We take a close look at the task of prospective time reproduction, wherein an individual is aware of the fact that she will subsequently be asked to reproduce a demarked duration. Our participants were either explicitly instructed not to count, or were allowed to count. When participants are allowed to count, their reproductions (R) tend to be a linear function of target duration (D). When instructed not to count, they exhibited a shorter log(class="EmphasisTypeItalic ">R) mean value than those who were allowed to count. Participants not counting are thus less veridical in time estimation. Given that for them β <1, this suggests that subjective time for them is not a linear function of physical time. We further contrast four major indices relating reproduced time to target duration: R/D, D/R, |R − D|, and |R − D|/D. While the D/R ratio score detected the difference between groups; this was not the case for the other measures.