Imagine photographing a statue in dim light. You face a dilemma: take one bright shot or a series of dim frames from different angles. In the first case, quality is limited by photon noise, the graininess that arises from a lack of photons, while in the second, details overlap, making it harder to reconstruct the 3D structure. X-ray microscanners (microtomographs) face the same problem: they illuminate the sample from various angles, collecting radiation attenuated by the object, and then compute the internal structure.

Scientists have derived a universal principle: regardless of the chosen imaging strategy (whether increasing the number of projections, accumulating signal per projection, or a combination) the minimum achievable resolution at any point in the reconstructed volume is determined solely by the total number of detected photons. This dependence follows a cube root: to double the sharpness, you need eight times more photons. The analogy with a photographer in twilight is apt but simplifies reality: in tomography, the 3D nature of the problem makes signal accumulation even more 'expensive'—each additional angle adds information but also spreads the photon budget over more measurements.

The practical value of this finding is that you can now calculate in advance the optimal balance between the number of projections and exposure for a specific tomograph. The researchers created an open calculator: the operator inputs the setup parameters (beam energy, geometry, detector characteristics) and gets a map of expected resolution. This helps avoid wasting time and resources and predict which sample details will be visible and which will remain beyond the physical limit.