High-flux solar simulators consist of lamps that mimic concentrated sunlight from a field of heliostats or parabolic dish. These installations are used to test promising solar-thermal technologies for technical viability. Ideally, the conditions in a given high-flux solar simulator closely approximate those in solar furnaces, experimental facilities that are driven by actual sunlight. During the characterization of a new high-flux solar simulator at the University of Colorado an artifact predicted in high-flux solar simulators, but not solar furnaces, was observed experimentally. Specifically, power measurements from the 18 lamp, 45kW(electric), device increased by 11.1% depending on the wider optical environment, which was either specularly reflective or diffusely absorbing. This calorimetry result was confirmed by a flux gauge, which showed that optical conditions could inflate incident flux by more than 20% at the radiation target. An "observer effect" in high-flux solar simulators was first suggested by Monte Carlo ray traces, and refined ray traces recapitulated the experimental results. Thus, solar-thermal designs may behave differently when evaluated in solar furnaces, versus high-flux solar simulators. Artifacts can be minimized by insuring that solar targets, including radiation measurement instrumentation, present diffusely absorbing surfaces to high-flux solar simulators. Under these conditions the new high-flux solar simulator featured a peak flux of 12.50 MW/m(2) and delivered 9.076 +/- 0.190 kW onto a empty set 10 cm target for a mean flux of 1.155 MW/m(2).