Backblaze's 2022 SSD Edition Report: Comparing the Failure Rates and Temperature Performance of SSDs and HDDs

Published by

Click here to post a comment for Backblaze's 2022 SSD Edition Report: Comparing the Failure Rates and Temperature Performance of SSDs and HDDs on our message forum
https://forums.guru3d.com/data/avatars/m/180/180832.jpg
Moderator
It's lonely at the top for Seagate again , who would have known ? 😀
https://forums.guru3d.com/data/avatars/m/242/242134.jpg
Not saying its related to a bad product, but all of the Seagate drives i bought for myself (in the beginning), or others (against my advice), died within 1y or 2 (2 within 2 month of use).
data/avatar/default/avatar28.webp
WhiteLightning:

It's lonely at the top for Seagate again , who would have known ? 😀
fry178:

Not saying its related to a bad product, but all of the Seagate drives i bought for myself (in the beginning), or others (against my advice), died within 1y or 2 (2 within 2 month of use).
The number to pay attention here is the AFR, and seagate is not doing worse than others :shrug: (Except for a model with only a handful or drives so it is not statistically relevant) (I don't give a s.. about seagate, just pointing out the right way to read the info)
https://forums.guru3d.com/data/avatars/m/259/259564.jpg
The Seagates are in line with the others, what are you guys talking about? The Seagate outlier is because they only had 3 drives and one failed. Doesn't tell you anything. The 1104 count Seagate has an AFR of .66%. The 554 count Seagate has an AFR of .96%, which is perfectly normal and better than 3 of the 4 other manufacturers in the test (Crucial, MIcron, WDC). Four of the Seagate models literally had a 0% failure rate. The only bad one is the outlier with 3 drives which is why their confidence interval for it is from .2-40.7. Guru3d method of reading a chart: *Stare blankly at it, out of focus, while considering your pre-existing beliefs* *Post pre-existing beliefs and pretend the chart confirmed them* *When someone challenges you, double down and act like they're a meanie*
data/avatar/default/avatar14.webp
Reardan:

The Seagates are in line with the others, what are you guys talking about? The Seagate outlier is because they only had 3 drives and one failed. Doesn't tell you anything. The 1104 count Seagate has an AFR of .66%. The 554 count Seagate has an AFR of .96%, which is perfectly normal and better than 3 of the 4 other manufacturers in the test (Crucial, MIcron, WDC). Four of the Seagate models literally had a 0% failure rate. The only bad one is the outlier with 3 drives which is why their confidence interval for it is from .2-40.7. Guru3d method of reading a chart: *Stare blankly at it, out of focus, while considering your pre-existing beliefs* *Post pre-existing beliefs and pretend the chart confirmed them* *When someone challenges you, double down and act like they're a meanie*
Nailed it. 😎
https://forums.guru3d.com/data/avatars/m/266/266074.jpg
They way i read these numbers is that Crucial for example has 488 units from which 5 failed and model ZA250CM10002 had 554 units from which 15 failed. Taking WDS250G2B0A with 89 units and 1 failed, if we hypothetically go for 5 times the number (kinda dumb but just to get an idea), it will get us to 445 units and 5 failed (in this dumb extrapolation, which obviously can be affected by other factors and not a guarantee of tendency). Anywho, it will get it to roughly Crucial numbers-ish, which look better to me than that model of Seagate. Micron looks pretty "great" as well lol. 😱 If i try to level it to a similar number of units for either ZA250CM10003 or that no name Seagate SSD entry, they still look kinda worse to me when comparing it in this fashion to either Crucial or WD. I'm mainly going for these models as they are listed with a bigger number of units. This is how my brain paints this to me. Those AFR numbers do not quite click to me, given this view, though I am ready to admit that there is a chance I may be leaving some things out of the picture. If anyone has anything to contribute to the topic, I legit am open to ideas as I might be making borked ideas which I do not quite wish to.
https://forums.guru3d.com/data/avatars/m/269/269912.jpg
It's funny, all I get out of this report is you probably don't want to install a 250gb ssd in your system.
https://forums.guru3d.com/data/avatars/m/56/56686.jpg
NewTRUMP Order:

It's funny, all I get out of this report is you probably don't want to install a 250gb ssd in your system.
that depend on which brand the WDC and micron look bad. but even that blow out portion like Seagate which has some highest amount if drive used. Every has there opinion and few of these things
data/avatar/default/avatar17.webp
Chart is a bit silly, really there's far too few drives being tested for any meaninful statistical comparison, particularly as they aren't being tested in a controlled enviroment with the same data, in same PC builds, etc. Mind you it wouldn't be a very exciting read if we made the chart only entries where there are 1000+ drives of one type to test.