Various notes about the TI 92

[Next][Index][Thread]

Various notes about the TI 92

These are just a few things I have noticed about the TI-92 and
suggestions about what could be changed.


It needs support for different number bases, I have to keep my TI-85
handy incase I need to use binary or hex.


I find it annoying how it moves square roots from the bottom to the
top.  e.g. the one over the square root of 2 becomes the square root
of 2 divide by 2.  It would be nice if this feature could be turned
off.


The TI-92 gives a incorrect result for the limit of 0^x as x tends
toward 0.  It gives the limit as 1 from the left, right and both
sides.  In fact the limit is 0 from the right, no limit from the left
and hence no limit from both sides.  It does give a 0^0 replaced by 1
warning, but that does not really make up for it.


Another feature that is missing from the TI-92 was the physical units
converter.  I know there is a program available for the 92 that will
do this but it is a bit limited e.g. you can't use the program in
conjunction with the solver to find what temperature has the same
value in both celsius and fahrenheit.  You could do that on the TI-85.


Another thing I miss is physical constants!  It is of course easy to
add them in but it would be nice if they were built in.  They could be
implemented like pi and e are in exact mode.


One small thing that would be nice is if the complex number symbol 'i'
could be changed to a 'j' at will.  It is only a little thing, but AC
electronics is hard enough with out having to convert j's to i's them
i's to j's.


Something I have noticed with polynomials and the Sigma (sum) is that
if you do it in a round about way it is faster than doing it directly.
e.g.
it is faster to do  (capital E represents sigma)
E(x^2+3x+4,x,1,b)
which gives b^3/3+2b^2+17b/3
then substitute in a value for 'b' and you get the answer.
For large b (>200) this is often quicker than entering 'b' directly
E(x^2+3x+4,x,1,100000) takes a long time, doing the above takes a lot
shorter time.


A feature I would like to see would be a normal distribution function.


Lastly the TI-92 needs more RAM!


Richard Gallagher
rwg1@wave.co.nz