Comparing Seagate vs Western Digital (WD), Toshiba and HGST hard disk failure rates and lifespans. 

Which drive is the best? or which maker? lol

Ty for saving me time bro

@@neoturfmasterMVS if you were looking for a second hand hdd, I prefer the HGST. However, some are selling this broken online without any notice from the buyer, so buy at your own risk.

exactly. It is like asking what car is best. Well wtf? Obviously, you would need to refine the spectrum to certain categories. Seagate makes the best NAS drives. WD makes the best basic computer storage. Toshiba makes the best paperweights.

​@@HappyBuddhaBoyd ur sure? Toshiba has the best overall durability over the others for NAS and even used it as surveillance storage even though it was not designed for that, Toshiba hdd has great quality of head performance and durability

Thank you for the comment. There is a lot more in the data also about specific models and I plan to look more closely at the larger capacity drives to see what they look like. Backblaze do not seem to deploy any NAS disks, probably for reasons I called out in my previous video, so there isn’t any data as yet on Iron Wolf, Iron Wolf Pro, or recent info on the WD Reds, though the Reds they deployed years ago seemed to fail pretty badly. So far I also had great experiences with the Exos disks, and IW Pros, but price wise right now Exos seem to be the pick.

This sounds like one of the worst stories I heard, and this would suck. I do believe Seagate can replace warranty returns with 'Reburbished' drives and I seem to remember seeing that in the T&Cs, but I believe these have a close proximity to new drives.They should however still provide the warranty period on these, and in fact in most countries, I think that is a legal obligation. It is possible the warranty date starts from the original warranty and you don't get 5 additional years however. It sounds very raw, and these are the kind of experiences that lead to never buying a vendor again. Which I completely understand. Sorry you had this experience, it really sucks to be in this position.

Thank you 🙏

@kerai129, thank you for the comment and the encouragement. It is appreciated. I think like many things, if you buy a certain brand and it serves you well then you tend to stick with what you know. If you have a bad experience, you are likely to change and less inclined to give them another chance if you move onto a good experience somewhere else. As long as pricing remains inline and the products are available when you need them of course. And thanks for the support on the channel. It is a slow and gradual process building up a channel with lots of other excellent content available. I will keep plugging away with it. 😎

bro cam u tell me does HDD enclosure may cause of hdd failure. i means can it badly impact hdd to eventually fail it.

Yeah, people become glued to and loyal to the brands they know. I have used Asus motherboards on all of my own personal computers since 2001, I will be changing over to either Gigabyte or MSI for my next computer due to Asus's anticonsumer behavior recently. I also used WD drives since 2001, and like you my only drive failures were with Seagate and I swore off them altogether in 2008 and have been using WD black or gold drives primarily for the last few years with a pair of Toshiba X300's in there because they were to cheap / cost effective to ignore. The X300's kind of opened my eyes in that they were 40$ cheaper than the WD equivalents, but seem to be holding up just as well, my only gripe is they are audibly louder when they spin up. Earlier today I bought a pair of Seagate Exos 18TB enterprise drive figure 16 years is a long enough time to circle back and try again.

I can’t talk to vibration and humidity as that data is not recorded. However, they temps are recorded and are not in a perfectly controlled environment as it varies over time. The data also implies there are multiple locations as different groups of disks are in obvious alignment for temps. But they said; the locations so all appear to be generally temp controlled and although I have not yet done a detailed analysis of this part of the data, I did graph it all and the temps seem to be within manufacturer bounds. At least generally. It might be possible to infer how the disks are distributed physically and how temp impacts reliability. And in the most recent data, Backblaze has started to record more details about locations and chassis disks are in. So could be some findings there with some digging.

@@sometechguy I'm not sure if it really matters if they are precisely in the same temperatures so long as the drives are equally represented and within standard deviation in each cluster. If there is equal representation in each data center and they are finding Xdrive by manufacturer A is failing moreso in those temps / that environment than YDrive by that same manufacturer then that is the type of info I care about when making purchasing decisions.

Hi, I don't know if you mean a single disk enclosure, like a caddy, or an enclosure that can store many disks in an array. But the only things that are likely to add to a risk of disk failure are excessive heat and movement, especially when the disks are unparked and being used. Sudden G-shocks can cause head crashes, rotational movement of the drive is not good due to the gyroscopic affect of spinning disks, and vibrations caused by other things in the enclosure, such as other disks. So it isn't the enclosure that would cause a failure, but the the environment the disk is subjected to while in the enclosure. If the disk is in use, making sure it is kept still on a stable surface will help. When disconnecting the drive, give it time to park and come to a rest before moving it.

The quality and stability of the 12V DC power supply is a major risk factor for HDD enclosures. Then it's just a matter of whether you can check the SMART data via the enclosure

It’s down in the description, but will add it here also. ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-UFDF39TRsl0.html

A pleasure, thanks for the feedback.

Hi, and thanks for sharing your experiences. I did some in depth analysis on the historical failure rates for HGST/WD Ultrastars and the Exos drives. The HGST/WD drives seem to have a reasonable edge and the AFR rates seems to remain consistently lower, that video is here: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-icNbexYV3M4.html . But I agree, in that I use both and have had good experiences. I ordered some Ultrastars and Exos disks this week for some additions and to replace some aging WD Reds which themselves have over 8 years on the clock. The statistical nature of failures, especially when they are down in the 0.5% to 1.5% range mean that some people will have a bad experience, but many won't. And its always those with bad experiences that share them, understandably. So good to have a positive and balanced piece of feedback.

I had a couple of IBM and then Hitachi drives. They were very good. It was very sad the first time when IBM sold its drives industries to Hitachi and then when Hitachi did the same to WD.

I released some other videos with more in depth model comparisons, but I think the summary is that consumer and enterprise drives don't have significant differences in failures from the same manufacturer, at least for the models in the data. Of course, enterprise drives have better warranties, and for larger capacities they are often cheaper per Tb. So personally, I buy enterprise disks, though they can be a little noiser if they will be running somewhere that is a problem. Overall, WD seems to make the most reliable disks. Toshiba seem a little ahead of Seagate but actually there is far less data on Toshiba drives, so I don't think thats a perfectly safe assumption. Watch the WD Blues, as there are quite a few SMR drives, which are not good for long writes. But they are the cheapest option for the 2-4Tb range. Overall, I would recommend WD currently based on all that analysis as they appear the most robust. Check this video for model and price comparisons for their disks. ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-QDyqNry_mDo.html Hope this helps.

@@sometechguy ok. Thanks for your opinion and advise. I think both Toshiba and WD are SMR type. I used these 3 brands of 2.5" sata drive before. And so far so good. Finally, I choose the Toshiba 3.5 sata hdd. It very hard to decide either purchase Toshiba or WD based on your review. I choose Toshiba because it cheaper quite a lot from WD. And they both offer 2 years warranty here.

Exos or Ironwolf Pro.

I did some in depth AFR analysis on the HGST drives, and they seem to go on and on. So second hand units could be great buys if they run nicely and pass a solid test when you get them. I am sure they will go on to serve you well.

they did. All HGST drives are properly working. Took a risky gamble since I don't have any cash to spare. That was a year ago. Now I bought anaother one a few days ago and no errors occured after serious testing. 👍

I also buy HGST second hand for active storage like games and downloads, or if I clone a drive from a laptop. In my experience I’ve noticed that they are really noisy and throw off some serious heat. I try to stay around the $30.00 price range which can be 3 to 4TB. A much better deal than 2TB for $47 plus tax new.

@@zeroturn7091 nice. As long as it doesn't create any clicking sounds, you are good. Its normal for hdd to heat up since you're using it as active storage, just try adding some pc fans it that bothers you. Unless the rpm of your drive is 5400 instead of 7200, it will likely create some noise and generate heat if connected to your pc.

Thank you, appreciate the comment.

I did another video directly comparing AFR rates from various drive lines from Toshiba, Seagate and Toshiba. For Tosh the data is all for MG07 and MG08 with data for around 55k drives. For Seagate, there was a great deal of data for the X18 line, but a lot of data for X12, 14 and 16. So unfortunately I can't compare those directly, but the data would likely give you a good idea about how each of these manufacturers are doing, and the direction of travel for reliability. The spoiler is that the Toshiba lines are a bit better, and more consistent over time also. You can check it out here : ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-icNbexYV3M4.html Usual caveat, more reliable doesn't always mean 'better' though its a really tangible data point. Price, availability and warranty/support will play a part in that also.

Hi, and thank you! The Backblaze data doesn't contain many NAS disks, its around 2100 in total out of the 450k total, and these were nearly all Normal WD Red disks, not Pro disks. These disks also suffered pretty high failure rates and were removed. So there isn't any comparative reliability data there for Pro vs Non-Pro unfortunately.

@@sometechguy thank you so much ! Huge help for me as I am planning on setting up a home NAS for family photos ;)

I have quite a few Exos, and I specifically have the X18 18Tb disks, and they have been great. I did some deeper analysis videos which you can find on my channel, and though WD Ultrastars (which I also own) appear to have better reliability statistically, I don't think either of the options is bad and if the Exos drives come at a better price point, I wouldn't hesitate to buy them. I hope they serve you well.

Thanks for the comment, and yes this gives an overview of drive deployments and failure rates but as I said there is a lot of data here and some deeper analysis is needed to arrive at firm conclusions. I did this, and here are three follow ups with in depth AFR analysis on the data to directly answer those questions. ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-icNbexYV3M4.html - Comparison of failure rates for Seagate, WD and Toshiba drives across a range of models. ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-IgJ6YolLxYE.html - Failure rate comparison of a range of 10tb - 16tb drives from these three. ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-l_YqdVGcC0o.html - Failure rate comparisons of 4tb, 6tb and 8th drives from these three with comparative failure rates for Desktop vs Enterprise disks over long periods. Hopefully that helps. There is so much data, that it needs slicing different ways to get specific insights from it.

so are they better to use now?

This is true, but interestingly the data shows that consumer drives are not significantly more likely to fail in most cases. And those consumer drives, often with only 2 years of warranty are surviving just as long as their enterprise counterparts, often over 7 years. I have a video that compares data on exactly this if its interesting : ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-l_YqdVGcC0o.html

It's interesting how the anecdotal experiences of people vary so much. I did a few videos looking at failures on the different manufacturers and drive models and despite the data, the comments are full of people who love one manufacturer and/or hate another. Their data is so important to people, that bad experiences with drive failures can have a significant affect on peoples opinions of the drive makers. But that aside, WD Gold drives are basically the same as Ultrastars with just branding differences and statistically they appear to be the most reliable. But there is still a 0.3% or so annual failure rate, and someone has to experience those failures. 😢

Hi, thanks for pointing this out. I can't notice this on speakers, but when using headphones there are some small (at least they seemed small to me) glitches as you say. I don't think its from the microphone, but seems to be an issue with some post processing. Maybe someone can give opinions. I am going to keep an eye out for this, and see if I can make some improvements. Getting the audio right does seem to take some experimentation and work. Thanks for bearing with it.

​@@sometechguy Those pops and glitches usually appear when the recording audio buffer size is too small (too short). Try increasing it in your recording software (e.g from 5ms to 20ms). Also try to switch so called Audio Device or Driver Type: MME/DirectX/Asio inside your recording software. Another reason may be your improper Windows audio settings. The third possible reason may be aggressive settings in your limiter/compressor/etc. module if any of these is used during the recording or editing.

The way a company treats customers when there is a problem and they need help is their true measure. I guess that's equally true of people generally. 🤔

Hi, and thanks for the thoughtful comment. I am not sure if you saw them, but I did release some follow up content to this with some detailed AFR analysis on 10-16Tb models, as well as 4-8Tb models which I think had some interesting detail on those manufacturers, but also called out some specific models which were stinkers. At least, based on the Backblaze data, as those could be influenced by shipping/handling issues. Though some of the drives that did badly had large numbers of units, so I think unlikely. What I did find was that consumer drives do not seem to have shorter lives or worse reliability than the enterprise variants. Though, shorter warranties of course. Even though the < 4Tb drives in the data set are mostly pretty old I think I will also cover those as it will probably crystallize some of the experiences people have had with these manufacturers from the past. I know Barracudas did not always have a good rep, so would be good to see if that is supported by data. Also, in one of those videos I think i saw some iterative improvements as Seagate moved though the X10, X12, X14 and X16 ranges, so I may also take a look to compare those (as well as the iterations also from Tosh and HGST/WD) to see if we are seeing general improvements as they consolidate incremental improvements in the drive manufacturing process. But I would skip Ironwolf/Pro personally, and go straight to Exos. They can be a bit noisier, but the price point and the warranty seems to make them a good economic choice and there is lots of reliability data on the models to avoid.

10 megabytes isn't huge unless you had old hard drives that come from a time people thought megabytes were a lot

@@ThtsWhaSheSaid64 It was huge back in the 80s.

@@orhunkabakli exactly what I was thinking but I also thought u might have made a typo

so you're saying i should get toshiba mg10 20tb instead of exos 20tb?

@@DeiLux Hi Our observations were based on build quality, internal components like preamp, head /spm etc and engineering design. Since this varies from model to model, we will have to analyze these 2 particular models in order to conclude. In general I will recommend Toshiba over others. Thanks

​@@kiranjoshi5267thank you. I ordered mg10's because they have 512mb cache over 256, extra 15 bucks for that hehe.

I did some other video that compare reliability between different models (4-8TB and 10-18TB) and WD generally seem to have the edge. But Seagate are often a bit cheaper and though not quite as reliable, the failure rates are still mostly low at 1.5-2% a year for most models. Seagate have a significant market share, and this also means more will complain about them, because more of their drives exist. But overall, price aside, WD seem the pick at the moment.

​@sometechguy didn't think about more units on the market means more failures, you're right i didn't thought about it

my Toshiba MG drive is made in Philippines...

I happen to own two HGST hdds, both product of malaysia, and second hand bought from online store (not marketplace). Only using them as storage for additional files and kept hidden away from elements like dust and humidity.

@@gregoryzet I also own toshiba hdd made in philippines. Nothing to worry as long as its not made in china.

I don't remember how it was 20 years ago. But in the last 5 years the only failures I've had is with Seagate. 3 in a row. Never buying Seagate again. They were the cheapest, and I wasn't thinking.

What model or type of disks where they and did they all fail early? I had a few WD Reds die (non Pro), but they were 60k hours and well beyond warranty. Not been a fan of WD for a while, but not actually due to product quality problems.

@@sometechguy It was almost +15 years ago and it was both internal and external disks. I don't remember the exact model since it was perhaps between 13 to 18 years ago. The one who failed fastest failed after 3 days and some of them survived perhaps a "whole" year but one/almost one year is still lousy. Never a WD in my house anymore.

@@birgerolovsson5203 product quality can change, but it shows that brand reputation is the hardest thing to repair and customers have long memories. 👍

​@@sometechguyi can't decide between 5tb western digital passport or 5tb seagate 😢 please help.

The WD My Passport would be a 2.5" disk in a caddy. So its most likely a WD Blue, or possibly a WD Black inside but you can't really be sure. I have not seen good reliability data for the 2.5" disks, and I have not really looked too hard for it. So I can't really say if they are 'good'. Given you never really know what disk you will get in a caddy drive, I normally buy based on brand reputation (or experience) and on price and use it for low usage cold storage, e.g. backups or carrying unimportant and bulky data when I am traveling. Also, I am generally suspicious of the reliability of caddy disks for intensive use. But I don't think any of this really helped you make a decision. 🤔

They did, and despite the evidence that they make great some great disks, I stopped buying their products for some years. And of course, it wasn't a one off, they made other poor choices around the way they treat their customers. And a pattern of behavior is a bigger red flag than a single piece of poor judgement. Thanks for commenting!

same i had been having bad experience with seagate, i will say like 3 HDs failed.

Yeah go with WD instead. Have WD drives for 8 years now and still at 100%.@@gizmo_therapy

Seagate mostly sucks, but their helium NAS drives are far better than WD.

I've owned 6 seagate hard drives and have used them for 20 years, including some lasting 12 years before decommissioning. Never had a failure.

I understand, thanks for the feedback. Here are a few of the problems here: For every person that complains that its too fast paced or dense, there are 5 people that complain its too long and detailed. I certainly don't try to add irrelevant details, and I try and keep the analysis relevant, but again different people want different things and whatever you do, someone will complain about it. As for the editing, I appreciate your feedback on this. Again, if its too slow paced, people complain and don't watch. At least if you want to spend a little more time analyzing the graphics, you can pause the video, or replay a section. I don't think there is an answer here that will make everyone happy. I am sure it is far from perfect, but I try to improve in each video. In this video I explored the data and looked for interesting trends, but I think some follow on videos did a better job at getting into more detail on specific models and comparing failure rates in more tangible ways. Its a really large dataset that contains a huge quantity of data, so extracting it efficiently and representing it so its actionable is the challenge. If you are interested in more, maybe check these follow ups which dive deeper into some specifics. But if you find them too hard to watch, I understand. Analysis on AFR rates for 10Tb - 16Tb drives: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-IgJ6YolLxYE.html Analysis on 4-8 Drives and Enterprise vs Desktop failure rates: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-l_YqdVGcC0o.html Seagate vs WD vs Toshiba for AFR rates: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-icNbexYV3M4.html Thanks for dropping by and commenting.