Ice friction is affected by various system and surface-related parameters such as ice temperature, ambient air temperature and humidity, relative sliding velocity, specific surface pressures and surface texture (waviness, roughness) as well as the macroscopic geometry of the samples. The influences of these parameters cannot be easily separated from each other. Therefore, ice friction is a very complex tribological system and it is challenging to draw sound conclusions from the experiments. In this work, ice friction experiments with stainless steel samples that have different isotropic surface roughness values were carried out. Two tribological experimental setups were used: (i) an inclined ice track where the sliding velocity of the freely sliding steel samples was determined and (ii) an oscillating tribometer, where the coefficient of friction was assessed. For both experimental setups, the environmental parameters such as air temperature, relative humidity and ice surface temperature as well as the test parameters such as normal load and surface pressure were kept as constant as possible. The results of the experiments are discussed in relation to the ice friction mechanisms and the friction regimes.