graphing IFTE with ROM 2.15
Wes
function using IFTE on his 50g. It worked fine on his (rom 2.09) but
kept failing on my 50g (rom 2.15).
Later, I tried it EMU48 using a "new" calculator using both ROMs. The
2.09 works fine, but the 2.15 gives the error message "| Error: Bad
Argument Type".
Interestingly, if you use a very simple function, such as 'IFTE
(X<1,2,3)' or 'IFTE(1,X,X^2)' it works. But something like 'IFTE
(X<1,X,X^2)' fails.
Can someone confirm if this is this a new bug introduced in 2.15?
-wes
Bicky-Bock
Bug in 2.15 confirmed, if entered as programme
<< X 2. > X << X SQ >> IFTE >>
no problem.
John H Meyers
Re 'IFTE(X<1,X,X^2)' STEQ ERASE DRAX DRAW PICTURE
> Bug in 2.15 confirmed ["IFTE Error: Bad Argument Type"]
Was already present in 2.10, with VER -> 4.20060919
How is it in BP's version?
VER in 2.09 is 4.20060602
What's VER in 2.15?
[r->] [OFF]
Wes
4.20060919 as well.
I may have to go back to 2.09 on this one. The ability to graph
piecewise functions is pretty important for me.
-wes
Bicky-Bock
HP 40gs, VER 4.20060602, IFTE(X>1,X,-X) graphs correctly.
John H Meyers
>> Re 'IFTE(X<1,X,X^2)' STEQ ERASE DRAX DRAW PICTURE
>>
>> > Bug in 2.15 confirmed ["IFTE Error: Bad Argument Type"]
>>
>> Was already present in 2.10, with VER -> 4.20060919
>> How is it in BP's version?
>> VER in 2.09 is 4.20060602
>> What's VER in 2.15?
> 4.20060919 as well.
>
> I may have to go back to 2.09 on this one.
> The ability to graph piecewise functions is pretty important for me.
None of the following "bugs out":
\<< 'X<1' 'X' 'X^2' IFTE \>>
\<< IF 'X<1' THEN 'X' ELSE 'X^2' END \>>
'(X<1)*X+(X\>=1)*X^2'
It's too bad, however,
that you have to choose a version,
according to which bugs bug you less :)
Since ROM updates are now pretty easy and fast,
you could also hop back & forth, like the bugs bunny :)
[r->] [OFF]
Wes
> None of the following "bugs out":
>
> \<< 'X<1' 'X' 'X^2' IFTE \>>
>
> \<< IF 'X<1' THEN 'X' ELSE 'X^2' END \>>
>
> '(X<1)*X+(X\>=1)*X^2'
>
> It's too bad, however,
> that you have to choose a version,
> according to which bugs bug you less :)
>
> Since ROM updates are now pretty easy and fast,
> you could also hop back & forth, like the bugs bunny :)
I have my TI-83+/84+ students use the '(X<1)*X+(X>=1)*X^2' trick.
Just seems a shame not to be able to use algebraic IFTE as intended.
The TI-Nspire has a true piecewise function entry with curly brackets
and all, like an algebraic CASE statement.
IFTE() functions also fail with TABLE as well with rom 2.15. Even
with 2.09, IFTE doesn't work with the grapher's F' command which
graphs the derivative, although taking the derivative on the stack
works properly.
The IFTE is an interesting function. If you define 'F(X)=SIN(X)',
then F(Z) evaluates to 'SIN(Z)'. But if 'F(X)=IFTE(X<1,2*X,X+3)',
then F(Z) will result in 'IFTE(Z<1,2*X,X+3)|(X=Z)' where the X's are
local variables (LAM). If you edit this expression with EQW with no
changes, the local X's change into global X's (ID X). Using <Left
Shift - Down Arrow> to edit instead also results in globals X's, but
with a different "coding". Look at it with EMACS or NOSY to see what
I mean.
-wes
Wes
> IFTE() functions also fail with TABLE as well with rom 2.15.
Interestingly, there appears to have been similar IFTE problems with
the hp49g. Doing a search for: IFTE table bug
pulls up some similar posts from 2002 and before.
http://groups.google.com/groups/search?&q=IFTE+table+bug+group%3Acomp.sys.hp48
Any connection?
-wes
hga...@xmission.com
>
> Any connection?
>
> -wes
Just enclose the algrebraic expression in program delimiters, i.e.,
\<< 'IFTE(X<1,X,X^2)' \>>
or append an EVAL for good measure, i.e.,
\<< 'IFTE(X<1,X,X^2)' EVAL \>>
and it works.
(Can't wait for the totally bug-free version 2.16, the mother of all
ROM updates, the last one we'll ever need!)
TW
> ROM updates, the last one we'll ever need!)
Hah?
It is extreeemly difficult to make any changes to the ROM. Most banks
are completely full *to the bit*. In other words, if you want to fix
something and to do so you need to add a single sysRPL comman (1
nibble), you will have to find a place nearby that you can edit to
save a nibble and that won't shift anything out of place. If you shift
it at all, the calculator will come crashing down.
There isn't lack of desire to fix and add things, but rather making
any changes takes WAY too long and quickly becomes an exercise in
frustration.
TW
hga...@xmission.com
I know. My comment was tongue-in cheek.
(Personally, I did go back to Ver. 2.09, though. Among other things,
I like the fact that, with that version, you can directly graph
algebraic expressions, without having to resort to work-arounds for
some cases).
TW
I thought so, but I would elaborate for anyone who might not
understand the issues. I will try and track down where the problem
happened at. I suspect it was to fix something else, but ended up
breaking another thing. . . :-(
TW
hga...@xmission.com
Thanks for all you do! (And this is not tongue-in-cheek).