What is the real resolution that can be realized in practical electron microscope operation? We have investigated this problem for TEM imaging in a systematic study on Ce-doped yttrium orthoaluminate, YAlO 3 . Employing a FEI Titan G3 50-300 (PICO) electron microscope equipped with the new CEOS C-COR+ corrector for spherical (C S ) and chromatic aberration (C C ) we studied the conditions to resolve the 57 pm Y-Y-atom pair distance in projection along [010]. Our successful imaging of this pair proves a record resolution for direct 200 kV TEM imaging [1].In a C S and C C corrected electron microscope the two major resolution limitations of previous instrument generations due to partial spatial and partial temporal coherence are significantly reduced. The spatial resolution of such an instrument is limited by an incoherent image spread [2]. This dampens linear and non-linear contributions at the same image frequency by an equal amount. This is an essential qualitative difference compared to the previously dominating partial temporal coherence effects, which in general cause a different dampening of linear and non-linear contributions. Besides the typical sources of image spread, such as mechanical vibrations of the specimen holder and electrical instabilities of beam deflectors, thermal magnetic field-noise, also known as Johnson noise, is currently considered as the major contribution to the image spread in the C-COR setup [3]. The C-COR allows to effectively reduce unwanted coherent axial aberrations of the imaging system up to the fourth order (C 4 ), and to obtain optimum phase-contrast transfer up to the information limit by adjusting also the fifth-order spherical aberration (C 5 ). The PICO microscope is equipped with an advanced version, C-COR+, of the image corrector. It is optimized for an as low as possible level of Johnson noise thus providing an improved resolution compared to that of C-COR [3] at 200 kV.In order to allow a comparison of the performance of different instruments resolution has inevitably to be defined universally and independent of sample parameters. In reality, however, these parameters may be decisive for the direct perception of resolution in practical work. On the other hand, today, where information limits in the range of 50 pm are characterizing electron microscopy, the dilemma of an insufficient image-signal separation due to sample parameters applies quite frequently and contrast simulations have become compulsory for image interpretation. These calculations are providing also an indirect access to the value of a microscope's spatial resolution. However, a necessary condition in order to be able to extract the instrument resolution based on image simulations is a quantitative comparison of experimental intensities with calculated values on the same absolute scale. This has just recently been achieved [4]. Within this numerical forward modeling of the imaging process in TEM the real atomic separation in the sample can be determined irrespective of whether this is directly visibl...