Statement of problem. An often-debated issue still exists concerning implant impression techniques, whether to splint impression copings. Different configurations are available for these copings for a variety of manufactures' implant systems. Purpose, This study evaluated and compared 4 impression techniques in terms of their dimensional accuracy to reproduce implant positions on working casts. Material and methods. A master model was designed to simulate ii clinical situation. Impressions were made using 4 techniques: (1) tapered impression copings trot splinted; (2) squared impression copings not splinted; (3) squared impression copings splinted with autopolymerizing acrylic resin; and (4) squared impression copings with a lateral extension on one side not splinted. Reference points machined onto the master model and onto special healing abutments were compared after abutments were transferred to casts using the 4 techniques. Measurements were made using a Reflex microscope, capable of recording in the x-, y-, and z-dimensions. Results. The dimensional accuracy was high and, although statistically significant (P=.022; power > 80%), a maximum distortion difference of only 0.31% was registered. Conclusion. The dimensional accuracy of all the techniques was exceptional and the observed differences can be regarded as clinically negligible.