Despite the claims regarding the potential disruptiveness of subtitling for audiovisual processing, existing empirical evidence supports the idea that subtitle processing is semi-automatic and cognitively effective, and that, in moderately complex viewing scenarios, dubbing does not necessarily help viewers. In this paper we appraise whether the complexity of the translated audiovisual material matters for the cognitive and evaluative reception of subtitled vs. dubbed audiovisual material. To this aim, we present the results of two studies on the viewers' reception of film translation (dubbing vs. subtitling), in which we investigate the cognitive and evaluative consequences of audiovisual complexity. In Study 1, the results show that a moderately complex film is processed effectively and is enjoyed irrespective of the translation method. In Study 2, the subtitling (vs. dubbing) of a more complex film leads to more effortful processing and lower cognitive performance, but not to a lessened appreciation. These results expose the boundaries of subtitle processing, which are reached only when the audiovisual material to be processed is complex, and they encourage scholars and practitioners to reconsider old standards as well as to invest more effort in crafting diverse types of audiovisual translations tailored both to the degree of complexity of the source product and to the individual differences of the target viewers.