>>12223702generally the meaningful way to do this is by the first method you mentioned : 1/0.5= Sginal to noise ratio of 2
notice that the phrase says. "signal" to "noise" is literally telling you that you need to consider the signal amplitud(1) with respect to the noise amplitud(0.5)
in this case is more meaningful to take the "base" reference of that (1)peak with respect to the 0.5 "noise floor"(actual term) and not with respect to the 0 base but you can perfectly make the reference to be 0 but in this case it would not be strictly speaking a "signal to NOISE" ratio
NOTE that this method is a "peak measurement" many times you want the RMS(root mean square) value and in that case things can become a lot more tricky depending on the case in this case the noise floor can be considered to be 0.5 RMS just the same as before but the peak value of that "sinusoid" would be more or less about 0.35 RMS WITH RESPECT TO THE NOISE FLOOR(noise floor is considered to be the 0 reference) as you can see all this depends a lot on what is the reference, a signal below the noise floor for all intent and practical purposes can be considered just part of the "noise" this is why is kinda point less to take the 0 as reference, the "noise floor" (0.5) is a much more meaninful reference
