Whats the status of your SSD?

By AimLXJ
in Hardware Hangout

 Share

Recommended Posts

Grand Master

Elliot B.

Posted
Posted

An honest question is why does this thread exists? Is they any proof that the numbers reported by SMART and this tool are reliable?

I certainly hope not!

Bought my Kingston SNV425-S2 64GB SSD in September 2010.

January 2011 (when I originally posted the following results in this topic)

SSD was 4 months old

Health was 85%

Work time: 2 months, 6 days

Powered on: 229 times

September 2011

SSD is 12 months old

Health is 62%

Work time: 6 months, 1 day

Powered on: 578 times

"Your drive health is in good condition and according to current use, estimated lifetime is August 2020."

October 2011

SSD is 13 months old

Health is 59%

Work time: 6 months, 15 days

Powered on: 646 times

"Your drive health is in good condition and according to current use, estimated lifetime is April 2020."

Mentor

coth

Posted
Posted

mine been 70% out of box with 0 hours

and it's still 70% (obviously this app do not support this samsung ssd)

http://online.hddlif...fa150a940922a98

but those unknown samsung smart parameters

EB

March 2011 (194 hours) : 7

September 2011 (5254 hours): 9

EC

March 2011 (194 hours) : 2

September 2011 (5254 hours): 55

ED

March 2011 (194 hours) : 15

September 2011 (5254 hours): 207

EE

March 2011 (194 hours) : 1196

September 2011 (5254 hours): 1196

Grand Master

Astra.Xtreme

Posted
Posted

I am sure the attraction to SSDs isn't their reliability, but their speed. I am not questioning the attraction to SSDs, but merely this topic. The point of posting reliability estimates would be to challenge those who see SSDs as less reliable than HDDs, otherwise why do it?

I have had more flash media die on me than anything other than Floppy Disks in my 13 years using PCs. This makes me wonder about the reliability of SSDs. As a result, looking at this thread aims to counter that, but offers nothing in the way of proof. I can find many people who are still running HDDs from over 5 years ago. A matter of fact, I have ones nearing 10 years old in use right now and they are still spinning fine. That doesn't mean HDDs never die. But if I told you my HDDs was going to last me another "3 years, 2 months, and 9 days" I should be showing why that number is so accurate and isn't just pulled out of someone backside.

Again, this isn't an attack on SSDs, but an inquiring as to what the point of posting the "status" of your SSD if the status is hogwash.

You make it sound like this program is randomly generating a date. You do realize that all drives (HHD or SSD) have a read/write rating of some sort, right? The manufacturers do actually test their product before they throw it out on the market, so they know exactly the amount of usage that an average user should get. So would you like to take a guess at how this program calculates that date?

Like I said before, both sides will obviously have failures. Shock, static charge, voltage spikes, temperature, etc. would all play a factor in the life of a drive. I don't have any proof to back this, but maybe SSD's are more commonly used in laptops since the majority of them come in the 2.5" form factor. If there's any truth to that, then obviously it would be prone to more outside forces. But if you put a SSD and HDD next to each other, in the same environment, and read/write the hell out of it, I'm willing to bet the SSD wins.

Grand Master

+LogicalApex MVC

Posted
  • MVC
Posted

You make it sound like this program is randomly generating a date. You do realize that all drives (HHD or SSD) have a read/write rating of some sort, right? The manufacturers do actually test their product before they throw it out on the market, so they know exactly the amount of usage that an average user should get. So would you like to take a guess at how this program calculates that date?

Like I said before, both sides will obviously have failures. Shock, static charge, voltage spikes, temperature, etc. would all play a factor in the life of a drive. I don't have any proof to back this, but maybe SSD's are more commonly used in laptops since the majority of them come in the 2.5" form factor. If there's any truth to that, then obviously it would be prone to more outside forces. But if you put a SSD and HDD next to each other, in the same environment, and read/write the hell out of it, I'm willing to bet the SSD wins.

It seems the program does pull its number out of thin air, so to speak.

Thanks to the link in Coth's post I was able to get a little insight into how the program generates its life date:

It's a simple mathematical problem: in order to calculate it at once, we need to know at least the date when you wrote data to the drive the first time, but unfortunately, drives do not provide this information. That is why we need some time after the first launch of SSDLife to monitor how intensively you use your SSD in order to determine its average load. And as soon as we can calculate the approximate average amount of data written to the drive per day, we will be able to determine the time it will take you to use the entire lifetime capacity of the drive, which means we will be able to calculate the date when the lifetime of the SSD will be over. Of course, this date will keep changing depending on how the intensity of the drive usage changes.

Source: http://ssd-life.com/eng/how.html

So the program seems to be of dubious value. It is only able to guess that your drive will last x days based on how many writes it is supposed to handle and how often it thinks you write to the drive. It can't read real wear on the drive or report any underlying problems with the drive that aren't detected by SMART. And we all know how useful SMART is...

Grand Master

Astra.Xtreme

Posted
Posted

It seems the program does pull its number out of thin air, so to speak.

Thanks to the link in Coth's post I was able to get a little insight into how the program generates its life date:

Source: http://ssd-life.com/eng/how.html

So the program seems to be of dubious value. It is only able to guess that your drive will last x days based on how many writes it is supposed to handle and how often it thinks you write to the drive. It can't read real wear on the drive or report any underlying problems with the drive that aren't detected by SMART. And we all know how useful SMART is...

Did you not read my post?... Thanks for basically re-emphasizing what I said...

And no, it does not pull the date out of thin air. It's an educated prediction just like any other prediction/forecast in the world. Did you honestly think that this program could magically predict the exact day your drive will die?

A drive is rated for a certain amount of read/writes. It will keep record of your usage over time and then that program will build a prediction using it's algorithm. It's common sense that it will change over time since sometimes it will be running for a month straight and sometimes it will be off for a week. The program is still correct for the usage you've put on thus far. It's not like a prediction is a bad thing...

In the end, there have been plenty of people on here that have used their SSD's for a long time and the program still says they have years and years of life left. If somebody only has a few hours put on it, then yeah it's not exactly valid, but that's definitely not the case here. The proof is all contained in this thread. You can believe what you want, but saying it's wrong is just non-sense.

Grand Master

+LogicalApex MVC

Posted
  • MVC
Posted

Did you not read my post?... Thanks for basically re-emphasizing what I said...

And no, it does not pull the date out of thin air. It's an educated prediction just like any other prediction/forecast in the world. Did you honestly think that this program could magically predict the exact day your drive will die?

A drive is rated for a certain amount of read/writes. It will keep record of your usage over time and then that program will build a prediction using it's algorithm. It's common sense that it will change over time since sometimes it will be running for a month straight and sometimes it will be off for a week. The program is still correct for the usage you've put on thus far. It's not like a prediction is a bad thing...

In the end, there have been plenty of people on here that have used their SSD's for a long time and the program still says they have years and years of life left. If somebody only has a few hours put on it, then yeah it's not exactly valid, but that's definitely not the case here. The proof is all contained in this thread. You can believe what you want, but saying it's wrong is just non-sense.

It is a very poor guess* as the program needs to run from Day 1 until the end to give a decent estimate, but even then it isn't guessing on anything more than estimated write capability from the drive maker. So it is really guessing how many estimated writes your drive has left not its "health".

That was my issue. If you look at the screenshots it implies that this product is able to read deeply into the SSD health and give an accurate estimate of how long it would last. That isn't the case. For even the one area it does monitor, wear level, there is no proof that their algorithm is even close to accurate.

* Poor guess as there is a lot of write activity it could miss due to formats, dual booting, virtual machines using hardware I/O virtualization, or any other scenario where this program can't record even write.

Grand Master

Astra.Xtreme

Posted
Posted

It is a very poor guess* as the program needs to run from Day 1 until the end to give a decent estimate, but even then it isn't guessing on anything more than estimated write capability from the drive maker. So it is really guessing how many estimated writes your drive has left not its "health".

That was my issue. If you look at the screenshots it implies that this product is able to read deeply into the SSD health and give an accurate estimate of how long it would last. That isn't the case. For even the one area it does monitor, wear level, there is no proof that their algorithm is even close to accurate.

* Poor guess as there is a lot of write activity it could miss due to formats, dual booting, virtual machines using hardware I/O virtualization, or any other scenario where this program can't record even write.

Why would the program need to run from day 1? I think you are really mistaken on how this program (and an SSD) works. The SSD itself is recording everything it does. All this program does is interpret that back-log and calculate a date for you.

There is no such thing as "wear" on an SSD. There are no mechanical parts, so the only thing that resembles "wear" is the completed read/write cycles.

That program truly is spot on since the SSD will die as soon as you use up those read/write cycles. That's just the physical nature of NAND memory. On the contrary, a mechanical HDD dies when the headers break or some other component fails. It's hard to predict the death of an HDD because it can still function as it goes "down hill". Whereas an SSD doesn't show symptoms; it just dies.

You should probably read their website to clear up your thoughts:

http://ssd-life.com/eng/how.html

Grand Master

+LogicalApex MVC

Posted
  • MVC
Posted

Why would the program need to run from day 1? I think you are really mistaken on how this program (and an SSD) works. The SSD itself is recording everything it does. All this program does is interpret that back-log and calculate a date for you.

There is no such thing as "wear" on an SSD. There are no mechanical parts, so the only thing that resembles "wear" is the completed read/write cycles.

That program truly is spot on since the SSD will die as soon as you use up those read/write cycles. That's just the physical nature of NAND memory. On the contrary, a mechanical HDD dies when the headers break or some other component fails. It's hard to predict the death of an HDD because it can still function as it goes "down hill". Whereas an SSD doesn't show symptoms; it just dies.

You should probably read their website to clear up your thoughts:

http://ssd-life.com/eng/how.html

I'm not sure I understand what you're saying...

Using the SSD (write cycles) causes wear on the drive. Every time you write to the drive you wear it out just a little more... They don't wear the same way mechanical HDDs do, but they wear nonetheless.

No, the program is not "spot on". I quoted their site earlier where they stated the drive only shows how many writes it has had at the moment they request it. The drive doesn't tell them when the first write occurred or how long it has been in use (so they have no way of knowing if the 45% "wear" on the drive occurred in 1 day or 10 years). They also admit that without this data their result is nothing more than an educated guess and can fluctuate as a result.

And as I said earlier, it only shows the drive's wear level really. It isn't representative of the drive's overall health as the screenshots imply.

Grand Master

Astra.Xtreme

Posted
Posted

I'm not sure I understand what you're saying...

Using the SSD (write cycles) causes wear on the drive. Every time you write to the drive you wear it out just a little more... They don't wear the same way mechanical HDDs do, but they wear nonetheless.

No, the program is not "spot on". I quoted their site earlier where they stated the drive only shows how many writes it has had at the moment they request it. The drive doesn't tell them when the first write occurred or how long it has been in use (so they have no way of knowing if the 45% "wear" on the drive occurred in 1 day or 10 years). They also admit that without this data their result is nothing more than an educated guess and can fluctuate as a result.

And as I said earlier, it only shows the drive's wear level really. It isn't representative of the drive's overall health as the screenshots imply.

Then how does the program display "Work Time"? I highly doubt that all these people have run this program before this thread was created. By your logic, everybody's work time would be 0 days in their screenshots. An SSD stores a log of what it does along with a time stamp. The program is only doing a simple calculation based on what it reads from the SSD. Nothing more, nothing less.

You still don't understand that with an SSD, health and wear level are the same thing. That is dictated by the read/write cycles, and that's it. It's very very simple.

You just need to look past your knowledge of how a mechanical HDD works, because the two are very different.

Grand Master

Astra.Xtreme

Posted
Posted

I just recently put an order in for an OCZ 120GB Vertex 3. Should be $180 after rebate.

Hope I don't regret it :/

I have that same one and it's blazing fast. I believe it's still the best rated SSD by a pretty large margin.

Be sure to update it to the latest firmware if possible. Sometimes it helps with stability and such.

  • 5 weeks later...
Grand Master

Elliot B.

Posted
Posted

Bought my Kingston SNV425-S2 64GB SSD in September 2010.

January 2011

SSD is 4 months old

Health is 85%

Work time: 2 months, 6 days

Powered on: 229 times

September 2011

SSD is 12 months old

Health is 62%

Work time: 6 months, 1 day

Powered on: 578 times

"Your drive health is in good condition and according to current use, estimated lifetime is August 2020."

October 2011

SSD is 13 months old

Health is 59%

Work time: 6 months, 15 days

Powered on: 646 times

"Your drive health is in good condition and according to current use, estimated lifetime is April 2020."

November 2011

SSD is 14 months old

Health is 56%

Work time: 6 months, 28 days

Powered on: 709 times

"Your drive health is in good condition and according to current use, estimated lifetime is June 2020."

Grand Master

primexx

Posted
Posted

Just noticed that the new Intel SSD toolbox has its own drive life estimate.

Bought my Kingston SNV425-S2 64GB SSD in September 2010.

January 2011

SSD is 4 months old

Health is 85%

Work time: 2 months, 6 days

Powered on: 229 times

September 2011

SSD is 12 months old

Health is 62%

Work time: 6 months, 1 day

Powered on: 578 times

"Your drive health is in good condition and according to current use, estimated lifetime is August 2020."

October 2011

SSD is 13 months old

Health is 59%

Work time: 6 months, 15 days

Powered on: 646 times

"Your drive health is in good condition and according to current use, estimated lifetime is April 2020."

November 2011

SSD is 14 months old

Health is 56%

Work time: 6 months, 28 days

Powered on: 709 times

"Your drive health is in good condition and according to current use, estimated lifetime is June 2020."

that seems like awfully quick wear =S

Proficient

Lord Zog

Posted
Posted

Bought my Kingston SNV425-S2 64GB SSD in September 2010.

January 2011

SSD is 4 months old

Health is 85%

Work time: 2 months, 6 days

Powered on: 229 times

September 2011

SSD is 12 months old

Health is 62%

Work time: 6 months, 1 day

Powered on: 578 times

"Your drive health is in good condition and according to current use, estimated lifetime is August 2020."

October 2011

SSD is 13 months old

Health is 59%

Work time: 6 months, 15 days

Powered on: 646 times

"Your drive health is in good condition and according to current use, estimated lifetime is April 2020."

November 2011

SSD is 14 months old

Health is 56%

Work time: 6 months, 28 days

Powered on: 709 times

"Your drive health is in good condition and according to current use, estimated lifetime is June 2020."

Theres something wrong with that drive..

Heres mine.....

Intel SSDSA2M080G2GC (Fw: 2CV102M3)

Total/free size: 80.0GB / 8.3GB

Work time: 8591 hours (11 months, 27 days, 23 hours)

Powered on: 124 times

Trime: supported, enabled

Health: 99%

Estimated lifetime: 8 Years, 8 months, 9 days

(T.E.C. date - July 24, 2020)

Data writen, GB: 3254.4

http://online.hddlif...21d01151899001a

This topic is now closed to further replies.
 Share
Go to topic listing

  • Recently Browsing   0 members

    • No registered users viewing this page.

  • Posts

    • Microsoft releases PowerToys v0.100.1, fixes a bug that made remapped keys misbehave

      By Ivan Jenic · Posted

      Microsoft releases PowerToys v0.100.1, fixes a bug that made remapped keys misbehave by Ivan Jenic Microsoft just released PowerToys v0.100.1, a patch update that addresses several stability and behavior issues found in v0.100.0. The v0.100.0 patch was a significant update for PowerToys, as it introduced all sorts of new features and additions, such as a rebuilt Shortcut Guide, a Command Palette Extension Gallery, webcam overlay support in ZoomIt, and more. However, the v0.100.0 version also introduced some bugs and stability issues. And now, Microsoft is addressing these issues in the new patch. The most impactful fix in this release perhaps is in Keyboard Manager, where remapped modifier keys were being delivered as system-key events, causing unexpected behavior in apps. The clearest example of this was Alt-to-Backspace remaps, deleting whole words instead of a single character. So, if you thought there was an issue with your keyboard, Microsoft just confirmed that it was PowerToys. Beyond the Keyboard Manager fix, v0.100.1 also addresses several other issues. It fixes a bug with Power Display that was preventing monitors from waking from standby correctly. Additionally, the new update patches Quick Access crashes on launch, and resolves a Shortcut Guide crash that occurred when switching between sidebar sections. Here’s the full changelog: Color Picker Fixed a bug where the main Color Picker window could appear inside the zoomed-in picker view Command Palette Fixed Run history initialization in AOT builds Fixed a bug where the Performance Monitor dock item could show ??? after restart Fixed the Hibernate command using the Sleep icon Limited the "pin to dock" dialog to displays where the dock is enabled Keyboard Manager Fixed modifier keys remapped to non-modifier keys being delivered as system-key events, which caused unexpected behavior in apps such as Alt-to-Backspace deleting whole words Power Display Fixed a bug where selecting On in the monitor power-state control did not wake a monitor from standby Fixed built-in display detection and brightness control on dual-GPU laptops where the internal panel is driven by the discrete GPU PowerToys Run Fixed VS Code Workspaces discovery after VS Code moved recently opened workspace data to shared storage Quick Access Fixed Quick Access flyout crashes caused by unhandled XAML exceptions during launch or page navigation Shortcut Guide Fixed a crash when navigating between Shortcut Guide sidebar sections Fixed number-key rendering in shortcut manifests and added a Postman shortcut manifest Updated bundled shortcut manifests to use the literal number-key token so number keys render correctly across apps ZoomIt Fixed a race condition in audio initialization for ZoomIt video recording You can download PowerToys v0.100.1 from the official GitHub releases page.
    • OBS Studio 32.2.0 Beta 2

      By Copernic · Posted

      OBS Studio 32.2.0 Beta 2 by Razvan Serea OBS Studio is software designed for capturing, compositing, encoding, recording, and streaming video content, efficiently. It is the re-write of the widely used Open Broadcaster Software, to allow even more features and multi-platform support. OBS Studio supports multiple sources, including media files, games, web pages, application windows, webcams, your desktop, microphone and more. OBS Studio Features: High performance real time video/audio capturing and mixing, with unlimited scenes you can switch between seamlessly via custom transitions. Live streaming to Twitch, YouTube, Periscope, Mixer, GoodGame, DailyMotion, Hitbox, VK and any other RTMP server Filters for video sources such as image masking, color correction, chroma/color keying, and more. x264, H.264 and AAC for your live streams and video recordings Intel Quick Sync Video (QSV) and NVIDIA NVENC support Intuitive audio mixer with per-source filters such as noise gate, noise suppression, and gain. Take full control with VST plugin support. GPU-based game capture for high performance game streaming Unlimited number of scenes and sources Number of different and customizable transitions for when you switch between scenes Hotkeys for almost any action such as start or stop your stream or recording, push-to-talk, fast mute of any audio source, show or hide any video source, switch between scenes,and much more Live preview of any changes on your scenes and sources using Studio Mode before pushing them to your stream where your viewers will see those changes DirectShow capture device support (webcams, capture cards, etc) Powerful and easy to use configuration options. Add new Sources, duplicate existing ones, and adjust their properties effortlessly. Streamlined Settings panel for quickly configuring your broadcasts and recordings. Switch between different profiles with ease. Light and dark themes available to fit your environment. …and many other features. For free. At all. OBS Studio 32.2.0 Beta 2 changelog: Beta 2 Changes Fixed a CI deployment issue. There are no application changes since Beta 1. 32.2 New Features Replaced add source dropdown with new dialog [Warchamp7] Improved FPS selector UX [jcm93] Added missing file support for filters [exeldro] Added ability for plugins to set custom icons for new source types [cg2121] Included .webp files when adding a directory to Image Slide Show source [TarunCore] Added copy paste functions to frontend API [exeldro] Added filter to compose SDR into HDR [jpark37] Added delete as a hotkey to delete sources on macOS [PatTheMav] Added dynamic bitrate support to multitrack video [lexano-ivs] 32.2 Changes Forced Intel-based installations to update to Apple Silicon version on macOS [PatTheMav] This change means that OBS Studio versions built for Intel-based Macs but running on Apple Silicon Macs will automatically update to OBS Studio built for Apple Silicon Macs. If an installation was using third-party plugins, those plugins will no longer load until replaced with Apple Silicon versions. Fixed audio mixer state getting out of sync when changing settings via websockets or plugins [Warchamp7] Added theming for checked QToolButtons [glikely] Improved OpenGL performance slightly on low-end machines [kkartaltepe] Set minimum size for color source to 1 pixel [exeldro] Added minimum width to spinboxes [Warchamp7] Disallowed overwriting the crash handler [sebastian-s-beckmann] Applied process mitigation policies for Windows [notr1ch] Adjusted description of multitrack video [jhnbwrs] Changed new capture devices to use fallback frame rate by default [PatTheMav] Improved DLL loading behavior on Windows [notr1ch] Limited multitrack video config to Custom service [PatTheMav] 32.2 Bug Fixes Fixed OAuth and dock state save corruption [PatTheMav] Fixed group bounds not resizing when removing items [howellrl] Fixed canvas mixes not being restored after video reset [dsaedtler] Fixed some erroneous crashes during shutdown [Warchamp7] Fixed display capture sometimes capturing black after a duplicator failure [ThrowTop] Fixed color of controls dock output buttons in System theme [shiina424] Fixed virtual camera reset failures [stephematician] Fixed potential crash when user discards changes in the settings window [suogesi] Fixed incorrect return value in virtualcam filter [xtfo] Fixed source toolbar buttons not working after dragging a source into a group [Warchamp7] Fixed properties hint icon spacing [Warchamp7] Fixed potential crash when a video device reconnects on macOS [jcm93] Fixed an issue where PipeWire could fail on NVIDIA GPUs [hoshinolina] Fixed obs_canvas_get_video_info returning incorrect framerate [dsaedtler] 32.2 Deprecations Deprecated obs_properties_add_button [sebastian-s-beckmann] Download: OBS Studio 32.2.0 Beta 2 | Portable | ARM64 | ~200.0 MB (Open Source) View: OBS Studio Homepage | Other Operating Systems | Screenshot Get alerted to all of our Software updates on Twitter at @NeowinSoftware
    • Microsoft is building an AI datacenter that "uses less water than a fast food restaurant"

      By Slugsie · Posted

      Is a fast food restaurant a good metric to compare against?
    • Grand Theft Auto VI pricing revealed alongside Ultimate Edition and pre-loading details

      By LoneWolfSL · Posted

      Grand Theft Auto VI pricing revealed alongside Ultimate Edition and pre-loading details by Pulasthi Ariyasinghe Last week, Rockstar revealed Grand Theft Auto VI pre-orders will be starting soon, and just a day ahead of that, now the studio has announced the official pricing for the highly anticipated game. This has been a hotly debated topic among fans and industry veterans for a long time, considering the game is expected to be the biggest entertainment product launch ever. The confirmed pricing for the Grand Theft Auto VI standard edition is $79.99, which Rockstar says gives access to the "single-player experience set in the biggest, most immersive evolution of the series yet." This follows what most of our readers thought would happen with the pricing too. At the same time, a $99.99 Grand Theft Auto VI: Ultimate Edition has been confirmed as well, which lands with "an exclusive collection of premium vehicles, weapons, apparel, and action threaded across all aspects of Jason and Lucia’s story." Pre-ordering will also give fans extra bonuses, including a Vintage Vice City Pack of cosmetic items as well as a free month of GTA+. Head to the official website of the game here to check out all the cosmetic rewards the Ultimate Edition and pre-orders bring. Interestingly, the studio does not mention Grand Theft Auto VI multiplayer at all in today's announcement. Perhaps this will arrive later, following the campaign launch, or the studio is keeping that reveal for a later date. Digital pre-orders for Grand Theft Auto VI will begin on June 25, 2026, at midnight local time across regions for Xbox Series X|S and PlayStation 5. The title is slated to launch on November 19 on those same platforms. Pre-loading for Grand Theft Auto VI will kick off on November 12, giving players a week to get the game ready on their consoles. As for the physical edition, Take-Two has confirmed that this will be available without a disc, with the box only containing a download code inside. This will be purchasable starting November 12, giving players who take this route time to pre-load the title as well.
    • Politically funny images of the World

      By +Edouard · Posted

  • Recent Achievements

    • One Year In
      OHI Accounting earned a badge
      One Year In
    • First Post
      Almohandis earned a badge
      First Post
    • Rookie
      DaviKar went up a rank
      Rookie
    • Dedicated
      HidekoYamamoto94 earned a badge
      Dedicated
    • One Month Later
      timbobit earned a badge
      One Month Later

  • Popular Contributors

    1. 1
      +primortal
      474
    2. 2
      +Edouard
      172
    3. 3
      PsYcHoKiLLa
      122
    4. 4
      Michael Scrip
      83
    5. 5
      Xenon
      72
    Show More

  • Tell a friend

    Love Neowin? Tell a friend!