Today, the variable regions of 16S rRNA gene (V3-V4) is widely used to profile microbial communities, however the limited read lengths (460pb for the V3-V4 region) combined with the complexity of microbial samples makes it difficult to accurately identify bacterial strains and their abundance. The arrival PacBio Sequel II allow us to generate longer reads named “HiFi reads” using the circular consensus sequencing (CCS) mode, with an accuracy similar to the one observed on short reads. Therefore, we wanted to evaluate the taxonomic resolution using longer target region sequencing, such as, the full-length 16S (1.5kb) and 16S-23S genes (operon 4.5kb) and compared them to 16S RNA V3-V4 region. First, we have evaluated and implemented an extraction method in order to produce high size DNA fragment for long read sequencing while maintaining a good representativeness. Then, we conducted a comparative study of three PacBio protocols, the “Barcoded universal primers protocol” (BUP), the “Barcoded target specific primers protocol” (BTSP) and the “Barcoded overhang adapters protocol” (BOA) in order to perform CCS sequencing of the 16S, and 16S-23S genes on a microbial Community standard (mock) as well as pig faeces samples. Finally, we compared the taxonomic resolution between the 16S-23S, the 16S full length, and the 16sRNA V3V4. This work have been carried out within the framework of the SeqOccIn project (Sequencing Occitanie Innovation), supported by Get-PlaGe and Genotoul Bioinfo core facilities and financed by FEDER funds (Programme Opérationnel FEDER-FSE_Midi-Pyrénées et Garonne 2014-2020).