‘I hit my laptop ’cause I got angry’

Topic

New Reply

LANGALIST By Fred Langa Here’s how to check your PC’s health after it’s suffered a physical impact — for any reason! Windows 10 has four built-in tool
[See the full post at: ‘I hit my laptop ’cause I got angry’]

3 users thanked author for this post.

Elly, abbodi86, woody

Reply | Quote

Replies

I’ve been looking after 120 or more Windows computers for about 18 years now. I have evolved a practice that tests drives as a routine part of my checkup procedures. There are two distinct types of tests that I do on drives: 1. the integrity of the data stored on the drive (chkdsk, sfc, etc) and 2. the integrity of the drive hardware.

I rarely see anyone doing physical drive testing. Drive manufacturers will not honour their warranty until the customer uses the manufacturers’ test software to prove a warranty claim. That software is good for testing drive hardware at any age. It is free. I use those tests routinely as part of my checkups. The two that are most always good are the following. WD for western digital and Seatools for Seagate. Most other brands honour one of these two.

“seatools for Windows” https://www.seagate.com/ca/en/support/downloads/seatools/
Data Lifeguard http://support.wdc.com/downloads.aspx?p=3

CT

2 users thanked author for this post.

Elly, rc primak

Reply | Quote

[rc primak]

Although I don’t do this out of anger, hitting a laptop or notebook PC (or banging it on a table corner, etc.) is yet another reason I opt for devices with SSD storage. Not completely bulletproof, but very stable in my experience.

-- rc primak

• This reply was modified 4 years, 10 months ago by rc primak.

Reply | Quote

I’ve read your article in AskWoody Plus Newsletter 17.21.0. Will this advice work for a SSD also?

Reply | Quote

A SSD does not have the sensitivity to impacts that a traditional hard drive does.  If it still works after the impact, you should not have any problems down the road.  For laptops, that’s one of the great advantages… SSDs are more durable by far, faster, use less power, and generate less heat.  For desktops, I use both together, SSD and HDD, but in laptops, I strongly prefer all SSD.

Dell XPS 13/9310, i5-1135G7/16GB, KDE Neon 6.2
XPG Xenia 15, i7-9750H/32GB & GTX1660ti, Kubuntu 24.04
Acer Swift Go 14, i5-1335U/16GB, Kubuntu 24.04 (and Win 11)

Reply | Quote

OK, but I am asking if Langa’s advice works for SSDs also.

Reply | Quote

I don’t really understand what you’re asking. His advice is all about what to do if your laptop has suffered an impact, something that a HDD is sensitive to, but a SSD is not.

If you’re asking if you can use the tools Fred Langa mentioned to check a SSD for errors… yes, you can do that, but whether you need to is another question.

In the example, there was an impact hard enough to cause the laptop to shut off, so some connection that isn’t soldered momentarily was lost, perhaps a momentary loss of contact between the battery connector and motherboard, and (just like turning off a computer forcibly when it’s not ready), that can cause soft errors on a SSD or HDD if there was a write going on at the moment.  That’s a damage to the data written, but not to the drive itself.  NTFS is a journaling file system, so generally, a standard chkdsk from the File Explorer would take care of that. The unexpected shutdown would set the “dirty bit” on the volumes that were in use at the time, and they should automatically be scanned and repaired (if necessary) at the next boot.  If not, for some reason, a manual scan (or scheduling it for the next boot) would work.

The way a hard drive can fail in that instance is different.  If there is a head crash, as the article mentions, it can physically scratch the platter surface and render that part of the hard disk permanently unusable (and since there are multiple heads and platters on some drives, there can be a number of head crashes on the same drive).  This is a hard error, something more than the journaling file system can detect or handle.  That’s why you do the surface scan… to look for, find, and mark such areas as damaged (so the drive never tries to use them again) to make the hard drive work normally once again.

You can do that scan on a SSD if you wish, but the reason for it (the possible head crash) can’t happen on a SSD, so there is really no need for it because an impact took place.

Dell XPS 13/9310, i5-1135G7/16GB, KDE Neon 6.2
XPG Xenia 15, i7-9750H/32GB & GTX1660ti, Kubuntu 24.04
Acer Swift Go 14, i5-1335U/16GB, Kubuntu 24.04 (and Win 11)

Reply | Quote

OK. Thanks for explaining what a head crash is and how SSDs are not susceptible to them.

This is my first SSD and I want to understand what I can do to be sure that it is working as it should be. I’ve heard that they wear out, if you do a lot of reading and writing, and do a lot of file deletions.

So, I was trying to find out whether I could extrapolate Langa’s advice for a SSD that has not been bashed around. [It’s entirely possible that I could have done something to cause a ‘soft error’]

Reply | Quote

WCHS wrote:

This is my first SSD and I want to understand what I can do to be sure that it is working as it should be. I’ve heard that they wear out, if you do a lot of reading and writing, and do a lot of file deletions.

Reading does not harm a SSD.  Writing will cause them to wear out in time, but that’s not something that most people would ever have to worry about.  The number of writes that a drive can tolerate is so high that the drive will probably be obsolete long before it reaches the wear-out limit.

The rated life remaining in a SSD is usually visible within the SMART data for that drive.  On my Samsung 840 Pro, attribute 177, “wear leveling count,” has a normalized value of 68 (this is one of the columns you will see if you get a SMART report for the drive).  That means that this particular drive has 68 percent of its rated life left.  That drive is five years old now, and I’ve subjected it to a level of use (or abuse) that most users never would.  Even so, a drive that old is only a third done with its rated life.  But even then, the rated life doesn’t mean the end!

A tech site called TechReport did a torture test to see how long SSDs really last.  They had PCs erase and write to various SSDs continuously until they all died.

The drives typically went way beyond their rated lives.  Their Samsung 840 Pro, a 256 GB model (mine is the 128GB), went 2 petabytes and change before it finally died. That’s 2,000 terabytes!

My 840 Pro is half the size of the one TechReports tested, so it has half the NAND cells. If it is equally as durable as the drive TR tested, I would expect it to go 1 petabyte, or 1,000 terabytes.  Its rated life, the life that it has 67% remaining, is 250 terabytes written. At the point that Samsung says the drive is worn out, it’s only a quarter of the way to truly being dead, if TR’s numbers hold.

By this standard, my drive would now have almost 92% of its remaining writes left, on a drive that has been used hard for 5 years. If I kept using it that hard, it would take 57.5 years to reach a petabyte of writes.

In reality, I do not expect it to last 58 more years, as by that time, I would expect some kind of failure that is not a result of the NAND cells being worn out.

the 840 Pro was the longest-lasting drive in TR’s test, but that was years ago, and newer drives have even longer rated life spans than those TR tested.  The lower-end Samsung Evo drives that are currently offered have longer rated lives than my 840 Pro.

How those rated lives translate to actual time before drive death is something we can only know if someone were to test the newer models to the point of failure.  Even then, it is likely that the numbers will vary quite a bit even for two supposedly identical drives. It’s reasonable to conclude, though, that it will be considerably longer than the rated life, as the manufacturers of the drives and NAND cells still have the same reasons for rating their products the way they do, below the absolute limits of endurance.

The lone exception to the “last even longer than rated” thing that I am aware of is with Intel drives (not sure if it is all of them or just some models, or if Intel even does this with new drives anymore), which will immediately go read-only when the rated life is gone, and they will brick themselves the next time they are shut down after that.

I think that’s absurd… all the data on the drive will be lost if you don’t rescue it in that narrow window between reaching the limit and the next shutdown.  Of course, I would hope any important data is backed up anyway, but that’s no reason to just shut it all down.  At the very least, they could have it be permanently in read-only mode from then on.  That allows the data to be rescued, but still pushes the user toward buying a new drive.  A SSD that is read-only isn’t very useful once the data on it has been rescued!

Dell XPS 13/9310, i5-1135G7/16GB, KDE Neon 6.2
XPG Xenia 15, i7-9750H/32GB & GTX1660ti, Kubuntu 24.04
Acer Swift Go 14, i5-1335U/16GB, Kubuntu 24.04 (and Win 11)

1 user thanked author for this post.

Reply | Quote

Ascaris wrote:

The rated life remaining in a SSD is usually visible within the SMART data for that drive. On my Samsung 840 Pro, attribute 177, “wear leveling count,” has a normalized value of 68 (this is one of the columns you will see if you get a SMART report for the drive).

Can you tell me how you access the SMART data for a SSD drive? And where I would look for attribute 177?

I am using Win10/Pro version 1909. Laptop is a Dell-Inspiron

Reply | Quote

See this article for SMART monitoring software.

https://www.askwoody.com/forums/topic/from-the-lounge-simple-and-cheap-data-backup-and-storage/

cheers, Paul

Reply | Quote

I wrote a reply to this post yesterday, but the HTTPS connection failed, with Waterfox telling me that the problem might be that I increased the minimum TLS level to 1.2, and offered to reduce it for me, which I refused.  I have been using these settings with AskWoody.com for a while, so I decided to try again later, and then forgot.  So, here it is, belatedly.

WCHS wrote:

Ascaris wrote:

The rated life remaining in a SSD is usually visible within the SMART data for that drive. On my Samsung 840 Pro, attribute 177, “wear leveling count,” has a normalized value of 68 (this is one of the columns you will see if you get a SMART report for the drive).

Can you tell me how you access the SMART data for a SSD drive? And where I would look for attribute 177?

I am using Win10/Pro version 1909. Laptop is a Dell-Inspiron

With Windows, I used CrystalDiskInfo for viewing SMART data. It’s free, and widely regarded as a good program. Here’s a page that has a link to where to download it and how to use it to see the SMART data.

The attribute 177 will show in the list of statistics for the drive. That’s for Samsung drives, though… if you have another brand of drive, it could be a different attribute number or have a different name. Most of my SSDs are Samsung, so that’s what I am most familiar with.  If you can’t find it, please post a screenshot of the SMART data and one of us will tbe happy to assist in finding the right attribute.

If your SSD is a Samsung, you can also use the Samsung Magician application to monitor drive health, show the SMART data, update the firmware, and other similar things. Most other SSD manufacturers have a similar application too.

Dell XPS 13/9310, i5-1135G7/16GB, KDE Neon 6.2
XPG Xenia 15, i7-9750H/32GB & GTX1660ti, Kubuntu 24.04
Acer Swift Go 14, i5-1335U/16GB, Kubuntu 24.04 (and Win 11)

Reply | Quote

WCHS wrote:

I’ve heard that they wear out

See this thread for more details on SSD life.

cheers, Paul

1 user thanked author for this post.

Reply | Quote