The ones showing in that chart are likely still not WD. That number is how many they had running in those particular years so they are probably a couple years old.
HGST Ultrastars are the only spin drives I am completely sold on. Refurbished at approximately 20 bucks a TB is a silly good deal as long as refurbished doesn't scare you. I have no experience with Deskstars though, which I assume BB uses since they are the non-enterprise, cheaper drives.
Had a seagate slim baracuda was it? for ages and still going, its remarkable good tbh with the access being completely quiet and smooth. Seems random to me for failure, long time ago I bought 2xWD raptors, those 10k speed ones and one failed after a year or so and the other is still going 6 years later or more.
You'll notice the model with the highest failure rate was a Western Digital, followed by a Seagate model, followed by a Toshiba model. However, all brands had most of their drives as very very reliable, with only isolated models here and there with issues.
I'd say that HGST is the clear winner in those charts. HGST was basically Hitachi's hard drive division, which was later bought by Western Digital.
Back in the late 90's or early 2k's IBM sold its hard drive technology to Hitachi and Hitachi continued to produce products just as reliable as IBM hard drives were back then. I guess they still sacrifice higher yields for reliability due to their outstanding QA standards inherited from IBM.
I must add that HGST hard drives usually have really good transfer rates with the lowest failure rates on the market but they are also pretty outdated in terms of acoustics. I love to use HGST hard drives in servers but those who are keen on quiet PC's should be cautious about picking an HGST bacause many of these hard drives produce pretty audible vibration and noise when they are not mounted accordingly. I guess that's a sacrifice you can make for great performance and reliability.
Here is another point -- Each system is different. The foibles of a specific system might sync with a vulnerability of a specific product and make that product less likely to work well // last long on that specific system. In general if you have an issue with a product, try something else if you have the option.
Seagate used to make good HDDs but I remember reading that something changed a number of years ago. I think they moved their operations somewhere cheaper.
Interesting enough, there was another study (also by Backblaze) about the effects of temperature on hard drives.
The found that, with few exceptions, hard drives weren't particularly picky about being cool, so long as they weren't operated outside of their advertised operating band. Even in the case of the Barracuda and Deskstar exceptions, the correlation between failure and temperature was fairly weak, and one failed more often under cooler conditions.
There was another study, by Google, where they came to a similar conclusion: "In our study, we did not find
much correlation between failure rate and either elevated
temperature or utilization. It is the most surprising result
of our study."
I can't recall the source, but there was another study dealing with temperature, and they came to a similar conclusion but had temperature over time data, and they determined that while the min-max temperature didn't have a big effect on failure rate, the rate of change of temperature did. Those experiencing rapid swings in temp were much more likely to fail than those that undergo gradual changes in temperature. I apologize I can't find the source for that study, but on the surface it makes sense.
SSDs may be an entirely different ballgame though (both with regard to reliability and temperature's effects on reliability), because that's an entirely different operating technology than platter drives that are being discussed here.
Kyleran: "Now there's the real trick, learning to accept and enjoy a game for what
it offers rather than pass on what might be a great playing experience
because it lacks a few features you prefer."
John Henry Newman: "A man would do nothing if he waited until he could do it so well that no one could find fault."
FreddyNoNose: "A good game needs no defense; a bad game has no defense." "Easily digested content is just as easily forgotten."
LacedOpium: "So the question that begs to be asked is, if you are not interested in
the game mechanics that define the MMORPG genre, then why are you
playing an MMORPG?"
Here is another point -- Each system is different. The foibles of a specific system might sync with a vulnerability of a specific product and make that product less likely to work well // last long on that specific system. In general if you have an issue with a product, try something else if you have the option.
I agree with this for the most part.
I usually go by the two strikes rule. Something going wrong once is an accident, and accidents happen and I understand that - that's why we have RMAs, warranties, and the like. The same thing going wrong twice in short succession is no longer an accident, it's a pattern.
A good story I like to tell, coincidentally related to hard drives. I purchased a set of 4 3TB WD Red NAS drives, for my 4 bay NAS. Within 2 months, my NAS reported a SMART error on a drive. I RMA'ed it, WD was great about the RMA, and it was replaced. Two weeks later, a different drive. Two weeks again, drive #3. FOur weeks later, the final of the original drives failed. I'm still running the RMA replacement drives today.
Similarly, I had also purchased a WD Mybook USB drive, just for Time Machine on my work computer. It failed to even turn in after a couple of months. I RMA'ed it, WD again was very good about the replacement. The second unit failed within a couple of months. I RMAed it again- WD again very good about it. The third unit, I left unopened, sold it on EBay as NITB (because it was), and went and bought something different.
So the takeaway from that.
The first story, I had not two, but four failed units. But I'm still using WD drives today. That's because the failures weren't because of a manufacturing flaw. I had purchased them originally from Newegg. At the time, Newegg was shipping OEM drives just wrapped in a layer of bubble wrap, and I didn't think anything of it when I received the drives, until all four of those original drives failed within a very short time span. The shipment caused all four drives to fail. WD honored the Warranty regardless, and the replacement drives are great. Newegg now ships all OEM drives in cushioned boxes, presumably because they had such a high failure rate.
The second, yeah, that looks like it was just a bad product. So I got out of that product after two strikes and got something different.
Totally unscientific but in all my decades of building and owning computers I have only ever had 1 HD fail... yes, it was a Seagate.
"Social media gives legions of idiots the right to speak when they once only spoke at a bar after a glass of wine, without harming the community ... but now they have the same right to speak as a Nobel Prize winner. It's the invasion of the idiots”
― Umberto Eco
“Microtransactions? In a single player role-playing game? Are you nuts?” ― CD PROJEKT RED
Older Seagate drivers with the stepping motor had serious problems (talking about late 90s, early 00s here). Current Seagate drives are just below-par with most other manufacturers (as the graph of @Ridelynn shows). None the less, my Seagate SATA2 boot drive (250Gb) comes from an old pre-made PC and is now about 10 years old and still running smoothly. Think OP just had REAL BAD luck with his drives.
@Ridelynn Interesting story about HGST. I know those BAD BAD BAD IBM/Hitachi drives (they made me loose my nearly completed BG1 game), though warranty was very good. Because of this I never touched Hitachi at all and when I heard PS4Pro was stock delivered with HGST drives, I feared the worst, but it looks like I was wrong about that. Looks like whenever I need new HDDS, I'll hop to HGST as well (I do have a 250Gb boot drive to replace soonish).
Hard drive failures are commonly mechanical failures rather than electrical. Hard drives also don't put out very much heat, at least as compared to CPUs and GPUs.
Hard drive failures are commonly mechanical failures rather than electrical. Hard drives also don't put out very much heat, at least as compared to CPUs and GPUs.
Well, mechanical failures can be caused by heat as well. The reason rapid changes are more fatal is that it causes uneven expansion across the platters and other head arms and such, which makes them more prone to unbalancing and/or head crashing because of the extremely tight tolerances and high rotational speeds.
Right, but the point is "rapid" temperature changes. I'm not sure how you would get a rapid increase in temp in a hard drive on a PC short of a bearing going bad or something along those lines.
"The surest way to corrupt a youth is to instruct him to hold in higher esteem those who think alike than those who think differently."
Right, but the point is "rapid" temperature changes. I'm not sure how you would get a rapid increase in temp in a hard drive on a PC short of a bearing going bad or something along those lines.
It would take a pretty extreme amount of heat. Edit: To add, enough heat to melt and destroy the device, they are pretty tolerant until such temperature extremes. As in a literal torch against the IC and hope you can read the drop before it melts. Not day to day use.
And this would only effect the encoder/decoder circuit of the drives. Or flip flop register circuit used to pre load the encoder/decoder.
The mechanical section and platter if kept balanced shouldn't wear too much from heat. The effect is minimal to the magnetic fields being used to read/write.
Unlevel surfaces can cause mechanical damage for sure. Just like a washing machine will over time. Not every building is perfectly level, sometimes intentionally for element reasons. Such as high winds, potential earthquakes etc. Level your desk out, shim it. Less drive failures.
At certain times there have been problems with a serie of drives, maybe the component had issues or maybe they got in some microscopic dustparticles where they shouldn't be. A bad PSU can also kill otherwise good drives for you.
Seagate is not the only one who had those problems, IBM had a bunch. Maxtors had a habit of overheating themselves, WD had some bad and so on...
I have got a bunch of drives that died on my during the years and more of those were Seagate but then the majority of the drives I owned have belonged to that brand so I don't think percentage wise they stick out for me. I owned around 50 drives, about 5 of them have died while the others been retired. It was over 10 years ago last a regular HD died on me, I still have 4x3 Tb Sata drives in my computer, 2 Seagate and 2 WD.
3 of those were system drives, 2 of them were in my server (and I think that PSU had it's issues), system drives run a lot more then a media drive so they die easier.
I also killed a SSD (Corsair Force 3), I had 9 of those (4 raided IBM X-80), 2 Samsung, 2 Corsair and my current 1 Tb Western digital.
Yeah, too low number for any really useful statics anyways but I think drives generally have been of better quality the last 10 years (just a theory, and I do have owned better PSUs the last 15 years which makes things more complicated).
Right, but the point is "rapid" temperature changes. I'm not sure how you would get a rapid increase in temp in a hard drive on a PC short of a bearing going bad or something along those lines.
Stick a Cheetah or a Raptor in a box with no ventilation and turn it on. That will heat it up in a hurry. The problem is made worse once you look at multiple drives in a small space, like RAID arrays, NASes, data centers, etc., even if you are just looking at more vanilla 7200 or 5400 rpm drives.
In a typical home computer - Poor case design, improper installation, or inadequate case ventilation will do it. Especially case ventilation, because it's not just the energy from the hard drive that's contributing to the hard drive heating up, it's the energy of everything inside the computer.
Unlevel surfaces can cause mechanical damage for sure. Just like a washing machine will over time. Not every building is perfectly level, sometimes intentionally for element reasons. Such as high winds, potential earthquakes etc. Level your desk out, shim it. Less drive failures.
This part, I don't think it necessarily applies to hard drives in the same way as it would washing machines.
Hard drives aren't supporting a lot of mass on a vertical bearing. They are more worried about rotational friction from high speed than from high mass. Hard drives are pretty happy spinning in just about any orientation you care to put them in. What they will not like is moving from that position once they are spinning, since they act like gyroscopes and they will generate their own counter-force.
Right, but the point is "rapid" temperature changes. I'm not sure how you would get a rapid increase in temp in a hard drive on a PC short of a bearing going bad or something along those lines.
Stick a Cheetah or a Raptor in a box with no ventilation and turn it on. That will heat it up in a hurry. The problem is made worse once you look at multiple drives in a small space, like RAID arrays, NASes, data centers, etc., even if you are just looking at more vanilla 7200 or 5400 rpm drives.
In a typical home computer - Poor case design, improper installation, or inadequate case ventilation will do it. Especially case ventilation, because it's not just the energy from the hard drive that's contributing to the hard drive heating up, it's the energy of everything inside the computer.
Plus how often to users clean the dust out of their systems?
Kyleran: "Now there's the real trick, learning to accept and enjoy a game for what
it offers rather than pass on what might be a great playing experience
because it lacks a few features you prefer."
John Henry Newman: "A man would do nothing if he waited until he could do it so well that no one could find fault."
FreddyNoNose: "A good game needs no defense; a bad game has no defense." "Easily digested content is just as easily forgotten."
LacedOpium: "So the question that begs to be asked is, if you are not interested in
the game mechanics that define the MMORPG genre, then why are you
playing an MMORPG?"
I used to help make SeaGate HD's back in the day. Actually did the chemical filming on the reader heads. Anyways, as DMKano already pointed out and this goes for anything consumable, such as food, vehicles, homes, cell phones, computers and so on. Anything can end up being a lemon. I've bought several different brands of computer parts and the only honest thing I've figured out is one brand is a lot cheaper than another and does the same thing with a possible increase chance of failure. Take AMD vs. Intel processors and most people will swear up and down that Intel is better. Problem is, it's not, especially when you compare price tags. On the flip side though when comparing graphics cards, I would agree nVidia is leagues better than Radeon. In the end though, if any computer component is not or has not been handled properly, you WILL fuck it up. Static is no joke in the computer hardware industry.
There is also what they used to call Infant Mortality on computer equipment that dies early.
Kyleran: "Now there's the real trick, learning to accept and enjoy a game for what
it offers rather than pass on what might be a great playing experience
because it lacks a few features you prefer."
John Henry Newman: "A man would do nothing if he waited until he could do it so well that no one could find fault."
FreddyNoNose: "A good game needs no defense; a bad game has no defense." "Easily digested content is just as easily forgotten."
LacedOpium: "So the question that begs to be asked is, if you are not interested in
the game mechanics that define the MMORPG genre, then why are you
playing an MMORPG?"
Comments
HGST Ultrastars are the only spin drives I am completely sold on. Refurbished at approximately 20 bucks a TB is a silly good deal as long as refurbished doesn't scare you. I have no experience with Deskstars though, which I assume BB uses since they are the non-enterprise, cheaper drives.
I must add that HGST hard drives usually have really good transfer rates with the lowest failure rates on the market but they are also pretty outdated in terms of acoustics. I love to use HGST hard drives in servers but those who are keen on quiet PC's should be cautious about picking an HGST bacause many of these hard drives produce pretty audible vibration and noise when they are not mounted accordingly. I guess that's a sacrifice you can make for great performance and reliability.
MAGA
The found that, with few exceptions, hard drives weren't particularly picky about being cool, so long as they weren't operated outside of their advertised operating band. Even in the case of the Barracuda and Deskstar exceptions, the correlation between failure and temperature was fairly weak, and one failed more often under cooler conditions.
There was another study, by Google, where they came to a similar conclusion: "In our study, we did not find much correlation between failure rate and either elevated temperature or utilization. It is the most surprising result of our study."
I can't recall the source, but there was another study dealing with temperature, and they came to a similar conclusion but had temperature over time data, and they determined that while the min-max temperature didn't have a big effect on failure rate, the rate of change of temperature did. Those experiencing rapid swings in temp were much more likely to fail than those that undergo gradual changes in temperature. I apologize I can't find the source for that study, but on the surface it makes sense.
SSDs may be an entirely different ballgame though (both with regard to reliability and temperature's effects on reliability), because that's an entirely different operating technology than platter drives that are being discussed here.
In the 80s maxtor drives sucked.
Epic Music: https://www.youtube.com/watch?v=vAigCvelkhQ&list=PLo9FRw1AkDuQLEz7Gvvaz3ideB2NpFtT1
https://archive.org/details/softwarelibrary_msdos?&sort=-downloads&page=1
Kyleran: "Now there's the real trick, learning to accept and enjoy a game for what it offers rather than pass on what might be a great playing experience because it lacks a few features you prefer."
John Henry Newman: "A man would do nothing if he waited until he could do it so well that no one could find fault."
FreddyNoNose: "A good game needs no defense; a bad game has no defense." "Easily digested content is just as easily forgotten."
LacedOpium: "So the question that begs to be asked is, if you are not interested in the game mechanics that define the MMORPG genre, then why are you playing an MMORPG?"
I usually go by the two strikes rule. Something going wrong once is an accident, and accidents happen and I understand that - that's why we have RMAs, warranties, and the like. The same thing going wrong twice in short succession is no longer an accident, it's a pattern.
A good story I like to tell, coincidentally related to hard drives. I purchased a set of 4 3TB WD Red NAS drives, for my 4 bay NAS. Within 2 months, my NAS reported a SMART error on a drive. I RMA'ed it, WD was great about the RMA, and it was replaced. Two weeks later, a different drive. Two weeks again, drive #3. FOur weeks later, the final of the original drives failed. I'm still running the RMA replacement drives today.
Similarly, I had also purchased a WD Mybook USB drive, just for Time Machine on my work computer. It failed to even turn in after a couple of months. I RMA'ed it, WD again was very good about the replacement. The second unit failed within a couple of months. I RMAed it again- WD again very good about it. The third unit, I left unopened, sold it on EBay as NITB (because it was), and went and bought something different.
So the takeaway from that.
The first story, I had not two, but four failed units. But I'm still using WD drives today. That's because the failures weren't because of a manufacturing flaw. I had purchased them originally from Newegg. At the time, Newegg was shipping OEM drives just wrapped in a layer of bubble wrap, and I didn't think anything of it when I received the drives, until all four of those original drives failed within a very short time span. The shipment caused all four drives to fail. WD honored the Warranty regardless, and the replacement drives are great. Newegg now ships all OEM drives in cushioned boxes, presumably because they had such a high failure rate.
The second, yeah, that looks like it was just a bad product. So I got out of that product after two strikes and got something different.
“Microtransactions? In a single player role-playing game? Are you nuts?”
― CD PROJEKT RED
@Ridelynn Interesting story about HGST. I know those BAD BAD BAD IBM/Hitachi drives (they made me loose my nearly completed BG1 game), though warranty was very good. Because of this I never touched Hitachi at all and when I heard PS4Pro was stock delivered with HGST drives, I feared the worst, but it looks like I was wrong about that. Looks like whenever I need new HDDS, I'll hop to HGST as well (I do have a 250Gb boot drive to replace soonish).
"The surest way to corrupt a youth is to instruct him to hold in higher esteem those who think alike than those who think differently."
- Friedrich Nietzsche
Edit: To add, enough heat to melt and destroy the device, they are pretty tolerant until such temperature extremes. As in a literal torch against the IC and hope you can read the drop before it melts. Not day to day use.
And this would only effect the encoder/decoder circuit of the drives. Or flip flop register circuit used to pre load the encoder/decoder.
The mechanical section and platter if kept balanced shouldn't wear too much from heat. The effect is minimal to the magnetic fields being used to read/write.
Unlevel surfaces can cause mechanical damage for sure. Just like a washing machine will over time. Not every building is perfectly level, sometimes intentionally for element reasons. Such as high winds, potential earthquakes etc. Level your desk out, shim it. Less drive failures.
Seagate is not the only one who had those problems, IBM had a bunch. Maxtors had a habit of overheating themselves, WD had some bad and so on...
I have got a bunch of drives that died on my during the years and more of those were Seagate but then the majority of the drives I owned have belonged to that brand so I don't think percentage wise they stick out for me. I owned around 50 drives, about 5 of them have died while the others been retired. It was over 10 years ago last a regular HD died on me, I still have 4x3 Tb Sata drives in my computer, 2 Seagate and 2 WD.
3 of those were system drives, 2 of them were in my server (and I think that PSU had it's issues), system drives run a lot more then a media drive so they die easier.
I also killed a SSD (Corsair Force 3), I had 9 of those (4 raided IBM X-80), 2 Samsung, 2 Corsair and my current 1 Tb Western digital.
Yeah, too low number for any really useful statics anyways but I think drives generally have been of better quality the last 10 years (just a theory, and I do have owned better PSUs the last 15 years which makes things more complicated).
In a typical home computer - Poor case design, improper installation, or inadequate case ventilation will do it. Especially case ventilation, because it's not just the energy from the hard drive that's contributing to the hard drive heating up, it's the energy of everything inside the computer.
Hard drives aren't supporting a lot of mass on a vertical bearing. They are more worried about rotational friction from high speed than from high mass. Hard drives are pretty happy spinning in just about any orientation you care to put them in. What they will not like is moving from that position once they are spinning, since they act like gyroscopes and they will generate their own counter-force.
Plus how often to users clean the dust out of their systems?
Epic Music: https://www.youtube.com/watch?v=vAigCvelkhQ&list=PLo9FRw1AkDuQLEz7Gvvaz3ideB2NpFtT1
https://archive.org/details/softwarelibrary_msdos?&sort=-downloads&page=1
Kyleran: "Now there's the real trick, learning to accept and enjoy a game for what it offers rather than pass on what might be a great playing experience because it lacks a few features you prefer."
John Henry Newman: "A man would do nothing if he waited until he could do it so well that no one could find fault."
FreddyNoNose: "A good game needs no defense; a bad game has no defense." "Easily digested content is just as easily forgotten."
LacedOpium: "So the question that begs to be asked is, if you are not interested in the game mechanics that define the MMORPG genre, then why are you playing an MMORPG?"
There is also what they used to call Infant Mortality on computer equipment that dies early.
Epic Music: https://www.youtube.com/watch?v=vAigCvelkhQ&list=PLo9FRw1AkDuQLEz7Gvvaz3ideB2NpFtT1
https://archive.org/details/softwarelibrary_msdos?&sort=-downloads&page=1
Kyleran: "Now there's the real trick, learning to accept and enjoy a game for what it offers rather than pass on what might be a great playing experience because it lacks a few features you prefer."
John Henry Newman: "A man would do nothing if he waited until he could do it so well that no one could find fault."
FreddyNoNose: "A good game needs no defense; a bad game has no defense." "Easily digested content is just as easily forgotten."
LacedOpium: "So the question that begs to be asked is, if you are not interested in the game mechanics that define the MMORPG genre, then why are you playing an MMORPG?"