During the years 1991-1994, the Norwegian Arthroplasty Register recorded 1,324 primary hip arthroplasties implanted with the Boneloc cement. We have compared the survival until revision due to aseptic loosening for charnley (n 955) and Exeter (n 172) prostheses. The Boneloc cemented hips were also compared with high viscosity cemented hips implanted during the same period. In the Boneloc cemented group, the estimated probability of survival at 4.5 years of a Charnley femoral component was 74% and for an Exeter femoral component 97% (p < 0.0001). Using a Cox regression model with adjustment for age, gender, type of cement, systemic antibiotic and stratified for diagnosis, an 8 times higher risk of revision was found in Boneloc cemented Charnley femoral components than in Exeter femoral components (p < 0.0001). For the acetabular components, the difference between the Charnley and Exeter components with Boneloc cement was not statistically significant. In both the Charnley and the Exeter prostheses, the high viscosity cemented components had significantly better survival than the Boneloc cemented components. The Cox regression model showed that a Boneloc cemented Charnley femoral component had a 14 times higher risk of revision than a high viscosity cemented component (p < 0.0001), and for Exeter femoral components a 7 times higher revision risk was found in the Boneloc cemented components (p = 0.003). Our results confirm the previously reported inferior results of Charnley prostheses implanted with Boneloc cement and inferior results of Boneloc cemented Exeter prostheses as well, but less pronounced than for Charnley prostheses.