Western Digital or Seagate

[xShadow]

A longer warranty is good to have. No arguments there. However, you can't use warranty to gauge the quality of parts. That's like saying "This item is cheaper than the other item, thus it is clearly of lesser quality."

No, it's different from that. If I was making that assumption, I would have been putting Samsungs (pre Seagate buyout) into the high risk bracket, because they were only 80 bucks. That obviously wasn't the case. They had good reviews and good street cred and were very good drives in general. For something that fails easily and (relative to other computer parts) often, warranty means a much different thing. It's pretty much a piecewise function. Basically, if a hard drive has good reviews, street cred, and reliability in general, a low warranty may not mean much. However, once the other factors start going the other way, one might question why something has such a bad warranty. Now, you may bring it up as a subjective factor, but that's obvious. Nothing about parts analysis these days is objective unless you do extensive testing involving thousands of the said part. No one has money for that.

<insert rest of your rant here>

The problem here is that hard drives and what you're ranting about are completely different parts by nature. I've powered decent systems on shitty power supplies that came with cases. I've used all kinds of PS's, even fucked up a port or two and they still work fine.

So, the factors that make you wrong:
1. Power supplies are much more durable and trustworthy parts. Fans aside, they don't really involve many variables like mechanical failure. Sure, there are shitty ones, but even those will work fine to a point. Assuming you look at a certain cut of power supplies the entire market is pretty much all low risk.
2. They're cheap. You can get a very reliable (ie chance of failure=about 0%) PSU for relatively low prices.
3. Most importantly, the criteria for reviews is different. Hard drives are 1 and 0. It either fails after X unit of time or it doesn't. The user doesn't need to have any technological expertise to report that.




Edit:
.... Why am I doing this? All I came here to do was advise upon a hard drive that was on sale.....

....

Unless you have some really compelling argument I think I'm gonna end this here...

[limefc]

No, it's different from that. If I was making that assumption

It was a comparison not an assertion that you're making this assumption.

1. Power supplies are much more durable and trustworthy parts. Fans aside, they don't really involve many variables like mechanical failure. Sure, there are shitty ones, but even those will work fine to a point. Assuming you look at a certain cut of power supplies the entire market is pretty much all low risk.

Where the hell are you getting this from or is it a joke? There is nothing low risk about the power supply market when the percentage of fake certifications, fake QC and fake specifications is the absolute highest of any type of part used for a PC. The statement that power supplies are universally durable without many variables involved is laughable at best, disgusting at worst. They are only durable as long as you don't try to use them. Incidentally if you barely ever use your HDD, it will also last longer.

Speaking of risk, if you buy a poor quality HDD, it will at worst break after a while.
Buy a shit PSU on the other hand, and you can lose a lot more than the 100 dollars a HDD could cost or the 15 dollars the PSU cost. Personally I've only lost 2 motherboards and 1 graphics card to shit power supplies (power supplies, of which one still works!) and none except the graphics card was more expensive than 50 dollars so I consider myself lucky. No idea how many of my HDDs I lost to that though, haven't tried identifying the failed part on them. I am guessing a whole lot have failed ICs since every single one I've opened up had pristine disks and the spindle motor was more often than not fine.

Hard drives are 1 and 0. It either fails after X unit of time or it doesn't. The user doesn't need to have any technological expertise to report that.

Nothing is black and white. Damaged during transit. Damaged SATA port on motherboard. Power failure from a shit power supply, can't supply adequate 12V to spin up the HDD - rare but possible for it to be accompanied by seemingly normal function of the computer itself.
Shock protection activating and disabling the HDD temporarily which user assumed to be permanent damage. Logic board getting ruined by shitty PSU. etc
You sure as hell won't run out of possible causes for HDD "death" that are irrelevant of the quality of said part. Besides, if you wanted to  you could break down PSUs the exact same way. Fails after x period or doesn't: Who gives a fuck about the reasoning, right? As long as fail rate is low enough, I am perfectly good with using the PSU even if it has no transient filtering or has a defective +5VSB design which gives 6.8V spikes on +5VSB after the low quality capacitors fail and slowly kills motherboards like clockwork.

The most important knowledge you should take out of this is that power supply failure can be transparent to the user. It will continue to run your computer while killing off other parts or causing stability issues that can be explained by another part failing, and fixed by replacing it (due to different out-of spec voltage tolerances, for example).

There is bad luck too. Speaking of newegg reviews, a lot of reviews are posted immidiately after purchase and installation and coloured by expectations. "After hearing that WD is the best thing since sliced bread, I bought a WD drive, but it arrived DOA. I think I just got unlucky and I should return it without making a fuss."

"After buying a Seagate drive it arrived DOA, my friend upon being told my story proclaimed that Seagate is a pile of shit and I should have expected that and I saw some people agreeing with him on the internet. I will thusly post a review saying that I got this DOA and replaced it with something else because it was a piece of shit."

Mass psychology!
If you want to take the opinion of newegg apes for any purpose other than information about possible problems and their solutions, then be my guest. You're however not going to convince me that they have any credibility. I am almost tempted to read HDD reviews to see what sort of retarded bullshit they write there while claiming to be tech experts.

Bottom line is, you're arguing for the credibility of idiots. I will grant you that you don't have a whole lot of other options than their feedback for assessing the quality of HDDs but that doesn't make me wrong. It just makes the only choice for decision making advice a bad one.
Both my statement that newegg feedback is full of raving morons and that warranty is not an indicator of quality are true.

[raandomer]

That's difficult to supply. Generally when I assess hard disks I sort the reviews list on newegg by ownership length and then look at how many people have them failing past one year. The only authoritative study on hard drive reliability (Google's) in recent history will pretty much tell you why. Just look at some of their graphs.

(click to show/hide)

It's not like there's anything quantitative to be taken out of that, though. There are simply far too many random variables involved in reviews, and trying to classify them into some standard deviation is probably something that would give me a headache; I'm a computer engineer, not a statistician. At best, you can take some ballpark estimate of a hard disk's general performance. Then there's also "the word on the street"... ie shit that people have probably read in some random tech magazine or hardware site (or a set of them). I don't trust those much, unless they've actually done some statistical studies or they've brought in a person who is seriously an expert in the field (PHD is the minimum). I just look at reviews and then mix in a bit of what I've heard on the streets (of the internet lol) when I make an assessment.

tl;dr:
Generally, it boils down 3 risk categories: high, medium, low. With at best mediocre reviews (which is for their 200 dollar line) and a one year warranty, Seagate naturally goes into the "high risk" pool. There isn't any way around this. I mean when When they have a 1 year warranty on their drives, that also says something about their expected quality. When I can get a WD Black for 145 bucks, which is guaranteed to at least give me some place to put my shit for 5 years (ie low risk category), why would I bother?

Ok, let me get this straight, you rate your hdds by these criterias:

• reviews on the newegg + web
• warranty period

ah, so it all boils down to being a sheep, since your saying seagates are shit because other people say its shit. The correct way to go about judging a product is to identify what properties make it "bad" (ie the 7200.11 firmware) not just going by word of mouth. And about the warranty: they still come from the same factory, with the same parts but since warranty has been cut then therefore they must be lower quality, reasoning ftw there. There is no correlation between warranty and reliability (just google it and you'll get plenty of papers on it), warranty is essentially a very powerful marketing tool that can easily fool people into thinking a product is reliable. My theory for the recent warranty cuts is that its now essentially become a oligopoly (off the top of my head i think only toshiba, wd and seagate are left) and wd probably made some close door deals with seagate.

TLDR:

• reviews/word of mouth are a bad indicator of reliability, everyone will rant about bad drives but very few (proportional to the bad reviews) will report having no problems, hence the data is skewed
• a better way of gauging reliability would be getting return rates from distros or better yet from the company themselves (but i doubt they'll reveal that figure to some random internet guy)
• warranty is also a bad indicator of reliability, since companies will most likely base warranty period on something like this: if profit from selling more units due to have advertised x warranty period outweighs warranty overhead for the x period then its all good

[Freedom Kira]

On the reliability vs. warranty thing, I want to point out one thing. Manufacturers typically select their warranty periods based on the expected percentage of their products dying by the end of the period, be it 1%, 2%, whatever.

So, at the very least, there is some correlation.

However, that correlation is not useful in determining a drive's reliability. The main reason is that the tolerance will differ between manufacturers. In the instance of Seagate vs. WD, it's entirely possible that Seagate only wants to cover 0.5% of their drives with their warranty, while WD is willing to cover 5%. Depending on this, it is entirely possible that Seagate's products are equally or more reliable than WD's (with the theoretical numbers, about 0.5% of Seagate's drives die within a year and 5% of WD's drives die within 5 years).

I understand the mentality of using a warranty period to gauge how much a manufacturer believes in their drives, though. Generally, the longer a manufacturer is willing to stand by its products, the longer those products are expected to last. A manufacturer slapping a 10-year warranty on stuff that dies every month is asking to go bankrupt. Thus, it's only natural to aim for stuff with solid warranties as opposed to stuff that's only guaranteed for a year or so.

[xShadow]

God dammit what did I get myself caught up in this time...

<wall of text responding to quote 1>

Lol. Where are you getting all of these horror stories from? I simply meant if you do proper research and pick out a certain subset (Good PS⊆All PS) of power supplies that has been proven to work well, you'll generally get something that serves you well for a long period of time... and for not much money. I don't think I'm wrong here, because this Corsair 650TX has been serving me for ages, and it wasn't terribly expensive. But who cares, right? All power supplies are guerrilla soldiers waiting to pop out with rifles and kill your entire system.

<insert rest of wall of text>

You're making a pretty moot point here. If you can hear it clicking and it's doing random shit that no other drive in your system is doing, it's probably failing. It's a 1 or 0 relationship, with some very small amount of standard deviation; most motherboards do NOT come with failing ports. Like you pointed out yourself, power supplies are transparent killers. Anyone would have trouble pointing out when their PS is fucking up. Of course their reviews would be sketchy.

No seriously you have some kinda trauma here. I'm sorry for your bad luck with PS's...

ah, so it all boils down to being a sheep, since your saying seagates are shit because other people say its shit. The correct way to go about judging a product is to identify what properties make it "bad" (ie the 7200.11 firmware) not just going by word of mouth. And about the warranty: they still come from the same factory, with the same parts but since warranty has been cut then therefore they must be lower quality, reasoning ftw there. There is no correlation between warranty and reliability (just google it and you'll get plenty of papers on it), warranty is essentially a very powerful marketing tool that can easily fool people into thinking a product is reliable. My theory for the recent warranty cuts is that its now essentially become a oligopoly (off the top of my head i think only toshiba, wd and seagate are left) and wd probably made some close door deals with seagate.

*Facepalm*
You won't know what factors make it bad without having other people test it for you! You'll end up going off word of mouth either way! Because guess what? You don't actually understand what's wrong with the firmware yourself. You can't just open it up yourself and go "oooooh, this line looks a bit off, I think I'll hold off on this hard drive purchase!" There's nothing but end-user experience to go off of it in the end. It's about being an "informed sheep". You need to take everything you see with a grain of salt, but take it into account nonetheless. That's the only information you have. You don't know half of what goes on in that microcontroller on your hard disk. You don't even know 25%. You don't even know 10%. Even if you wanted to, the manufacturers won't let you know. The only reliable method is massive end-user testing and then reporting the results (see: Google paper). What you can do with reviews in front of you is something halfway similar.

reviews/word of mouth are a bad indicator of reliability, everyone will rant about bad drives but very few (proportional to the bad reviews) will report having no problems, hence the data is skewed

That's wrong. Just look at any hard drive's review page and you'll see it's wrong. None of them would be rated 4 stars if that was the case.

I understand the mentality of using a warranty period to gauge how much a manufacturer believes in their drives, though. Generally, the longer a manufacturer is willing to stand by its products, the longer those products are expected to last. A manufacturer slapping a 10-year warranty on stuff that dies every month is asking to go bankrupt. Thus, it's only natural to aim for stuff with solid warranties as opposed to stuff that's only guaranteed for a year or so.

Bingo.

[Dhruv]

I'll have to agree with Kira about the warranty issue. The longer the manufacturer is ready to stand by it's products it's likely more reliable they are. Take an example, i currently have a Seagate 2TB FreeAgent GoFlex Desk and seagate gives me a 3 year warranty for it. It's working perfectly and it turns out that was the last faultless batch manufactured by seagate before they started fucking up their drives around December 2011. Mine was manufactured in October 2011.
But this can't be taken as the only factor too. Some drives in the fucked up batch work good too like one of those my friend has. It was manufactured in January 2012 and is working good till now.

But agreed... consumer reviews are practically shit as they know nothing about the drives and it's attributes etc. They can't be relied upon completely.

[xShadow]

But agreed... consumer reviews are practically shit as they know nothing about the drives and it's attributes etc. They can't be relied upon completely.

And I have to reclarify here (not responding to you directly) before anyone else brings it up.

What I'm not doing is telling you to pay attention to what the reviews says. Like "LOLOLO GOOD DRIVE LO SHIT DAWG". That doesn't matter. What you need to do is pay attention to the number of 1 egg votes and note how long they say the drive lasted. It's very simple.

*1 egg generally means the drive went bad, regardless of their tech expertise.
*Time of failure says something about hard drive quality, based on the google study.

That's data you have to weigh yourself to get an idea of hard disk reliability. What people actually say in reviews is irrelevant. You're not reading them for that.

[kitamesume]

the only things that customer reviews contribute is the sheer amount of RMA/DOA, they're helpful to indicate if that certain line of product is heavily unstable. theres also some rare occasions that other products has zero negative feedback, its either they didn't care about posting DOA/RMA or the customer service is top notch that they don't get the customers disappointed.

as what xshadow says, if 99% of the feedback has gone into 1egg saying DOA or RMA after #days that would mean that line of product is ridiculously suspicious.

[raandomer]

or, dont look at that crappy skewed data on newegg and maybe talk to a distro. They're always happy to provide me info on when an abnormal amount of drives are being returned. Or if you arent on friendly terms with any distro managers just talk to a few local shops, granted the sample size is smaller but you'll still get a better indication of reliability than reading off newegg.

I'll state this again, you cant force people to post bad or good reviews on a product, but in most cases people are more likely to complain about a bad product than a good one (there are plenty of surveys on this, so dont just outright say i'm wrong, READ up on it, like those academic papers on warranty reliability correlation). Hence your data is wrong even before you start analysing it.

[xShadow]

I'll state this again, you cant force people to post bad or good reviews on a product, but in most cases people are more likely to complain about a bad product than a good one (there are plenty of surveys on this, so dont just outright say i'm wrong, READ up on it, like those academic papers on warranty reliability correlation). Hence your data is wrong even before you start analysing it.

1. You take the results with a grain of salt.
2. Honestly that doesn't even matter. It only would if I was looking for a RATIO of good to bad. I've stated this like a million times before, but that's not really what I'm doing.

Though if a drive had like 5% bad reviews out of 1000 votes.. considering what you just said it'd be a damn good one. You don't analyze every data set the same.

[Tatsujin]

I'll have to agree with Kira about the warranty issue. The longer the manufacturer is ready to stand by it's products it's likely more reliable they are. Take an example, i currently have a Seagate 2TB FreeAgent GoFlex Desk and seagate gives me a 3 year warranty for it. It's working perfectly and it turns out that was the last faultless batch manufactured by seagate before they started fucking up their drives around December 2011. Mine was manufactured in October 2011.
But this can't be taken as the only factor too. Some drives in the fucked up batch work good too like one of those my friend has. It was manufactured in January 2012 and is working good till now.

But agreed... consumer reviews are practically shit as they know nothing about the drives and it's attributes etc. They can't be relied upon completely.

You're right and you're also wrong. When you do your research, do proper research on those who writer more details about what happened, how is it fairing, etc. Not just "shit flopped this product sucks". You can also go to some geek websites and google the product name "review" and see what you can come up with.

If you want a reliable and durable HDD, get the WD black edition. Period.

[Dhruv]

I'll have to agree with Kira about the warranty issue. The longer the manufacturer is ready to stand by it's products it's likely more reliable they are. Take an example, i currently have a Seagate 2TB FreeAgent GoFlex Desk and seagate gives me a 3 year warranty for it. It's working perfectly and it turns out that was the last faultless batch manufactured by seagate before they started fucking up their drives around December 2011. Mine was manufactured in October 2011.
But this can't be taken as the only factor too. Some drives in the fucked up batch work good too like one of those my friend has. It was manufactured in January 2012 and is working good till now.

But agreed... consumer reviews are practically shit as they know nothing about the drives and it's attributes etc. They can't be relied upon completely.

You're right and you're also wrong. When you do your research, do proper research on those who writer more details about what happened, how is it fairing, etc. Not just "shit flopped this product sucks". You can also go to some geek websites and google the product name "review" and see what you can come up with.

If you want a reliable and durable HDD, get the WD black edition. Period.

Eh.... Didn't you tell me to get WD Elements?

[limefc]

xShadow, because you aren't putting much effort into your posts, neither will I.

Lol. Where are you getting all of these horror stories from?

I only posted one, which involved a double transistor 5VSB design used in a couple of Bestec power supplies that were utilized by OEMs. 5VSB had no SP, OCP and killed motherboard when the caps failed. I know of plenty more though.

Rest is just theory as to why a hard drive might be dead which will earn the HDD a 1 egg rating but is irrelevant of the HDD quality/age.

But who cares, right? All power supplies are guerrilla soldiers waiting to pop out with rifles and kill your entire system.

Straw man.

You're making a pretty moot point here. If you can hear it clicking and it's doing random shit that no other drive in your system is doing, it's probably failing.

Only one drive I ever owned failed from "clicking". Seagate back when it was considered to be the absolute best by the way. Rest still spin up today and act as if nothing is wrong.

Customers don't know why shit fails. They can write an elaborate article and could have owned it for 20 years and still say nothing useful.

By the way at around the cut you're talking about which is Corsair TX, you can still buy trash. Although it is pretty rare these days, thank god.
Assuming your cut starts at around 80 dollars for a PSU that is.
Below that you're swimming in trash.

And I don't have bad luck with power supplies. I KNOW why they killed my equipment. It wasn't unlucky, it was inevitable and the only factor I did not have at hand was the time - when will it happen.

One more thing, if a massive reliability problem plagues drives then you will hear about it and it won't be from the idiots at newegg feedback section because data on return rates is published. Something like a 2.5-10% return rate from a single brand or model is going to be noticed.
That's why everyone knows that Hitachi shit lives up to the Deathstar name right now.
And that WD Black 2TB (and all other 2TB drives) had high return rates... except Samsung EcoGreen and surprisingly WD20EARS.

[Tatsujin]

I'll have to agree with Kira about the warranty issue. The longer the manufacturer is ready to stand by it's products it's likely more reliable they are. Take an example, i currently have a Seagate 2TB FreeAgent GoFlex Desk and seagate gives me a 3 year warranty for it. It's working perfectly and it turns out that was the last faultless batch manufactured by seagate before they started fucking up their drives around December 2011. Mine was manufactured in October 2011.
But this can't be taken as the only factor too. Some drives in the fucked up batch work good too like one of those my friend has. It was manufactured in January 2012 and is working good till now.

But agreed... consumer reviews are practically shit as they know nothing about the drives and it's attributes etc. They can't be relied upon completely.

You're right and you're also wrong. When you do your research, do proper research on those who writer more details about what happened, how is it fairing, etc. Not just "shit flopped this product sucks". You can also go to some geek websites and google the product name "review" and see what you can come up with.

If you want a reliable and durable HDD, get the WD black edition. Period.

Eh.... Didn't you tell me to get WD Elements?

Oh that's stupid of me. Yes, the WD Elements is for external. As for the WD black edition, that's for internal usage.

[Dhruv]

Eh.... Didn't you tell me to get WD Elements?

Oh that's stupid of me. Yes, the WD Elements is for external. As for the WD black edition, that's for internal usage.

Whew.... relieved... i almost started doing the calculations and stuff to buy that Disk....
What are these Guys saying about WD Black Having a high return rate? I Thought Black Along with Velociraptor were the best Western Digital internal HDD.

[limefc]

2TB, 3TB drives sold between October 2011 and April 2012 had high return rates within 6 months of being sold. Ones that didn't have insanely high return rates were 2TB WD Greens and Samsung EcoGreens.

http://www.behardware.com/articles/862-6/components-returns-rates-6.html

Data is directly from a French etailer. Only high capacity drives, 1.5TB and more have abnormally high return rates.
According to that, Samsung is the most reliable with both Seagate and WD trailing it and from my experience with Samsung drives.. I can't disagree. Like 90% of the drives around my place are Samsung and they've been consistently good, although my server which is the only system in 24/7 use has Hitachi Deathstars and a single WD blue.

Back when I bought the Deathstars, Hitachi had an overall returns rate of only 0.92% for a #1 spot. Oh how the mighty have fallen.

[Dhruv]

Ohhh... that is a Deskstar.... i wonder why you people call it a Deathstar
Looks like no drive is reliable and durable enough.... But i don't see WD Black name on the list.
And As far as WD Elements Goes... tatsujin has had an extremely good experience with it.... he has 4 or 6 of those drives and none has failed till now.

[limefc]

2TB WD Black is on the list.

If you want to buy something under 1.5TB internal, buy something from WD, Seagate or Samsung and as long as you don't get really bad luck you'll get what you want. If it's bigger or equal to 1.5TB then go for Samsung.

There is no reason to prefer WD Blacks over Samsungs or Seagate XTs at 1TB level when the concern is reliability.
There is reason to prefer Samsung over anything at 1.5TB and higher capacities.

WD for externals is fine. WD elements is only up to 1TB anyway and that's not where the massive failers are (that's 2TB capacity drives).

Elements is most likely a WD Blue or WD Green by the way. My external has a WD Scorpio Blue in it, for instance. Blues and Greens in WD based externals are dime in a dozen and blues can last for a very long time.
Externals are nothing more than a SATA&SATA power->USB2 or USB3 (optionally with a power brick) adapters and a plastic/metal box to store that and a 2.5 or 3.5 inch internal SATA drive in.

[Dhruv]

I am going by the name not the model number... That way I don't see Black written over there... But I take your word for it.

If you have read the first post of the thread I want to buy a disk with a capacity of 2TB or 3TB and as far as elements goes it comes upto 3TB in capacity... Till 1TB it is portable and after that it is a desktop drive. I am not aware of its constituent drive. Probably tatsujin can tell us that.

Looking at that link it seems that Seagate Barracuda is one of the worst drives.

[limefc]

It's there - 2TB Caviar Black. Also the bigger externals are NOT WD Elements - elements is only up to 1TB. The ones you're talking about are - I believe, WD My Book Essentials and those contain in them a WD Green.
WD green is listed. 3TB model is in the over 5% return rate section and two models of the 2TB are in the 2TB section.

Anyway you'll get below average reliability at 2TB and above level even if you buy Samsung which is by far the best. So my point stands, get a Samsung. Or if you can't, stay the hell away from Seagate and Hitachi and go with a WD Green based external. Just don't ask for solid reliability, that's asking for the impossible.