This is a quote from samsung' website regarding the latest 980 pro drives with 7GB/s speeds on ssd:
And here is my stats for 512gb m1 Air:
Smartctl is having error when reporting "Percentage Used: 3%" because then it would mean that TBW for M1 Air 512gb is 4000 TBW which is obviously incorrect as Apple doesn't have secret UFO technologies apparently.
Clearly Apple nowhere and never presented TBW for its' ssd drives, that is why smartctl is making up some number for that. But if I go with the Samsung numbers, then it would mean that i used 119TB(out of 300TBW) in exactly 1 year for my 512gb, meaning that it could last me 2.5-3 years only.
It gets even worse if you have 256 gb ssd with 150 TBW...
That is what the drive is
warranted for not how long it will actually last - that is why I linked to
the results of ~250 TB drives in 2014 test article. In fact one of those was Samsung 840 Series (regular and pro) and while regular had a few hiccups at 300 TB it sailed on to
800 TB where it started having issues again and effectively died at 900TB.
The Samsung 840 pro is warranted
73 TB TBW and "Reallocated sectors started appearing in volume after
600TB of writes. Through
2.4PB, the Pro racked up over 7000 reallocated sectors totaling 10.7GB of flash. Samsung’s Magician utility gave a clean bill of health, though, and the used-block counter showed ample reserves
to push past 2.5PB" But after leaving the drive off for a week it would not read but the noted this:
"Before moving on to the performance analysis on the next page, I should note that the 840 Pro exhibited a curious inflation of writes associated with the power outage after
2.1PB. The SMART attributes indicate an extra 38TB of host writes during that period, yet Anvil’s logs contain no evidence of the additional writes. Weird.
Maybe the SMART counter tripped up when the power cut out unexpectedly."
The rule of thumb is the bigger the SSD and the less you keep on it the longer it will live. Using
Samsung's own figures everything else being equal going from 128GB/256GB to 512GB/1TB
doubles the TBW. Using the
worst values we get
1200 TBW until problems set in.
Since the tests show a linear progression using 119TB/year gives us
10.1 years. As I said before you have to look at
all the data.
smartctl doesn't "make up" anything - it simply reports SMART attributes for the drive. Sure with older (especially SATA) SSDs you could even test and calculate it yourself. Here's a paper by Samsung itself on how to estimate write amplification and SSD lifetime from attributes from an actual test:
https://image-us.samsung.com/SamsungUS/b2b/resource/2016/05/31/WHP-SSD-SSDSMARTATTRIBUTES-APR16J.pdf
Note that this is NOT what smartctl does, it is simply reporting what it sees in the attribute fields. Since we're talking about NVMe drives, the older definition for some of the attributes no longer apply. Here's a newer NVME spec from JEDEC:
https://www.nvmexpress.org/wp-content/uploads/NVM_Express_1_2b_Gold_20160603.pdf - attribute 05h returns "percentage used" with an internal vendor calculation and this is what smartctl is reporting i.e. it isn't doing any calculations itself.
View attachment 1948593
I was pointing out nearly from the get go the "smartmontools must be reading the wrong value" (
Apr 7, 2021)
According to a
2021 Micron Technology page
Smart isn't the "standard" everyone thinks it is and these number are showing that what is in that article is effectively correct as no one would guarantee a 1 TB SSD for 6880 TBW.
"Percentage Used: Contains a
vendor specific estimate of the percentage of NVM subsystem life used based on the actual usage and
the manufacturer’s prediction of NVM life. A value of 100 indicates that the estimated endurance of the NVM in the NVM subsystem has been consumed, but may not indicate an NVM subsystem failure. The value is allowed to exceed 100. Percentages greater than 254 shall be represented as 255. This value shall be updated once per power-on hour (when the controller is not in a sleep state)." -
SMART Attribute Details
In fact, though I can't find it right now there was a report that the smartmontools tool everybody was using early on was using out of date pointers resulting in incorrect values for
both TBW and percentage used. GIGO.
"Eliminate the impossible, and what ever remains, no matter how improbable, is the truth."
As I said before you can't pick and choose what part of the data you use and that is why I am "dredging up this nonsense again". Also if you are really that paranoid just boot up from an external or cut down on using a ridiculous number of tabs in the browser (as well as turn off the disk cache).