Next-generation sequencing (NGS), through the implementation of metagenomic protocols,has led to the discovery of thousands of new viruses in the last decade. Nevertheless, these protocolsare still laborious and costly to implement, and the technique has not yet become routine for everydayvirus characterization. Within the context of CRESS DNA virus studies, we implemented twoalternative long-read NGS protocols, one that is agnostic to the sequence (without a priori knowledgeof the viral genome) and the other that use specific primers to target a virus (with a priori). Agnosticand specific long read NGS-based assembled genomes of two capulavirus strains were compared tothose obtained using the gold standard technique of Sanger sequencing. Both protocols allowed thedetection and accurate full genome characterization of both strains. Globally, the assembled genomeswere very similar (99.5–99.7% identity) to the Sanger sequences consensus, but differences in thehomopolymeric tracks of these sequences indicated a specific lack of accuracy of the long reads NGSapproach that has yet to be improved. Nevertheless, the use of the bench-top sequencer has provento be a credible alternative in the context of CRESS DNA virus study and could offer a new range ofapplications not previously accessible.