Performance differences in protein and proteome characterization achieved by data-independent acquisition (DIA) LC/MS(E) and data-dependent acquisition (DDA) LC/MS/MS approaches were investigated. LC/MS(E) is a novel mode of generating product ion data for all coeluting precursors in parallel as opposed to LC/MS/MS where coeluting precursors must be serially fragmented one at a time. During LC/MS(E) analysis, alternating MS scans of "normal" and "elevated" collision energy are collected at regular intervals, providing nearly a 100% duty cycle for precursor detection and fragmentation because all precursors are fragmented across their full chromatographic elution profile. This is in contrast to DDA-based MS/MS where serial selection of precursor ions is biased toward interrogation and detection of the highest abundance sample components by virtue of the intensity-driven interrogation scheme employed. Both modes of acquisition were applied to a simple four-protein standard mixture with a 16-fold dynamic range in concentration, an in-gel digest of the Arabidopsis thaliana protein FLS2 purified by immunoprecipitation, and a solution-digested tomato leaf proteome sample. Dramatic improvement for individual protein sequence coverage was obtained for all three samples analyzed by the DIA approach, particularly for the lowest abundance sample components. In many instances, precursors readily detected and identified during DIA were either interrogated by MS/MS during DDA at inopportune points in their chromatographic elution profiles resulting in poor quality product ion spectra or not interrogated at all. Detailed evaluation of both DDA and DIA raw data and timing of the MS-to-MS/MS switching events clearly revealed the fundamental limitations of serial MS/MS interrogation and the advantages of parallel fragmentation by DIA for more comprehensive protein identification and characterization which holds promise for enhanced isoform and post-translational modification analysis.