Estimates for Electricity Waste of Windows 10

I have a pet peeve with Windows 10, in that it defaults to automatically updating by waking up in the middle of the night (if the computer is in sleep or hibernate mode) in order to update; and usually remains awake afterwards.

Now, there are several solutions to this problem, but most of these are only ever implemented by a savvy user, and even then, most of those fixes are reverted after Windows 10 undergoes a significant update. Presumably, this means there are many other users that may be dealing with this issue, either knowingly or otherwise.

In order to make a point, I'd like to calculate exactly how much waste this results in and if, at any point, it would become worth it to purchase a separate OS, just in power savings alone. Notable variables that I can think of to the calculations would be as follows:

• (C): average number of computers a household uses


• (T): typical amount of time from wake to user putting it back to sleep if unused


• (V(s)): average power usage of the most common type of windows 10 computer in kWh/year while in sleep mode


• (V(w)): average power usage of the most common type of windows 10 computer in kWh/year while awake

So the goal would be to calculate the total extra usage per year (with extra points for cost conversions).
The formula would presumably be something like

f(x) = C * T *( V(w) - V(s) )

---

Noting that the time while updating should not count as time wasted, and that I'm not looking for the most precise data, only reasonable estimates, for conservation purposes.

I'm also counting time that a laptop is closed and unplugged, but still awake due to its post-update awakened status, as wasted.

I also expect that I'm missing some variables in the equation so please don't assume that my given formula or even my hypothesis are correct!

Steve Rindsberg also pointed out that the default amount of time between idling and returning back to sleep mode should be a consideration as well. (I believe this defaults to 30 minutes; so that could be an adequate substitution for V(w)-V(s).)