The objective of this experiment was to evaluate acid-insoluble ash (AIA) and indigestible NDF (iNDF) as intrinsic digestibility markers in comparison with total fecal collection (TC) in dairy cows fed corn silage- and alfalfa haylage-based diets. The experiment was part of a larger experiment, which involved 8 Holstein cows [102±28.4 d in milk, 26.4±0.27 kg/d of dry matter (DM) intake, and 43±5.3 kg/d milk yield]. The experimental design was a replicated 4×4 Latin square with the following treatments: metabolizable protein (MP)-adequate diet [15.6% crude protein (CP); high-CP], MP-deficient diet (14.0% CP; low-CP), and 2 other low-CP diets supplemented (top-dressed) with ruminally protected Lys or Lys and Met. Data for the 3 low-CP diets were combined for this analysis. Total feces were collected for 5 consecutive days during each period to estimate total-tract apparent digestibility. Digestibility was also estimated using AIA (digestion with 2 N HCl) and iNDF (12-d ruminal incubation in 25-μm-pore-size bags). Significant diet × digestibility method interactions were observed for fecal output of nutrients and digestibility. Fecal output of nutrients estimated using AIA or iNDF was lower compared with TC and fecal output of DM, organic matter, and CP tended to be higher for iNDF compared with AIA for the high-CP diet. For the low-CP diet, however, fecal output of all nutrients was lower for AIA compared with TC and was higher for iNDF compared with TC. Data from this experiment showed that, compared with TC, AIA underestimated fecal output and overestimated digestibility, particularly evident with the fiber fractions and the protein-deficient diet. Compared with TC, fecal output was overestimated and digestibility of the low-CP diet was underestimated when iNDF was used as a marker, although the magnitude of the difference was smaller compared with that for AIA. In the conditions of the current study, iNDF appeared to be a more reliable digestibility marker than AIA in terms of detecting dietary differences in apparent digestibility of some nutrients, but significant diet × marker interactions existed that need to be considered when estimating total-tract digestibility using intrinsic markers.