Why is shutting down your computer with the actual power button worse than the software based one...

why is shutting down your computer with the actual power button worse than the software based one? is this just an urban legend?

Attached: pc-power-button-960x540.jpg (960x540, 105K)

Other urls found in this thread:

sqlite.org/howtocorrupt.html#_disk_drives_that_do_not_honor_sync_requests
twitter.com/NSFWRedditVideo

Data Corruption/loss is a big reason why you use software button. Software - gives os time to write all data to disk/stop network services from using the disk/data. "says, ok it's now safe to shutoff now". Power button - there is no grace period, its off no mater if the drive is in use or not.

When you shut things down in software then the software knows that it's shutting down and can stall until its ready to shut down. Say the OS is in the middle of re-creating an important cache file after storing its progress in memory. You hard stop its progress in some unknown state and when it wakes up it has this half-written file that it has no idea what to do with. Does it get left there? Does it see the file exists and think it's complete and then throw an error in the middle of an operation? Who knows.

And straight up cutting the power to a hard drive probably isn't the most healthy thing to do for its longevity.

Attached: 1504200077221.png (743x895, 1.21M)

Urban myth, I've yet to have an issue with holding the power button down in my 25 years of using computers.

This. Even back before power buttons (ala Windows 3.1), you still had to close your program manager, at which point the screen would go black and tell you "It is now safe to power off your computer."

There's 0% chance you've never had random software problems in 25 years of using a computer period and there's 0% way to know if the problem was caused by improper shut-down or just from bad logic or other reasons that data might get corrupted. I have no idea why you would even make that boast other than to display your own ignorance.

It depends on how your hardware handles the power button. Some give it to the operating system, only forcing it off if held for ~10 seconds, letting the OS shut down properly. Others just power off the computer without letting the OS sync its files and whatnot, causing corruption and other issues.

This was more an issue with old storage writing methods, where you could risk data corruption during such process. Nowadays, there are failsafes in place that make it a lot less likely to happen.

Pressing once? No. Holding down till the power is cut? Yes. If a file is being written (and they always are) then that file can be damaged.

Data corruption was still a thing with HDDs, because they need to spin down in order to write all of the data down.
With SSDs this problem is long gone because no moving parts.

Software button for modern people is a relict of the past, I haven't had any problems flipping the switch on the power outlet since I've gone full samsung SSD.

Why do the hardware button at all, then? Why not just rip out the power cord?

What do I know? Ask Tim Cook & Co. Maybe because wireless charging is coming up and ripping out a cord isn't possible anymore and the designers thought of this?

Wrong. Problem described by can still occur, no matter what kind of hardware you're using.

well then you have to plug it back in retard

If you run any raid array you'll quickly get hit by the infamous "Rebuild" message upon restart if you use the power button. Due to the controller not having enough time to properly write all data to disk/tell the os "ok I'm done, you can turn off now" Depending on array size a rebuild can take days to complete not to mention data getting fucked up. So all the more reason to use the software button

Because it's true?

Only if you're handling sensitive hardware. I had million dollar computers fault just because of cold shutdowns and boots. General consumer hardware, however, are made for Redditors aka retards, so you should be fine with handling it either way, although you should be weary of your harddrives if you do force-shutdowns

Well I have had problems, so I guess it's 50/50 now

ALL of this is very true. all programs should be put into a safe shutdown option through a program. Or the actual hardwire shutdown should be mapped to the software based shutdown and the manual should be to a holed down for 5 seconds and forced should be 10 seconds.

Attached: a9-5.jpg (1512x1512, 749K)

this is actually a good idea

...
It said that because those early computer power supplies couldn't be turned off through motherboard. It was pretty much the same as Raspberry Pi is nowadays, you can't tell your phone charger to turn itself off.

Not if you had a proper computer. You always powered the Amiga off by just pressing the button, instant off and on,

Attached: 1274950338.jpg (400x533, 48K)

fpbp
/thread

I thought it's a lasagna from the thumbnail.

>Power button - there is no grace period, its off no mater if the drive is in use or not.
>Or the actual hardwire shutdown should be mapped to the software based shutdown

This has been a thing for like 20 years.

Attached: Windows_10_Define_Power_Button_3_(1).png (502x400, 100K)

When you press the button, the OS gets a signal to shut down. If you just press it and wait it's perfectly safe.
Holding it down causes a hard power cutoff, and that's usually not a good idea. Only reserved for emergencies.

At least someone with sense

you're moving the goalposts
sure, you can redefine the function, but we're talking about strictly shutting the PC down by using hardware button

In modern computers where the front power switch merely sends a signal to the motherboard that the user shut it down, there's no harm in using the front power button to shut off the computer, it's the same as selecting it through the software menu.

The only exception being if you hold down the power button for several seconds until the computer powers off. That's a hard shutdown, you don't properly make sure everything is wound down and committed to the disk, etc., so you want to avoid hard shutdowns unless the operating system is totally unresponsive.

>Others just power off the computer without letting the OS sync its files and whatnot, causing corruption and other issues.
This has not been the case for about 20 years.

Most of this is due to caching if everything has been written to disk most of the time there is no issue.

I don't know what (You)'re talking about m8, but it sure seems like the first 23 posters ITT had no idea the power button actually acts just like the software button. OP didn't say anything about holding the button down for 5 seconds.

I hate to be the contrarian asshole here (lol no i dont), but even with a software-based (ACPI) shutdown, many consumer storage devices (specifically SSDs) will lose and corrupt data depending on outstanding write cache buffers on the device and the timing of events. Its actually a really faggy problem to have to deal with when you are writing things like databases:
>sqlite.org/howtocorrupt.html#_disk_drives_that_do_not_honor_sync_requests

Being such a fucking idiot you don't realize momentarily pressing the power button is a software shutdown
Don't hold hold the button down you tremendous faggot

If losing cached contents was such a widespread problem with consumer SSDs, Windows would've had tons of problems.

Some OSes, Windows 2000 in particular, have a different type of software shutdown when the button is pressed, killing all user processes without warning and only shutting down the core properly.
Some OSes, specifically anything using the latest Gnome, cannot be shut down using the power button at all because the developers are tremendous faggots.

Attached: Capture.png (610x93, 9K)

Pressing and HOLDING to power off dude.
Maybe it should be phrased as ripping out the power cable form the PSU so you get it

Why should windows have problems? How likely is it that critical shit in system32 is being written to actively at the time you are trying to shutdown? There are safeguards at the OS level to deal specifically with things like this as well (as evidenced by the fact that most people dont see issues). Do not assume that because something just works really well its not because of decades of engineering and trial-and-error.

Because the SQLite article linked says that drives lie about writes being completed at the hardware level.

Right, and I specifically indicated that the OS itself has mitigations for these exact kinds of problems. Read the entire post and relevant context, please.

Then how the fuck is this problem even relevant to the thread?

This is the stupidest thing I've ever seen being discussed on Sup Forums, worse than which cap color is better for your mech keyboard, worse than homescreen threads and worse than those who actually argue with macfags. And you are actually talking about something so basic