PURPOSE: The 3 min all-out test has significantly expedited the process of determining the power asymptote (critical power; CP) and the curvature constant (W') of the hyperbolic power-duration relationship for severe intensity exercise. It was hypothesized that selective glycogen depletion of the type I and type II muscle fibres would result in alterations to the CP and W'. Specifically, glycogen depletion of type II muscle fibres was hypothesized to result in a reduction in W' with no appreciable impact upon CP, whereas glycogen depletion of type I muscle fibres was hypothesized to result in a reduction in CP, with no reduction in W'. METHODS: Eight physically active males performed the 3 min all-out cycling test on three occasions: the control condition (C); the day after completing 3-hrs of cycle exercise at 30% VO2max (Type I depletion; T1); and the day after completing 10 x 1-min intervals (5 min rest) at a power output equivalent to 120% VO2max (Type II depletion; T2). The mean power output over the final 30 s of the test was used to estimate the CP; W' was estimated as the power-time integral above end-test power. Data were analyzed using one-way ANOVA. RESULTS: No significant differences were observed between C, T1, or T2, for the estimated CP (225 ± 44, 226 ± 55, 220 ± 46 W respectively) or W' (21.1 ± 2.2, 19.2 ± 4.8, 20.8 ± 3.4 kJ respectively). Blood lactate concentration immediately post the 3 min all-out test was significantly lower following T1 compared to control (9.72 ± 2.32, 8.23 ± 2.31, and 9.44 ± 2.04 mM.L-1 for C, T1 and T2 respectively). CONCLUSIONS: Selective glycogen depletion of either type I or type II muscle fibers had no significant effect on the estimated CP or W'. It is possible that the 3 min all-out test is not of sufficient duration to produce the expected reduction in the CP following T1. The unchanged W' following T2 may be considered surprising, and questions the importance of anaerobic glycolysis to this parameter. Whilst these data are in contrast to our hypotheses, they may have some important implications for the practitioner working in applied sports science. The robustness of the parameters of the 3-min all out cycling test indicates that athletes may complete this test for a valid determination of the CP and W', despite having completed strenuous glycogen-depleting exercise in the preceding 24h.