I leave my PCs on. While I’m asleep, they check for updates, get backed up, and more.
Good point. And if anyone feels guilty about leaving their machine on all night and wasting ALL that power, they can sign up for SETI-at-home or any number of other resource-sharing programs.
I have an HP all-in-one PC (TouchSmart iq504) with Vista 64-bit. I can’t turn the display off separately.
I have been using sleep mode when I go to bed at night, and it works great. But I recently added an APC UPS with their Powerchute software, which will shut down the PC gracefully in the event of a power loss. The only problem is that when the PC goes into sleep mode, the USB port that the UPS connects to loses power, the connection to the UPS is lost and it renders the software useless. So, if we lose power at night while I’m asleep and don’t realize it, the UPS battery will run out and the PC will shut down hard anyway.
So, I’ve stopped using sleep mode and just keep it on all night. I’ll have to see if it affects the PC in any way.
If this was really a problem (it isn’t) you could just schedule your machine to wake up in the middle of the night, do it’s duties, and shut down again.
Leaving it on all night to take care of these tasks is just a waste.
What do you think caused the failures? Isn’t it hard to tell? If you never turn yours off, you’d know something else caused the failure, but based on your earlier posts I’d guess you ARE turning it on and off.
I think it is pretty hard to tell whether staying on or cycling daily is more of a problem, and to do so really requires lots of experimentation and research, or else lots more of just one or the other. Consistent with this is the idea that neither one is much worse than the other, and it does not matter much.
Without question, energy saving when the thing is off is a small but measureable and undeniable advantage to turning it off.
Similarly, doing work like running updates and security sweeps overnight is a small but undeniable advantage to keeping it on.
Incidence of real-world hard drive failures show little correlation with activity level or temperature. Conversely they do show a significant correlation with drive model, manufacturer and vintage (Pinheiro et al. 2007). High power cycling (on/off switching) in the study was correlated with a 2% increase in failure rate, but was cited as being due to the change in ‘population mix’ and the effect (rather than cause) of troubled machines that required ‘many repair iterations to be fixed’.
Why waste the limited power-on hours of a drive doing little to nothing? And running updates or virus checks doesn’t necessitate the need to keep it on every night. Billions of kWh are not ‘small’ in terms of CO[sub]2[/sub] or cost. With the money an individual saves within a matter of months, they could easily afford to buy a spare hard drive to back up properly, or replace any number of ‘noisy or faulty fans’ anecdotally suggested by Cazzle.
As far as the power consumption angle is concerned, how about expanded on another poster’s 54 cents a day figure for one computer. OK, I have two that run all night and every day and never get shut down unless there’s a problem. And then there’s trickles of power everywhere: two cellphone chargers plugged in night and day, automatic coffemaker, electric shaving charger, etc, etc…how much does all that stuff cost? How can anyone reliably tell?
It is hard to tell. But both failures were due to head-related (as opposed to spindle-related) problems, and one occurred in the middle of a work session. I would expect that a power-cycling failure would tend to occur at power-up.
In years past, most drive failures seemed to be spindle-bearing related. Now most failures I see are I/O failures. Spindle failures are definitely correlated with the number of power-on hours.
Deserving in a thread of itself, and I wouldn’t be surprised if it’s been covered already. This provides quite a useful individual context.
I should have mentioned in my previous post that the 2% increase in failure rate was after 3 years.
Really, all in all, it comes down to choice. Occasionally I will leave my computer on all night. It will be doing virus scans, defragging, backing up etc. and I don’t want to sit there and watch it till it is done so I can go to bed. Usually if I am not using it for a few hours though, it is off. Personal choice is what it all boils down to. How you will decide will be your own in the end.
How to tell is easy: do the experiment.
My mother’s power bill had a savings of several € (I don’t remember the exact amount but between 10 and 20€ per month) when she started switching off the TV and its attendant machinery completely, unplugging chargers when not in use and switching off the computer’s power bar.
When your income is below 1000€/month, four more cups of coffee can mean a lot.