When considering the replacement of incandescent lamps by screwbase integrated compact fluorescent lamps (CFLs), we must ensure that the CFL will provide satisfaction to the users in order to avoid the hark-back to the inefficient technology of incandescent lamps. One of the possible reasons of the users' disappointment is the perception of the quantity of the luminous flux of these lamps. The first cause of this bad perception is the lamp photometric distribution which can affect the illuminance of the task areas and the background environment. The second is the time it takes before the nominal flux of the lamp is reached. The first objective of this paper is to analyze the time required for different CFL to reach their nominal flux. While some lamps start rapidly (1.5 minutes to reach 90% of the nominal flux), others require a very long delay to reach their nominal flux and to stabilize (more than 15 minutes to reach 90% of the nominal flux). This delay is difficult to foresee and differences appears even for same type of lamp (model and brand), for different powers. The second objective of this paper is to determine the photometric distribution of various types of CFLs. While quad and three tubes lamps have very different photometric curves (radiating more horizontally than vertically) compared to incandescent bulbs, the photometry of CFL with bulbs is very similar to the photometry of incandescent lamps. Simulations of these lamps placed in a room were carried out and they did not show great variations of the room illuminance, for the different lamps.