We present fitted UBVRI-ZY and u ′ g ′ r ′ i ′ z ′ magnitudes, spectral types and distances for 2.4 M stars, derived from synthetic photometry of a library spectrum that best matches the Tycho2 B T V T , NOMAD R N and 2MASS JHK 2/S catalog magnitudes. We present similarly synthesized multi-filter magnitudes, types and distances for 4.8 M stars with 2MASS and SDSS photometry to g < 16 within the Sloan survey region, for Landolt and Sloan primary standards, and for Sloan Northern (PT) and Southern secondary standards.The synthetic magnitude zeropoints for B T V T , UBV RI, Z V Y V , JHK 2/S , JHK M KO , Stromgren uvby, Sloan u ′ g ′ r ′ i ′ z ′ and ugriz are calibrated on 20 calspec spectrophotometric standards. The UBV RI and ugriz zeropoints have dispersions of 1-3%, for standards covering a range of color from −0.3 < V − I < 4.6; those for other filters are in the range 2-5%.The spectrally matched fits to Tycho2 stars provide estimated 1σ errors per star of ∼0.2, 0.15, 0.12, 0.10 and 0.08 mags respectively in either UBV RI or u ′ g ′ r ′ i ′ z ′ ; those for at least 70% of the SDSS survey region to g < 16 have estimated 1σ errors per star of ∼0.2, 0.06, 0.04, 0.04, 0.05 inThe density of Tycho2 stars, averaging about 60 stars per square degree, provides sufficient stars to enable automatic flux calibrations for most digital images with fields of view of 0.5 degree or more. Using several such standards per field, automatic flux calibration can be achieved to a few percent in any filter, at any airmass, in most workable observing conditions, to facilitate intercomparison of data from different sites, telescopes and instruments.Internal relative flux calibration of single or repeated data sets are routinely achieved to 0.2% or better, including during non-photometric conditions, by reference to non-variable stars in the observed field. But significant questions arise about effective cross calibration of observations from different epochs. The situation is particularly complicated for time-domain science, where multiple sites, telescope apertures, filters and sets of instrumentation become involved, or when time constraints or observing conditions preclude traditional calibrations. Offsets between otherwise very accurate data sets can be much larger than expected. This can introduce significant uncertainty in multi-observation analysis, or obscure real variations.Cross comparisons can be facilitated by parallel wide-field observations along the line of sight to each image, as in the CFHT skyprobe facility (Cuillandre et al 2004) 1 , or other "context" camera systems. But having multi-filter standards within each digital image offers many simplifying advantages. The ideal scenario for calibration of optical imaging data would be if there were all-sky stars of adequately known brightness on standard photometric systems, and present in sufficient numbers to provide a few in every digital image.Precursors to such standards include i) the Sloan Digital Sky Survey (SDSS) of 360 million objects, observed with a 2.5m telesco...