Models of fluid flows in porous materials commonly fall short because they fail to capture the effects of the puzzling underlying microscopic dynamics. These flows are very common: examples range from groundwater flow and H₂ storage in underground rocks to water discharge in fuel cells. The fluctuating, microscopic dynamics are poorly understood because they have so far been inaccessible in the 3D labyrinths formed by pore geometries, hampered by the optical opacity of the materials. In FLOWSCOPY, I will cause a paradigm shift by resolving this inaccessibility, enabling the measurement of unsteady 3D flows inside opaque porous materials. First, I will enable the inspection of flow fields in all their µm-scale complexity by creating a method that tracks tracer particles flowing through the pores with 3D X-ray imaging. To achieve the required millisecond imaging times - up to 3 orders of magnitude faster than my state-of-the-art preliminary results – the new approach will retrieve tracer locations in each of the many radiographs that conventionally make up a single tomographic time frame. Then, I will untangle the upscaling problem, building the first method that can measure flow maps averaged on a sliding scale from nano- to centimetres. Finally, I will apply the methods’ transformative capabilities to two pertinent problems in arguably some of the most complex porous media: geological materials. First, I will investigate how two fluids, such as water and H₂, displace each other in porous rocks, lifting the veil on capillary fluctuations that deviate from current models. Second, I will unriddle flows of viscoelastic fluids, such as those to clean up polluted sediments, which exhibit a poorly understood transition from steady to chaotic dynamics. Beyond this, the new techniques will be applicable to a wide range of natural and engineered microstructures, from arteries to building materials.