After this discovery of lower bounds by Stark and Odlyzko, some people searched for number fields with relatively small discriminants. Also some others studied algorithms which will produce number fields with the smallest absolute value of the discriminant for a fixed degree n. However, this algorithmic approach does not work well if the extension degree n is greater than seven or eight. And it is not clear whether current computers can finish those algorithms in a reasonable amount of time for large n. So we are justified in our approach for this subject. In this paper we will report the results of our computations. Some number fields we found are better than previously known examples. Others are rediscoveries of known examples. However, there is theoretically nothing new in our approach to this subject.