Overclock.net banner
Cancel
Create thread New Forums

SSD near to die?

Jump to Latest
21 - 40 of 42 Posts
#21 ·
Quote:
Originally Posted by KobaltRock View Post

Not true, a SSD can fail at any time.
Heck, any storage device can fail at any time, with no warning at all.
Click to expand...
You're not understanding me. What you just said goes without saying. We all know that.

You might be interested in this:

http://techreport.com/review/27909/the-ssd-endurance-experiment-theyre-all-dead

So, unexpected failures ASIDE, you'd have to beat the living hell out of a solid state drive in order to kill it - just as this test proves. Again, it goes without saying that things can fail at anytime without warning. It doesn't need to be stated.
 
#22 ·
Quote:
Originally Posted by TwoCables View Post

Quote:
Originally Posted by KobaltRock View Post

Not true, a SSD can fail at any time.

Heck, any storage device can fail at any time, with no warning at all.
Click to expand...
You're not understanding me. What you just said goes without saying. We all know that.

You might be interested in this:

http://techreport.com/review/27909/the-ssd-endurance-experiment-theyre-all-dead

So, unexpected failures ASIDE, you'd have to beat the living hell out of a solid state drive in order to kill it - just as this test proves. Again, it goes without saying that things can fail at anytime without warning. It doesn't need to be stated.
Click to expand...
It is one thing to say that you don't have to worry about wearing out the NAND (well, assuming normal consumer workloads), and another thing to say that you don't have to worry about it failing.

I don't think most people know that a SSD can fail, they just assume since it is a SSD, and it has no moving parts, that it will outlast a HD.
I have a pile of SSDs in front of me that are all dead, from all different companies.
Not one of these died because of NAND wear, which has to do with endurance. Most of these failed without notice.

That is why I said to the OP, just backup, and keep using the drive.
 
#23 ·
Good grief, man. Let's relax a little bit. I think what I said was understood just fine without you coming in to take control. Like I said, what you said GOES WITHOUT SAYING. You didn't have to come in here and state the obvious like that as though you are CORRECTING me.
 
Discussion starter · #24 ·
#25 ·
Quote:
Originally Posted by zhup View Post

Hello TwoCables,

I asked the Stablebit support about the meaning of SSD Life value (in %) reported by Stablebit Scanner and they said that only 33% of the life left.

http://community.covecube.com/index.php?/topic/2853-ssd-life-value/

Click to expand...
Yeah of course they're going to say that because that's what their software says. They're just explaining what their software is saying. However, again, look here:

http://crystalmark.info/software/CrystalDiskInfo/manual-en/HealthStatus.html

You have a Micron drive, and it says that for Micron drives, the percentage is for "Percentage Of The Rated Lifetime Used". Since all of your S.M.A.R.T. data comes up saying that the drive is in very good health and you're seeing nothing but blue across the board, I am far more inclined to believe CrystalDiskInfo than this Stablebit program (which I've never heard of, by the way).

Then there's this:

http://techreport.com/review/27909/the-ssd-endurance-experiment-theyre-all-dead

Seriously, read this carefully. You will see that you're worrying way too much about your drive (their test proves that you have to send a solid state drive to hell and back several times before it even begins to start dying). I think the only reason you would have for replacing it is running out of space (or if you just need a much faster drive).
 
Discussion starter · #26 ·
Quote:
Originally Posted by TwoCables View Post

Yeah of course they're going to say that because that's what their software says. They're just explaining what their software is saying. However, again, look here:

http://crystalmark.info/software/CrystalDiskInfo/manual-en/HealthStatus.html

You have a Micron drive, and it says that for Micron drives, the percentage is for "Percentage Of The Rated Lifetime Used". Since all of your S.M.A.R.T. data comes up saying that the drive is in very good health and you're seeing nothing but blue across the board, I am far more inclined to believe CrystalDiskInfo than this Stablebit program (which I've never heard of, by the way).

Then there's this:

http://techreport.com/review/27909/the-ssd-endurance-experiment-theyre-all-dead

Seriously, read this carefully. You will see that you're worrying way too much about your drive (their test proves that you have to send a solid state drive to hell and back several times before it even begins to start dying). I think the only reason you would have for replacing it is running out of space (or if you just need a much faster drive).
Click to expand...
Hello TwoCables,

What do think about this:



https://hardforum.com/threads/ssd-near-to-die.1932485/#post-1042992434
 
#27 ·
Quote:
Originally Posted by zhup View Post

Hello TwoCables,

What do think about this:



https://hardforum.com/threads/ssd-near-to-die.1932485/#post-1042992434
Click to expand...
He's saying the same thing I am. If it's reading the percentage of the rated lifetime used, then it's saying that only 33% of the rated lifetime has been used. This is why you are seeing nothing but blue, which means everything is going very very very well for your drive. I think it would be best to just relax and enjoy your drive which is still in excellent condition.
 
#28 ·
At this time I am going to agree with 2 cables but remain skeptical. According to the crystal disk info manual linked several times the percentage for a micron drive is life used.

Every time I have ran a smart check on a brand new SSD it has been 100%, good. or just good.
I just ran a smart check on my Adata SSD in my machine I am currently on and it doesn't even give a percentage, just states "good"
Ran it on my laptop with an m.2 drive and it also says good at 100% but that is a transcend ssd not a micron. I have a crucial (micron) 128GB SSD in my second gaming rig at home that i could check later if you like and compare it.
 
Discussion starter · #29 ·
Quote:
Originally Posted by XanderTheGoober View Post

At this time I am going to agree with 2 cables but remain skeptical. According to the crystal disk info manual linked several times the percentage for a micron drive is life used.

Every time I have ran a smart check on a brand new SSD it has been 100%, good. or just good.
I just ran a smart check on my Adata SSD in my machine I am currently on and it doesn't even give a percentage, just states "good"
Ran it on my laptop with an m.2 drive and it also says good at 100% but that is a transcend ssd not a micron. I have a crucial (micron) 128GB SSD in my second gaming rig at home that i could check later if you like and compare it.
Click to expand...
It would be good to compare it. Thank you in advance.
 
#30 ·
I installed stablebits software to compare to Crystal Disk Info on stablebits scanner it says 8% used for my corsair force GT 120 i got back in 2009 on Crystal Disk Info it says i have 92% left so their software is reading smart correctly just how it lists the results are differnt i have 32tb writes and 52tb reads on my drive.
 
Save
Reply Quote Rep+ Rep+
Discussion starter · #31 ·
Quote:
Originally Posted by Bal3Wolf View Post

I installed stablebits software to compare to Crystal Disk Info on stablebits scanner it says 8% used for my corsair force GT 120 i got back in 2009 on Crystal Disk Info it says i have 92% left so their software is reading smart correctly just how it lists the results are differnt i have 32tb writes and 52tb reads on my drive.
Click to expand...
Your Corsair force GT (SandForce):
Crystal Disk Info: 92% left
Stablebir Scanner: 8% used

So, it is matching.
 
#32 ·
Well after reading this:

Technical Note - M500 SSD Firmware MU02 SMART Attributes Reference



Looks like CrystalDiskInfo mislabels the 202 CA attribute, it's not "Percent Lifetime Used" but "Percent Lifetime Remaining".

The raw value in OP's SSD is hex 43 = dec 67. Thus 100-67 = 33.

The paper explains that the raw value is the quotient of: Average erase count for a super block (stripe of blocks) / erase count for which the part is rated (block life).

So it sounds like remaining lifetime is 33%, based on erase count ratio (and not on power on hours).

Now the weird thing here is that if ~10TB of writes have eaten 67% of the lifetime, total SSD lifetime would be a meager ~15TB.. At least under OPs particular conditions of usage.
 
#33 ·
I forgot that the PNY ssd from my main rig transitioned to the second when i upgraded the main to a sammy SSD. I don't think I have a Crucial SSD at my disposal anymore. Sorry bud.
 
Discussion starter · #34 ·
Quote:
Originally Posted by TELVM View Post

Looks like CrystalDiskInfo mislabels the 202 CA attribute, it's not "Percent Lifetime Used" but "Percent Lifetime Remaining".
Click to expand...
I reported this topic to CrystalDisk developer. Let's hope he answer.

Quote:
Originally Posted by TELVM View Post

Now the weird thing here is that if ~10TB of writes have eaten 67% of the lifetime, total SSD lifetime would be a meager ~15TB.. At least under OPs particular conditions of usage.
Click to expand...
This is really weird:
- this ssd is mostly free (78,6GB free of 111GB),
- 685 working days and about 11TB writes,
- no special actions or tasks was done on it,
- it is in NAS server as OS drive (Windows Server 2012).
 
Discussion starter · #36 ·
Quote:
Originally Posted by TELVM View Post

Crucial says 72TBW endurance for the M500. And it uses MLC so its NAND should last a while.

Interesting post on the subject from a Crucial rep.

There's a Crucial utility for its SSDs that shows SMART data.
Click to expand...
Hello Telvm,

Thank you for your involvement.

The Crucial Storage Executive software shows that 67% Lifetime is used.

 
#37 ·
IMHO as long as "reallocation event count" remains at zero the NAND is healthy (as far as can be said from SMART data). I'd only worry (start thinking about replacing the SSD) if and when that number begins to creep up.

Then if and when "unused reserved block count" countdown approaches zero, I'd go panic mode (replace SSD ASAP!).


How NAND dies by a thousand cuts: The SSD Endurance Experiment: They're all dead.
 
Discussion starter · #38 ·
Quote:
Originally Posted by TELVM View Post

IMHO as long as "reallocation event count" remains at zero the NAND is healthy (as far as can be said from SMART data). I'd only worry (start thinking about replacing the SSD) if and when that number begins to creep up.

Then if and when "unused reserved block count" countdown approaches zero, I'd go panic mode (replace SSD ASAP!).


How NAND dies by a thousand cuts: The SSD Endurance Experiment: They're all dead.
Click to expand...
I will replace this ssd with new one next week.

I have two possibilities:
- Intel SSD 535 - SandForce 2281, 16nm MLC NAND, 5 years warranty.
- Adata Premier Pro SP920 - Marvell 88SS9189, 20nm MLC NAND, 5 years warranty.

Which should I choose?
 
21 - 40 of 42 Posts
About this Discussion
41 Replies
8 Participants
Last post:
OrangeRaptor
Overclock.net
posts
28M
members
562K
Since
2004
The premier forum for overclocking experts and enthusiasts. Discuss hardware optimization, custom builds, benchmarking, cooling solutions, and pushing the boundaries of computing performance. From beginner guides to extreme overclocking, join our technical community to master system tuning.
Full Forum Listing
Explore Our Forums
Intel CPUs Hardware News NVIDIA PC Gaming Water Cooling