what has happened to hard drive reliability?

[cadder]

Are mechanical hard drives less reliable than they were 3-5 years ago?

My current desktop machine was built with a WD Black 640GB drive. When win7 came out I swapped in a WD Black 750 and installed Win7 on it. I recall researching drives on newegg and these 2 drives had very good reliability. My methodology is to pick a product, look at the reviews on newegg, and look at the percentage of reviews that are 3 eggs and above. I'm thinking that a good product will have 90% or more of its reviews at 3 eggs or higher. The older WD Black drives fit this. I've researched other products this way and many times I can find a product I'm interested in that has high percentages of good reviews.

Now I'm interested in upgrading hard drives in my desktop to meet new storage needs, maybe in the 3TB range, so I start researching drives again. Using my methodology of looking at review percentages, the results are appalling. Among 3TB drives with good numbers of reviews, the best that I can find is only 83%, and many of the name brand drives range from 76% down to 54%. I don't want to buy a product where one out of 4 buyers didn't like it, I certainly don't want to buy a product where almost half of the buyers hated it. Why can't I find a product that is closer to 90%? Are these numbers not as meaningful as I think they are? (Maybe there are a lot of people that buy good drives and never enter reviews for them.)

I was also interested in external drives in 3TB range. My rating method gives a high of 77% down to 43%. I don't want to buy any of these.

 

[kanewolf]

If you paid the same $/GB for a new disk that you did for the WD 640GB disk you should expect similar reliability. Should you expect similar reliability if you pay 1/10th per GB ? That is a difficult question. Buying the drives with a 5 year manufacturers warranty is about the best you can do. The WD Red Pro disks are intended for continuous usage in a NAS environment and in other ways similar to the WD Blacks. Maybe the Red Pro drive is where you should look.

 

[USAFRet]

Review percentages are pretty useless.

People will complain about anything.
People who have issues are far more likely to write a bad review than someone who has no issue is likely to write a good review.

 

[cadder]

If you paid the same $/GB for a new disk that you did for the WD 640GB disk you should expect similar reliability. Should you expect similar reliability if you pay 1/10th per GB ? That is a difficult question. Buying the drives with a 5 year manufacturers warranty is about the best you can do. The WD Red Pro disks are intended for continuous usage in a NAS environment and in other ways similar to the WD Blacks. Maybe the Red Pro drive is where you should look.

My first hard drive was a 20MB, at a cost of $22.50/MB. If I but a new 3TB drive for the equivalent price of $67.5 million, then it better be a VERY reliable drive!

And unfortunately some of the red drives have pretty low ratings too.

Review percentages are pretty useless.

People will complain about anything.
People who have issues are far more likely to write a bad review than someone who has no issue is likely to write a good review.

I understand this but I think it is still valid to compare reviews for one model vs. reviews for another model. If one model has 25 people out of 100 that complain while another has only 5 people out of 100 to complain, shouldn't this mean something? What I would like to know though is for the drive model that has complaints from 25 people out of 100, how many other people are there that are happy with their drives and don't enter reivews? There are likely more than 75 out of 100 drives that are good, but are there 90 out of 100, or 95 out of 100? That would make me feel better about buying a product that has less than impressive reviews.

 

[USAFRet]

There is no real way to know how many buyers did not write a review, good or bad.

For reviews, I tend to look at the meat of what they wrote rather than percentages.

It is a known fact that people are paid to write glowing reviews, skewing the ratio.
I'd be VERY surprised to find that the opposite did not also happen.