OOPS 7 point. Basically followed this http://escher3d.com/pages/wizards/wizarddelta.php and this https://www.youtube.com/watch?v=tDLbqLve128 . It is all manual as you use manual controls/ gcode to lower the head to the plate and use paper as a thickness gauge. I choose 85 mm as my print radius. Enter the results for each of the 7 points remaining z-height when you have the appropriate z height clearance. The result will suggest changes to delta rod length, delta radius, tower angle changes and end stops. I use it the first time and make suggested changes. The second time, it seems to get worse. So what I do is keep measuring the nozzle height error ( i.e. remaining z-height ) and calculate the end stop steps required to make the x,y,z tower positions equal. When you have the 3 tower points ok, usually the in-between tower points are ok and just correct the center. Delta rod length effects this. Too long of a rod causes a dip in the center ( appears that the bed has a dent in it. see http://boim.com/DeltaUtil/CalDoc/Calibration.html

Clear as mud. I can explain better if you like.