Are Seagate HDD's fine?

[Nonco]

Ive used them in build and never had any issues, however I have had friends who will just rage when they see them in builds as they assume they just die.

 

Whats the general consensus on this?

[Aytex]

seagate baracuda OR

WD Black/blue

[VioDuskar]

yeah, i like them.

[TheN00bBuilder]

They are all sorts of nope.

 

https://www.backblaze.com/blog/hard-drive-reliability-update-september-2014/

[Agazed]

Why wouldnt they be? Just as reliable as any other WD or Hitachi..

[TheN00bBuilder]

Look at le link I post above.

[W-L]

They are all sorts of nope.

 

https://www.backblaze.com/blog/hard-drive-reliability-update-september-2014/

 

They are perfectly fine and of course blackblaze did not properly test their drives and give proper sample sizes so it's a skewed result. It's been discussed before.

Edited July 3, 2015 by W-L

[cesrai]

@terrytek

---

I personally prefer WD cause I've used them my whole life and they've been great.

[Gofspar]

they are loud as a motherfucker, get WD Blues

[t-scene]

Some can be louder than others but I like them. Have 3 one 2.5 SSHD and two 3.5 HDD's.

[TheN00bBuilder]

I have also had 4 Seagates myself fail. Never had a WD or Hitachi Deskstar fail, even the 40GB Deathstar. Actually still have a WD Caviar Black 80GB drive from 2003 still kicking.

[doctorraichu]

I do not notice a performance or noise difference from WD to Seagate, although the Seagate more or less had a chernoybl meltdown.

[Arc_Jester]

I use prefer WD but Seagate works good too from my experience, although I have seen numerous complaints of failure.

[Guest Philosophy]

Yes. They are fine. 

[dfsdfgfkjsefoiqzemnd]

They are perfectly fine and of course blackblaze did not properly test their drives and give proper sample sizes so it's a skewed result. It's been discussed before.

They were used in the same enterprise environment as all the other non-enterprise drives, so the whole "not properly tested" excuse is just plain wrong. On a level playing field they did far worse than WD and HGST, so it makes sense to assume that they in fact ARE far worse.

As for proper sample sizes, if you have OVER FOUR THOUSAND drives of a particular model and around a third of those die within the first 3 years, that's a major issue.

The rest of the Seagate models also had a 6 to 10% failure rate, which is 3 to 5 times more than what the thousands of HGST (also non-enterprise) drives were doing in the same environment.

The current models seem to be doing better, but there's a lot of the slightly older rubbish ones in the stores still.

[W-L]

They were used in the same enterprise environment as all the other non-enterprise drives, so the whole "not properly tested" excuse is just plain wrong. On a level playing field they did far worse than WD and HGST, so it makes sense to assume that they in fact ARE far worse.

As for proper sample sizes, if you have OVER FOUR THOUSAND drives of a particular model and around a third of those die within the first 3 years, that's a major issue.

The rest of the Seagate models also had a 6 to 10% failure rate, which is 3 to 5 times more than what the thousands of HGST (also non-enterprise) drives were doing in the same environment.

The current models seem to be doing better, but there's a lot of the slightly older rubbish ones in the stores still.

 

It's been discussed before the drives were used in poorly designed drive cages which are not for consumer drives. The main issue with sample size is the fact the drive were not the same drives throughout testing and not applied at the same time, some drives were internal storage drives others were taken from external hard drive and had their casing removed. 

 

Note at 6-10% failure rate they would go bankrupt as said before also.

[alpenwasser]

They were used in the same enterprise environment as all the other non-enterprise drives, so the whole "not properly tested" excuse is just plain wrong. On a level playing field they did far worse than WD and HGST, so it makes sense to assume that they in fact ARE far worse.

I honestly don't recall all the specifics, but if I remember right, one of the complaints

against Backblaze was indeed that there was actually not a level playing field. Not all

enclosures were the same, some were of a newer design, some of an older design, and the older

enclosures had a disproportionally high amount of Seagate drives in them. I think the main

difference between the two enclosures was the improved vibration dampening for drive protection

on the newer ones, which might actually make a significant difference for survival rates.

It's been a while since this has been last discussed, so I must admit I'm a bit hazy on the

details at this point. Also, I'm definitely no expert in statistics. However, as best as I

can tell, that would precisely be what is required in this situation. Somebody who properly

understands the whole statistics thing, and can adequately process the data so that we can

actually draw meaningful and realiable conclusions from it. As it stands, what we have

is merely raw data, and while I'm no statistics guru, I have among other things been taught

that drawing conclusions from raw data without proper care can lead to terribly incorrect

conclusions (it doesn't necessarily have to, but it can, or at least so I've been taught).

I fully agree with you that something smells fishy here with that high of a failure rate,

and it should be investigated. Maybe there was indeed something wrong with the drives,

but just going from the raw data without any additional info (What was the average load

on those drives? Load spikes, how frequent and how high? Average temperates? Temperature

spikes, how frequent and how high? Vibrations? I'm sure there's more) shouldn't be taken

as a be-all-end-all indicator of actual drive quality just like that. There are just way too

many unknown variables in play for such conclusions to be reliable.

I'm not saying field data cannot be useful and only data from a laboratory experiment

should be used to draw conclusions from, but when you do field experiments, you need to

document every possible environmental variable with utmost care and then take that into

account when drawing conclusions from your dataset (again, going by my admittedly very

basic knowledge on statistics). And as far as I know, that has not been done here.

That's what I make of the whole thing anyway. I honestly appreciate that Backblaze

release the data, but I take it with at least two grains of salt.

EDIT: In the end, I'm not saying nothing went/is wrong here, I'm just saying that as

best as I can tell, we lack the information (all those unknown/undocumented variables)

to reliably draw the conclusion from that data that Seagate's drives are indeed horrible.

Backblaze basically didn't document their experiment properly enough for that, as far

as I can tell. If any expert on statistics ever comes across this, I would be very

interested in what they have to say. Maybe my conclusions are horribly misguided, but

with my limited knowledge on statistics, this is the best I can do, so to speak.

/EDIT

Also, this (see also quote below):

Note at 6-10% failure rate they would go bankrupt as said before also.

Lastly, this is my post from a thread about the Backblaze thing from a while back. It

compares the data with return rates from a large French online retailer (based on another

member's post, see link in quote), and things don't look too bad there, all HDD manufacturers

come out about the same:

 

- snip -

Anyway, I shall just post the link to a thread in which return rates for various hardware

components were given to a French hardware site by a French retailer.

The HDD section specifically (numbers in parentheses are numbers from previous dataset):

 

- Toshiba 1,15%

- Seagate 1,44% (vs 1,65%)

- Western 1,55% (vs 1,44%)

- Samsung 2,24% (vs 1,30%)

- Hitachi 2,40% (vs 3,45%)

Data is from 2013.

Link to thread: http://linustechtips.com/main/topic/108284-huge-list-of-failure-rates-on-pc-components-french-but-i-translated-nearly-everything/

Look at it this way: If there really was a serious problem with any one HDD manufacturer's

quality, they would long ago have lost all OEM contracts (just imagine Dell or HP buying

a few million drives and having a quarter of them fail, that HDD company would lose that

contract faster than they could say "lawsuit"), and since that's where the big money is,

they would have gone out of business sooner rather than later. Since that hasn't happened

so far, we can quite safely assume that for the most part there are no major faults with

any specific manufacturer. They might on occasion screw up, but by and large, there really

isn't that much of a difference between manufacturers these days.

Link to thread: http://linustechtips.com/main/topic/347320-who-makes-the-most-reliable-hard-drives/

EDIT 2:

This is not at all a properly documented experiment either, since it's many different

drives and many different users, but as a side note: In the 10 TB thread, we currently

have 1,110 drives, 366 of which are Seagate. We've had one or two users who have indeed

reported bad experiences with them, but we also have users who run tons of them without

issues. So far it doesn't seem like we come anywhere near an annual failure rate for them

of 10%~40% (depending on model, roughly the range reported by Backblaze).

Now, as said, this is not a properly documented experiment at all, however if Seagate

drives were indeed so horrible (i.e. if the conclusion that they are horrible were reliable),

I think I would expect more failures (then again, just because I expect something doesn't

mean it has to happen ). Of course not everyone will update their info and all that, so

who knows what's really happened, but I thought I'd add this for further pondering or

something like that.

Edited July 3, 2015 by alpenwasser

[Average Joe]

I had a bunch of Seagate Barracudas and I had never any issue with them.

[joaosambro]

Seagates hard drives are fine ., Used mine for close to 4 years now and no issue outside the usual wear and tear. So I can recommend their product for you..xx

[Revan654]

I had to toss four Seagate drives out today. Whens theirs a choice I would go with Western Digital/HGST over Seagate. Go with WD Black or Blue.