Any real basis to SSD reliability claims?

Soldato
Joined
20 Oct 2008
Posts
12,082
When people are recommending SSDs it’s quite common for better reliability to be mentioned as a selling point.

When someone claims that for example the Samsung 830 is more reliable than <insert drive of your choice> is there any real published data to back it up, or is it just gut feeling and repeating what’s already been written?
 
Depends what you mean, people claim the Samsung is maybe more reliable than other SSD's largely because if you look across various tech forums the number of posts about dead Samsung SSD's is low, relatively the same for Crucial and OCZ drives drop like flies.

This is in huge part because OCZ tend to jump on the latest controller before anyone else, which leads to teething problems and loads of returned drives.

as for ssd's being sold with reliability as a feature, its over HDD drives. Simply put electronic circuits fail at a lower rate than mechanical in the vast majority of situations. HDD's have mechanical moving parts that WILL wear out eventually and they have circuits, a controller and memory chip. SSD's have no mechanical parts so won't wear out in the same way, electronics can still fry but HDD's basically have an entire extra set of things that can go wrong.

SSD's are more reliable than HDD's but that doesn't mean you won't end up buying the one that dies.

IN reality SSDs/HDD's have low failure rates, its more important in business where downtime can cost you lots of money, at home, meh.

Reliability is pretty much the lowest priority reason to buy an SSD. Performance, noise, power, reliability in that order really.

Though also keep in mind for performance you'll not always notice a hell of a lot, its best in the worst situations HDD's have, unrar something while running a AV scan, and or running an encode, a HDD will crawl to a halt and just opening a webpage becomes a challenge, an ssd laughs at that.

Game loading, 5 seconds to load a level of 6, you won't really notice in most situations.
 
I was really referring to individuals making (unchallenged) claims about the relative reliability of the various SSDs without any real evidence to back it up.

Faster, cheaper, prettier – all fairly easy to prove. More reliable is difficult to know or prove without real statistical data.

In the case of the Samsung 830 making claims for its reliability when it hasn’t even been in the market place for 12 months is a bit brave. They could all get to two years old and start dropping like flies (hopefully not as I’ve got one on the way at the moment).
 
Personally I think it's more to do with the fact that the M4 and Samsung drives don't use the sandforce controller which has had endless issues with reliability, stability etc.
 
I've read various statistics and plausible arguments on how SSD's are more reliable than mechanical drives, but I've yet to experience it. I've seen quite a few SSD's, the most reliable of which has been the M4 in my home PC. However the majority of SSD's (various brands) we've had delivered in laptops at work have failed within 6 months. Thats far less life than we typically get out of mechanical drives. Perhaps we've just been unlucky, idk, but the evidence is mounting.
 
SSD vs mechanical is a different argument (although equally hard to prove).

On personal experience SSDs are infinitely more reliable because all of mine are still working. Nothing older than a fairly early model C300, so there isn't anything particularly amazing about that.
 
btw samsung produce their own flash. they can cherry pick the best and bin the rest to other manufacturers. I presume this reduces the chance of their drives failing.
As for general reliability, ssd's are not suited for highly sequential write environments.
 
I think without knowing what the actual numbers of drives in service out there are, the forum chatter about failure rates is just guess work.

The reports of failing drives of whatever make are self selecting samples, ie, you never hear about drives that have no issues!

I have Crucial M4 and OCZ Agility drives in my systems, and fingers crossed, I haven't had any failures.

Without the sales data, and failure rates per 1000 or similar, anybody posting about failure rates is just guessing or passing on their own experiences, neither of which are much help in working out what brand or product is more reliable than another.
 
personally, i use two ssd's, both kingston, and both are quite solid drives in the two years i've had them.

on an semi related note, i have had some harddrive failures across a few brands (seagate, maxtor, WD) but as of yet, i haven't had any samsung drives fail.
 
personally, i use two ssd's, both kingston, and both are quite solid drives in the two years i've had them.

on an semi related note, i have had some harddrive failures across a few brands (seagate, maxtor, WD) but as of yet, i haven't had any samsung drives fail.

Give it time, they'll all fail eventually. If you're lucky they get taken out of production before it happens.

I wonder if there are still any IBM GXPs out there? That was a line of drives that failed in a fairly spectacular way. Most problems since have paled in comparison.
 
SSD vs mechanical is a different argument (although equally hard to prove).
A year or so back there was an article on Toms Hardware where they had got stats from some big disk users (server farms etc) who were experimenting with moving to some use of SSDs and were surprised to find that there was very little difference between failure rates on SSD and mechanical drives. That said, I suspect in that environment they'd be using the "higher quality" mechanical drives.
 
Yes SandForce had some issues in the past with BSOD, but havn't seen any problems for almost a year. Guess its all good now. SandForce is the faster compared to M4 and samsung, But samsung is a good buy as well.
 
All I can say is I have had an SSD drive in this PC over 2 years. The PC is a server/workstation and does video editing along with being on 24/7. No reliability problems. I now plan to use SSD's in all new and home work computers.
 
Back
Top Bottom