Given the following information, what is the percent error for the results of this titration?

1. The molarity of the NaOH being used to do the titration is 1.650M


2. The volume of NaOH used in the titration was 16.00mL


3. The formula for calculating the molarity of the unknown solution being titrated is 1.650 * 16.00/10 = 2.64


4. The actual molarity of the unknown solution was 2.64M.


5. The percent error for the procedure should be equal to zero: abs(2.64-2.64)/2.64, but Perl calculates the percent error as
   1.6821560979169e-14
   
   
6. This can be traced to the fact that Perl indicates the difference between the experimental value calculated in step 3 above and the value given in step 4 is actually
   -4.44089209850063e-16 rather than 0.

What's the best way to deal with this in Lon-Capa? Bear in mind that we are dealing here with experimental results, so they could vary quite a bit. It does not seem practical therefore to set the tolerance to an absolute value, but in the case where the actual result should be zero, only a percent tolerance of 100% will include zero. Is there a practical method to force an answer, when it turns out to be the result of the computer's floating point inaccuracies, to be rounded off to what is reasonable?