No it isn’t with modern computers. Modern computers are like TVs , DVD players etc - shutting them down places them in standby mode where they can be easily turned back on.It doesn’t actually turn them off.
A TV in standby mode can be turned on from the remote control whereas one that is turned off cannot.
A computer in standby mode can be turned on in a number of different ways depending on the BIOS settings - button on the front, modem or lan signal. One computer I have that gets shut down but not actually turned off is set to turn on when a mouse button is clicked. When the computer actually gets turned off it cannot be turned back on that way.
The difference in power usage between a computer in standby mode and one that is right off is about the same as between one that is in hibernate mode compared to turning it right off - since the only difference is what gets saved to disk first.
the switch to completely turn off a computer is a two way switch, not a push button one.
The estimates on the amount of power people waste by leaving their tv and computer in standby mode rather than physically turning it off at the switch (and on a computer that switch if there is one is on the power supply) is quite large.
It doesn’t matter whether the switch is at the wall, on a powerboard, on on the power supply at the back of the computer. All of them do the same job of actually turning the computer right off.
Shutting down or using hibernate place the computer into a standby mode where it uses the minimum of power until powered back up using one of a variety of means - modem, lan, mouse, etc - or the push button on the front of the computer (which always works to wake the computer back up even if it has been physically turned off and back on - which would disable the other ways to wake it up.
I’m never off the computer for longer than the 8 hours or so that I sleep. So I leave the hard drive on and let it do routine maintenance stuff (scan, defrag). But I turn off the peripherals.
I’ve heard that it takes more electricity to restart a computer than it does to leave it on. Although I imagine this is comparible to “florescent lights use more to turn on than to leave on”.
As far as lights, it’s true if the light is going to be off for less than roughly 10 minutes (i.e. startup uses the amount of electricity required to run 10 minutes). But I have no idea what the figures would be like for computers. Hopefully more modern computers are more energy efficient, but who knows.
Actually guys I have a surge protector extension lead plugged into another extension lead and i just unplug that from the other and thats it done! So it’s just one plug im unplugging if you like
For lights the figure is actually closer to 10 seconds rather than 10 minutes.
With modern computers the various components can get shut down when not in use so you can have the hard drive and even the processor turn themselves off when they aren’t actually needed. That would allow you to keep the power usage down even without actually turning the computer off. With all the components set to turn themselves off when not required you can effectively have the computer put itself into sleep mode so many minutes after it finishes off all the tasks it needs to do. While sleep mode still uses more power than shutdown/hibernate mode does since it still needs to keep power to the memory that is still fairly minimal compared to leaving everything running at full power.
There is less power usage through actually turning a computer off and back on as compared to shutdown/sleep since shutdown/sleep mode still has minimal power running so as to be able to tell when you want to start using the computer. The extra power required when you start using a computer from off/shutdown/hibernate compared to sleep mode would be that required to boot the computer and either load the operating system ready to start or to reload the daa saved when going into hibernate mode. With hibernate mode you’d also have the extra power that was required to save the current state in the first place but that ought to be less in total than would be required to close and reopen all the applications.
I hope you aren’t using “unplug” as an off switch. The computer should do various “clean up” tasks before powering down and you’re likely to end up with file fragments and incomplete registry changes. You can probably get away with this for the most part but eventually it will bite you. I had to do a complete re-install (only once - so far) after a hang forced me to kill the power that way.
@felgall 10 seconds? Well that throws the leave them on to save energy argument out the window. Hmmm, maybe leave them on to extend lifetime would be an acceptable argument?
You should always power down or hibernate a computer BEFORE turning it off at the on/off switch or unplugging it.
Exactly. and that’s for the ones that have a high energy use when first starting. Most modern ones will have saved energy if off for even one second.
There is of course still the argument that leaving them on extends their life since most bulbs blow due to the power surge when turning them on. Some of Edisons original bulbs which have been left on since they were first installed (such as in his house in the USA and some in use in caves to the west of Sydney here in Australia) and will probably continue in use for many years or centuries to come as long as no one turns them off. That then comes down to how much power they use by being left on unnecessarily compared to the cost of replacing them. In the case of the bulbs in use in the caves the cost of replacing them would be very high due to the inaccessibility of where they are located and the risk to the life of the person getting to that location to be able to change the bulb.
This may be true of older computers. But for modern computers, where the power-efficiency of booting up and shutting down has been improved, this may not longer be true.
Any Energy Star certified computer with Wake-on-LAN enabled is required to use less than 1.7 watts when shut down or in standby. With Wake-on-LAN disabled, it’s required to use less than 1 watt in standby.
That reasoning is junk science. Power cycling does virtually nothing destructive to light bulbs. Destructive power cycling is directly traceable to wild speculation not tempered by knowledge of how light bulbs degrade and not tempered by well proven numbers.
For example, how to make a light bulb last longer. A conventional 500 hour light bulb at full voltage will last 4,100,000 hours if operating at half voltage. IOW maybe 400 years.
If power cycling was destructive, then the first light bulb to fail is that flashing orange light in stop signals. Reality. That bulb that power cycles most often also has the least number of operational hours. Therefore it lasts longest.
Is power cycling harmful to computers? Yes. And then we apply numbers. The lowest power cycle number I even saw was on an IBM disk drive - 20,000 cycles. That means power cycling the computer seven times every day (including weekends and holidays) for … almost 8 years. IOW, nobody cares. Most other components are rated for twice or five times that number.
So yes, power cycling is destructive - as long as one is making claims without numbers. Once we add the numbers, then power cycling does not matter. Turn it off when done.
BTW, hibernating is not the same as sleep. Sleep leaves power applied. Hibernate turns power off.
It’s really more of a scale. None of the ACPI states (hibernate/sleep/standby/shutdown) correspond to the machine using no power. Even in hibernate or shutdown, there can be up to a watt powering the USB and network controllers for wake-on-USB and wake-on-LAN functions. In sleep, most components are powered down other than the CPU and RAM.
You are talking garbage with that comment as the evidence I have accumulated provides scientific proof of what I said.
Why is it (if power cycling does nothing destructive as you claim) that bulbs always fail immediately when being turned on and never fail during the time that they are on. Every one of the thousands of light bulbs that I have seen fail has failed while someone’s finger was still on the on button having just tried to turn the light on. There is a flash of light as it comes on and the filament burns out.
With no examples of bulbs that have failed while actually on I think that is a sufficiently large sample to prove what I said. The number of bulbs that fail while on is too small a percentage for me to have ever come across one.
Most situations where they want the lights to be able to be turned on and off and not burn out now use LEDs rather than bulbs in order to avoid the problem of the initial power surge eventuually burning out the filament in a conventional bulb.
THIS is junk science. That the bulb fails when powered on is not evidence that powering on the bulb is what shortens the life of the bulb.
What causes a bulb to burn out is the evaporation of tungsten from the filament. This evaporation occurs due to heat, so is greatest after the bulb has warmed up and is at full temperature. The evaporating filament causes it to become thinner and thinner over time.
It was evaporation at full temperature, last time you had the bulb on, that cause it to fail. It finally became too thin to be able to stand a surge of current when cold. Then when you next tried to turn it on, it broke. It’s not the turning it on that wore out the bulb, it was the evaporation over time when it was operating.
This is why the lifetime of a bulb is rated in operating hours, not power cycles.
What we call ‘no power’ is how the machine sits when completely turned off. Only hibernate duplicates that ‘near zero’ power consumption.
Moving on to felgall’s post. Your criticism is quite restrained. Initial power surge doing bulb failure is junk science. Did he first learn how things work using science? Or just speculate from observation? Always required homework means first discovering well published equations - understood generations ago - that define what causes bulb failure.
Observation alone is junk science. Always. And especially when a conclusion is without numbers. Unfortunately so many so hate science and logic as to automatically know from observation, subjective reasoning, and what others ‘ordered’ them to believe. Same logic that proved Saddam’s WMDs. We are expected to learn from history – ie why Saddam’s WMDs existed due to junk science reasoning.
LEDs have longer life expectancy because of how LEDs operate. If power surge is destroying light bulbs, then power surge is also destroying LEDs - for the same reason.
Let’s move on to transistors. When a transistor switches, switching is so violent as to emit a burst of infrared light. Therefore switching transistors have shorter life expectancy due to violent and destructive switching. Using that wild speculation based only on observation, then leaving a computer running (even when not doing anything) severely shortens computer life expectancy. Another conclusion based in speculation – because all facts came from observation. Conclusion made without numbers and not tempered by how things work.
Light bulb life expectancy is well understood. Voltage (filament temperature) and hours of operation are the relevant parameters. If power cycling was destructive, then explain why orange traffic light bulbs last so much longer? Why do numbers contradict observation? Because numbers were ignored – used classic junk science reasoning. It scares me why a majority ignore (violate) what was taught in junior high science to conclude using only observation - as felgall has.
felgall - this is not personal. This is an example of why so many are so easily deceived by overt propaganda - do not ask damning questions - even ignore what was taught in junior high science so that lies and myths can be exposed. Observation not tempered by fundamental facts, well proven concept, and numbers (ie what is necessary to construct a valid hypothesis) is classic junk science. Too many forget. No numbers is a first indication of junk science reasoning - which is why so many assume leaving a computer on is better.
To get back to computers specifically (I doubt many have tungsten filaments), AFAIK the major enemy of life expectency is excessive heat i.e compromized ventillation. True it’ll be hotter running than when off, but sinks and fans should handle that no? As there’s so many parts, is there any data as to what component is most likely to fail first? Resistors? A mechanical part as opposed to an electrical?
and i just unplug that from the other and thats it done! So it’s just one plug im unplugging if you like 