Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Do you leave your iMac on 24/7? or turn off at night
- Thread starter gagaliya
- Start date
- Sort by reaction score
Thanks for taking the trouble to respond. However it seems the problem is a
bug in late 2012 iMacs. 21" have a fix, 27" don't yet, see:
https://forums.macrumors.com/threads/1509682/
I leave it on 24/7. No sleep. I do put the display to sleep when I walk away though. I set up the top left corner as a hot corner that sleeps the display. So I just swipe the mouse pointer to that corner as I get up and it goes off
Same here but bottom right!
----------
Not if you have a Fusion Drive. The spinning part will only be used when information is needed from it (well thats my understanding anyway!)
Oh right, I can't say I've noticed a problem since I've changed the sleep to 1 hour, never had to enter a password without expecting it so far
I always turn it off when I am done using it at night. I usually only get on the computer for a couple of hours each night. If it's during the day and I'll be coming back to it later, I will let the computer go to sleep and wake it up when I need to get back online.
Walk away and it dozes off after a while.
Tap a key and it's on.
No effort is a lot easier than a little effort.
The electrical use is negligible when the screen is off.
Thermal & electrical shock of complete power off/on cycles is the greater cause of premature aging of electronics.
As a simplistic analogy: the simple act of turning on a light bulb uses more power than leaving that bulb on for two hours. Considering how often conscientious (and ignorant) "keep it off when not in use" types turn lights off & on, that's wasting a whole lot more power - and burning bulbs out a lot sooner - than "wasteful" people who leave lights on most of the time.
Ditto computers, just on a different & more complicated scale.
Not to derail the thread, but I call your bluff. Prove it.
There is no way a 60W light bulb uses 120W when you first turn it on for it to start producing light. What you are saying is full of $!@# and I ask that you enlighten us with a credible source proving your point.
Actually, you are both somewhat right and somewhat misinformed.
The discussion seems to be focused on power, which is where incorrect conclusions are reached. Power is an instantaneous parameter that, in the case of an incandescent light bulb, is indeed higher when it it first turned on due to the time it takes for the filament to come up to operating temperature. But this time is only a fraction of a second. You are not billed for instantaneous power, but rather for energy used. Energy is power times the amount of time it exists. That is why your electric bill is for the number of KW-Hrs (an energy unit) rather than for KW (a power unit).
Using the 60 W lightbulb as an example, the power when first turned on might be say 200 W but only for perhaps 0.3 seconds and then the steady state 60 W lasts for the duration of the time it is operating. If you pay 10 cents per KW-Hr for electricity, the cost for the turn-on high power operation would be 0.002 cents, and the cost for operating the bulb for each hour thereafter would be 0.6 cents. Thus, the actual utility cost of the turn-on transient is indeed inconsequential. (This is a simplified example for ease of illustration. In actuality the turn-on transient will have an exponential waveform which must be integrated to calculate the energy. But the result will be similar.)
So, from a purely utility billing point-of-view, the cost is minimized by turning off the bulb when not in use.
However, this cost is not the only criterion in the case of light bulbs. You may have noticed that lots of times an incandescent bulb will fail immediately when it is turned on rather than later. This is because the turn-on condition with the high instantaneous power is indeed a very high stress condition for the filament of the bulb, and is perhaps the primary failure mode for incandescent bulbs. So, from a reliability viewpoint, failures can indeed be reduced by minimizing the number of times the light is turned on and off.
Now, lets turn our attention to computers rather than lightbulbs. The failure mode discussed above does not exist with a well designed computer power supply - and the power supply in your Mac is indeed a very good one. There are no filaments to burn out! The Mac power supply should handle tens of thousands of turn-on cycles without any appreciable reliability reduction. So, in my opinion, how you decide to operate your Mac in terms of keeping it on or turning it on and off should be decided by your own assessment of operating cost vs. convenience and not by any concern about reliability.
For my own iMac, I have it set to turn off the display after 1 hour (reduces power consumption by about half) and to go to sleep after 3 hours (very substantial power reduction).
The discussion seems to be focused on power, which is where incorrect conclusions are reached. Power is an instantaneous parameter that, in the case of an incandescent light bulb, is indeed higher when it it first turned on due to the time it takes for the filament to come up to operating temperature. But this time is only a fraction of a second. You are not billed for instantaneous power, but rather for energy used. Energy is power times the amount of time it exists. That is why your electric bill is for the number of KW-Hrs (an energy unit) rather than for KW (a power unit).
Using the 60 W lightbulb as an example, the power when first turned on might be say 200 W but only for perhaps 0.3 seconds and then the steady state 60 W lasts for the duration of the time it is operating. If you pay 10 cents per KW-Hr for electricity, the cost for the turn-on high power operation would be 0.002 cents, and the cost for operating the bulb for each hour thereafter would be 0.6 cents. Thus, the actual utility cost of the turn-on transient is indeed inconsequential. (This is a simplified example for ease of illustration. In actuality the turn-on transient will have an exponential waveform which must be integrated to calculate the energy. But the result will be similar.)
So, from a purely utility billing point-of-view, the cost is minimized by turning off the bulb when not in use.
However, this cost is not the only criterion in the case of light bulbs. You may have noticed that lots of times an incandescent bulb will fail immediately when it is turned on rather than later. This is because the turn-on condition with the high instantaneous power is indeed a very high stress condition for the filament of the bulb, and is perhaps the primary failure mode for incandescent bulbs. So, from a reliability viewpoint, failures can indeed be reduced by minimizing the number of times the light is turned on and off.
Now, lets turn our attention to computers rather than lightbulbs. The failure mode discussed above does not exist with a well designed computer power supply - and the power supply in your Mac is indeed a very good one. There are no filaments to burn out! The Mac power supply should handle tens of thousands of turn-on cycles without any appreciable reliability reduction. So, in my opinion, how you decide to operate your Mac in terms of keeping it on or turning it on and off should be decided by your own assessment of operating cost vs. convenience and not by any concern about reliability.
For my own iMac, I have it set to turn off the display after 1 hour (reduces power consumption by about half) and to go to sleep after 3 hours (very substantial power reduction).
Me too. I shut my iMac down more often with the Fusion drive. No point in it being a power vampire when not being used.
With that said, I do let it sleep some of the day when I'm at home then shut it down at night.
You don't know how an incandescent bulb turns on and subsequently functions, do you?
When it's off, it's cold. The filament is just a wire shorting + to -.
When you turn the switch on, there is no electrical resistance. You've just shorted 110V of darned near infinite current source to ground. That's a LOT of power running thru the bulb. That initial power surge is huge.
With all that power running thru it, it heats up. As it heats, resistance increases - the filament gets hot, and starts glowing, which is how the the device makes light. After a few milliseconds the current flow and resistance balance out and your light is on.
So...when you turn the light on, the process of heating the filament up enough to glow uses a lot of power. Not for long, but a lot.
A simple experiment evaluating this is here: http://physicsed.buffalostate.edu/pubs/TPT/TPTDec99Filament.pdf
Conclusion/measurement is that at turn-on a 60W bulb draws about 7A. Given power equation P=IV, that means the light uses 840W turning on.
How long this takes, of course, is the real question. colodane takes a good stab at this. Factor in wear of repeated cycling, bulb cost, burnout frequency, AC power complications, etc. and you'll work out what it costs.
I suggest you not insist someone "is full of $!@#" when you don't know how something as simple as a light bulb works.
Wake quickly when needed.
I used to have it awake all the time with the display sleeping.
My goal was to be able to access it quickly at any time, even remotely.
I have it set to sleep on inactivity and wake on USB or LAN. Goal accomplished.
Turns out by having a web server on it, it sleeps very little. It is constantly searched by robots/spiders/crawlers (e.g. google, yahoo) and attacked by hackers. And each time it wakes.
Same is true when I turned on SSH (remote login). I get constant port 22 attacks which wakes the machine. Turned that off and turned on Back-to-My-Mac, which doesn't have the same problem.
Waking on LAN used to be a bother because it also waked the display.
I used to have it awake all the time with the display sleeping.
My goal was to be able to access it quickly at any time, even remotely.
I have it set to sleep on inactivity and wake on USB or LAN. Goal accomplished.
Turns out by having a web server on it, it sleeps very little. It is constantly searched by robots/spiders/crawlers (e.g. google, yahoo) and attacked by hackers. And each time it wakes.
Same is true when I turned on SSH (remote login). I get constant port 22 attacks which wakes the machine. Turned that off and turned on Back-to-My-Mac, which doesn't have the same problem.
Waking on LAN used to be a bother because it also waked the display.
Yes, I do understand how a filament works. I do realize more power is used when it first burns than when it is already burning, but you are claiming that it uses more power in the first few seconds of lighting up than it does by leaving it on for hours. That is simply not true, and you my friend are the one who is uninformed.
The way I look at it:
I shut the display on my rMBP, and it goes to sleep... so: I'll sleep my iMac too.
I shut the display on my rMBP, and it goes to sleep... so: I'll sleep my iMac too.
Not that we follow 'the book' because most of you rip apart your machines and add got knows what to them. But it actually says in the manual that comes with your iMac that you should really only shut it down if you are going to be away from it for more than a couple of days.
. But it actually says in the manual that comes with your iMac that you should really only shut it down if you are going to be away from it for more than a couple of days.