be done in an emulator, to reconstruct the card in the emulator. Objects in R/O (ROM) cards can be recalled using XRCL from the HP48 Hack library: http://www.hpcalc.org/details.php?id=1817 http://www.hpcalc.org/hp48/programming/hacker/hack.zip This is, IIRC, exactly how I got my HP48 Hacker's ROM Rev. B (May 1996, EduCALC #GD10C) into my emulator. A card image for emulator use (Emu48) must, in any case, be one nibble (4 bits) of data per byte of stored file, i.e. the stored file will contain 256K bytes per 128K port of the original card (64K bytes for a 32K original card?) Once you have any card transferred to an emulator, you can send the computer file to any other emulator user, who can plug the card into the emulator's port 2 (on-calc card copying programs Copy and COPY2TO1, found below, can then copy one entire port to another within the emulator). Here are some card copy utilties for HP48: http://www.hpcalc.org/search.php?query=card+copy&hp48=1 It appears that more than one program for copying entire cards to PC (even ROM cards) is already there, e.g. Copy-ROM and Woopi, though I don't see whether those create the format needed by the emulator (otherwise what do you do with the output?) Does convert.exe for ROM dumps perform that kind of task? === Subject: Re: Size of Flash memory in 50g <6bG8l.114765$_03.78698@reader1.news.saunalahti.fi> posting-account=sOAX1QkAAAC-FcySTSbz29Uk8huUtFRz CLR 1.1.4322; .NET CLR 2.0.50727),gzip(gfe),gzip(gfe) > This has been discussed here in the past so a Google search should > give you in depth answers. When, where, what? Or search your own memory :) S.C. === Subject: Re: Size of Flash memory in 50g THAT was what I was looking for... I need to upgrade or enlarge my brain cell count... > This has been discussed here in the past so a Google search should > give you in depth answers. When, where, what? Or search your own memory :) S.C. === Subject: Re: Size of Flash memory in 50g > 766 KB is correct because the O.S. is using two additional FlashBanks > with the recent O.S. versions (>= 2.00). This has been discussed here in the past so a Google search should > give you in depth answers. When, where, what? John H. Meyers explains that additional memory is used by the newer ROMs for the ARM9 CPUs. The emulation also consumes 128K of the 512K S.C. === Subject: 2009 HHC HP Calendar To Be Ready Next Week Just FYI, the 2009 HP calendars (sponsored by HHC2007/HHC2008) will be returning from the printers in time to start being mailed next week. ALL ATTENDEES OF HHC2007 (in San Diego) WILL AUTOMATICALLY BE SENT A CALENDAR, since their conference registrations financed the printing of this year's calendar. For those who were there in 2007 but aren't sure, please check http://www.pahhc.org/2009/Calendar.htm for calendar information, including a list of registrants. If your mailing address has changed since then or you want to make sure I have your correct address, please follow the directions there about contacting me regarding where your calendar should be mailed. (If it hasn't changed, you don't have to do anything.) For those who did not attend HHC2007 but wish to purchase the calendar, more information on that will be available next week also at Jake Schwartz === Subject: Blue calc bottom right ? (was : 2009 HHC HP Calendar To Be Ready Next Week) Can anyone tell what it that blue calc bottom right on the picture ? Never heard of it... Herve Jake Schwartz a .8ecrit : > > Just FYI, the 2009 HP calendars (sponsored by HHC2007/HHC2008) will be > returning from the printers in time to start being mailed next week. > ALL ATTENDEES OF HHC2007 (in San Diego) WILL AUTOMATICALLY BE SENT A > CALENDAR, since their conference registrations financed the printing > of this year's calendar. For those who were there in 2007 but aren't > sure, please check http://www.pahhc.org/2009/Calendar.htm for calendar > information, including a list of registrants. If your mailing address > has changed since then or you want to make sure I have your correct > address, please follow the directions there about contacting me > regarding where your calendar should be mailed. (If it hasn't changed, > you don't have to do anything.) > > For those who did not attend HHC2007 but wish to purchase the > calendar, more information on that will be available next week also at > > Jake Schwartz === Subject: Re: Blue calc bottom right ? (was : 2009 HHC HP Calendar To Be Ready Next Week) Can anyone tell what it that blue calc bottom right on the picture ? > Never heard of it... Herve > http://www.hp.com/hpinfo/newsroom/press kits/2009/calculators2009/?mtxs=rss-corp-combined === Subject: Re: Blue calc bottom right ? (was : 2009 HHC HP Calendar To Be Ready Next Week) > http://www.hp.com/hpinfo/newsroom/press kits/2009/calculators2009/?mt... They appear to have the same sloped keys as the 20b (is it too hopeful to assume tactile feedback of the 35s?). Unfortunately, none of them are RPN machines... S.C. === Subject: Re: 2009 HHC HP Calendar To Be Ready Next Week Just FYI, the 2009 HP calendars (sponsored by HHC2007/HHC2008) will be > returning from the printers in time to start being mailed next week. > ALL ATTENDEES OF HHC2007 (in San Diego) WILL AUTOMATICALLY BE SENT A > CALENDAR, since their conference registrations financed the printing > of this year's calendar. For those who were there in 2007 but aren't > sure, please checkhttp://www.pahhc.org/2009/Calendar.htmfor calendar > information, including a list of registrants. If your mailing address > has changed since then or you want to make sure I have your correct > address, please follow the directions there about contacting me > regarding where your calendar should be mailed. (If it hasn't changed, > you don't have to do anything.) For those who did not attend HHC2007 but wish to purchase the Jake Schwartz Will there ever be an HHCxxxx for those of us of meager financial means who can't traverse the continent to The Golden State? === Subject: Re: 2009 HHC HP Calendar To Be Ready Next Week posting-account=K31tcwgAAABVnT_oG3A7BQonvJCmOPn4 For those who did not attend HHC2007 but wish to purchase the Jake Schwartz Will there ever be an HHCxxxx for those of us of meager financial > means who can't traverse the continent to The Golden State? Well, first thing, you haven't indicated where you live (fwiw, I'm in southern NJ). Secondly, it looks like HHC2009 will likely be in Vancouver, Washington, where the HP Calcs R&D manager works (since he can get us the facility for the conference to meet for no cost). In order for a conference to be anywhere , a volunteer must come forth to indicate that he or she will organize the thing. (There really is no formal organization here...just the remnants of the old PPC Calculator Club which existed from 1974 to the early 1990s.) There is no funding source from which to draw, so it's all volunteer and hard work. Back in 2001 and 2005, the Chicago-area CHIP HP calculator group sponsored conferences in their back yard, and that is the farthest eastward things have gotten in the recent several years. For a complete list of all the conferences and their locations, check out http://holyjoe.net/hhc2008/conflist.htm on the web. There were conferences in Providence, Orlando, Atlanta, Philadelphia, Rockville MD and Reston VA in the past in the 1980s and in each of those cases, a group of locals did all the work and organized it all. Nothing happens automatically, unfortunately. Jake === Subject: Re: 2009 HHC HP Calendar To Be Ready Next Week posting-account=5jQj0AoAAAAGAGJcqkkpunLMBpVi1N5o 2.0.50727; .NET CLR 3.0.04506.30; InfoPath.2),gzip(gfe),gzip(gfe) > For those who did not attend HHC2007 but wish to purchase the > Jake Schwartz Will there ever be an HHCxxxx for those of us of meager financial > means who can't traverse the continent to The Golden State? > Well, first thing, you haven't indicated where you live (fwiw, I'm in > southern NJ). Secondly, it looks like HHC2009 > will likely be in Vancouver, Washington, where the HP Calcs R&D > manager works (since he can get us the facility > for the conference to meet for no cost). In order for a conference to > be anywhere , a volunteer must come forth > to indicate that he or she will organize the thing. (There really is > no formal organization here...just the remnants > of the old PPC Calculator Club which existed from 1974 to the early > 1990s.) There is no funding source from which > to draw, so it's all volunteer and hard work. Back in 2001 and 2005, > the Chicago-area CHIP HP calculator group sponsored > conferences in their back yard, and that is the farthest eastward > things have gotten in the recent > several years. For a complete list of all the conferences and their > locations, check outhttp://holyjoe.net/hhc2008/conflist.htm > on the web. There were conferences in Providence, Orlando, Atlanta, > Philadelphia, Rockville MD and Reston VA in the > past in the 1980s and in each of those cases, a group of locals did > all the work and organized it all. Nothing > happens automatically, unfortunately. Jake === posting-account=2sgSMAoAAADHaHf0CqF6w39rFfvaJpp9 2.0.50727; .NET CLR 3.0.04506.648; .NET CLR 3.5.21022; .NET CLR 1.1.4322; MS-RTC LM 8; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729; Zune 3.0; WWTClient2),gzip(gfe),gzip(gfe) run across cost more than I'm willing to pay. Has anyone documented the connector pinout, data transfer protocol, etc. for rolling your own memory card? Alternatively, has anyone tried building an adapter that would allow someone to use an SD memory card in the gx? === >run across cost more than I'm willing to pay. Has anyone documented >the connector pinout, data transfer protocol, etc. for rolling your >own memory card? Many years ago I did it for my HP48SX. http://www.criseis.ruhr.de/hp48/hp48b.html They will work in the GX, too. BTW: I am still using the cards without any problem. I only had to exchange the battery a few years ago. >Alternatively, has anyone tried building an adapter >that would allow someone to use an SD memory card in the gx? It is possible to make a card reader for SD and use it with your own driver. But it is not possible to use a SD-Card like a ramcard, because SD are to slow. Oh..and did you realy think you need more than 1MByte in your pocketcalculator? I think it is not so complicate to make 128kB Ram and also SD/microSD Reader on the same card. But nobody did it, because nobody need it. Another funny idea that sometimes come to my brain, building a card with a FPGA on it. This can be memory, soundcard, cardreader, math-coproz and many more things. Than I feel lazy soon and stop this kind of idea. :-) Olaf === > run across cost more than I'm willing to pay. Has anyone documented > the connector pinout, data transfer protocol, etc. for rolling your > own memory card? Alternatively, has anyone tried building an adapter > that would allow someone to use an SD memory card in the gx? http://www.hpcalc.org/details.php?id=3530 maybe? === Subject: Re: Disable interrupts > ML is very difficult. I need to study more. > If you're interested in learning more on ML, i would recommend you reading the book : Introduction to Saturn Assembly Language by Gilbert Fernandes which can be downloaded on HPCalc.org. http://www.hpcalc.org/details.php?id=1694 This book alone learned me all i needed to know for programming ASM. === Subject: noob question: 50g ignores X in functions - what to do? posting-account=_HGaIAoAAAB6Jn3lrST3pff5_EdBru_y Gecko/2008120122 Firefox/3.0.5,gzip(gfe),gzip(gfe) I just recently bought a 50g and am kind of overwhelmed by the sheer number of functions it comes with. Now what I'm trying to do is get the graph of, say, a function of x of 4th order. So when I enter (in ALG mode) PLOT(1/4*x^3) press ENTER and select DRAW at the PLOT SETUP screen it just shows a plane coordinate system. When I then press ON to return to the screen showing what I entered it reads :PLOT(1/4*x^3) 0. If I add a constant to the function and try sth like PLOT(1/4*x^3+1.5) the coordinate system just shows me a straight line, so to say y=1.5 Returning to the initial screen I find :PLOT(1/4*x^3+1.5) 1.5 It just looks like it ignores any symbols.. dno if this is of relevance, but the head line reads: DEG XYZ HEX R= 'X' ALG What can I do? thx === Subject: Re: noob question: 50g ignores X in functions - what to do? > :PLOT(1/4*x^3) Did you mean 'X' (capitalized?) > What can I do? Do the WIN operation on the top line of keys, press NXT then RESET (all). Do 2D/3D (or rerun PLOT by pressing ENTER when the above-quoted line is on the bottom display line, or highlight it and press ENTER, either of which opens the same window) and make sure Indep variable is the same name and case as in the formula, then ERASE DRAW. Works here (weather permitting :) === Subject: Re: noob question: 50g ignores X in functions - what to do? posting-account=_HGaIAoAAAB6Jn3lrST3pff5_EdBru_y Gecko/2008120122 Firefox/3.0.5,gzip(gfe),gzip(gfe) > :PLOT(1/4*x^3) Did you mean 'X' (capitalized?) What can I do? Do the WIN operation on the top line of keys, press NXT then RESET (all). Do 2D/3D (or rerun PLOT by pressing ENTER when the above-quoted line > is on the bottom display line, or highlight it and press ENTER, > either of which opens the same window) > and make sure Indep variable is the same name and case > as in the formula, then ERASE DRAW. Works here (weather permitting :) okay, but how do i clear X from having the value 0? wherever i put X the calc behaves like i put a zero -.- === Subject: Re: noob question: 50g ignores X in functions - what to do? >> > :PLOT(1/4*x^3) >> >> Did you mean 'X' (capitalized?) >> >> > What can I do? >> >> Do the WIN operation on the top line of keys, press NXT then RESET (all). >> >> Do 2D/3D (or rerun PLOT by pressing ENTER when the above-quoted line >> is on the bottom display line, or highlight it and press ENTER, >> either of which opens the same window) >> and make sure Indep variable is the same name and case >> as in the formula, then ERASE DRAW. >> >> Works here (weather permitting :) >> >> okay, but how do i clear X from having the value 0? >wherever i put X the calc behaves like i put a zero -.- 'X' PURGE Damir === Subject: Re: noob question: 50g ignores X in functions - what to do? Okay, got it working again, thank you all very much, especially John, for the clear and helpful explanation! It was exactly the way you described it and now after purging X I can use it again as expected. philippe === Subject: Re: noob question: 50g ignores X in functions - what to do? > I just recently bought a 50g and am kind of overwhelmed by the sheer > number of functions it comes with. Now what I'm trying to do is get the graph of, say, a function of x of > 4th order. So when I enter (in ALG mode) > PLOT(1/4*x^3) > press ENTER and select DRAW at the PLOT SETUP screen it just shows a > plane coordinate system. When I then press ON to return to the screen > showing what I entered it reads > :PLOT(1/4*x^3) > 0. If I add a constant to the function and try sth like PLOT(1/4*x^3+1.5) > the coordinate system just shows me a straight line, so to say y=1.5 > Returning to the initial screen I find > :PLOT(1/4*x^3+1.5) > 1.5 It just looks like it ignores any symbols.. dno if this is of relevance, but the head line reads: > DEG XYZ HEX R= 'X' ALG > What can I do? > thx Use the plotter menu ([LS][F1] (Y=)). Then ADD (F2) your expression, press ERASE (F5) to erase any residue plots from previous graphs, and DRAW (F6). Since you bought the 50g (and not an ALG-only calculator), you might as well switch the 50g to RPN mode because it is really meant to be used that way. ALG mode is like an afterthought on the 49/50 series and does not work nearly as well as the native RPN mode does. Plus, RPN is much, much faster and more efficient once you get the hang of it (which takes less than an hour). *In RPN mode, to access the Y= menu, you need to *hold* down LS while pressing F1. S.C. === Subject: Re: noob question: 50g ignores X in functions - what to do? posting-account=_HGaIAoAAAB6Jn3lrST3pff5_EdBru_y Gecko/2008120122 Firefox/3.0.5,gzip(gfe),gzip(gfe) I just recently bought a 50g and am kind of overwhelmed by the sheer > number of functions it comes with. Now what I'm trying to do is get the graph of, say, a function of x of > 4th order. So when I enter (in ALG mode) > PLOT(1/4*x^3) > press ENTER and select DRAW at the PLOT SETUP screen it just shows a > plane coordinate system. When I then press ON to return to the screen > showing what I entered it reads > :PLOT(1/4*x^3) > 0. If I add a constant to the function and try sth like PLOT(1/4*x^3+1.5) > the coordinate system just shows me a straight line, so to say y=1.5 > Returning to the initial screen I find > :PLOT(1/4*x^3+1.5) > 1.5 It just looks like it ignores any symbols.. dno if this is of relevance, but the head line reads: > DEG XYZ HEX R= 'X' ALG > What can I do? > thx Use the plotter menu ([LS][F1] (Y=)). Then ADD (F2) your expression, > press ERASE (F5) to erase any residue plots from previous graphs, and > DRAW (F6). Since you bought the 50g (and not an ALG-only calculator), you might > as well switch the 50g to RPN mode because it is really meant to be > used that way. ALG mode is like an afterthought on the 49/50 series > and does not work nearly as well as the native RPN mode does. Plus, > RPN is much, much faster and more efficient once you get the hang of > it (which takes less than an hour). *In RPN mode, to access the Y= menu, you need to *hold* down LS while > pressing F1. S.C. thank you, but that doesn't really work out well.. the way you told me i get to draw it, but when I select edit at the plotter menu and confirm it again just returns the numbers, x disappears. about RPN i'm still feeling a bit unsure.. was using a simple scientific Casio for the last ten years throughout school and now i found out that i need a graphing one for a test, which is going to be in about 2 weeks, so i'm a bit worried if i can work into a new calculator, especially one that's this difficult, within so little time. Actually it looks like I assigned the number 0 to X , accidentally while playing around with it: That would explain the strange behaviour about leaving out powers of x as they'd have the value 0 and just get left out, also when I divide a number by x it returns Indefinite Result How do I clear that? === Subject: Re: noob question: 50g ignores X in functions - what to do? > Actually it looks like I assigned the number 0 to X, > accidentally while playing around with it. Merely having a value pre-stored in 'X' would not normally interfere with plotting a formula involving 'X', due to the ingenious design of the (original) plotter, which creates a completely new variable anyway, temporarily, during the DRAW operation of plotting, but I think that you will find that even your _formula_ (in the 'EQ' variable) has become just a zero (or constant), which is the actual immediate reason why you got a flatline. This happened because your PLOT command (which is probably not so frequently used) first _evaluated_ your formula, which, because it contained a variable whose value already exists, produced in this case the number zero (or a constant), then it dutifully first stored that result into 'EQ' (where the actual formula to be plotted is stored) before even showing you the Plot setup screen, which you should clearly see on the second line of that screen, as EQ:0 (or a constant), rather than the original formula which you had supplied as the argument to PLOT. You can always return to the Plot setup screen via the 2D/3D operation in the top row of keys, by the way. What happened to you in this exercise is one of the many consequences of a basic design philosophy of the CAS (and ALG mode), in which it was decided that all expressions would always be evaluated (and variables thus assigned their current value) all the time, despite this not always being a desirable thing, when purely symbolic, literal formulas are intended. This is the very same reason that many CAS functions will automatically demand that you first purge a variable, before evaluating the function -- the plotter, however, does not remind you of this, hence you fall into the sand trap of having variables in formulas evaluated by the CAS anyway, even if flag -3 is clear, which never happened in the original HP48 series, until the CAS and ALG mode were added. To purge the variable 'X' (in ALG mode) press TOOL PURGE X then ENTER, then start over (I think that it must also be purged from any higher directories, but in this case you don't have any, so you are spared the effort :) Note that you can re-run your original PLOT command (if it's still on the stack) by just pressing ENTER when it's on the lowest line of the display, or by using the HIST key or cursor up key to highlight it before pressing ENTER -- this is one of the particularly nice things about the ALG mode user interface, demonstrating that nothing is all bad, after all :) === Subject: Re: a GOSUB for SysRPL program Le dimanche 04 janvier 2009, Yann a .8ecrit : >:: > Beggining of Program > CODEM GOTO .MySubroutine ENDCODE ( Call to sub-routine ) > This is sentence 2 > COLA End of Program CODEM > *.MySubroutine > A=PC A+10.A PC=(A) > ENDCODE >:: This is subroutine sentence ; > ; This could be written a little shorter, I guess: :: begin CODEM A=PC *Here LC(5) MySub-Here A+C.A PC=(A) ENDCODE end SEMI LABEL MySub :: Subroutine ; ; I'm not sure about the syntax, though, but you should get the idea. === Subject: Re: a GOSUB for SysRPL program posting-account=eF2f0AoAAAB2spBRiZOs91ItDKLGDCIk Gecko/2008120122 Firefox/3.0.5,gzip(gfe),gzip(gfe) This could be written a little shorter, I guess: :: > begin CODEM > A=PC > *Here > LC(5) MySub-Here > A+C.A > PC=(A) > ENDCODE end > SEMI LABEL MySub > :: > Subroutine > ; > ; I'm not sure about the syntax, though, but you should get the idea. well. The size advantage however change if there are several calls, as the first code has a very short gosub sequence repeated. === Subject: Issue with printing from HP50g Hi all, I was playing with the USB serial port of my HP50g, ROM 2.09. I successfully sent and received files with the XMODEM protocol. As a printer, I use a homemade HP82249 simulator: http://pagesperso-orange.fr/kdntl/hp49/hp82240_emulator.tar.gz This simulator is a simple program (for UNIX) which can read the serial port and outputs the picture of what a real printer would print. Also note that the real printer is a read-only device (one-way communication), and so is the simulator. I tried to print things via wire (flag -33 is off, flag -34 is on). Most of the printing commands (PR1, PRST, PRSTC) works as expected. I didn't test them deeply, though. I encountered issues with the PRLCD command (which prints the screen): - When flag -38 is on (no linefeed), the calc sends no data. - When flag -38 is off (add linefeed), the calc sends wrong data. Please see: http://pagesperso-orange.fr/kdntl/misc/lcd.png I found three issues: too many linefeeds sent, missing data and unexpected data. You can take a look at the hexadecimal representation of data received by the printer simulator here: http://pagesperso-orange.fr/kdntl/misc/lcd.txt This file describes the issues. - When the ROM 2.09 is emulated by Emu48, PRLCD works as expected. I suspect a bug in PRLCD on ARM-based calcs. I reproduced the issues with two computers (so, my USB ports work). There is possibly a bug in the USB-serial driver of Linux. I don't think so, my setup works flawlessly with my HP50g when using other I/O commands. But who knows? Could you please try to reproduce the issue mentionned before? You would just have to get a dump of the data sent by the PRLCD command on your HP49g+/50g. Under Linux, just type the following command: cat < /dev/ttyUSB0 > datadump (remove the double quotes). Then, run the PRLCD command on your calc, turn it off and disconnect it. You could either take a look at your datadump file (using the hexdump command, or similar) or send it to me (my email is kdntl at yahoo dot fr). I don't know the procedure on other operating systems. Khanh-Dang === Subject: Issue with sat_pop_real() Cc: asrhael@msn.com posting-account=EnsxjQoAAAAC4ntxvhot7w8LkhBoqyQP But i have a precision trouble with the command that allow me to pop a real into my code. i use the next code, but when i put values big in the stack, the final result of kk have a lot of errors in all the decimals numbers. I think that sat_pop_real() is kidding me and rounded some decimals. all the comunity!! #include #include int main(void) { long double AE,EI,L,DX,DY,kk; sys_slowOn(); DY = sat_pop_real(); DX = sat_pop_real(); L = sat_pop_real(); EI = sat_pop_real(); AE = sat_pop_real(); kk = (DX*DX*(L*L)*AE+12*DY*DY*EI)/(L*L*L); sat_push_real(kk); return(0); } === Subject: Re: Blue calc bottom right ? (was : 2009 HHC HP Calendar To Be Ready Next Week) sc_usenet@hotmail.com a .8ecrit : >> http://www.hp.com/hpinfo/newsroom/press_kits/2009/calculators2009/?mt... > > They appear to have the same sloped keys as the 20b (is it too hopeful > to assume tactile feedback of the 35s?). Unfortunately, none of them > are RPN machines... > > S.C. Actually the 20b IS at your choice an alg or RPN machine... === Subject: Re: Blue calc bottom right ? (was : 2009 HHC HP Calendar To Be Ready Next Week) <49652167$0$24202$426a74cc@news.free.fr> posting-account=sOAX1QkAAAC-FcySTSbz29Uk8huUtFRz CLR 1.1.4322; .NET CLR 2.0.50727),gzip(gfe),gzip(gfe) > They appear to have the same sloped keys as the 20b (is it too hopeful > to assume tactile feedback of the 35s?). Unfortunately, none of them > are RPN machines... S.C. Actually the 20b IS at your choice an alg or RPN machine... Yes I know; I meant that none of the three new calcs (SmartCalc 300s, EasyCalc 100, OfficeCalc 300) are RPN machines. S.C. === Subject: Re: Blue calc bottom right ? (was : 2009 HHC HP Calendar To Be Ready Next Week) Are they real HP calcs or relabeled trash > They appear to have the same sloped keys as the 20b (is it too hopeful > to assume tactile feedback of the 35s?). Unfortunately, none of them > are RPN machines... S.C. Actually the 20b IS at your choice an alg or RPN machine... Yes I know; I meant that none of the three new calcs (SmartCalc 300s, EasyCalc 100, OfficeCalc 300) are RPN machines. S.C. === Subject: Re: Blue calc bottom right ? (was : 2009 HHC HP Calendar To Be Ready Next Week) > Are they real HP calcs or relabeled trash > Kinpo? Casio? For example, the HP 10s looks like a rebranded Casio... Are any of the modern HP calcs real ? What qualifies as real ? Corvallis? ACO? S.C. === Subject: Re: Blue calc bottom right ? (was : 2009 HHC HP Calendar To Be Ready Next Week) Rebranded Casio is Casio Corvallis, ACO, Hydrix, eveb KinpOS are HP > Are they real HP calcs or relabeled trash > Kinpo? Casio? For example, the HP 10s looks like a rebranded Casio... Are any of the modern HP calcs real ? What qualifies as real ? Corvallis? ACO? S.C. === Subject: Re: noob question: 50g ignores X in functions - what to do? posting-account=sOAX1QkAAAC-FcySTSbz29Uk8huUtFRz CLR 1.1.4322; .NET CLR 2.0.50727),gzip(gfe),gzip(gfe) >okay, but how do i clear X from having the value 0? >wherever i put X the calc behaves like i put a zero -.- 'X' PURGE Damir > Or, in ALG mode: PURGE('X') (returns NOVAL) <-- no value S.C. === Subject: Re: noob question: 50g ignores X in functions - what to do? > I just recently bought a 50g and am kind of overwhelmed by the sheer > number of functions it comes with. > > Now what I'm trying to do is get the graph of, say, a function of x of > 4th order. So when I enter (in ALG mode) > PLOT(1/4*x^3) > press ENTER and select DRAW at the PLOT SETUP screen it just shows a > plane coordinate system. When I then press ON to return to the screen > showing what I entered it reads > :PLOT(1/4*x^3) > 0. > > If I add a constant to the function and try sth like PLOT(1/4*x^3+1.5) > the coordinate system just shows me a straight line, so to say y=1.5 > Returning to the initial screen I find > :PLOT(1/4*x^3+1.5) > 1.5 > > It just looks like it ignores any symbols.. > > dno if this is of relevance, but the head line reads: > DEG XYZ HEX R= 'X' ALG > > > > What can I do? > thx The problem is that your graphing variable is, by default, an upper case, or capital X, not a lower case x, so the calculator does recognize 'x' as a variable when trying to graph. One of several things: 1: Simplest: replace lower case 'x' by upper case 'X' in '1/4*x^3' 2: execute STOVX(x) 3: left-shift D down-arrow down-arrow enter ALPHA LEEFT_SHIFT x === Subject: Re: noob question: 50g ignores X in functions - what to do? > Okay, got it working again, thank you all very much, especially John, > for the clear and helpful explanation! > It was exactly the way you described it and now after purging X I can > use it again as expected. philippe Note that ALL the variables in above directories are also visible So sometimes, when using subdirectories we need to think variable name selections, too Perhaps you should search CleanSolver from www.hpcalc.org === Subject: Re: Issue with sat_pop_real() posting-account=KxEIGwoAAAALMxy6hykxLK1wPdiyuNkK Gecko/2008120122 Firefox/3.0.5,gzip(gfe),gzip(gfe) But i have a precision trouble with the command that allow me to pop > a real into my code. i use the next code, but when i put values big in the stack, the final > result of kk have a lot of errors in all the decimals numbers. I think that sat pop real() is kidding me and rounded some decimals. all the comunity!! #include > #include int main(void) > { > long double AE,EI,L,DX,DY,kk; sys slowOn(); DY = sat pop real(); > DX = sat pop real(); > L = sat pop real(); > EI = sat pop real(); > AE = sat pop real(); > kk = (DX*DX*(L*L)*AE+12*DY*DY*EI)/(L*L*L); sat push real(kk); return(0); } I don't think this is a sat pop real() problem. Instead, I think that you're overflowing the precision of a long double in hpgcc/ARM. Keep in mind that the intermediate results must be represented as double longs, so if DX, L and AE are very large, then DX*DX*L*L*AE is likely to be REALLY big.... You can check this by computing the result using the large numbers by hand and observing the value of the intermediate values. You can fix this problem, or at least avoid it a little better, by re- arranging the calculations: // (DX*DX*(L*L)*AE+12*DY*DY*EI)/(L*L*L) == DX*DX*AE/L + 12*DY*DY*EI/(L*L*L) // Now compute it this way and alternate multiplication and division to reduce the chance of overflow: DX = DX/L*DX*AE + 12*DY/L*DY/L*EI/L; Dave === Subject: Re: Issue with sat_pop_real() posting-account=Q2CEjQoAAACue5ZDDhUeRzv1w0u8hxZE rv:1.9.1b2) Gecko/20081201 Firefox/3.1b2,gzip(gfe),gzip(gfe) > I don't think this is a sat pop real() problem. Instead, I think that > you're overflowing the precision of a long double in hpgcc/ARM. Yes. Please read the whole page, (or at least this section) from the fount of all knowledge for an important summary: http://en.wikipedia.org/wiki/Floating-point#Accuracy problems There is a problem with sat push real() that I have seen before where the number is pushed to the stack correctly, but displays a leading 0. I haven't been able to figure it out yet, and haven't seen it in quite a while, and it doesn't cause any problems with either mp math routines, or the hpgcc stack interaction stuff. Basically you get a number on the stack displayed as 09.1234 or something like that. TW === Subject: Re: Issue with printing from HP50g posting-account=zYTuBQoAAAC_bXzGjGVT5rxv8bOnpefP rv:1.9.0.5) Gecko/2008120121 Firefox/3.0.5,gzip(gfe),gzip(gfe) > I was playing with the USB serial port of my HP50g, ROM 2.09. I > successfully sent and received files with the XMODEM protocol. As a > printer, I use a homemade HP82249 simulator: http://pagesperso-orange.fr/kdntl/hp49/hp82240 emulator.tar.gz Great start. Very nice. Feature request. When using conn4x with Windows I can grab a screen shot via USB or Serial (IIRC) without using PRLCD, I can just ask for a screen shot from conn4x. Can you add a flag to hp82240 to just snag the screen on connect and exit? Unsure how this works with conn4x, I do not have a sniffer for serial, but there are USB software sniffers. > - When flag -38 is on (no linefeed), the calc sends no data. Works for me. However I am not using Linux or USB. I am using OS/X and Serial (keyspan USB to serial). I need to do more research with OS/X to see if it has a generic USB/Serial driver/module like Linux does. > - When flag -38 is off (add linefeed), the calc sends wrong data. Please > see: http://pagesperso-orange.fr/kdntl/misc/lcd.png Same here. I get the extra lines. I do not recall having to change this flag with a 82240B, perhaps when an 82240B gets graphics it ignores the extra linefeed. You could probably do the same. My output is here: http://sense.net/~egan/82240E/ OS/X command: ./hp82240 /dev/cu.KeySerial1 foo2.pmb Ctrl-C when done. === Subject: Re: Issue with printing from HP50g Le jeudi 08 janvier 2009, datajerk@gmail.com a .8ecrit : >> http://pagesperso-orange.fr/kdntl/hp49/hp82240_emulator.tar.gz Great start. Very nice. Feature request. When using conn4x with > Windows I can grab a screen shot via USB or Serial (IIRC) without > using PRLCD, I can just ask for a screen shot from conn4x. Can you > add a flag to hp82240 to just snag the screen on connect and exit? I've just added an optional third argument to specify how many lines the program should print before exiting. Just use 8 for 131x64 screens (because 64=8*8) or 10 for 131x80 screens (because 80=10*8). Please download the new version at the same location. The documentation was also updated. I'm not sure the screenshot feature of conn4x would work. I believe it won't. IIRC, to use this feature, one has to press [ON]+[PRG] on HP48S/SX, [ON]+[1] on HP48G/GX or HP49G and [ON]+[up arrow] on HP49G+/50g. Could someone confirm or correct this? I tried to get the part of the ROM that does the job of printing with the [ON]+[] way. Obviously, these keys are not handled like normal keys, because the [ON] key triggers an interruption and, surely, it is a lot more complicated. Anyone has some information? > My output is here: http://sense.net/~egan/82240E/ Well, it works as expected with your setup :) I really don't know what to blame, my HP50g or the linux usb-serial driver? > OS/X command: ./hp82240 /dev/cu.KeySerial1 foo2.pmb Ctrl-C when done. You need no sniffer to reverse-engineer the screenshot via conn4x way. (Actually, it's not reverse engineering because I'm pretty sure it's documented somewhere, at least in Conn4x's source code -- available on hpcalc.org, as always.) The following command should enable you to get a dump: $ cat foo.dump === Subject: 49g+ sometimes freezes Hi like in the object my 49g+ suffers of a strange behaving. In fact sometimes when I press a button it seem to freeze and to think then it does the action of the just-pressed button. I find this thing annoying and dangerous because if you don't spot this behaving during calculations you can miss someting with wrong results and you loose time. My rom is 2.09and the libraries installed are: tgviewer2, unitman, chemlab 2.7, vigag 4.1, truss, sectn49, fem49 and equation library and data have you any hint?Can you help me? Duncan === Subject: Re: 49g+ sometimes freezes Hiding in the (not distributed with calculator) AUR's description of the MEM command is this small hint about what may be happening: Before it can assess the amount of memory available, MEM must remove objects in temporary memory that are no longer being used. This clean-up process (also called garbage collection ) also occurs automatically at other times when memory is full. Since this process can slow down calculator operation at undesired times, you can force it to occur at a desired time by executing MEM. In a program, execute MEM DROP. More thoughts, plus a program FMEM to cause GC to be more frequent, but correspondingly more brief, thus less noticeable or intrusive: 50G - sporadic pauses - garbage collection? The discussion and programs in the above thread pertain to all models in the HP48/49/50 series. === Subject: Re: 49g+ sometimes freezes posting-account=sOAX1QkAAAC-FcySTSbz29Uk8huUtFRz CLR 1.1.4322; .NET CLR 2.0.50727),gzip(gfe),gzip(gfe) > Before it can assess the amount of memory available, > MEM must remove objects in temporary memory > that are no longer being used. This clean-up process > (also called garbage collection ) also occurs automatically > at other times when memory is full. Since this process > can slow down calculator operation at undesired times, > you can force it to occur at a desired time by executing MEM. > In a program, execute MEM DROP. If you happen to press a key at the same time as a garbage collection, your keypress will not be processed until you press *another* key. Thus the keystroke was not missed by the calculator; its response is simply delayed. Sometimes when this happens to me I assume it was a missed keystroke or that I did not press the key all the way down and will repeat the keystroke, only to find that it has now doubled because of the garbage collection issue. A minor inconvenience, but just keep your eye on the display, and use mental estimation (from the slide rule days) to assess if the calculator's answer is reasonable before accepting it. S.C. === Subject: Re: 49g+ sometimes freezes > If you happen to press a key at the same time as a garbage collection, > your keypress will not be processed until you press *another* key. That's another good thought -- I think it's when you press a key at exactly the time that the (ARM) processor is going into an idle state, however, rather than a (Saturn, emulated) GC. If the final response to a keypress is only delayed, but doesn't wait until the next button is pressed, then it's more likely due to a GC, but if the response never comes at all, until precisely when the next key is pressed, then it's more likely the idle bug (or Busy bug ) Busy Bug? [Dec 2006] That is also a very old bug (but still present in recent HP50G, AFAIAA), and likely resides in the Kinpo internal emulator: ROM 2.01 build 83 ? [Dec 2005] Joe Horn points out, in great detail, in the last post of the last thread (remainder quoted from Joe Horn): > Would the delayed stroke eventually execute, if you simply wait. Two possibilities: IF it's delayed by a garbage collection, THEN the answer is definitely yes: the busy annunciator will be ON, nothing will seem to be happening (during the garbage collection process), and then the delayed keystrokes (up to 15 of 'em) get executed. During ordinary operations, garbage collection is extremely fast on the 49g+, but the more different unreferenced things there are on the stack, the longer it takes. IF it's delayed by The Busy Bug, THEN the answer is barely yes: the busy annunciator will be OFF, nothing will in fact be happening, and then when the next system interrupt occurs the key will execute. The next system interrupt would ordinarily be caused by the next keystroke, but if you don't press another key and just wait, then the next system interrupt might take a long time to occur (depending on your TOFF setting, alarms, clock display, and so on). [End] === Subject: stress concentration factors Hi some time ago for an exam I found the Mollier diagram. Now for another exam I have paper tables for stress concentration factors for bending and torsion of a round bar with a u groove and of a stepped round bar with a shoulder fillet for studying fatigue, but they aren't handful so I'd like to know if there is something about that for a 49g+. === Subject: Re: stress concentration factors InfoPath.1; .NET CLR 2.0.50727; .NET CLR 3.0.04506.30; .NET CLR 1.1.4322; .NET CLR 3.0.04506.648),gzip(gfe),gzip(gfe) >stress concentration factors for bending and > torsion of a round bar with a u groove and of a stepped round bar with a > shoulder fillet for studying fatigue, but they aren't handful so I'd like to > know if there is something about that for a 49g+. The best place to look, as always, is www.hpcalc.org. Try the sections for civil engineering and physics/structural. This one might be interesting: http://www.hpcalc.org/details.php?id=4515 Bill === posting-account=2sgSMAoAAADHaHf0CqF6w39rFfvaJpp9 2.0.50727; .NET CLR 3.0.04506.648; .NET CLR 3.5.21022; .NET CLR 1.1.4322; MS-RTC LM 8; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729; Zune 3.0; WWTClient2),gzip(gfe),gzip(gfe) > Many years ago I did it for my HP48SX. http://www.criseis.ruhr.de/hp48/hp48b.html They will work in the GX, too. up!) I may end up going that route but I was hoping for gx-style bank switching, which from what I understand your design doesn't do. Quick question, when user memory is erased due to a bad syseval or user reset, I'm under the impression all the data from ram cards is erased as well. Is that correct? > >Alternatively, has anyone tried building an adapter > >that would allow someone to use an SD memory card in the gx? It is possible to make a card reader for SD and use it with your own > driver. But it is not possible to use a SD-Card like a ramcard, > because SD are to slow. I haven't investigated it at all, but I figured it would require some significant circuitry between the SD card and the 48, possibly even an additional microprocessor to handle the data transfers. > Oh..and did you realy think you need more than > 1MByte in your pocketcalculator? Nope, not at all... unless you think 1 MB isn't enough for me to download the internet. ;) I was thinking about SD simply because it is so cheap and readily available. > Another funny idea that sometimes come to my brain, building a card > with a FPGA on it. This can be memory, soundcard, cardreader, > math-coproz and many more things. Than I feel lazy soon and stop this > kind of idea. :-) Well get off your rear and get back to work! ;) === >switching, which from what I understand your design doesn't do. Quick >question, when user memory is erased due to a bad syseval or user >reset, I'm under the impression all the data from ram cards is erased >as well. Is that correct? If the card is used to expand main memory it is correct, if it used like a memory card for data and libary nothing is lost. >I haven't investigated it at all, but I figured it would require some >significant circuitry between the SD card and the 48, possibly even an >additional microprocessor to handle the data transfers. It depends on your need for speed. :-) I think the easiest possible interface is a driver like 74HC573 and a switchable buffer for reading HC245. In this case you have to program each line of SPI. This will work, but it is slow. A little bit easier is to use a device that will translate 8bit memory writing/reading to SPI. Unfortunatly I dont know a single device that can do it when you buy it. But of course it is easy to implement something like this in a small CPLD like a XC9572. For example you can connect a CPLD like this to the HP48 bus and a 128kb ram on the other side. Than it is a ramcard you can use in slot1. In your Verilog program it will look for reading for some special behavier (for example read from byte1 and than byte0) and it will connect one adress to your SPI-uart. But before you are thinking about it seriosly you should check the amount of power a CPLD will need. It is possible the battery will last shorter than the usual year you got today. :-) Oh..when I am thinking about it, it is possible that there are some old UART-device that can provide syncronous transfer, too. You can check for this, too. Unfortunatly the old one are made for 5V and you also can not use 40pin DIP. >download the internet. ;) I was thinking about SD simply because it >is so cheap and readily available. You can think about buying an old smartmedia card from ebay. Perhaps 64MB or 128MB is enought for you, too. This card have a 8Bit bus (data and adress multiplexed) I think it is very easy to connect it to the HP48. Some years ago I have seen a page in internet where someone connect it to a Sharp E500 pocketcalculator for the same reason. >Well get off your rear and get back to work! ;) Sorry, on my desktop are waiting now five old Palmpilot from Ebay for some tuning. :) I am still the owner of three HP48SX (one at home, one at company, one for emergency cases) but in the last years I only use them for calculating. Olaf === > When user memory is erased due to a bad syseval or user reset, > I'm under the impression all the data from ram cards > is erased as well. Is that correct? As best my dim data store recalls: User reset (whether ON+C, paper clip in reset hole, or even ON+A+F to erase all user memory): No -- I believe any damage from a reset done to recover from a Memory Lost is already done by the bad program, see below); the basic design intended that even erasing memory via ON+A+F should not affect cards, Bad program (OS wipeout and Memory Lost, that is, since other mere hangs may be more benign): GX switched banks in slot 2 are mapped in memory space only when necessary, _if_ slot 1 is occupied by a card, hence slot 2 is more protected, whereas in slot 1, either port 1 or merged port 0 is always mapped, but then: Write-protect mechanisms exist (if used) to protect each slot, and HP48GX system software is designed to somewhat isolate even writeable ports in slot 2, but there's a lot of detail involved in analyzing this, for which you can directly Ask Dave (Arnett, who designed the circuitry for the GX), about the intricacies of the original HP48, years before joining HP to create the HP49G (and is still reading and contributing here): === posting-account=2sgSMAoAAADHaHf0CqF6w39rFfvaJpp9 2.0.50727; .NET CLR 3.0.04506.648; .NET CLR 3.5.21022; .NET CLR 1.1.4322; MS-RTC LM 8; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729; Zune 3.0; WWTClient2),gzip(gfe),gzip(gfe) run across that yet. > http://www.hpcalc.org/details.php?id=3530maybe? === posting-account=n58npQkAAACrXE24A7MGIDa1g25mTFtC GTB5; .NET CLR 1.1.4322; .NET CLR 2.0.50727),gzip(gfe),gzip(gfe) > run across cost more than I'm willing to pay. Has anyone documented > the connector pinout, data transfer protocol, etc. for rolling your > own memory card? Alternatively, has anyone tried building an adapter > that would allow someone to use an SD memory card in the gx? I've asked for the same kind of thing about a year ago but couldn't find the ressources to make the project. I'm sure it's possible to drive a SD card in SPI mode (in and out) with a 48gx but the problem is to communicate through the HP card port and I don't know how to achieve this. val === >I'm sure it's possible to drive a SD card in SPI mode (in and out) >with a 48gx but the problem is to communicate through the HP card port >and I don't know how to achieve this. You can use two register/bus driver and make an 8bit port for reading and writing. This is easy, but it is very slow. Olaf === posting-account=n58npQkAAACrXE24A7MGIDa1g25mTFtC GTB5; .NET CLR 1.1.4322; .NET CLR 2.0.50727),gzip(gfe),gzip(gfe) >I'm sure it's possible to drive a SD card in SPI mode (in and out) > >with a 48gx but the problem is to communicate through the HP card port > >and I don't know how to achieve this. You can use two register/bus driver and make an 8bit port for reading > and writing. This is easy, but it is very slow. Olaf This is exactly what I was looking for ! Do you know how to do it on a card port ? val === >This is exactly what I was looking for ! Do you know how to do it on a >card port ? You can do it easiely by the two circuit I told you in last news. But perhaps it is a little bit more easier to use an 82C55 if you can find one in SMD. It will give you three 8Bit port you can read and write. Olaf === Subject: Re: 49g+ sometimes freezes > When your hp49g+ freezes it does something knows as the Garbage > collector: the system deletes objects which are not uses anymore... > The hp49g+ works like this: it creates objects in ram and when ram > is full the hp49g+ makes cleaning: some objects are useless so the system > put them in trash (to free some memory...) and it could takes > times... So there is no solutions except do the garbage collector when you > want... I find this behaving very strange because when it was new it didn't happen. Moreover it's strange because it happens quite often and also using only the stack for calculations. I don't understand how do the garbage collection when I want === Subject: Re: 49g+ sometimes freezes reply-type=response > I find this behaving very strange because when it was new it didn't > happen. Moreover it's strange because it happens quite often and also > using only the stack for calculations. I don't understand how do the > garbage collection when I want The more full your calculator's memory is, the more frequently garbage collection occurs, so this makes sense if you are using a lot more of your calculator's memory now than you did before. You can try purging things you don't need (or move them to flash) to reduce the frequency of garbage collection. You can perform a garbage collection on demand by running MEM. This collects garbage and reports the amount of free memory in bytes (which you can DROP). Eric Rechlin === Subject: Re: stress concentration factors Bill Markwick ha scritto nel messaggio >stress concentration factors for bending and > torsion of a round bar with a u groove and of a stepped round bar with a > shoulder fillet for studying fatigue, but they aren't handful so I'd like > to > know if there is something about that for a 49g+. The best place to look, as always, is www.hpcalc.org. Try the sections for civil engineering and physics/structural. This one might be interesting: http://www.hpcalc.org/details.php?id=4515 Bill anything useful. Moreover I don't know if I explained myself well, but fem48 isn't helpful === Subject: Re: Syntax of all HP48 HP50g commands <629nm4pblbeltang0405gmk7qnv15vgmum@4ax.com> posting-account=ZnpnxQoAAAAXzWo26i-7z19miWsTIM3P AppleWebKit/525.19 (KHTML, like Gecko) Chrome/1.0.154.36 Safari/525.19,gzip(gfe),gzip(gfe) Hey Guys I had a HP48g and now a 50g. >Sometimes I missed a Command Syntax Help to the calculator. >Something that shows all commands AND the syntax of each command. A >kind of HELP. >This is because several times I just can«t remember the syntax of each >command. And during my classes, I don«t have access to the Manual. This should be great even if it was just a TXT file saved in memory. >In new 50g we have the CATALOG menu, but there are just some commands >inside. Do you know something about this? > http://www.hpcalc.org/details.php?id=5973 Damir I was looking for this software for a long time! === Subject: Re: Issue with sat_pop_real() posting-account=KxEIGwoAAAALMxy6hykxLK1wPdiyuNkK Gecko/2008120122 Firefox/3.0.5,gzip(gfe),gzip(gfe) > I don't think this is a sat pop real() problem. Instead, I think that > you're overflowing the precision of a long double in hpgcc/ARM. Yes. Please read the whole page, (or at least this section) from the > fount of all knowledge for an important summary: http://en.wikipedia.org/wiki/Floating-point#Accuracy problems There is a problem with sat push real() that I have seen before where > the number is pushed to the stack correctly, but displays a leading > 0. I haven't been able to figure it out yet, and haven't seen it in > quite a while, and it doesn't cause any problems with either mp math > routines, or the hpgcc stack interaction stuff. Basically you get a > number on the stack displayed as 09.1234 or something like that. TW Looking at the code and assuming that ARM uses IEEE floating point, it's possible that sat double2real() doesn't properly handle the case when the number approaches zero. I'm reaching back 20+ years to grad school here, but as I recall, if you keep dividing a floating point double by 2, you eventually reach the minimum exponent, and then the number denormalizes. As you keep dividing by two, the exponent remains the same, but you loose the implicit 1 to the left of the radix point and the mantissa shifts right giving you smaller and smaller numbers with fewer and fewer significant digits until you reach zero. === Subject: Re: LINSOLVE function I am using OS version 2.09 and algebraic mode with numeric, approx and complex all un-checked. When I enter LINSOLVE([x+3*y=1,2*x+4*y=1],[x,y]) I get the following returned {{[x+3*y=1 2*x+4*y=1] [x,y]} Specific: 2 -2 2 1.000} [x= -1/2 y=1/2]} What is Specific: 2 -2 2 1.000 for and what does it mean? === Subject: Re: 2009 HHC HP Calendar To Be Ready Next Week posting-account=K31tcwgAAABVnT_oG3A7BQonvJCmOPn4 1.1.4322; .NET CLR 2.0.50727; InfoPath.1; .NET CLR 3.0.04506.30; .NET CLR 3.0.04506.648; .NET CLR 3.5.21022; MS-RTC LM 8),gzip(gfe),gzip(gfe) Just FYI, the 2009 HP calendars (sponsored by HHC2007/HHC2008) will be > returning from the printers in time to start being mailed next week. > ALL ATTENDEES OF HHC2007 (in San Diego) WILL AUTOMATICALLY BE SENT A > CALENDAR, since their conference registrations financed the printing > of this year's calendar. For those who did not attend HHC2007 but wish to purchase the > calendar, more information on that will be available next week also at Jake Schwartz I just wanted to say that the calendars have begun shipping to the HHC2007 attendees, starting on Tuesday 1/13. So far, 16 have gone out and a handful is being mailed each day until the whole lot is exhausted. So, look for yours in the mail soon. (Again, if you don't remember whether you were there at the 2007 conference, consult http://www.pahhc.org/2009/Calendar.htm at the bottom of the page.) Jake === Subject: addrATTNFLG, addrTEMPENV, etc. posting-account=eF2f0AoAAAB2spBRiZOs91ItDKLGDCIk Gecko/2008120122 Firefox/3.0.5,gzip(gfe),gzip(gfe) i just had a hard time trying to build a source code which would stay valid accross all HP4x versions, starting with HP48S Rom A. One of my issue was ATTNFLG, which was not correctly interpreted on 48S. Hopefully, there is addrATTNFLG, which, as stated by the documentation, holds the address of ATTNFLG in a version-independant manner . But this didn't work. Up to the point i discovered this documentation : http://www.hpcalc.org/hp48/docs/programming/ramptrs.txt So , according to this documentation, you need to use addrATTNFLG+2. +2 ? How come ? And it's not limited to addrATTNFLG. Many other entries need also to be modified manually with +2 : addrTEMPENV, addrTEMPTOP, adrTIMEOUTCLK, etc. However, i can't help wondering : What's the logic behind this +2 ? Why not providing the right address immediately in the entry list for the compilator ? === Subject: Re: addrATTNFLG, addrTEMPENV, etc. =addrATTNFLG is a location in ROM, which actually points to a D0=(5) =ATTNFLG Since the address field of this opcode is preceeded by 1B (for D0=(5)), one has to read from a address which is two nibs higher to get the actual =ATTNFLG address as coded in the ROM. Therefore: D0=(5) (=addrATTNFLG)+2 C=DAT0 A * Now the actual address of =ATTNFLG is in C[A] . Again, most these addr... entries point to the first nib of an opcode, not to the address. A few years ago I also made intensive use of these version-independent locations, but if you want to have one source file for all platforms, the real HP-48 and the doorstop calcs, you could consider using conditional assembly;-) HTH Raymond Yann schrieb im Newsbeitrag i just had a hard time trying to build a source code which would stay > valid accross all HP4x versions, starting with HP48S Rom A. > One of my issue was ATTNFLG, which was not correctly interpreted on > 48S. Hopefully, there is addrATTNFLG, which, as stated by the > documentation, holds the address of ATTNFLG in a version-independant > manner . But this didn't work. Up to the point i discovered this documentation : > http://www.hpcalc.org/hp48/docs/programming/ramptrs.txt So , according to this documentation, you need to use addrATTNFLG+2. +2 ? How come ? And it's not limited to addrATTNFLG. Many other entries need also to > be modified manually with +2 : addrTEMPENV, addrTEMPTOP, > adrTIMEOUTCLK, etc. However, i can't help wondering : > What's the logic behind this +2 ? > Why not providing the right address immediately in the entry list for > the compilator ? === Subject: Re: addrATTNFLG, addrTEMPENV, etc. posting-account=eF2f0AoAAAB2spBRiZOs91ItDKLGDCIk Gecko/2008120122 Firefox/3.0.5,gzip(gfe),gzip(gfe) By the way, is there any documentation on conditionnal assembly available ? I found the SASM doc file a bit too short to properly understand how it works. === Subject: Re: addrATTNFLG, addrTEMPENV, etc. Le vendredi 16 janvier 2009, Yann a .8ecrit : > By the way, is there any documentation on conditionnal assembly > available ? > I found the SASM doc file a bit too short to properly understand how > it works. In your source code, just use (you need hptools >= 3.0.3): code for hp49 #ELSE code for other #ENDIF Note that the #ELSE directive is optional. You may use code for hp49 #ENDIF code for hp48g/gx #ENDIF code for hp48s/sx #ENDIF command line (and you have to set it, eventually to zero). Now, if you want to compile your code for HP49, run sasm with: If you are targetting HP48G, use this instead: And so on. You can use the #IF #ENDIF directives in rplcomp's code too, because rplcomp won't process them. It only outputs them verbatim in the assembly file to be read later by sasm. Of course, using tools such as Makefile greatly facilitate the development process. === Subject: Re: addrATTNFLG, addrTEMPENV, etc. posting-account=eF2f0AoAAAB2spBRiZOs91ItDKLGDCIk Gecko/2008120122 Firefox/3.0.5,gzip(gfe),gzip(gfe) > =addrATTNFLG > is a location in ROM, which actually points to a > D0=(5) =ATTNFLG > Since the address field of this opcode is preceeded by 1B (for D0=(5)), > one has to read from a address which is two nibs higher to get the actual > =ATTNFLG address as coded in the ROM. > This is much clearer, and makes sense now. As it is an opcode, i though for a while that addrATTNFLG was supposed to be used this way : GOSBVL =addrATTNFLG but this is not correct. So i guess this is only a part of a larger program, most probably the SysRpl ATTN? call. As a consequence, there is really no other way to use this field than Dn(5) =addrATTNFLG+2 > This is from the post > where you will also find a hint by Mika Heiskanen how to test if you > are running on a SX or GX. You always seem to have the right link to the right information Andreas, even very distant ones. A few years ago I also made intensive use of these version-independent > locations, but if you want to have one source file for all platforms, > the real HP-48 and the doorstop calcs, > you could consider using conditional assembly;-) > Yes, you're right Raymond. I still have to make my first foray into conditionnal assembly though. I'm mostly embarassed by Debug4x breakpoint limitations when using INCLUDED files, so trying to keep a single source file with a different header per project (per target calc) is still source for concerns for me. === Subject: Re: addrATTNFLG, addrTEMPENV, etc. posting-account=57eBeAkAAABkNMhsZnDBzGQsstJRmvKD Gecko/2008120122 Firefox/3.0.5,gzip(gfe),gzip(gfe) Hello Yann, I can«t help you with the +2. table. GX (**) SX NIB LENGTH: PURPOSE 807F7**70679=ATTNFLG 5: ATTN flag; ON (CANCEL) was pressed The GX address is also valid for the 49G series. This is from the post where you will also find a hint by Mika Heiskanen how to test if you are running on a SX or GX. HTH, Andreas http://www.software49g.gmxhome.de === Subject: creating tables and charts posting-account=Iz6FlwoAAABdQTI08xdVoDk-iYgiEfif .NET CLR 2.0.50727; Media Center PC 5.0; .NET CLR 3.0.04506; InfoPath.2),gzip(gfe),gzip(gfe) i have finding a way to write some tables and charts that are located in the back of my chemistry text. Unfortunately some of the have multiple submenus. I don't understand the manual whatsoever. Is there anyone who may help me with this? John === Subject: Re: creating tables and charts posting-account=eF2f0AoAAAB2spBRiZOs91ItDKLGDCIk Gecko/2008120122 Firefox/3.0.5,gzip(gfe),gzip(gfe) What do you mean ? You are trying to create menus which call menus ? === Subject: Eneloops posting-account=7ypjXwoAAAA6Phe0dE2NwavGiqMYGDdw Presto/2.1.1,gzip(gfe),gzip(gfe) I've been using Eneloops for my camera exclusively and have been very happy with them, so I decided to get some for my 50g. For some reason, though, it won't work with them. It doesn't even turn on. If I replace one Eneloop with a normal NiMH, it works. I haven't broken them in yet or anything, so trying discharging and charging them a few times. Anyone here have luck with them? Christoph === Subject: Re: Eneloops maybe one of the battery springs doesn't apply the needed pressure? I think there was a discussion about early units having too short battery terminals... HTH Raymond Christoph Koehler schrieb im Newsbeitrag I've been using Eneloops for my camera exclusively and have been very > happy with them, so I decided to get some for my 50g. For some reason, > though, it won't work with them. It doesn't even turn on. If I replace > one Eneloop with a normal NiMH, it works. > I haven't broken them in yet or anything, so trying discharging and > charging them a few times. Anyone here have luck with them? === Subject: Re: Eneloops maybe one of the battery springs doesn't apply the needed pressure? I think there was a discussion about early units having > too short battery terminals... HTH Raymond I think there's plenty of pressure. The batteries don't seem loose. Plus I can switch out one Eneloop for a standard NiMH and it works every time. I bought the calc about 1 year ago, so it's not all that old. Christoph === Subject: Re: Eneloops Hi Christoph, do all of the Eneloops have the needed voltage? I don't know the minimum voltage the 50g can work with, but since it seems to accept rechargeables, it should work with about 4.8V onwards. HTH Raymond Christoph Koehler schrieb im Newsbeitrag maybe one of the battery springs doesn't apply the needed pressure? I think there was a discussion about early units having > too short battery terminals... HTH Raymond I think there's plenty of pressure. The batteries don't seem loose. Plus I can switch out one Eneloop for a standard NiMH and it works every time. I bought the calc about 1 year ago, so it's not all that old. Christoph === Subject: Re: Eneloops posting-account=7ypjXwoAAAA6Phe0dE2NwavGiqMYGDdw Presto/2.1.1,gzip(gfe),gzip(gfe) > Hi Christoph, do all of the Eneloops have the needed voltage? I am not sure, don't have a multimeter. I assume so, they have been freshly charged and have the same supposed voltage as the substitute that does work: 1.2V. Christoph === Subject: HP48 + Arduino Duemilanove posting-account=EjMqYwoAAADPG-eKZyudrVG4uwsv15vp .NET CLR 2.0.50727; .NET CLR 3.0.04506; Tablet PC 2.0; .NET CLR 3.5.21022; OfficeLiveConnector.1.3; OfficeLivePatch.0.0),gzip(gfe),gzip(gfe) Duemilanove. I wired the serial port, but when I send AB for example, I get }{ on my calculator. In fact, there are some chars that I've found searching for a translation : REC DEC_R BIN_R BIN_S SENT BIN_S y 121 1111001 1000011 C 0000110 w 119 1110111 1000100 D 0001000 u 117 1110101 1000101 E 0001010 s 115 1110011 1000110 F 0001100 q 113 1110001 1000111 G 0001110 o 111 1101111 1001000 H 0010000 So, if I send CDEFGH into my HP48g, I get ywusqo with << 6 SRECV >> More about: The CDEFGH string looks like this in bin: 100001110001001000101100011010001111001000 And the received string looks like this: 111100111101111110101111001111100011101111 If we apply NOT to the received binary string we get: 000011000010000001010000110000011100010000 If you look carefully it is like almost the original string, but: M000011_000100_000101_000110_000111_0010000 M stands for a missing bit, and _ is a 0 that must be a 1. Why can this happen? I tried with another arduino, my oldest decimila model and I have connected the HP48g directly to my PC to test it, and with Hyperterminal works very well, so isn't a hardware problem. Can this experience be done with a HP50g and the newest serial port? === Subject: Re: HP48 + Arduino Duemilanove posting-account=KxEIGwoAAAALMxy6hykxLK1wPdiyuNkK Gecko/2008120122 Firefox/3.0.5,gzip(gfe),gzip(gfe) > Duemilanove. I wired the serial port, but when I send AB for example, I get }{ > on my calculator. In fact, there are some chars that I've found > searching for a translation : REC DEC R BIN R BIN S SENT BIN S > y 121 1111001 1000011 C 0000110 > w 119 1110111 1000100 D 0001000 > u 117 1110101 1000101 E 0001010 > s 115 1110011 1000110 F 0001100 > q 113 1110001 1000111 G 0001110 > o 111 1101111 1001000 H 0010000 So, if I send CDEFGH into my HP48g, I get ywusqo with << 6 SRECV More about: > The CDEFGH string looks like this in bin: > 100001110001001000101100011010001111001000 And the received string looks like this: > 111100111101111110101111001111100011101111 If we apply NOT to the received binary string we get: > 000011000010000001010000110000011100010000 If you look carefully it is like almost the original string, but: > M000011 000100 000101 000110 000111 0010000 M stands for a missing bit, and is a 0 that must be a 1. Why can this happen? I tried with another arduino, my oldest decimila model and I have > connected the HP48g directly to my PC to test it, and with > Hyperterminal works very well, so isn't a hardware problem. Can this experience be done with a HP50g and the newest serial port? > It looks to me like you have a signal inverted somewhere. Maybe you've swapped the signal and ground pins in your cable connector somewhere? Double check the pin outs. With the signal inverted, the start bit gets missed, then the transition to the first data bit gets interpreted as the start bit, then the transition to the second data bit gets interpreted as the first data bit, only reversed because the signal is inverted etc. Dave === Subject: Re: HP48 + Arduino Duemilanove posting-account=EjMqYwoAAADPG-eKZyudrVG4uwsv15vp .NET CLR 2.0.50727; .NET CLR 3.0.04506; Tablet PC 2.0; .NET CLR 3.5.21022; OfficeLiveConnector.1.3; OfficeLivePatch.0.0),gzip(gfe),gzip(gfe) Duemilanove. I wired the serial port, but when I send AB for example, I get }{ > on my calculator. In fact, there are some chars that I've found > searching for a translation : REC DEC R BIN R BIN S SENT BIN S > y 121 1111001 1000011 C 0000110 > w 119 1110111 1000100 D 0001000 > u 117 1110101 1000101 E 0001010 > s 115 1110011 1000110 F 0001100 > q 113 1110001 1000111 G 0001110 > o 111 1101111 1001000 H 0010000 So, if I send CDEFGH into my HP48g, I get ywusqo with << 6 SRECV More about: > The CDEFGH string looks like this in bin: > 100001110001001000101100011010001111001000 And the received string looks like this: > 111100111101111110101111001111100011101111 If we apply NOT to the received binary string we get: > 000011000010000001010000110000011100010000 If you look carefully it is like almost the original string, but: > M000011 000100 000101 000110 000111 0010000 M stands for a missing bit, and is a 0 that must be a 1. Why can this happen? I tried with another arduino, my oldest decimila model and I have > connected the HP48g directly to my PC to test it, and with > Hyperterminal works very well, so isn't a hardware problem. Can this experience be done with a HP50g and the newest serial port? > It looks to me like you have a signal inverted somewhere. Maybe > you've swapped the signal and ground pins in your cable connector > somewhere? Double check the pin outs. With the signal inverted, the start bit gets missed, then the > transition to the first data bit gets interpreted as the start bit, > then the transition to the second data bit gets interpreted as the > first data bit, only reversed because the signal is inverted etc. Dave- Ocultar texto de la cita - - Mostrar texto de la cita - I managed to solve the problem. HP now has a small translator and the messages arrives correctly: For further info on this issue, my custom translator is: << -> S Q << S 1 S SIZE START DUP NUM 255 - NEG 2 / CHR Q SWAP + 'Q' STO TAIL NEXT DROP Q >> >> So, you receive the data with << BUFLEN DROP SRECV DROP >> === Subject: Re: HP48 + Arduino Duemilanove Your tests seem to be pointing towards a synchronization issue. In this case, solution is a bit difficult : you may have some results by caring about DCE/DTE role : one of the devices must provide the synchronisation, the other one must follow. ) === Subject: A Game for Hp 50g like games please test it and give me your suggestions. You can download a prealfa version from: http://mnavarro.fortunecity.es/BMprealfaen.htm === Subject: Re: Eneloops . For some reason, > though, it won't work with them. It doesn't even turn on. If I replace > one Eneloop with a normal NiMH, it works. I had the same thing happen twice. The first time was with regular alkalines, and the problem was that one cell was completely open (it was ohmless!). The second time was with regular NiMH cells and the problem was the battery contacts in the calculator - although the cells seemed to fit perfectly, the calc's positive contact wasn't quite long enough for certain brands of batteries (it worked fine with most alkalines). The cure was to pry out the positive contact a little bit with a tiny screwdriver - I've had no trouble since. It's pretty silly when a high-tech company can't manage to make a reliable batteryholder. But that's how it goes these days - I bought a box of nails and half of them had the heads on the wrong end. :-) === Subject: Re: Eneloops . For some reason, > though, it won't work with them. It doesn't even turn on. If I replace > one Eneloop with a normal NiMH, it works. I had the same thing happen twice. The first time was with regular alkalines, and the problem was that one cell was completely open (it was ohmless!). The second time was with regular NiMH cells and the problem was the battery contacts in the calculator - although the cells seemed to fit perfectly, the calc's positive contact wasn't quite long enough for certain brands of batteries (it worked fine with most alkalines). The cure was to pry out the positive contact a little bit with a tiny screwdriver - I've had no trouble since. It's pretty silly when a high-tech company can't even manage to make a reliable batteryholder. But that's how it goes these days - I bought a box of nails and half of them had the heads on the wrong end. :-) === Subject: Re: Eneloops > I had the same thing happen twice. The first time was with regular > alkalines, and the problem was that one cell was completely open (it > was ohmless!). The second time was with regular NiMH cells and the > problem was the battery contacts in the calculator - although the > cells seemed to fit perfectly, the calc's positive contact wasn't > quite long enough for certain brands of batteries (it worked fine with > most alkalines). The cure was to pry out the positive contact a > little bit with a tiny screwdriver - I've had no trouble since. Bill, positive end, had a small plastic ledge between the two contact wires. Scraped that off and bent the contacts out a bit, and it works now. Wow, that's an odd issue. === Subject: Re: Eneloops I use Eneloops exclusively on my HP 50g. Never had problems. Been using them for about a year and a half now. Try recharging the eneloops. Might also be one faulty battery or pack. On 16 ene, 14:20, Christoph Koehler I've been using Eneloops for my camera exclusively and have been very > happy with them, so I decided to get some for my 50g. For some reason, > though, it won't work with them. It doesn't even turn on. If I replace > one Eneloop with a normal NiMH, it works. > I haven't broken them in yet or anything, so trying discharging and > charging them a few times. Anyone here have luck with them? === Subject: HP50g won't graph I had a bunch of equations saved on my 50g. I didn't know how to clear them so I deleted twelve files in my Home directory from Y1=, Y2=,. . .Y12=. Now I can't graph anything. When I press LShift+F1 or go to Equation Entry from APPS, I get Error: Too Few Arguments . Can anyone help me? Stella === Subject: Re: HP50g won't graph I had a bunch of equations saved on my 50g. I didn't know how to clear > them so I deleted twelve files in my Home directory from Y1=, > Y2=,. . .Y12=. Now I can't graph anything. When I press LShift+F1 or > go to Equation Entry from APPS, I get Error: Too Few Arguments . Can > anyone help me? Stella Quick solution: press ON+A+F, then F. (Libraries on port 2 won't be deleted) === Subject: Re: HP50g won't graph Hi On 2009-01-19 09:40:13 +1100, Avatar_e said: > > Quick solution: press ON+A+F, then F. (Libraries on port 2 won't be > deleted) A bit drastic don't you think ? === Subject: Re: HP50g won't graph Quick solution: press ON+A+F, then F. (Libraries on port 2 won't be > deleted) I'm curious, what does the command ON+A+F, then F do? Clear something? Is that little trick in the manual? Apparently there are a lot of tricks and quirks that one needs to know. Has anyone compiled a list somewhere? === Subject: Re: HP50g won't graph I'm curious, what does the command ON+A+F, then F do? Clear > something? Is that little trick in the manual? Apparently there are > a lot of tricks and quirks that one needs to know. Has anyone > compiled a list somewhere? Page 842 (G-3) of the HP 50g User's Guide (PDF on the CD): ON+A+F: Cold restart: all memory erased ON+B: Cancel ON+C: Warm restart: memory preserved ON+D: Self-Test ON+E: Automatic self-test (ON+C to exit) === Subject: Re: HP50g won't graph I'm curious, what does the command ON+A+F, then F do? Clear > something? Is that little trick in the manual? Apparently there are > a lot of tricks and quirks that one needs to know. Has anyone > compiled a list somewhere? Page 842 (G-3) of the HP 50g User's Guide (PDF on the CD): ON+A+F: Cold restart: all memory erased > ON+B: Cancel > ON+C: Warm restart: memory preserved > ON+D: Self-Test > ON+E: Automatic self-test (ON+C to exit) S.C. these things. When the manual says all memory erased surely that doesn't include the operationg system? What do they mean by all memory? And what memory is preserved with warm restart? Variable values? Flags? Graphing equations such as Y1(x), Y2(x), etc.? === Subject: Re: HP50g won't graph Also take a look at these very helpful unofficial documentations: http://www.hpcalc.org/hp49/docs/misc/ Cold restart returns everything to factory conditions. Warm restart seems leaves home director, settings and flags untouched, and deletes graphing equations except Y1. === Subject: Re: HP50g won't graph > Also take a look at these very helpful unofficial documentations:http://www.hpcalc.org/hp49/docs/misc/ Cold restart returns everything to factory conditions. Warm restart > seems leaves home director, settings and flags untouched, and deletes > graphing equations except Y1. === Subject: Re: HP50g won't graph posting-account=sOAX1QkAAAC-FcySTSbz29Uk8huUtFRz CLR 1.1.4322; .NET CLR 2.0.50727),gzip(gfe),gzip(gfe) I'm curious, what does the command ON+A+F, then F do? Clear > something? Is that little trick in the manual? Apparently there are > a lot of tricks and quirks that one needs to know. Has anyone > compiled a list somewhere? ON+A+F: reset ON+C: warmstart (less drastic than ON+A+F) S.C. === Subject: Re: HP50g won't graph posting-account=hlK3EAoAAADTMH46dcKhe8xdTh6-UarT Gecko/2008103100 SUSE/3.0.4-3.1 Firefox/3.0.4,gzip(gfe),gzip(gfe) Hi everyone, I had a bunch of equations saved on my 50g. I didn't know how to clear > them so I deleted twelve files in my Home directory from Y1=, > Y2=,. . .Y12=. Now I can't graph anything. When I press LShift+F1 or > go to Equation Entry from APPS, I get Error: Too Few Arguments . Can > anyone help me? Stella Quick solution: press ON+A+F, then F. (Libraries on port 2 won't be > deleted) In the future, is this the how I should go about clearing the saved equations? And lastly, where may I learn all the shortcuts and key combinations? === Subject: TIMER1 decremented 16 times/s posting-account=HQcyKwkAAAAfyVM-3k2cuaNBENqZAwPy Gecko/2008120122 Firefox/3.0.5,gzip(gfe),gzip(gfe) :: ' CODE DELAY EQU 4 GOSBVL =SAVPTR D1=(5) =TIMERCTRL.1 LC(1) 4 DAT1=C 1 D1=(2) =TIMER1 LC(1) (DELAY)-1 DAT1=C 1 D1=(2) =TIMERCTRL.1 - SHUTDN C=DAT1 B ?CBIT=0 3 GOYES - GOVLNG =GETPTRLOOP ENDCODE xTEVAL ; on 50g: 0.26_s on 49G: 0.26_s The definition: --> TIMER1 decremented 16 times/s ... if so, the program is 4/16 = 0.25_s At this point, it is ok! -------------------------------------- Now the table, shows some DELAY values: -------------------------------------- DELAY ExpVal 50g 49G 1 0.0625 0.0708 0.0540 2 0.1250 0.1335 0.0901 3 0.1875 0.1959 0.2025 4 0.2500 0.2585 0.2339 5 0.3125 0.3175 0.3192 6 0.3750 0.3805 0.3790 7 0.4375 0.4421 0.4305 8 0.5000 0.5087 0.4902 9 0.5625 0.5669 0.0244(?) 10 0.6250 0.6335 0.0247(?) 11 0.6875 0.6918 0.0247(?) 12 0.7500 0.7550 0.0249(?) 13 0.8125 0.8174 0.0250(?) 14 0.8750 0.8804 0.0251(?) 15 0.9375 0.9419 0.0253(?) 16 1.0000 1.0059 0.0254(?) ExpVal: The expected value: DELAY/16 Are bugs on 49G the values with (?). Can someone help me to understand this point? - Gaak - === Subject: Re: TIMER1 decremented 16 times/s > xTEVAL > ; > > 11 0.6875 0.6918 0.0247(?) > 12 0.7500 0.7550 0.0249(?) > 13 0.8125 0.8174 0.0250(?) > 14 0.8750 0.8804 0.0251(?) > 15 0.9375 0.9419 0.0253(?) > 16 1.0000 1.0059 0.0254(?) > > ExpVal: The expected value: DELAY/16 Sounds like a TEVAL issue to me ... JY -- They who would give up an essential liberty for temporary security, deserve neither liberty or security (Benjamin Franklin) === Subject: Re: TIMER1 decremented 16 times/s Hi On 2009-01-19 09:34:54 +1100, Gaak said: > > Are bugs on 49G the values with (?). > Can someone help me to understand this point? Is this a 49G or a 49G+ ? -- They who would give up an essential liberty for temporary security, deserve neither liberty or security (Benjamin Franklin) === Subject: Re: TIMER1 decremented 16 times/s <4973d238$0$15293$426a34cc@news.free.fr> posting-account=HQcyKwkAAAAfyVM-3k2cuaNBENqZAwPy Gecko/2008120122 Firefox/3.0.5,gzip(gfe),gzip(gfe) > Hi On 2009-01-19 09:34:54 +1100, Gaak said: Are bugs on 49G the values with (?). > Can someone help me to understand this point? Is this a 49G or a 49G+ ? > -- > They who would give up an essential liberty for temporary security, > deserve neither liberty or security (Benjamin Franklin) HP49G (Saturn) ROM2.09 - Gaak - === Subject: Re: TIMER1 decremented 16 times/s Hi On 2009-01-19 12:31:54 +1100, Gaak said: > HP49G (Saturn) > ROM2.09 Where did you get 2.09 for the 49G ? JY -- They who would give up an essential liberty for temporary security, deserve neither liberty or security (Benjamin Franklin) === Subject: Re: TIMER1 decremented 16 times/s <49746140$0$7869$426a74cc@news.free.fr> posting-account=57eBeAkAAABkNMhsZnDBzGQsstJRmvKD Gecko/2008120122 Firefox/3.0.5,gzip(gfe),gzip(gfe) > HP49G (Saturn) > ROM2.09 Where did you get 2.09 for the 49G ? I described first how you can obtain the latest ROM from the emulator for the FHB, the last Saturn machine: Of course this is unsupported but HP never made a recent ROM for the 49G ! process: Andreas http://www.software49g.gmxhome.de === Subject: Re: New fast version of PTLIB for HP-48 (+49/50) posting-account=sOAX1QkAAAC-FcySTSbz29Uk8huUtFRz CLR 1.1.4322; .NET CLR 2.0.50727),gzip(gfe),gzip(gfe) > What are the 50g results for the above > fomulas using the HP version of PTLIB? > The HP version Library 229 gives the correct values of 46.0688 g/mol and 114.2302 g/mol, respectively. Your version of MOLWT seems to work properly on the ARM machines as long as there are no subscripts. For instance CCHHHHHOH MOLWT returns 46.0688 g/mol. Also C MOLWT returns 12.011 g/mol, but C2 MOLWT returns 8439096.99422 g/mol. If it is not possible to test out your 49g+/50g version on an emulator, I will be happy to try out specific cases for you. S.C. === Subject: Re: New fast version of PTLIB for HP-48 (+49/50) What are the 50g results for the above > fomulas using the HP version of PTLIB? The HP version Library 229 gives the correct values of 46.0688 g/mol > and 114.2302 g/mol, respectively. Your version of MOLWT seems to work properly on the ARM machines as > long as there are no subscripts. For instance CCHHHHHOH MOLWT returns 46.0688 g/mol. Also C MOLWT returns 12.011 g/mol, but C2 MOLWT returns 8439096.99422 g/mol. If it is not possible to test out your 49g+/50g version on an > emulator, I will be happy to try out specific cases for you. S.C. H MOLWT gives 1.0079 H0 MOLWT gives 7081.646506 (HP version: 0) H1 MOLWT gives 708164.660679 H2 MOLWT gives 708164.670758 H3 MOLWT gives 708164.680837 H4 MOLWT gives 708164.690916 H1,H2,H3,H4,... differ by a common difference of 0.010079, which is 1/100 of the weight of an H atom. Perhaps this gives some insight? S.C. === Subject: Re: New fast version of PTLIB for HP-48 (+49/50) I just checked my PTLIB with an 50g emulator, and it returns the correct results. So I assume it's a flag issue only. Could you send me your flag settings? Raymond schrieb im Newsbeitrag What are the 50g results for the above > fomulas using the HP version of PTLIB? The HP version Library 229 gives the correct values of 46.0688 g/mol > and 114.2302 g/mol, respectively. Your version of MOLWT seems to work properly on the ARM machines as > long as there are no subscripts. For instance CCHHHHHOH MOLWT returns 46.0688 g/mol. Also C MOLWT returns 12.011 g/mol, but C2 MOLWT returns 8439096.99422 g/mol. If it is not possible to test out your 49g+/50g version on an > emulator, I will be happy to try out specific cases for you. S.C. H MOLWT gives 1.0079 H0 MOLWT gives 7081.646506 (HP version: 0) H1 MOLWT gives 708164.660679 H2 MOLWT gives 708164.670758 H3 MOLWT gives 708164.680837 H4 MOLWT gives 708164.690916 H1,H2,H3,H4,... differ by a common difference of 0.010079, which is 1/100 of the weight of an H atom. Perhaps this gives some insight? S.C. === Subject: Re: New fast version of PTLIB for HP-48 (+49/50) ) I just checked my PTLIB with an 50g emulator, > and it returns the correct results. So I assume it's a flag issue only. Could you send me your flag settings? > Raymond > RCLF { # 180008203E10FB0h # 100000000000000h # 98100402000000h # 0h } (System flags -22, -34, -40, -56, -57, -90, -99, -109, -116, -117, -120, -128 set) S.C. === Subject: Re: New fast version of PTLIB for HP-48 (+49/50) I checked with your flag settings on the 50g emulator, and got the correct results. So it's not the flag settings...hmm. No idea what causes the strange results on your 50g, however I'd like the issue resolved, of course. Can someone also check the results for the chemical formulas C2H5OH (which should give 46.0688) and C8H18 (which should give 114.2302) on their 50g's , and tell me the results? Raymond schrieb im Newsbeitrag I just checked my PTLIB with an 50g emulator, > and it returns the correct results. So I assume it's a flag issue only. Could you send me your flag settings? > Raymond > RCLF { # 180008203E10FB0h # 100000000000000h # 98100402000000h # 0h } (System flags -22, -34, -40, -56, -57, -90, -99, -109, -116, -117, -120, -128 set) S.C. === Subject: Re: New fast version of PTLIB for HP-48 (+49/50) Hi all, now I know where the problem is (or was) So far I only checked the interactive mol weight calculator inside the table, but never the external command, which is very mode-sensitive in the published version of my PTLIB for the 49/50. I just made a new version of my PTLIB where the MOLWT command works as expected. This version will be posted in a few hours. Raymond Raymond Del Tondo schrieb im Newsbeitrag I checked with your flag settings on the 50g emulator, > and got the correct results. So it's not the flag settings...hmm. No idea what causes the strange results on your 50g, > however I'd like the issue resolved, of course. Can someone also check the results for the chemical formulas > C2H5OH (which should give 46.0688) and > C8H18 (which should give 114.2302) on their 50g's , and tell me the results? > Raymond > schrieb im Newsbeitrag >> >> I just checked my PTLIB with an 50g emulator, >> and it returns the correct results. >> >> So I assume it's a flag issue only. >> >> Could you send me your flag settings? >> >> >> Raymond >> > RCLF { > # 180008203E10FB0h > # 100000000000000h > # 98100402000000h > # 0h } (System flags -22, -34, -40, -56, -57, -90, -99, -109, -116, -117, > -120, -128 set) S.C. > === Subject: Re: New fast version of PTLIB for HP-48 (+49/50) Can the program be debugged using an emulator? this problem seems to be 49gplus/50g specific. On the HP-48 and on the fhb (49g) with ROM 1.19-6, > the results are correct. C2H5OH gives 46.0688 and > C8H18 results in 114.2302 . Since I don't have a 50g (and no ROM 2.09) here, > I can't check whether there are differences > related to the different hardware or software. What are the 50g results for the above > fomulas using the HP version of PTLIB? > Raymond schrieb im Newsbeitrag >> >> Wow! The table is really fast; much much faster than the HP version. >> >> I'm having a little trouble with your MOLWT function though... >> >> I tried ethanol (C2H5OH) and got 9147278.70252 g/mol instead of >> 46.0688 g/mol. Then I tried octane (C8H18) and got -649760625.629 g/ >> mol instead of +114.2302 g/mol. >> >> I tried entering formulas as strings and as variables and both yield >> the same (incorrect) results. Can anyone confirm this? > Sorry, I forgot to write that I am using a 50g with ROM 2.09C. S.C. > === Subject: help with ITE program almost the same as the example below. If somebody could help me. :: DUP % 1000. %<= ITE :: A ; :: % 1000. %/ mA ; ; @ Example found in the forum that works. :: DUP %7 %= ITE :: %3 %+ ; :: %3 %- ; ; @ Fabrizio === Subject: Re: help with ITE program posting-account=ky6NnQoAAAAl8HjjF10EUMKbzXiIKhTR Gecko/2008120122 Firefox/3.0.5,gzip(gfe),gzip(gfe) > almost the same as the example below. If somebody could help me. > :: > DUP > % 1000. > %<= > ITE > :: > A > ; > :: > % 1000. > %/ > mA > ; > ; > @ Example found in the forum that works. > :: > DUP > %7 > %= > ITE > :: > %3 > %+ > ; > :: > %3 > %- > ; > ; > @ Fabrizio I don't think the code you provided is the source of the problem you're experiencing. I found that your code results in no problems for me on my 50g, ROM 2.09, BTW you don't need a secondary if there's only one argument: DUP % 1000. %<= ITE A :: % 1000. %/ mA ; ; @ So basically what I'm suggesting is that you take a look at some of your code outside of this snippet, I'm no expert but I don't see anything wrong with the code you provided. Hope you find the culprit :) Jacob === Subject: FAO: Jean-Yves Avenard Jean-Yves, I can't find an email address for you. Would you contact me please at the address below. It's nothing too serious but not for public consumption. To everyone else: sorry for the tease. :-) -- Bruce Horrocks Surrey England (bruce at scorecrow dot com) === Subject: Re: Jean-Yves Avenard AAARRGGGGHH!!! It is surely something VERY exiting! :-( > Jean-Yves, I can't find an email address for you. Would you contact me please at the > address below. It's nothing too serious but not for public consumption. To everyone else: sorry for the tease. :-) -- > Bruce Horrocks > Surrey > England > (bruce at scorecrow dot com) === Subject: Update to PTLIB 9.01 available... Hi all, sorry it took some more hours due to free time restrictions;-) As written earlier, I fixed the mol weight calculation bug. The new version should be on www.hpcalc.org soon, but as long as the online archive is not updated you can also get the new PTLIB v 9.01 from me directly. Have fun:-) Raymond === Subject: Re: Update to PTLIB 9.01 available... > Hi all, sorry it took some more hours due to free time restrictions you mean this: time is money so it's not free - time, I mean... thank U 4 the program === Subject: a GOSUB for SysRPL program posting-account=eF2f0AoAAAB2spBRiZOs91ItDKLGDCIk Gecko/2008120122 Firefox/3.0.5,gzip(gfe),gzip(gfe) After reading some code from Detlef Muller, author of RPL48, I noticed some use of GOTO ASM statements right in the middle of SysRPL codes, which seems unstandard to say the least. This was merely about calling a distant ASM code source, but then i wondered if this idea could be used for SysRPL too. It happens that i'm currently needing a typical ASM jump statements, namely GOSUB...RTN, for a SysRPL program. This is usefull if you have a routine which is called from various positions within your program. Of course, you could simply break the code into several parts, assign a global ID name to each of them, and then call the routine using its ID name whenever you need it; This is the standard UserRPL way. But this has 2 drawbacks : - Performance wise : each call to a Global ID trigger a search, which is time consuming. If you have many calls, this will make a large performance difference. The larger the number of VAR in directory, the longer the search. - Presentation wise : this is one thing to provide a single file, this is another to hand over several ones. You may prefer to keep your program as a single file for user simplicity. So, now, we figure we just want that, keep everything within a single file. At this stage, a Library could also fit the bill, but i'm willing to test another direction, and keep the program as a single file typically into HOME directory. How then to handle all these calls to internal sub-routines ? First possibility : Program your code as a recursive one, which call back itself using an argument in order to branch to the correct sub-routine. This is easy to implement (case statement), and will work for userRPL. You keep your code as a single file, but you still have the Global ID search performance problem. And you need to know by which name your program is saved. Second possibility : with an RPLCOMP compiler. This method looks good. beginning of your program, and should not change it afterwards, not even a single additionnal Local variable. This could be a limitation Third possibility : This is largely experimental, but I've been trying an ASM GOTO statement as a way to handle such calls to internal subroutines. And it works well also. Here is a short and hopefully explicative example : :: Beggining of Program CODEM GOTO .MySubroutine ENDCODE ( Call to sub-routine ) This is sentence 2 COLA End of Program CODEM *.MySubroutine A=PC A+10.A PC=(A) ENDCODE :: This is subroutine sentence ; ; If you copy paste this code and compile it, you'll get the subroutine sentence string before sentence 2. Note the DOCOL :: which is absolutely mandatory. When the program hits the SEMI ; it just goes back to previous RPL position, in this case sentence 2 string. So you essentially made a GOSUB. Well, i'm not so sure, and that's the objective of this post, to discuss about it. else. There is also *maybe* a memory advantage when not storing subroutines The drawback is that mixing ASM & SysRPL code this way does not look standard. Well, is that really a problem ? Do we have other issue to consider ? Any comment and source of information is welcomed. === Subject: Disable interrupts Hello! I wish everyone a happy new year! I«m making a program ( in UserRPL) to protect the access to calculator (50g rom 2.09) using a password. The program is protected against ON +C, but no against [ON]+[C] and [<-] ( drop ). I tried to find a rpl entry to disable interrupts , but I find only ML entries for that. I don't know ML language, but I read a few of the book Introduction to Saturn Assembly Language and also saw some ML programs and tried to compile something using MASD: !RPL !NO CODE CODE ST=0 15 ENDCODE @ !RPL !NO CODE CODE GOSBVL =DisableIntr ENDCODE @ !RPL !NO CODE CODE GOSBVL 26791 ENDCODE @ compiled but not run! Can somebody help me? === Subject: Re: Disable interrupts posting-account=eF2f0AoAAAB2spBRiZOs91ItDKLGDCIk Gecko/2008120122 Firefox/3.0.5,gzip(gfe),gzip(gfe) Hummm, i'm not sure that DisableInterrupt is valid outside of ASM Code. Note that, if you want your code to work, you need to add a RPL statement within the code, for example : CODE GOSBVL =DisableIntr RPL ENDCODE An even proper way would be : CODE SAVE GOSBVL =DisableIntr LOADRPL ENDCODE As a beginner advise , always start your MASD code with SAVE, and end it with LOADRPL. Now, even if this code run properly, i'm not sure that DisableIntr will stay valid outside of the MASD code itself... === Subject: Re: Disable interrupts posting-account=T6_JoAoAAAD5X1NpIb7ZcQrmFCd8Otqa Gecko/2008120122 Firefox/3.0.5 (.NET CLR 3.5.30729),gzip(gfe),gzip(gfe) And, If I use two programs PRG1 (ML) and PRG2 (user). how can I call the PRG2 from the PRG1 ? I don«t know the sintax in ML to execute a global name. CODE SAVE GOSBVL =DisableIntr ??? PRG2 ??? LOADRPL ENDCODE === Subject: Re: Disable interrupts posting-account=eF2f0AoAAAB2spBRiZOs91ItDKLGDCIk Gecko/2008120122 Firefox/3.0.5,gzip(gfe),gzip(gfe) And, If I use two programs PRG1 (ML) and PRG2 (user). > how can I call the PRG2 from the PRG1 ? This is alas not possible (well, not easily). What you can do is have an overall PRG3, which calls PRG1(ML) then PRG2(user). Specifically, if you plan to trigger =DisableIntr from within ML1, it will no longer be active as soon as you leave ML1 code. You have to keep in mind that when you leave your ML program, you lose everything regarding its context/state/value. So you cannot come back . Well, this is assembler, so it is only normal that a higher-level language such as RPL makes use of it too. Each SysRPL/UserRPL instruction can be seen as sequence of SysRPL instructions, which can themselves be also sequence of SysRPL, up to the point that you cannot go further : at its core level, the fundamental SysRPL instructions are each one an Assembler Code. So, of course, they make use of registers/flags available, and there is nothing left from your own PRG1. So what you can do is leave an ML program to start an RPL one, then start another ML program. You can, in between, keep some sort of context info into the stack, so that the second ML program will be able to use some results from the first ML program. === Subject: Size of Flash memory in 50g posting-account=vJkpuAoAAAB8gwi_HMDl-BrUbyqyvmJy .NET CLR 2.0.50727; .NET CLR 3.0.04506.648; .NET CLR 3.5.21022),gzip(gfe),gzip(gfe) I have a few days HP50g purchased. When I go to the filer, then I see that Port2 (Flash), only 766 KB free space, although nothing is stored. It is not that the flash Sreicher 1 MB must show? Or am I mistaken? === Subject: Re: Size of Flash memory in 50g See what's inside: equation libraries if you don't need them, delete them > I have a few days HP50g purchased. When I go to the filer, then I see > that Port2 (Flash), only 766 KB > free space, although nothing is stored. It is not that the flash > Sreicher 1 MB must show? > Or am I mistaken? === Subject: Re: Size of Flash memory in 50g > See what's inside: equation libraries > if you don't need them, delete them I have a few days HP50g purchased. When I go to the filer, then I see > that Port2 (Flash), only 766 KB > free space, although nothing is stored. It is not that the flash > Sreicher 1 MB must show? > Or am I mistaken? - Zitierten Text anzeigen - There is absolutely nothing on the calc, except the OS === Subject: Re: BZ2 compression posting-account=eF2f0AoAAAB2spBRiZOs91ItDKLGDCIk Gecko/2008120122 Firefox/3.0.5,gzip(gfe),gzip(gfe) After quite some testings, and still no bug found, BZ2 is now released in a stable format. Compare with initial b.90ta, there is little new : - a new Max compression settings, which squeeze even more performance - generic speed improvements, valid for all settings So, you can basically select 3 different speed/compression combinations, between fast, normal & max. An optional decoder, UBZ2, is also provided, which is about 40-50% faster than original BZ decoder. But you are not obliged to use it. older BZ version, including the one integrated into HP49 ROM. More details (benchmark, doc, and the program itself) can be found at website : http://BZ2.webhop.org Hope this can be usefull to anyone === Subject: Important update to PTLIB 9.011 available... Hi all, yes, another important update;-) PTLIB v9.01 crashed the calc when trying to copy a property to the stack using the >STK menu key. This has been fixed in PTLIB v9.011 . As usual, the update should show up on hpcalc.org soon, and the people who mailed me automatically get the new version via mail. Have fun:-) Raymond === Subject: Re: BZ2 compression posting-account=57eBeAkAAABkNMhsZnDBzGQsstJRmvKD Gecko/2008102920 Firefox/3.0.4,gzip(gfe),gzip(gfe) Hello Yann, I will surely give it a try. If it keeps it promissed I would be interested in including you compressor in my 'Compression Aid' Lib. Would that be O.K. for you ? 'Compression Aid' is free and auto-simplifies the compression of a library. More information can be found at http://www.software49g.gmxhome.de/EnglishSeite01.htm#Komp Andreas === Subject: Re: BZ2 compression posting-account=eF2f0AoAAAB2spBRiZOs91ItDKLGDCIk Gecko/2008120122 Firefox/3.0.5,gzip(gfe),gzip(gfe) > I would be interested in including you > compressor in my 'Compression Aid' Lib. Would that be O.K. for you ? Hello Andreas Yes, Compression Aid is a really excellent tool. I could only wish it exist the same for HP48. Of course, you are welcomed to use BZ2 with it if it fits your need. Yann === Subject: Re: BZ2 compression posting-account=57eBeAkAAABkNMhsZnDBzGQsstJRmvKD Gecko/2008102920 Firefox/3.0.4,gzip(gfe),gzip(gfe) Hello Yann > Yes, Compression Aid is a really excellent tool. > I could only wish it exist the same for HP48. Well, the only 49G specific is the InformBox (IfMain2). It should be quite easily to port it to DoInputForm with debug4x. Let me think of it... > Of course, you are welcomed to use BZ2 with it if it fits your need. Aid. Andreas === Subject: Re: BZ2 compression Marvelous BZ2 for both series and now even Compression Aid - ultimately to old 48-series, too 50g user > Hello Yann >> Yes, Compression Aid is a really excellent tool. > I could only wish it exist the same for HP48. > Well, the only 49G specific is the InformBox (IfMain2). It should be > quite easily to port it to DoInputForm with debug4x. Let me think of > it... > Of course, you are welcomed to use BZ2 with it if it fits your need. > Aid. Andreas > === Subject: Re: Size of Flash memory in 50g posting-account=57eBeAkAAABkNMhsZnDBzGQsstJRmvKD Gecko/2008102920 Firefox/3.0.4,gzip(gfe),gzip(gfe) 766 KB is correct because the O.S. is using two additional FlashBanks with the recent O.S. versions (>= 2.00). This has been discussed here in the past so a Google search should give you in depth answers. HTH, Andreas http://www.software49g.gmxhome.de === Subject: Re: Size of Flash memory in 50g 766 KB is correct because the O.S. is using two additional FlashBanks > with the recent O.S. versions (>= 2.00). This has been discussed here in the past so a Google search should > give you in depth answers. When, where, what? === Subject: Re: Size of Flash memory in 50g posting-account=vJkpuAoAAAB8gwi_HMDl-BrUbyqyvmJy .NET CLR 2.0.50727; .NET CLR 3.0.04506.648; .NET CLR 3.5.21022),gzip(gfe),gzip(gfe) 766 KB is correct because the O.S. is using two additional FlashBanks > with the recent O.S. versions (>= 2.00). This has been discussed here in the past so a Google search should > give you in depth answers. HTH, > Andreashttp://www.software49g.gmxhome.de === Subject: Re: Disable interrupts ML is very difficult. I need to study more. === Subject: Re: How about a little hint of what's in the works. SV1) ; .NET CLR 1.1.4322; .NET CLR 2.0.50727; .NET CLR 3.0.04506.30; .NET CLR 3.0.04506.648),gzip(gfe),gzip(gfe) >... > * A new printer? How about a printing calculator with IR access to > the printer, to be used with the usual IR enabled calcs? How about > printing to a full-size, logical-page file (hp's own PCL-3 format) > inside the calculator, to later beam or transfer the file to a PCL- > compatible printer? >... About the printer: As far as I know, the hp 82240B printer is discontinued. It is nice to have, but slow, lo-res, with only two fonts, and unusable without a supported calculator, not to mention expensive. Now that hp has IrDA- capable calculators plus a new line of calculators that includes a small printing one, an enhanced version of such calculator could be designed. It should have a better printhead (higher resolution and speed, able to print on paper rolls as well as labels), an LCD display for calculator-mode results and slave-printer settings, USB connectivity for recharging, and for PC usage either as a numeric keyboard extension and/or printer. It could even be a stand-alone printing solution for the recently-announced Mobile Calculating LAB (MCL). By ading more functionality, the printer could be marketed for different groups, not just calculator users. There you have it. My round table participation in hp's new product brainstorming session. === Subject: Very fast Sudoku solver for HP 49g+ and 50g posting-account=KxEIGwoAAAALMxy6hykxLK1wPdiyuNkK Gecko/2008120122 Firefox/3.0.5,gzip(gfe),gzip(gfe) Hello all, To highlight the speed of the ARM processor in the 49g+ and 50g, I ported my C++ sudoku solver to the calculator using hpgcc. It's available at http://www.hpcalc.org/details.php?id=7062 It takes less than 3 seconds to solve any puzzle with 1 solution on my 50g. The vast majority of that time is spent in some wrapper UserRPL code that converts a matrix to a string for input the solver, and converts an output string back to a matrix for output. The actual C++ part solves the puzzle almost instantly. I'm just learning SysRPL now, but I'll convert the slow code eventually. === Subject: Memory card reader I am looking for a standalone reader able to read the HP/EPSON (SMI, ultimate goal is to transfer the data to a flash card. I have heard of an Epson adapter for a Needham's programmer but the company has === Subject: What is the last year of fabrication of the HP 32SII ? What is the last year of fabrication of the HP 32SII ? Denis === Subject: Re: What is the last year of fabrication of the HP 32SII ? > What is the last year of fabrication of the HP 32SII ? Denis In 2002, I was idly looking around an electronics store when I saw a dusty shelf full of calculators, with an HP32SII at the bottom. A clerk said, I didn't even know we had that. They don't make those any more. And he offered it to me for $30. After I picked myself up off the floor, I left with my brand-new-but-old HP32SII. Amazing Fact #2: it's still running on the original batteries. Bill === Subject: Re: What is the last year of fabrication of the HP 32SII ? >In 2002, I was idly looking around an electronics store when I saw a >dusty shelf full of calculators, with an HP32SII at the bottom. A >clerk said, I didn't even know we had that. They don't make those >any more. And he offered it to me for $30. After I picked myself up >off the floor, I left with my brand-new-but-old HP32SII. Simple finds are great! I recently found an HP-41C at a swap-meet; because it had no battery and couldn't be tested, the seller gave it to me free if I bought its $5 furry case. And of course it still works well.... -- Chris. === Subject: Re: What is the last year of fabrication of the HP 32SII ? 1991-1999 Denis | What is the last year of fabrication of the HP 32SII ? | | Denis | | === Subject: Re: 49g+ sometimes freezes > Hi like in the object my 49g+ suffers of a strange behaving. In fact > sometimes when I press a button it seem to freeze and to think then it does > the action of the just-pressed button. I find this thing annoying and > dangerous because if you don't spot this behaving during calculations you > can miss someting with wrong results and you loose time. My rom is 2.09and > the libraries installed are: tgviewer2, unitman, chemlab 2.7, vigag 4.1, > truss, sectn49, fem49 and equation library and data > have you any hint?Can you help me? > Duncan > > When your hp49g+ freezes it does something knows as the Garbage collector: the system deletes objects which are not uses anymore... The hp49g+ works like this: it creates objects in ram and when ram is full the hp49g+ makes cleaning: some objects are useless so the system put them in trash (to free some memory...) and it could takes times... So there is no solutions except do the garbage collector when you want... === Subject: Re: LINSOLVE function > I am using OS version 2.09 and algebraic mode with numeric, approx and > complex all un-checked. > When I enter LINSOLVE([x+3*y=1,2*x+4*y=1],[x,y]) I get the following > returned > {{[x+3*y=1 2*x+4*y=1] [x,y]} Specific: 2 -2 2 1.000} [x= -1/2 y=1/2]} > What is Specific: 2 -2 2 1.000 for and what does it mean? According to the AUR, page 4-49 (PDF page 393), that is the list of pivot points. Specific: is just a label for this list and doesn't mean anything in particular. Eric Rechlin === Subject: Re: LINSOLVE function <6t5h1jF8s1mmU1@mid.individual.net> posting-account=GT4frAoAAADMvAtJu2WIOkYTwgy0O29T 1.1.4322; InfoPath.2),gzip(gfe),gzip(gfe) > I am using OS version 2.09 and algebraic mode with numeric, approx and > complex all un-checked. > When I enter LINSOLVE([x+3*y=1,2*x+4*y=1],[x,y]) I get the following > returned > {{[x+3*y=1 2*x+4*y=1] [x,y]} Specific: 2 -2 2 1.000} [x= -1/2 y=1/2]} > What is Specific: 2 -2 2 1.000 for and what does it mean? According to the AUR, page 4-49 (PDF page 393), that is the list of pivot > points. Specific: is just a label for this list and doesn't mean anything > in particular. > Eric Rechlin === Subject: Re: LINSOLVE function > I wonder what a pivot point is? Calculator is a tool, but does not teach the underlying subject. Curiosity can be satisfied by reading texts or other sources of theory. http://physics.gmu.edu/astro/classes/c80197/l04/lecture4.html don't play with your loop index [extra credit topic] === Subject: Re: LINSOLVE function posting-account=GT4frAoAAADMvAtJu2WIOkYTwgy0O29T 1.1.4322; InfoPath.2),gzip(gfe),gzip(gfe) > Calculator is a tool, but does not teach the underlying subject. I agree John but there seems to be a group of people at that concept. I think it would be interesting to see what happened if help. === Subject: Sequences on the 50g posting-account=_HGaIAoAAAB6Jn3lrST3pff5_EdBru_y Gecko/2008120122 Firefox/3.0.5,gzip(gfe),gzip(gfe) How can I evaluate the values of a certain term of a sequence? Like a(7) for a(n) = a(n-1) + a(n-2) with a(1) = 1 , a(2) = 1 as start? couldn't actually plot sequences, if I got that right, but that is not quite helpful as it's not what I'd actually need -.- === Subject: Re: Sequences on the 50g InfoPath.1; .NET CLR 2.0.50727; .NET CLR 3.0.04506.30; .NET CLR 1.1.4322; .NET CLR 3.0.04506.648),gzip(gfe),gzip(gfe) > How can I evaluate the values of a certain term of a sequence? Like a(7) > for a(n) = a(n-1) + a(n-2) Here's yet another way, using the old APPLY command from the HP48 command set. You can find it in CAT or by typing 93 MENU (which gives you the older SYMB menu, another archaeological bit from the 48). Enter: { n } 'a' and run APPLY to get -> a(n) Store any desired value (7, say), as n. Rewrite your expression as 'a(n)=(n-1)+(n-2)' Press EVAL and get -> a(7)=11 Bill === Subject: Re: Sequences on the 50g posting-account=l4FfmwoAAACYV9A7drprQGL6FQTZEt0C Gecko/2008120122 Firefox/3.0.5,gzip(gfe),gzip(gfe) > How can I evaluate the values of a certain term of a sequence? Like a(7) > for a(n) = a(n-1) + a(n-2) with a(1) = 1 , a(2) = 1 as start? couldn't actually plot sequences, if I got that right, but that is not > quite helpful as it's not what I'd actually need -.- > If you look at the programming examples in chapter 2 of the HP49g+ Advanced User's Reference ( available from www.hpcalc.org), there is an example that covers this exact subject. It shows both a recursive and an iterative program for generating Fibonacci numbers (which is what your example does). These programs can be easily adapted to just about any similar sequence. This manual is actually for the previous model, but there is no software difference between the two, and the manuals for the 50g frankly aren't much good. BTW, if you want to plot such a sequence, you can save the values of the sequence in the Statistics Matrix and use the built-in plotting commands to display it. HTH, John === Subject: Re: Sequences on the 50g > How can I evaluate the values of a certain term of a sequence? > > Like a(7) > for > > a(n) = a(n-1) + a(n-2) > > with a(1) = 1 , a(2) = 1 > > as start? In RPL mode, type in the following algebraic and leave it on level 1 of the stack: 'A(N)=IFTE(A<=2,1,A(N-1)+A(N-2))' Then run the command DEFINE (left-shift 2). This creates a variable in the current directory, called A, that contains your function and the code to run it. To evaluate A(7) just put 7 on the stack and press the menu key labelled A. Alternatively, you can use 'A(7)' in an expression. -- Bruce Horrocks Surrey England (bruce at scorecrow dot com) === Subject: Re: Sequences on the 50g | > How can I evaluate the values of a certain term of a sequence? | > | > Like a(7) | > for | > | > a(n) = a(n-1) + a(n-2) | > | > with a(1) = 1 , a(2) = 1 | > | > as start? | | In RPL mode, type in the following algebraic and leave it on level 1 of | the stack: | 'A(N)=IFTE(A<=2,1,A(N-1)+A(N-2))' 'A(N)=IFTE(N<=2,1,A(N-1)+A(N-2))' Denis | | Then run the command DEFINE (left-shift 2). This creates a variable in | the current directory, called A, that contains your function and the | code to run it. To evaluate A(7) just put 7 on the stack and press the | menu key labelled A. Alternatively, you can use 'A(7)' in an expression. | | -- | Bruce Horrocks | Surrey | England | (bruce at scorecrow dot com) === Subject: HHC2008 HP Calc Conference Video Now Available posting-account=K31tcwgAAABVnT_oG3A7BQonvJCmOPn4 1.1.4322; .NET CLR 2.0.50727; InfoPath.1; .NET CLR 3.0.04506.30; .NET CLR 3.0.04506.648; .NET CLR 3.5.21022; MS-RTC LM 8),gzip(gfe),gzip(gfe) The 14-and-a-half-hour video from the HHC2008 HP Calculator Conference (held 27-28 September 2008 in Corvallis, Oregon) is now available. Please check http://www.pahhc.org/video.htm if interested. Jake Schwartz P.S. - There are plenty of year-2009 HP Calc Calendars left.... see === Subject: Re: HP48 + Arduino Duemilanove posting-account=EjMqYwoAAADPG-eKZyudrVG4uwsv15vp .NET CLR 2.0.50727; .NET CLR 3.0.04506; Tablet PC 2.0; .NET CLR 3.5.21022; OfficeLiveConnector.1.3; OfficeLivePatch.0.0),gzip(gfe),gzip(gfe) > > Duemilanove. > I wired the serial port, but when I send AB for example, I get }{ > > on my calculator. In fact, there are some chars that I've found > > searching for a translation : > REC DEC R BIN R BIN S SENT BIN S > > y 121 1111001 1000011 C 0000110 > > w 119 1110111 1000100 D 0001000 > > u 117 1110101 1000101 E 0001010 > > s 115 1110011 1000110 F 0001100 > > q 113 1110001 1000111 G 0001110 > > o 111 1101111 1001000 H 0010000 > So, if I send CDEFGH into my HP48g, I get ywusqo with << 6 SRECV > More about: > > The CDEFGH string looks like this in bin: > > 100001110001001000101100011010001111001000 > And the received string looks like this: > > 111100111101111110101111001111100011101111 > If we apply NOT to the received binary string we get: > > 000011000010000001010000110000011100010000 > If you look carefully it is like almost the original string, but: > > M000011 000100 000101 000110 000111 0010000 > M stands for a missing bit, and is a 0 that must be a 1. > Why can this happen? > I tried with another arduino, my oldest decimila model and I have > > connected the HP48g directly to my PC to test it, and with > > Hyperterminal works very well, so isn't a hardware problem. > Can this experience be done with a HP50g and the newest serial port? > It looks to me like you have a signal inverted somewhere. Maybe > you've swapped the signal and ground pins in your cable connector > somewhere? Double check the pin outs. With the signal inverted, the start bit gets missed, then the > transition to the first data bit gets interpreted as the start bit, > then the transition to the second data bit gets interpreted as the > first data bit, only reversed because the signal is inverted etc. Dave- Ocultar texto de la cita - - Mostrar texto de la cita - I managed to solve the problem. HP now has a small translator and the messages arrives correctly: For further info on this issue, my custom translator is: << -> S Q > << S 1 S SIZE START > DUP NUM 255 - NEG 2 / CHR > Q SWAP + 'Q' STO TAIL NEXT DROP > Q >> So, you receive the data with << BUFLEN DROP SRECV DROP >> > - Mostrar texto de la cita - More info about this experience: Using my translator, only 7 bits chars are received. Even the XLAT functions of the HP won't work because the transference is encoded using that NOT( char << 1). So the problem is that you can't send ASCII 127-255, special chars, even if you encode this with -> notation. My solution is to use the SYSEVAL of the XLAT function, so: 1. Receive the encoded data from Arduino: // Arduino code void setup() { Serial.begin(9600); pinMode(P LED,OUTPUT); blink(5000); blink(100,20); // Inicializaci.97n Serial.print(0x00,BYTE); Serial.print( #131 #64 BLANK PICT STO { #0 #0 } PVIEW ); pause(); Serial.print(0x00,BYTE); Serial.print( << HOLA MUNDO 34 CHR SWAP + 34 CHR + MSGBOX >> + + OBJ-> 'mundo' STO ); pause(); Serial.print(0x00,BYTE); Serial.print( PICT { #2 #2 } Inicializando... 1 ->GROB REPL 1000 0.2 BEEP 2000 0.1 BEEP 1 WAIT ); } // pause() is just a led blinking routine 2. Translate the received data in the HP: << -> S Q << S 1 S SIZE START DUP NUM 255 - NEG 2 / CHR Q SWAP + 'Q' STO TAIL NEXT DROP Q >> 3. Re-translate special HP chars << #107d SYSEVAL + >> @HP48 4. Evaluate the code, its like having a dual-core arduino :D for instance you can do even more interesting things: Serial.print(0x00,BYTE); Serial.print( 1 1000 PRIMESEARCHMACRO XMIT ); And the HP will send the results to arduino of your 'PRIMESEARCHMACRO'. Pretty cool huh? === Subject: Re: BZ2 compression posting-account=57eBeAkAAABkNMhsZnDBzGQsstJRmvKD Gecko/2008120122 Firefox/3.0.5,gzip(gfe),gzip(gfe) Hello Yann, could you contact me by PM please. I«d like to discuss some things about Compression Aid with you privately. Andreas === Subject: Re: creating tables and charts posting-account=Iz6FlwoAAABdQTI08xdVoDk-iYgiEfif .NET CLR 2.0.50727; Media Center PC 5.0; .NET CLR 3.0.04506; InfoPath.2),gzip(gfe),gzip(gfe) > i have finding a way to write sometablesandchartsthat are located > in the back of my chemistry text. Unfortunately some of the have > multiple submenus. I don't understand the manual whatsoever. Is > there anyone who may help me with this? > John i don't know but that sounds about right. An example is the periodic table it has both the atomic number as well as the atomic mass. if i was able to do something like that i would be in good shape. === Subject: Re: creating tables and charts posting-account=Q2CEjQoAAACue5ZDDhUeRzv1w0u8hxZE rv:1.9.1b2) Gecko/20081201 Firefox/3.1b2,gzip(gfe),gzip(gfe) i have finding a way to write sometablesandchartsthat are located > in the back of my chemistry text. Unfortunately some of the have > multiple submenus. I don't understand the manual whatsoever. Is > there anyone who may help me with this? > John i don't know but that sounds about right. An example is the periodic > table it has both the atomic number as well as the atomic mass. if i > was able to do something like that i would be in good shape. Have you looked at this: http://www.hpcalc.org/details.php?id=7065 Or maybe this: http://www.hpcalc.org/details.php?id=4851 I suspect what you are looking for is already there. TW === Subject: Debug4x Build 136 Available http://www.debug4x.com has a new version with revised font that will display all the characters on both XP and Vista. -- - - - - - - - - - - - - - - - - Bill Graves RKBA! bgraves@ix.netcom.com === Subject: Re: HP50g won't graph <497461e8$0$7869$426a74cc@news.free.fr> posting-account=EjMqYwoAAADPG-eKZyudrVG4uwsv15vp .NET CLR 2.0.50727; .NET CLR 3.0.04506; Tablet PC 2.0; .NET CLR 3.5.21022; OfficeLiveConnector.1.3; OfficeLivePatch.0.0),gzip(gfe),gzip(gfe) > Hi On 2009-01-19 09:40:13 +1100, Avatar_e said: Quick solution: press ON+A+F, then F. (Libraries on port 2 won't be > deleted) A bit drastic don't you think ? -- > They who would give up an essential liberty for temporary security, > deserve neither liberty or security (Benjamin Franklin) When the user don't knows what they do, ON+A+F is the best solution. I know because in my own webpage www.ried.cl I receive about 2 daily complains about HP calcs (all solved by ON+A+F and a please read the manual ) === Subject: Re: HP50g won't graph > I had a bunch of equations saved on my 50g. > I didn't know how to clear them Highlight equations and use the DEL menu key in the plot - function (Y=) screen. > so I deleted twelve files in my Home directory from Y1=, > Y2=,. . .Y12=. Now I can't graph anything. When I press LShift+F1 or > go to Equation Entry from APPS, I get Error: Too Few Arguments . Try also purging the 'EQ' variable, which contains references to all the others. === Subject: Re: TIMER1 decremented 16 times/s <49746188$0$7869$426a74cc@news.free.fr> posting-account=HQcyKwkAAAAfyVM-3k2cuaNBENqZAwPy Gecko/2008120122 Firefox/3.0.5,gzip(gfe),gzip(gfe) > xTEVAL > ; 11 0.6875 0.6918 0.0247(?) > 12 0.7500 0.7550 0.0249(?) > 13 0.8125 0.8174 0.0250(?) > 14 0.8750 0.8804 0.0251(?) > 15 0.9375 0.9419 0.0253(?) > 16 1.0000 1.0059 0.0254(?) ExpVal: The expected value: DELAY/16 Sounds like a TEVAL issue to me ... JY > -- > They who would give up an essential liberty for temporary security, > deserve neither liberty or security (Benjamin Franklin) TEVAL is only for demonstrative purpose. The real objective is a delay (one second)... maybe :: ONE dowait ; is the solution but I need this behavior from ML (without GetTime process) The more simple solution is repeating: 2x(0.5 s). - Gaak - === Subject: Re: TIMER1 decremented 16 times/s <49746140$0$7869$426a74cc@news.free.fr> posting-account=HQcyKwkAAAAfyVM-3k2cuaNBENqZAwPy Gecko/2008120122 Firefox/3.0.5,gzip(gfe),gzip(gfe) > Hi > On 2009-01-19 12:31:54 +1100, Gaak said: HP49G (Saturn) > ROM2.09 Where did you get 2.09 for the 49G ? JY Extracted from Debug4x (50g). This pack contains the latest ROM version. 1. change the 49G KMLscript pointing to ROM2.09 (ROMG+.49G). 2. Load the 49G emulator. 3. Send ROM via ROMUPLOAD command. This version is very stable, the best after ROM 1.19-6. To me, ROM 2.10-7 is buggy (commands like ^WRITEMENU). - Gaak - === Subject: Re: TIMER1 decremented 16 times/s Hello together, in the HP49G case is this the normal behavior of the Yorke chip timer hardware. The CPU is awakend from the sleep state when the WKE bit of the =TIMERCTRL.1 register is set and the MSB of the =TIMER1 value is 1. This is the same behavior like in the earlier chips 1LP2 Centipede (HP28C), 1LR2 Lewis (HP42S), 1LR3 Sacawajawea (HP32SII) and 1LT8 Clarke (HP48SX). In result this mean that the HP50G timer emulation inside the Saturnator is buggy comparing to the behavior of the original Yorke chip. Emulated calculators on Emu48 and Emu48+ behave like working on the orginal Yorke chip, but with less accuracy. Christoph Gaak schrieb im Newsbeitrag > :: > ' > CODE > DELAY EQU 4 > GOSBVL =SAVPTR > D1=(5) =TIMERCTRL.1 > LC(1) 4 > DAT1=C 1 > D1=(2) =TIMER1 > LC(1) (DELAY)-1 > DAT1=C 1 > D1=(2) =TIMERCTRL.1 > - SHUTDN > C=DAT1 B > ?CBIT=0 3 > GOYES - > GOVLNG =GETPTRLOOP > ENDCODE > xTEVAL > ; on 50g: 0.26_s > on 49G: 0.26_s The definition: > --> TIMER1 decremented 16 times/s > ... if so, the program is 4/16 = 0.25_s At this point, it is ok! -------------------------------------- > Now the table, shows some DELAY values: > -------------------------------------- > DELAY ExpVal 50g 49G > 1 0.0625 0.0708 0.0540 > 2 0.1250 0.1335 0.0901 > 3 0.1875 0.1959 0.2025 > 4 0.2500 0.2585 0.2339 > 5 0.3125 0.3175 0.3192 > 6 0.3750 0.3805 0.3790 > 7 0.4375 0.4421 0.4305 > 8 0.5000 0.5087 0.4902 > 9 0.5625 0.5669 0.0244(?) > 10 0.6250 0.6335 0.0247(?) > 11 0.6875 0.6918 0.0247(?) > 12 0.7500 0.7550 0.0249(?) > 13 0.8125 0.8174 0.0250(?) > 14 0.8750 0.8804 0.0251(?) > 15 0.9375 0.9419 0.0253(?) > 16 1.0000 1.0059 0.0254(?) ExpVal: The expected value: DELAY/16 Are bugs on 49G the values with (?). > Can someone help me to understand this point? - Gaak - === Subject: Re: TIMER1 decremented 16 times/s posting-account=HQcyKwkAAAAfyVM-3k2cuaNBENqZAwPy Gecko/2008120122 Firefox/3.0.5,gzip(gfe),gzip(gfe) > In result this mean that the HP50G timer emulation inside the Saturnator is > buggy comparing to the behavior of the original Yorke chip. One thing is certain, the Saturnator work fine :-D Documentation says: TIMER1 -> 1 nibble timer -> decremented 16 times/s ... if so, is only 1/2 nibble timer for Saturn. Anyone can make this process with TIMER2 ? (delay one second) - Gaak - === Subject: Re: TIMER1 decremented 16 times/s Timer2 works in the same way on the Yorke chip (just replaced the 1 by 2): The CPU is awakend from the sleep state when the WKE bit of the =TIMERCTRL.2 register is set and the MSB of the =TIMER2 value is 1. What's correct or not, my Clarke hardware description document definitely say: WAKE - If the corresponding timer's MSB is a one AND WAKE is set AND the CPU is asleep then NCD is pulled low. This bit is cleared on the first data strobe after a wakeup. So the Yorke chip works exactly like the hardware document say. Christoph Gaak schrieb im Newsbeitrag > In result this mean that the HP50G timer emulation inside the Saturnator > is > buggy comparing to the behavior of the original Yorke chip. One thing is certain, the Saturnator work fine :-D Documentation says: TIMER1 -> 1 nibble timer -> decremented 16 times/s ... if so, is only 1/2 nibble timer for Saturn. Anyone can make this process with TIMER2 ? (delay one second) - Gaak - === Subject: Re: TIMER1 decremented 16 times/s posting-account=KxEIGwoAAAALMxy6hykxLK1wPdiyuNkK Gecko/2008120122 Firefox/3.0.5,gzip(gfe),gzip(gfe) > -------------------------------------- > DELAY ?ExpVal ?50g ? ? 49G > ?1 ? ? 0.0625 ?0.0708 ?0.0540 > ?2 ? ? 0.1250 ?0.1335 ?0.0901 > ?3 ? ? 0.1875 ?0.1959 ?0.2025 > ?4 ? ? 0.2500 ?0.2585 ?0.2339 > ?5 ? ? 0.3125 ?0.3175 ?0.3192 > ?6 ? ? 0.3750 ?0.3805 ?0.3790 > ?7 ? ? 0.4375 ?0.4421 ?0.4305 > ?8 ? ? 0.5000 ?0.5087 ?0.4902 > ?9 ? ? 0.5625 ?0.5669 ?0.0244(?) > 10 ? ? 0.6250 ?0.6335 ?0.0247(?) > 11 ? ? 0.6875 ?0.6918 ?0.0247(?) > 12 ? ? 0.7500 ?0.7550 ?0.0249(?) > 13 ? ? 0.8125 ?0.8174 ?0.0250(?) > 14 ? ? 0.8750 ?0.8804 ?0.0251(?) > 15 ? ? 0.9375 ?0.9419 ?0.0253(?) > 16 ? ? 1.0000 ?1.0059 ?0.0254(?) You're storing DELAY-1 in TIMER1 and I can't help noticing that the values go bad when the nibble stored changes from 7 to 8, which is the same as changing from 7 to -8 if the value is interpreted as a signed two's complement value. So could the problem be that the counter is interpreted as two's complement? === Subject: Re: TIMER1 decremented 16 times/s ----- > DELAY ?ExpVal ?50g ? ? 49G > ?1 ? ? 0.0625 ?0.0708 ?0.0540 > ?2 ? ? 0.1250 ?0.1335 ?0.0901 > ?3 ? ? 0.1875 ?0.1959 ?0.2025 > ?4 ? ? 0.2500 ?0.2585 ?0.2339 > ?5 ? ? 0.3125 ?0.3175 ?0.3192 > ?6 ? ? 0.3750 ?0.3805 ?0.3790 > ?7 ? ? 0.4375 ?0.4421 ?0.4305 > ?8 ? ? 0.5000 ?0.5087 ?0.4902 > ?9 ? ? 0.5625 ?0.5669 ?0.0244(?) > 10 ? ? 0.6250 ?0.6335 ?0.0247(?) > 11 ? ? 0.6875 ?0.6918 ?0.0247(?) > 12 ? ? 0.7500 ?0.7550 ?0.0249(?) > 13 ? ? 0.8125 ?0.8174 ?0.0250(?) > 14 ? ? 0.8750 ?0.8804 ?0.0251(?) > 15 ? ? 0.9375 ?0.9419 ?0.0253(?) > 16 ? ? 1.0000 ?1.0059 ?0.0254(?) You're storing DELAY-1 in TIMER1 and I can't help noticing that the > values go bad when the nibble stored changes from 7 to 8, which is the > same as changing from 7 to -8 if the value is interpreted as a signed > two's complement value. ?So could the problem be that the counter is > interpreted as two's complement? No, the syntax is not the problem, neither the ROM. All 49G has the same problem, and you can replace DELAY-1 by n (number). - Gaak - === Subject: debug4x garbled screen and can't load 50g on emu48 1.47 I just downloaded and installed debug4x, and noticed that pressing on-d on emu-50g that comes with it shows the bottom screen garbled, i also tried but couldn't figure out a way to enter the rom-update menu, and i tried using the cristophs's stand-alone emu48 latest release (1.47) to load the debug4x kml files for the 50g and was unable to. I'm frustrated now :-( Any hints? Steve Sousa === Subject: Re: debug4x garbled screen and can't load 50g on emu48 1.47 > I just downloaded and installed debug4x, and noticed that pressing on-d on > emu-50g that comes with it shows the bottom screen garbled, I did 0n-D here with no problem. Whay do you mean by bottom screen ? What are you trying to do? -- - - - - - - - - - - - - - - - - Bill Graves RKBA! bgraves@ix.netcom.com === Subject: Re: debug4x garbled screen and can't load 50g on emu48 1.47 posting-account=itRlSAoAAADIJvOzSGM8mjsK7gvlk1wg Gecko/20080201 (CK-PC-WELT) Firefox/2.0.0.12,gzip(gfe),gzip(gfe) Hello Steve, when having installed debug4x, you just have to start the HP50g- Emulator from windows start menu. I did it some days ago and it worked. If you do not need the whole debug environment, you have to copy the folder EMU to a separate place and de-install debug4x. As far as I understand you can edit the kml-file to choose the rom version manually. Peter === Subject: PARTF not doing the job? posting-account=GVUxYAoAAAAdOTkvz69xDfu_dTC2GGXu Gecko/2008120122 Firefox/3.0.5,gzip(gfe),gzip(gfe) I wanted to have simple fractions for 2/(1-x^2) so I thought PARTF was the function to use. However the output was 2/-((x-1)(x+1)) which is not even a simple fraction. Of course I was expecting 1/(1+x) + 1/(1- x). Isn't PARTF what I should be using, or is there anything that I didn't set up well? PS: I have an HP 50g. === Subject: Re: PARTF not doing the job? > I wanted to have simple fractions for 2/(1-x^2) so I thought PARTF was > the function to use. However the output was 2/-((x-1)(x+1)) which is > not even a simple fraction. Of course I was expecting 1/(1+x) + 1/(1- > x). > Isn't PARTF what I should be using, or is there anything that I didn't > set up well? > > PS: I have an HP 50g. On my 49+ it is PARTFRAC, and works properly. Try doing CASCFG first. === Subject: Re: PARTF not doing the job? posting-account=Q2CEjQoAAACue5ZDDhUeRzv1w0u8hxZE Gecko/2008120122 Firefox/3.0.5,gzip(gfe),gzip(gfe) > Isn't PARTF what I should be using, or is there anything that I didn't > set up well? Run CASCFG (which resets default CAS configuration)and try again. TW === Subject: Re: PARTF not doing the job? posting-account=GVUxYAoAAAAdOTkvz69xDfu_dTC2GGXu Gecko/2008120122 Firefox/3.0.5,gzip(gfe),gzip(gfe) OK, so I tried that and it works. I realized that PARTF doesn't do that simplification if complex mode is OFF. Honestly, I don't see why... === Subject: WHIRL Program posting-account=HQcyKwkAAAAfyVM-3k2cuaNBENqZAwPy Gecko/2008120122 Firefox/3.0.5,gzip(gfe),gzip(gfe) Gustavo Portales presents: The new Whirl program (v1.2) available for 49 series. http://www.gaak.org/hp/whirl/ Enjoy it! - Gaak - === Subject: Re: WHIRL Program Nice looking, and very fluid on HP48SX, quite a great achievement Gustavo ! === Subject: HP50g wireless starts on by itself The wireless on my HP50g periodically turns on after pressing the ON key. Sometimes the wireless turns off after restarting the calculator. Sometimes wireless stays on after restart, and I get the massage Warning: LowBat (S). Wireless stays on after warm and cold reboots. After waiting several several minuits, the calculator turns on with wireless off. Can anyone help me diagnose/fix the situation? === Subject: HP50g wireless starts on by itself The wireless on my HP50g periodically turns on unintentionally. Wireless turns on after switching windows or just a plain restart. Sometimes the wireless turns off after restarting the calculator. Sometimes wireless stays on after restart, and I get the massage Warning: LowBat (S). Wireless does not turn off after warm and cold reboots. After waiting several several minuets, the calculator again turns on with wireless off. Can anyone help me diagnose/fix the situation?