The simple noise factor: F = m^x

The simple noise factor expression gives these impairment contours (spaced at 1 dB intervals). Note: that, as with the germanium APD curves, these form parallel lines.

Using McIntyre's Expression

McIntyre's expression gives less pessimistic predictions for deviations from the optimum gain where this is low and more pessimistic predictions for optimum gains greater than around 20 (which is the region where these receivers are likely to operate).

Comparing predictions of the two models

The above contour plot shows the difference (in dBs) obtained from the two sets of data by subtracting those obtained using the simpler expression from those obtained using McIntyre's expression; note the two regions - the lower gain one where McIntyre is more optimistic and the higher gain region where the simpler (and less accurate) model predicts lower sensitivity to gain errors.
An isometric plot of the same data is given below:

Conclusions

In the case of silicon APD receivers, the errors involved in using the less exact expression for avalanche noise may result in a predicted lower sensitivity of gain errors to loss of performance than is the case in reality. Under these circumstances, the McIntyre expression should be used.