Intel SSD Reliability

[Georgeisdead]

This is a note to address several articles I have come across lately that state intel's reputation for quality and reliability in the SSD market as if it is a given. These comments are from my personal experience with intel's drives. I have owned 3 intel solid state drives, one X25-M G1, and two X25-M G2's. The X25-M G1 failed after 2 years while one of the G2 drives failed after 2.5 years. Now, I am not an expert on MTBF and reliability, but in my opinion this is a pretty poor track record. It is entirely possible that this is a coinicidence, however both drives failed in the same manner, from the same problem (determined by a third party data recovery specialist): Bad NAND flash.

As best I understand it as it was descibed by the company that analyzed these failed drives, a block of NAND flash either went bad or became inaccessible by the controller rendering the drives useless and unable to be accessed by normal means of hooking it up to a SATA or USB port. Two drives, different NAND (50 nm for the G1 and 34 nm for the G2), same failure mode.

Once again, this is not definitive, just my observations but to me, I think review sites need to be a little more cautious about how they qualify intel's reputation for quality and reliability because from my perspective, intel has neither and I have since began using crucial SSD's. Hopefully, I will see much longer life from these new drives.

 

[ShintaiDK]

Owning a G1 80GB, G2 80GB and 2 320 300. No failures. the G1 got a poweron time of 23336 hours and 10.4TB writes.

Are you sure the fault aint related to yourself? Excessive writes? Maybe some defrag running or other similar utilities or applications.

The drives all got extra NANDs to replace any failed. And a drive wont die just because of a portion of failed NAND cells. It works the exact same way on HDs.

Looking at the SMART data. For example using Intels SSDToolBox will show you how much reserved space there is and the wearout indicator.

 

[Burner27]

I have to agree with ShintaiDK. We're these all attached to the same controller port on the same mobo? Did you ever update the FW on any of these drives? Did you ever update the Bios on the mobos these drives were attached to? We're these boards used while overclocking? Did you increase the voltage on any of these mobos? Can you list the complete specs of the systems these failed SSDs were in please?

 

[Georgeisdead]

I'm not here to argue and I'm not looking for solutions or explanations. This is an informative thread so that people in the market for an SSD do not blindly believe that just because they buy an intel drive, they are safe. It is simply not true and in my opinion intel should not be able to command the price premium that they charge based on reliability.

Yes I am certain it is not my fault. The failures were in 2 separate systems with 2 separate usage patterns, both WELL within tolerance. I am a very technically capable user who has been building systems for over 15 years and the SSD implementations I have done have all been thoroughly researched and set up to prevent the things you mention such as excessive writes, defragging etc.

 

[ShintaiDK]

I'm not here to argue and I'm not looking for solutions or explanations. This is an informative thread so that people in the market for an SSD do not blindly believe that just because they buy an intel drive, they are safe. It is simply not true and in my opinion intel should not be able to command the price premium that they charge based on reliability.

Yes I am certain it is not my fault. The failures were in 2 separate systems with 2 separate usage patterns, both WELL within tolerance. I am a very technically capable user who has been building systems for over 15 years and the SSD implementations I have done have all been thoroughly researched and set up to prevent the things you mention such as excessive writes, defragging etc.

Problem is you basicly claim something thats not backed up in real world statistics.

Intels own controller is backed up by years of RMA rates to match it. And the biggest single vendor deployment.

While I wouldnt buy Intel anymore since they dropped their own controller. Thats a whole other matter. But Intels G1, G2 and G3(320) is still the undisputed king of reliability. Today I would buy C400/M4.

If you actually wanted to post an informative thread. Then add the SMART numbers from your drives. That can still be read even tho the NAND cells are exausted.

 

[Zap]

failed in the same manner, from the same problem (determined by a third party data recovery specialist): Bad NAND flash.

I'm not sure some 3rd party data recovery place is able to do FA (failure analysis) on NAND. Have you considered that they are just saying something that sounds likely in order to placate the customer? It's like telling you that your HDD had a head crash which damaged a platter. Sounds plausible, right?

It is the mindset of the dealings between customers and fixers, no matter what the product is.

For instance if my car breaks down and it is religiously maintained, I might get in the service manager's face and demand to know why. Why did it break? Why did I have to pay so much? What can be done to prevent it? So, the service manager has a choice.

A) Explain thoroughly to the customer in detail that while these water pumps have only a 5% failure rate over a span of 3 years, statistically there exists the chance that it will happen to you due to no fault of your own, nor the fault of the manufacturer because components are built to tolerances using materials chosen for properties of acceptable reliability as well as cost and ease of manufacture, and a small chance of failure is accepted and unavoidable in order to hit the price point because no amount of money can make a part 100% reliable with absolutely 0% chance of failure and last forever, thus trade-offs must be made.

B) Tell customer, "yeah, we've seen that happen before with the OEM pump. That's why we put in an aftermarket unit for you."

Problem is you basicly claim something thats not backed up in real world statistics.
...
But Intels G1, G2 and G3(320) is still the undisputed king of reliability.

This is pretty much proven fact. I believe the stats put them at under 2% failure rate overall, which is pretty good for mass storage (both SSDs and HDDs).

Of course who knows where their reliability will be after they have been in use for 5 years (their warranty span). Then again, I don't think I'd trust a 5 year old HDD either.

 

[Makaveli]

Well said both Zap and ShintaiDK.

I own two G2 drives 160GB and no issues at all, they have been very reliable for me and i'm in the same boat Shintaidk. When I finally do decide to upgrade them i'm weary of not being able to purchase an Intel SSD without an intel controller I won't touch a sandforce drive even the intel branded one.

As for the OP what is this post suppose to warm me or others about. You had two bad drives it happends. I would be more interested in seeing an RMA report from intel on what happends with these drives than some third party.

 

[Burner27]

I'm not here to argue and I'm not looking for solutions or explanations. This is an informative thread so that people in the market for an SSD do not blindly believe that just because they buy an intel drive, they are safe. It is simply not true and in my opinion intel should not be able to command the price premium that they charge based on reliability.

Yes I am certain it is not my fault. The failures were in 2 separate systems with 2 separate usage patterns, both WELL within tolerance. I am a very technically capable user who has been building systems for over 15 years and the SSD implementations I have done have all been thoroughly researched and set up to prevent the things you mention such as excessive writes, defragging etc.

No one is here to argue with you nor are we here to question your technical abilities. What we were trying to do was assess why you have failures with your drives when we don't have all the facts. I know of another member here who has had failures with his G1 and G2 Intel SSDs as well and he couldn't explain it either. But, like others have stated, these drives have the lowest failure rates backed up by statistics. All companies can have a 'bad batch' now and again and you were just unlucky. Hopefully your drives are still under warranty and you can get them replaced.

 

[tweakboy]

I would think its a firmware issue. Intel is very reliable in the SSD field... Or bad batch like gurus said...

 

[Old Hippie]

I know of another member here who has had failures with his G1 and G2 Intel SSDs as well and he couldn't explain it either.

That may be me.

I RMA'd three 80GB G2s.

Mine would just drop from the BIOS for no particular reason and Intel never would/could give me a reason.

Obviously Intel has a great reputation for reliability and I would have no problem buying another, but you're always going to have some failures.

 

[tweakboy]

Maybe Intel should just stick with CPU's .....

 

[jwilliams4200]

intel has neither and I have since began using crucial SSD's. Hopefully, I will see much longer life from these new drives.

I could not help laughing at this. You claim the failure is due to bad flash. But Crucial SSDs use Micron flash, which is nearly identical to Intel flash, since they both manufacture their flash with the same processes, developed in their IMFT partnership.

 

[Georgeisdead]

Maybe Intel should just stick with CPU's .....

Agreed.

 

[Coup27]

I could not help laughing at this. You claim the failure is due to bad flash. But Crucial SSDs use Micron flash, which is nearly identical to Intel flash, since they both manufacture their flash with the same processes, developed in their IMFT partnership.

You beat me to it.

Both Intel and Crucial use flash exclusively from IMFT, their joint flash production company. If you are concerned about SSDs dying due to poor flash quality, then switching from Intel to Crucial will achieve nothing.

 

[Georgeisdead]

I think pretty much everyone has missed the point here.

Definition of insanity = doing the same thing over and over and expecting a different result.

Why then would I continue to buy intel drives given my history with them?

You can quote failure rates until you are blue in the face but the fact remains, I have had two drive failures with relatively low life. Intel is horrible to deal with to correct these problems by issuing warranty replacements (only 1 out of 2). So based on my experience with their company and their product, I choose not to deal with them.

The moral of my little story here is basically you can have the lowest failure rates of anyone in the industry, but that does not mean you are buying a quality and reliable product that deserves a price premium over the competition. I only hope that my experience will make perspective SSD owners aware of some of the problems they could potentially face even when buying drives that are claimed to be the most reliable.

 

[Georgeisdead]

You beat me to it.

Both Intel and Crucial use flash exclusively from IMFT, their joint flash production company. If you are concerned about SSDs dying due to poor flash quality, then switching from Intel to Crucial will achieve nothing.

I said "hopefully" NOT "I expect to..." (also, the crucial drives use 25nm flash, which I have not tried yet)

Once again, I'm not here to argue. I just want to post my experience for others to consider.

 

[jwilliams4200]

Definition of insanity = doing the same thing over and over and expecting a different result.

So are you saying you are insane, since you said you bought a Crucial SSD which uses the same IMFT flash that you said caused your problem before, and yet you hope for a different result this time?

If you are convinced that your problems are due to bad IMFT flash (I'm not convinced, BTW) then the rational thing to do would be to try an SSD with different flash. Say a Plextor M3 (Toshiba flash) or a Samsung 830 (Samsung flash).

 

[Georgeisdead]

So are you saying you are insane, since you said you bought a Crucial SSD which uses the same IMFT flash that you said caused your problem before, and yet you hope for a different result this time?

Pretty sure they never made an X25-m with 25nm flash.

 

[jwilliams4200]

Pretty sure they never made an X25-m with 25nm flash.

Pretty sure they never made an Intel 320 or 520 with anything but 25nm IMFT flash.

So what is your rationale again?

 

[Georgeisdead]

Pretty sure they never made an Intel 320 or 520 with anything but 25nm IMFT flash.

So what is your rationale again?

Wow. You're so smart, let me just bask in the glory of your IQ for a moment.

 

[jwilliams4200]

Wow. You're so smart, let me just bask in the glory of your IQ for a moment.

Are you arguing yet?

 

[ShintaiDK]

Wow. You're so smart, let me just bask in the glory of your IQ for a moment.

You aint exactly helping your thread with those comments.

And since you are already in the "Intel should never have made SSDs" corner. Then the entire thread just aint looking good.

 

[Phynaz]

Once again, I'm not here to argue.

And yet you are :\

 

[Georgeisdead]

could a moderator go ahead and delete this thread. Now I know not to post anything on these forums since it is full of people who know everything.

 

[Coup27]

Georgeisdead you are really making less and less sense as this thread continues. Especially with comments like this:

The moral of my little story here is basically you can have the lowest failure rates of anyone in the industry, but that does not mean you are buying a quality and reliable product that deserves a price premium over the competition.

A failure rate is a metric of data which helps determine the quality of a product. By having the lowest failure rates of a product within its sector, means you have one of the more reliable products in its sector.