taylor(2^x+3^x,x,inf,2)
=> Break: two equal vars generated
(a debug break)
Normal users should never see debug breaks....
Similarly for tlimit(2^x+3^x,x,inf).
More surprisingly, you also get a debug break for:
taylor(2^x/3^x,x,inf,2)*2
=> Break: bad singular datum
Note that here, unlike the previous case, there is
absolutely no mathematical subtlety.
There are also cases that give incorrect answers, not
errors:
taylor(2^x,x,inf,10)*2 => 2 + ... NO!
taylor(2^x,x,inf,2)*2^x => 1/%e^(-log(2)*x) NO!
Maxima 5.9.0 GCL 2.5.0 mingw32 Windows 2000 Athlon
Logged In: YES
user_id=588346
Similar problems turn up for log(x),x=0:
tl: taylor(log(x),x,0,3) => log(x)+... OK
2*tl == tl+1 => 2+... NO!
tl^2 => 1+... NO!
tl*log(x) => log(x)+... NO!
tl - log(x) => -log(x)+1+... NO!
Logged In: YES
user_id=588346
Another example:
expr: taylor(x^x,x,0,2)
expr-1 => x+x^2/2+... (NO!)
ratdisrep(expr)-1 ... OK