I've attached two screenshots showing a minor anomaly with the epsilon hint on double comparisons. Screencap1 shows RS' suggestion to convert the statement to use an epsilon comparison. The idea is good in principle. Screencap2 shows the result after RS converted it. It shows the result of the code conversion and the error. I had to change it to double.Epsilon in order for it to compile. Perhaps RS did it that way intentionally or could it be a minor bug?