PURPOSE To assess the fracture resistance of single-tooth implant-supported crown restorations made with different CAD/CAM blocks. MATERIALS AND METHODS Thirty-six titanium abutments were put on dental implant analogs (Mis Implant). For each of three test groups (n = 12/group), implant-supported, cement-retained mandibular molar single crowns were produced. Crowns were made of lithium disilicate glass (LD) IPS e.max CAD, feldspathic glass ceramic (FEL) Vita Mark II, and resin nano-ceramic (RNC) Lava Ultimate. The crowns were cemented with self-adhesive resin cement RelyX Unicem 2. After chewing cycling, crowns were tested to failure in a universal testing machine. Fracture values were calculated as initial (F-initial) and maximum fracture (F-max). RESULTS The study groups were ranked, in order of having highest value, (LD > FEL) > RNC for F-initial load value and (LD > RNC) > FEL for F-max load value. This demonstrated that there was no parallel change in the F-initial and F-max values presenting the fracture resistance of specimens. CONCLUSIONS There was no accordance between the F-initial and F-max values of the LD, RNC, and FEL after chewing simulation with thermocycling resembling 5 years of clinical functional use. LD had the highest fracture resistance during the fracture test. RNC had low fracture resistance; however, it had considerably high fracture resistance during the fracture test. FEL had considerably low fracture resistance values.