A43 This may be the problem. I don't own a 48, but I think it doesn't have hardware flow control on IR either (the 49 doesn't have it on wire, that's for sure). Try XON/XOFF (software). HTH There are a couple of commands which will display text, one of which is not in the manual (a late entry). 1. DISP n;text This displays the text on line n (of 7) on the screen. It is quick and easy but very annoying because it wipes the entire line before writing the text. This means that you can't use it to (for example) label a graph as it destroys the image you're writing on. You also can't write part of a line and then come back later and add to it. 2. MSGBOX This pops up a message box containing the text. It is quick and easy but the font is not adjustable. eg MSGBOXPress ENTER to finish: 3. Store the image as a GROB and edit. (NOT recommended) This method is VERY slow and clumsy. If you're working on the 38G or want your program to be portable to a 38G you have to use it as the next command (DISPXY) is not available on the 38G. Basically you capture the image to a GROB (graphics object G0..G9), convert your text to another GROB, superimpose the two and then redisplay the new image. Eg. REPLACE Page;(Xmin,Ymax);G1: Suppose M=2 and C=3. This program will capture the PLOT screen last displayed, create a GROB G1 containing the equation Y=2X+3 and superimpose it onto the plot image then redisplay on the screen. Note the QUOTE( ) around Y and X so that the symbols are used instead of the values. 4. DISPXY This was a command that JYA added to the calculator too late to go into the manual although it does appear in the CMDS menu. It was something that I'd been pleading with him to add all along and I think he either took mercy or more likely did it to get me to finally shut up :-). This command is really great as it allows you to write text directly onto an (x,y) coordinate on the screen (coords determined by current plot settings of active aplet) using a couple of different fonts. It is VERY easy to use and makes the aplets/programs run much faster too. Here's the same code as the previous example, done using DISPXY. DISPXY Xmin;Ymax;1;y=Mx+C: You can find more information on DISPXY on my FAQ page at http://members.iinet.net.au/~ccroft . See question #64. H216ctor ( The fields in minifont ) IfMsgGetFieldGrob OVER#=case :: DROP FLASHPTR IfSetGrob TRUE ; DROPFALSE This is interesting. It appears you may have stumbled upon a real rarity : a bug in the 48-R . The problem stems from the fact that xPi is not a number. Rather, it is a small secondary that determines ( based on the system flags 2 and 3 ) whether it should push the real number %PI to the stack or the symbolic SYMBOL xPi ;. This causes no problems when you hit the keys for Pi because in that case the secondary is evaluated. But when you enter Pi into a list at the command prompt it is not evaluated so that when x* encounters xPi in the list it errors out because a secondary is not a valid argument type. neat how you can put either of these into an equation with EQW, in function notation =) I'll have to try, a little later, to is. One thing I'd like to do though, is be able to put in the function (f+g), as in the subject where 'x' is given to it. This is a notation my calculus teacher used recently; it's equivalent to 'f(x)+g(x)', and I imagine that more interesting operations than addition can happen when you build a function in that manner. EQW doesn't seem to accept it though. Is there a simple way to put (f+x)(x) into EQW, in a manner that would evaluate correctly? I'm thinking of having an algebraic-object- modifying function 'Q' that would translate (f+g)*(x) to 'f(x) + g(x)', but I'd like to know if there was another way. (www.samsoncables.com !) You're in the kill files eg. NG filters and now those who might know the answer can't see your question AND even if they do - just might not wan't to answer... So maybe polite behaviour is a good thing after all Sorry that I (who reads everything here) couldn't answer this interesting (& repeated) question of yours :-( When I try to STOre tagged object in variable: << M : 5 @ any tagged object 'A' STO in variable A will be only 5 (M will bot be stored). Is it a bug? So, how can I store tagged object in variable? Impossible? Would you settle for {:tag:object } STOred in a variable? object. How about a double Choose, which is used in the HP 39/40 models??? .. X 101.6 also A Bor[g]land [humor for Nick] Object Pascal Compiler Environment I know that, but thats not the point. Shouldn't the BASIS command depend on flag -98, so that if you express your vectors vertically, then you should get r vectors with n elements, and if you like the vectors horizontally, you get r vectors with m elements. A steep easier/simpler would be: a*(5*b^9*c^13)^(1/5) Or run the little ASMs from Keyman: There you have the choice to assign LSH SPC r~¡ ? XYZ~R