Optical mapping is a fluorescence-based physiological method to image spreading of action potential in excitable tissues, such as the heart and central nervous system. Because of the requirements for high speed imaging in low light conditions, highly sensitive high-speed cameras together with an optical system with maximum photon efficiency are required. While the optimization of these two components is relatively straightforward, the choice of the perfect light source is less simple; depending on the other (usually fixed) components, various parameters may acquire different weight in decision-making process. Here we describe the rationale for building an optical mapping setup and consider the relative advantages and disadvantages of three different commonly available light sources: mercury vapor lamp (HBO), xenon lamp (XBO), and light emitting diode (LED). Using the same optical system (fluorescence macroscope) and high-speed camera (Ultima L), we have tested each of the sources for its ability to provide bright and even illumination of the field of view and measured its temporal fluctuations in intensity. Then we used each in the actual optical mapping experiment using isolated, perfused adult mouse heart or chick embryonic heart to determine the actual signal to noise ratio at various acquisition rates. While the LED sources have undergone significant improvements in the recent past, the other alternatives may still surpass them in some parameters, so they may not be the automatic number one choice for every application.