let's talk HDD's

[typographie]

I had 2 Seagate drives fail. Wester Digital Master Race

Every Seagate drive I've ever had has failed.  Never had a WD drive fail.  I'm with you on this one.  Got 10 WD drives right now.

 

Boy, the Storage forum sure is fun. So objective and informative.

 

For what it's worth I've only had one drive actually fail in about 20 years of PC use and it was a Samsung Spinpoint F3, which Seagate replaced in a timely manner almost a year outside of it's original warranty. In my experience, and looking at raw failure rates, they are overwhelmingly likely to outlive their usefulness.

[Lingrah]

I still recommend going Seagate over WD at all costs.

Why?  Because you had one WD drive fail?

[LarsReviews]

Boy, the Storage forum sure is fun. So objective and informative.

 

For what it's worth I've only had one drive actually fail in about 20 years of PC use and it was a Samsung Spinpoint F3, which Seagate replaced in a timely manner almost a year outside of it's original warranty. In my experience, and looking at raw failure rates, they are overwhelmingly likely to outlive their usefulness.

 

I was just sharing my personal experience, I don't remember the model number, but alot of people were experienced the same issue. First the drive was silence and within 8 month it became louder and eventually made sounds like a pump running dry. 

[Blade of Grass]

I'll just leave this right here:

Hard Drive Reliability Update – Sep 2014

Blackblaze's data is not accurate at all, and should be taken with a grain of salt. Their methodology is extremely flawed and they have not given enough information to be able to make an informed conclusion.

http://www.tweaktown.com/articles/6028/dispelling-backblaze-s-hdd-reliability-myth-the-real-story-covered/index.html

[handruin]

 

I'll just leave this right here:

Hard Drive Reliability Update – Sep 2014

Blackblaze's data is not accurate at all, and should be taken with a grain of salt. Their methodology is extremely flawed and they have not given enough information to be able to make an informed conclusion.

http://www.tweaktown.com/articles/6028/dispelling-backblaze-s-hdd-reliability-myth-the-real-story-covered/index.html

 

 

I still see no evidence to make claim that Backblaze's data is not accurate at all.  It may not be perfect, but it's got some value.  I'll need more than a crap Tweaktown article (if I dare call it that) to make the claim that there data isn't accurate.

 

Tweaktown offered nothing concrete in their dismissal of data BackBlaze shared.  They babbled on about how BB tries to acquire drives on the cheap and then roughly tries to tie together a correlation as to how that contributes to the failure rates?  They make empty statements regarding vibrations yet offers no proof that this was even a contributor to early failure.  Tweaktown even makes vague open statements of "many others" without citing any kind of source.  Even if vibration or temperature were contributing factors that means some consumer drives are better at managing these extremes than others.  I realize it's not that simple to conclude without more data but Tweaktown certain offers no useful additional information.   I'm aware that BB wasn't performing scientific testing but their data has have some relevance.  They have a huge sample size and with a bit more refinement, better conclusions can be drawn.  Yes, more questions come from this but they certainly have some great data points.

 

Tweaktown makes a bold statement about driving a truck through BB's methodology but they don't even realize the huge hole they left open in their almost useless rebuttal.

[Dimwitted]

Actually these are follow on articles from the original Backblaze posts. It became obvious very quickly that they screwed up the Seagate drives. Backblaze started up right at the time of the Thailand floods. When both Seagate and WD shutdown they were left hanging in getting drives. They bought external consumer drives, deshelled them and put them into racks. Even worse, they ran out of racks and custom made their own. Heat and vibration issues abounded. It was a total shitshow and the collected data on the relative merits of a particular drive is very suspect.

 

The bitch is that there's no one else doing anything. It's very aggravating. Bad data is better than no data I think. As Backblaze gets their shit together hopefully we'll see better, more accurate results.