I've now compared the accuracy of my LamW formula with that of Maxima 5.39.x and Wolfram alpha. LamW is in perfect agreement with Maxima 5.39.x , lambert_w() function ; however there is a discrepency between LamW and ProductLog at the 17th decimal digit . You might also notice that after 12 zooms or so the resolution DETERIORATES . Sciwise