#### roieco

hi all,

so the last system I've had ran straight (no shut downs, sleep or hibernation) for about 10 years, and it remained in such great condition that I've now passed it to another to use.
it was used daily for work, games and miscellaneous tasks.

with the new system I've bought I've employed a policy of putting it to sleep if it is not used for more than 30 minutes.

I'd like to discuss the following two topics:
1. which of those two policies is better for the longevity and health of the system's components (electricity bill be damned)
2. how much do I actually save (in kwph) if we assume the computer sleeps for 8 hours every night and ~3-4 more hours during the day
thanks!

#### alceryes

##### Splendid
What about the third option of turning it off when not in use?
I've read papers on the effects of electron migration degradation on components kept on 24x7x365. The takeaway is that it doesn't really matter. You are exponentially more likely to upgrade (and pass on the computer) or suffer failure due to an unknown defect (or just because 'reasons') than to be able to experience the extra longevity from letting your computer sleep. However, given the exact same full set of hardware (down to the microscopic level - which is an impossibility), exact same power and environmental experience (another impossibility outside of a lab environment), and the exact same usage (another impossibility as operating systems and programs do things like updates, etc. at semi-random intervals), your componets that sleep a third of the day will last longer, yes.

The answer to your questions about energy/money savings will vary widely depending on usage, energy cost, and how long the computer sleeps. For24x7x365 usage, let's take a crude average guess of 180W (.18kW). If you pay 10 cents per killowatt hour you'd be looking at just over 43 cents a day. With 8 hours of sleeping you would save a little over 14 cents a day in this instance.

#### lvt

##### Commendable
so the last system I've had ran straight (no shut downs, sleep or hibernation) for about 10 years, and it remained in such great condition that I've now passed it to another to use.
it was used daily for work, games and miscellaneous tasks.
No computer system can run 24/7 for 10 years, it might have been in some power states like S3 (aka suspend to RAM). In this mode everything is shutdown except the RAM and input devices like mouse, keyboard. The whole power consummation during this time is about 2W. You can wake the system up by simply click the mouse button or press a key on the keyboard.

#### roieco

@lvt - the computer did run 24/7/365 for that time, and was never shut down except for maintenance. the only components that i had to replace over the years were hard drives, and even those not for the last few years.

#### roieco

i heard that turning the computer on causes a small spike that can damage components over time. can anyone refute/verify if that's true?

#### lvt

##### Commendable
i heard that turning the computer on causes a small spike that can damage components over time. can anyone refute/verify if that's true?
It shouldn't happen if you have a good & reliable PSU, PSUs usually have a voltage regulator circuit to maintain a steady current.

Motherboards and other components have their own voltage regulator as well.

I have several old HDDs with over 3,000 hours of power on count, it means that there is no danger of "spike" despite the fact that the computer has been turned on/off thousands of times down the years.

Moderator

roieco

thanks all

#### hotaru.hino

##### Respectable
i heard that turning the computer on causes a small spike that can damage components over time. can anyone refute/verify if that's true?
Inrush current, while a thing, is a well known issue that any electronics designer worth the air they breath would account for. It's basically a non-issue.