A comparison of silver and molybdenum microfocus X-ray sources for single-crystal structure determination was conducted using six model compounds with varying absorption factors. The experiments were performed on two 30 W air-cooled Incoatec I μS microfocus sources with multilayer optics mounted on a Bruker D8 goniometer with a SMART APEX II CCD detector. All data were analyzed, processed, and refined using standard Bruker software. The results showed that Ag Kα radiation can be beneficial when heavy elements are involved. A numerical absorption correction based on the positions and indices of the crystal faces was found to be of limited use for highly focused microsource beams. Empirical corrections implemented in SADABS, although originally intended for absorption correction, also corrected variations in the effective volume of the crystal irradiated well. In three cases, the final SHELXL R1 against all data after applying empirical corrections was below 1%. Further correction for the 2θ dependence of absorption was required, and SADABS used the transmission factor of a spherical crystal with a user-defined value of μr. The results suggest that the IUCr publication requirement for numerical absorption correction for strongly absorbing crystals may need revision. The study compared molybdenum and silver microsource data for various crystals with significant absorption. The results showed that Ag Kα data were generally less affected by systematic errors than Mo Kα data. The precision of the corrected intensities was comparable for both sources, but the accuracy was higher for the silver source due to reduced absorption. The empirical corrections performed remarkably well despite the unfavourable combination of highly focused beams and relatively high absorption. The study concluded that the Ag Kα microsource is preferable for strongly absorbing crystals.A comparison of silver and molybdenum microfocus X-ray sources for single-crystal structure determination was conducted using six model compounds with varying absorption factors. The experiments were performed on two 30 W air-cooled Incoatec I μS microfocus sources with multilayer optics mounted on a Bruker D8 goniometer with a SMART APEX II CCD detector. All data were analyzed, processed, and refined using standard Bruker software. The results showed that Ag Kα radiation can be beneficial when heavy elements are involved. A numerical absorption correction based on the positions and indices of the crystal faces was found to be of limited use for highly focused microsource beams. Empirical corrections implemented in SADABS, although originally intended for absorption correction, also corrected variations in the effective volume of the crystal irradiated well. In three cases, the final SHELXL R1 against all data after applying empirical corrections was below 1%. Further correction for the 2θ dependence of absorption was required, and SADABS used the transmission factor of a spherical crystal with a user-defined value of μr. The results suggest that the IUCr publication requirement for numerical absorption correction for strongly absorbing crystals may need revision. The study compared molybdenum and silver microsource data for various crystals with significant absorption. The results showed that Ag Kα data were generally less affected by systematic errors than Mo Kα data. The precision of the corrected intensities was comparable for both sources, but the accuracy was higher for the silver source due to reduced absorption. The empirical corrections performed remarkably well despite the unfavourable combination of highly focused beams and relatively high absorption. The study concluded that the Ag Kα microsource is preferable for strongly absorbing crystals.