Question When should I power off my computer?

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

jamesdsimone

Senior member
Dec 21, 2015
616
164
116
There isn't a general computing forum so posting this here. I've read that powering a computer on and off causes more wear than leaving it on. Is this true and if I have a computer that is idle for days at a time should I power that off?
 

bononos

Diamond Member
Aug 21, 2011
3,883
142
106
Note I said SSD, I have no spinners running in this house. Yes, those die, but SDDs ??? No failures yet, but many spinners have died.

Do mechanical harddisks have to be kept on at all times? I think consumer harddisks have been auto sleeping and auto parking for many years now so they are actually 'switched off' when idle. Not sure if the wdidle tool still works for current drives but it can manually disabIe sleep mode, don't know if its a good idea doing turning it off for current drives.
 

Markfw

Moderator Emeritus, Elite Member
May 16, 2002
25,482
14,434
136
Do mechanical harddisks have to be kept on at all times? I think consumer harddisks have been auto sleeping and auto parking for many years now so they are actually 'switched off' when idle. Not sure if the wdidle tool still works for current drives but it can manually disabIe sleep mode, don't know if its a good idea doing turning it off for current drives.
Since all my computers are on 24/7/365, I can't answer that. BUT spinners will die way faster than SSDs IMO
 

bba-tcg

Senior member
Apr 8, 2010
569
277
136
computerguyonline.net
It would be hasty to make such an assumption. For example, the ground circuit could be open and there would be no obvious problem, yet in that situation there would be no reason at all to leave the cord plugged in, and the danger of doing so while handling the machine is increased.
This, along with the switch being off, would, in effect, be the same as unplugged.
 

bba-tcg

Senior member
Apr 8, 2010
569
277
136
computerguyonline.net
Are you really saying that having an ungrounded device plugged in has an identical condition to the same device being not plugged in?
What I'm saying is that, in the current context, with the switch off, that there is no practical difference in having it plugged in with no ground, or having it unplugged. Either way, there is no path to ground and no danger to the operator or the equipment. It does negate the reason that I leave it plugged in but really poses no added risk to working on the machine.

If you believe it does, then go ahead and unplug it. As I've already stated, if the power supply doesn't have an on/off switch, I'll unplug it. This is because on modern systems the board will have power unless the switch is off or it's unplugged.

Edit: I will state that having an ungrounded electronic device is not good practice, whether it's on or off. Assuming that said device doesn't use negative voltages, of course.
 
Last edited:

mikeymikec

Lifer
May 19, 2011
17,577
9,268
136
My main computer goes into sleep mode so I usually don't power it off. I have a server that I might not access for several days at a time sometimes a week. I have it set so the hard drives go into standby mode after 3 hours of inactivity. Xigmanas reports 0% CPU activity but the fans don't shut down.

The fans ought to be powering down. It might be worth checking the BIOS for the kind of sleep mode it offers: S1 is the ancient standby mode which did very little, S3 has been the standard for nearly twenty years. S3 should make the computer sound like it is completely switched off, which is what I'd recommend you use.

If your computer is making regular use of S3 sleep mode then to answer your OP question then S3 is very similar to the computer being completely switched off; the memory is the main component still receiving power (to maintain the contents of memory so the system can resume from sleep where it left off). Fixed storage and the processor are off. USB can be powered in sleep mode or not, it's usually configurable. Network devices can be allowed to wake the system or not.
 
  • Like
Reactions: coercitiv

bigboxes

Lifer
Apr 6, 2002
38,486
11,943
146
My main rig and my home server do 24/7/365 duty. I turn off all power saving settings. Everything runs through a UPS. I do power off my HTPC since it's rarely used these days. Funny, I used it yesterday to watch House of the Dragon.
 

bba-tcg

Senior member
Apr 8, 2010
569
277
136
computerguyonline.net
define "negative voltage", with respect to the hot/neutral/ground ac wiring in a US 115V outlet, please.
Negative DC voltage in the equipment. It's not possible, as far as I know, to have a negative AC voltage. Generally ground is ground with both the DC and AC normally tied to ground. The DC side can't be tied to ground with a negative voltage involved. One way to accomplish that would be to eliminate ground. Thus the statement.
 

Markfw

Moderator Emeritus, Elite Member
May 16, 2002
25,482
14,434
136
define "negative voltage", with respect to the hot/neutral/ground ac wiring in a US 115V outlet, please.
Here is the way it works, here and in Europe(pretty sure). The house is given 2 "hot" AC lines, 120 volts each and a ground. For 220v appliances (open, dryer, stove) they are given these 3 lines, and the difference is 220v between the 2 120v hot lines. For the US, each 110v circuit is driven off one or the other input hot lines, and then there is neutral and ground. In the breaker box, the white 110v hot (neutral) is connected to ground and the hot is the actual 120v hot.
 

bba-tcg

Senior member
Apr 8, 2010
569
277
136
computerguyonline.net
Here is the way it works, here and in Europe(pretty sure). The house is given 2 "hot" AC lines, 120 volts each and a ground. For 220v appliances (open, dryer, stove) they are given these 3 lines, and the difference is 220v between the 2 120v hot lines. For the US, each 110v circuit is driven off one or the other input hot lines, and then there is neutral and ground. In the breaker box, the white 110v hot (neutral) is connected to ground and the hot is the actual 120v hot.
There's still really no negative voltage wherever you place the probes, it's still just an AC voltage, just different values.
 

VirtualLarry

No Lifer
Aug 25, 2001
56,227
9,990
126
That was precisely my point. You said, "unless it uses negative voltages", with respect to disconnecting the AC wiring prior to working on a PC (ALWAYS a good idea). I was inquiring as to how/which "negative voltages" were present on AC wiring. (Hint, not really negative, just potentially out-of-phase.)
 
  • Haha
Reactions: bba-tcg

Markfw

Moderator Emeritus, Elite Member
May 16, 2002
25,482
14,434
136
There's still really no negative voltage wherever you place the probes, it's still just an AC voltage, just different values.
exactly, that is what I was trying to explain. AC current and the word negative don't go together. AC is alternating + and - .
 
  • Like
Reactions: bba-tcg

bba-tcg

Senior member
Apr 8, 2010
569
277
136
computerguyonline.net
That was precisely my point. You said, "unless it uses negative voltages", with respect to disconnecting the AC wiring prior to working on a PC (ALWAYS a good idea). I was inquiring as to how/which "negative voltages" were present on AC wiring. (Hint, not really negative, just potentially out-of-phase.)
I believe that I cleared up any ambiguity in my expanded explanation. "It" being the equipment. I used to work with POTS systems that did use negative DC voltage.

Edit: And although I did still disagree with your assertion about unplugging equipment, you putting always in all caps has swayed me. lol. /s

Edit2: The statement was: "I will state that having an ungrounded electronic device is not good practice, whether it's on or off. Assuming that said device doesn't use negative voltages, of course." So, really, there was no ambiguity until you changed device to it.


 
Last edited:

VirtualLarry

No Lifer
Aug 25, 2001
56,227
9,990
126
The statement was: "I will state that having an ungrounded electronic device is not good practice, whether it's on or off.
And this is incorrect.

There is no benefit to a PC chassis and components being at actual earth ground while working on them, none whatsoever.

You use an anti-static wrist strap, to equalize the potential voltage differences between you and components and the chassis. Whether that's at earth ground is irrelevant.
 

bba-tcg

Senior member
Apr 8, 2010
569
277
136
computerguyonline.net
And this is incorrect.

There is no benefit to a PC chassis and components being at actual earth ground while working on them, none whatsoever.

You use an anti-static wrist strap, to equalize the potential voltage differences between you and components and the chassis. Whether that's at earth ground is irrelevant.
Again, the statement said nothing specific about when working on them. But regardless of what you think, any circuit with electronic equipment connected should be a grounded circuit. For what should be obvious reasons. The added benefit of it being grounded while working on it is just that: an added benefit.
 

aigomorla

CPU, Cases&Cooling Mod PC Gaming Mod Elite Member
Super Moderator
Sep 28, 2005
20,839
3,174
126
Steve and he'll have someone over with the parts later that day.


You know when you hear steve make clicky noises in the background, he's really not paying attention to you, and is actually playing Halo online.

I caught many Steve's doing that, you can stump them and make then go back the first question asked if you do the proper sequence of counter questions.
 

Timur Born

Senior member
Feb 14, 2016
277
139
116
50-55W idle for my 13900K system and 65W+ idle for my 5900X system, after optimizing them somewhat for idle power consumption, so it may be more for you. 20W out of this is the Palit NVidia RTX 2070 Super GPU, though, which is the real shame. I re-enabled Windows' fast boot and push the button when I leave the room/house for longer times. BIOS also is set to make the PSU not use any power in soft-off (modern PSUs draw a lot of wattless/reactive current when off, though).

I don't trust Windows' (automatic) standby, because too many drivers still cause issues after waking up after all this time of it being a thing and many mainboard do stuff like run fans at 100% rpm and stuff like that.
 

mikeymikec

Lifer
May 19, 2011
17,577
9,268
136
I don't trust Windows' (automatic) standby, because too many drivers still cause issues after waking up after all this time of it being a thing and many mainboard do stuff like run fans at 100% rpm and stuff like that.

When reading your post it feels like you're taking me back to the year ~2006 (re driver maturity for sleep mode), I think your mistrust is extremely out-of-date. The only reason I can think of for any fans still running while in sleep mode is that you haven't got S3 sleep mode enabled in the BIOS. I can't remember the last time in my line of work that I encountered any PC running any fans during sleep mode (probably because S3 is the default sleep mode setting in the BIOS these days).
 

Timur Born

Senior member
Feb 14, 2016
277
139
116
Fans running at 100% after waking from sleep was what I meant in my example.

I also feel like 2006, because mainboard manufacturers and driver developers still don't have their act together when it comes to standby/sleep. My own Gigabyte Z790 Aero G just looses all sensors/software control over its fans after wakeup. They don't run at 100% but do run at a higher rpm than what they are set to during idle load.

Doesn't inspire confidence. And there are still drivers out there that cannot deal properly with standby/sleep mode. It's a shame, but fortunately fast boot *does* work properly.

Before Standby:
1673174330356.png

After Standby:
1673174373969.png

Restarting HWinfo (fans completely vanish, even Gigabyte's own software cannot control them anymore):
1673174396916.png
 

Khanan

Senior member
Aug 27, 2017
203
91
111
There isn't a general computing forum so posting this here. I've read that powering a computer on and off causes more wear than leaving it on. Is this true and if I have a computer that is idle for days at a time should I power that off?
This is not true and just a rumour. Power relays, capacitors etc in the pc have all a specific life time, once it is reached things start to die off, it will not die off if you switch the pc on off more often. And PCs who are powered on more will also die quicker than PCs that are often shut down. It’s like a car that is running constantly, obviously that’s not an upside.
 

mikeymikec

Lifer
May 19, 2011
17,577
9,268
136
Fans running at 100% after waking from sleep was what I meant in my example.

I also feel like 2006, because mainboard manufacturers and driver developers still don't have their act together when it comes to standby/sleep. My own Gigabyte Z790 Aero G just looses all sensors/software control over its fans after wakeup. They don't run at 100% but do run at a higher rpm than what they are set to during idle load.

Doesn't inspire confidence. And there are still drivers out there that cannot deal properly with standby/sleep mode. It's a shame, but fortunately fast boot *does* work properly.

Before Standby:
View attachment 74200

After Standby:
View attachment 74201

Restarting HWinfo (fans completely vanish, even Gigabyte's own software cannot control them anymore):
View attachment 74202

For the builds I do I stick to Asus, never had this trouble. IMO you've got two problems, Gigabyte and a less mature platform. I wonder whether your driver problems might be due to Gigabyte as well; do you need to install any from them specifically?
 

Timur Born

Senior member
Feb 14, 2016
277
139
116
For the builds I do I stick to Asus, never had this trouble. IMO you've got two problems, Gigabyte and a less mature platform. I wonder whether your driver problems might be due to Gigabyte as well; do you need to install any from them specifically?
While my fans are a specific standby related problem on my system the whole "standby still doesn't work reliably" (or rather wakeup) is a general issue that other people suffer from just the same. Someone posted in 11/2022 that his 5800X3d reproducibly uses higher clocks after wake from sleep than before (plus higher accompanying temps). You do you and I do fast boot until these reports finally die out and at least my own system doesn't show oddities (and I know how to test that properly).
 

Khanan

Senior member
Aug 27, 2017
203
91
111
Standby is just a mess I stopped using it many years ago and never looked back. If I’m using the pc it’s on if not I shut it down if I know I won’t go back. In the era of ssds it’s really not needed to use standby.

edit: I should add my monitors are set to power off after 5 minutes.