The current second language (L2) instruction research has raised great motivation for the use of both processing instruction and meaningful output instruction tasks in L2 classrooms as the two focus-on-form (FonF) instructional tasks. The present study investigated the effect of structured input tasks (represented by referential and affective tasks) compared with meaningful output tasks (implemented through text reconstruction cloze tasks) on the acquisition of English nominal clauses (NCs). The study sought to investigate if (1) both input and output instruction would lead to significant gains of knowledge in acquiring NCs, and (2) there were any significant differences between learners' receptive and productive knowledge of nominal clauses. First-year undergraduate students studying at four intact university classrooms participated in the study. The effectiveness of the tasks was determined by a noun-clause recognition test and a sentence combination production test administered both as the pretest and posttest. The results revealed that both processing instruction and meaningful output instruction helped the learners improve their receptive knowledge of grammar effectively nevertheless, the processing instruction group did not significantly outperform the meaningful output group in their gains of receptive knowledge of grammar. The findings further illustrated that meaningful output instruction group significantly outperformed processing instruction group in their productive knowledge of grammar.