The performance of any quantitative instrument depends absolutely on the signal-to-noise ratio. A star in the night sky can be seen, and its intensity measured with a suitably equipped telescope, but during the day it is invisible within the noise of the high background light intensity. The signal-to-noise ratio is zero in day light due to the noise. The terrestrially based astronomer is also in trouble at night if there is cloud about. In this case, the signal will be zero and its doesn't matter how low the noise is, you simply will not see anything. If the sky at night is absolutely clear and the telescope is functioning perfectly to its design specification you will be disappointed if you are trying to observe below the detection limit. However, failure to see anything doesn't necessarily mean that there is nothing there.

These analogies are not only readily appreciated but also very close to flow cytometry where performance depends on the correct and integrated functioning of all the constituent parts of the instrumentation, light sources, fluidics, mechanics, optics, photodetectors, electronics and computers. Performance also depends on the biological preparation and the question you are trying to answer. The latter is particularly pertinent if the assay involves working close to the detection limit.

Even if all these components are functioning correctly and you are well above the detection limit with a ‘good’ cell preparation the performance may still be sub-optimal due to design specification or software deficiencies which do not allow the user either to collect or extract from the data base all the information/that may be needed.