How much longer should I keep my C drive? (pics)

Scannerman · Sep 14, 2025

The way I say it silicon drives are pretty much designed to burn out. Storage can be kept indefinitely by cloning, back ups, and air gapping. Spinners (mechanical hard drives) are useful in this respect for storing media and files. Operating systems work most efficiently on silicon. I would definitely ensure that I had a clone of my OS when the drive it is installed on has already reached low health. When the drive finally craters I would replace it with the clone, update my OS, and write back my files from backups and storage if need be. Archiving files once a week is also prudent. There are dozens of different ways to back up data and it's always a good practice not to rely on just one method.

PerpetualCycle · Sep 14, 2025

MisterEd said:
I think you are misunderstanding what the health means. The drive is not going to stop working when the health reaches 0%. The health is only a probability of its lifespan. I assume it is based on average lifespans for SSDs. However, like any drive it will eventually fail. The older it is and the more it is used the probability that it will fail increases. In any case you should be making regular backups and be prudent and replace it before it fails.

I am not misunderstanding. It is not based on some "average" lifetime of the SSD model. To write to a block on an SSD the block first has to be erased. Each block only has a fixed number of erase cycles before it cannot be used anymore (around a thousand or so depending on the type of NAND chips used). After that many erases it is toast - it cannot reliably store data. There are statistics involved as each block may vary slightly in the maximum number of erase cycles. There are complex algorithms in the SSD that maintain a pool of pre-erased blocks for new writes and that balance the writes across blocks for even wear for a maximal lifetime. One or more core processors on the SSD running in the background are dedicated to this.

The smart statistic "Percentage remaining" is the number of erase cycles that remain out of the total possible. It is very accurate and it does depend on usage. In general, if you write more than the TB/yr rating of the drive, the erase cycles will accelerate due to write amplification (too long an explanation for that here).

tldr; the SMART percentage remaining statistic is not based on some average lifetime of that model of SSD. It is based on the internal data of erase cycles used on that particular SSD and accurately predicts, in real-time, its life expectancy. It will stop working around zero percent at full capacity - it may last a bit longer at reduced capacity as the blocks start failing, but I would not go there.

jimbo45 · Sep 14, 2025

Amazing how some of "these old wives tales" still persist. SSD's etc don't wear out any quicker than spinners any more -- at least not within normal lifetimes of your PC. !!!

In any case even if a drive etc DOES fail so long as you've got a retrievable backup - then what's the problem.

Also why do people still bother with ludicrous things like defrag disks - especially when they are ssd's / nmvme's -- shows they don't understand the difference between how a "spinner" accesses a piece of data or how an ssd type does -- and in any case modern "spinners" with 7200 rpm and very fast cache's don't need defragging either.

Cheers
jimbo

cereberus · Sep 14, 2025

The key issue here is few people understand what 23% means.

With every SSD, the vendor states how many NOMINAL GB can be written to a drive. So 23% just means you have written 67% of the nominal writes a drive could have.

Now here is the thing - the nominal writes is an abitrary figure and in reality, the drive will probably handle 2 or 3 times as much.

So why are the nominal writes so low? It is (imo) a vendor warranty scam i.e. if you exceed the nominal writes in the warranty period, the vendor says you have exceeded the write linit and then refuls to payout. Its a bit like the car warranties that say your car is insured for 3 years or 60000 miles whichever come first.

In the end, you need to review other metrics to properly assess things.

In fact modern ssd/nvme drives usually fail for more mundane reasons e.g. heat affecting electronic components more than reaching end of write life.

Of course, nobody truly knows the life of a drive - some will last a long time, most will last a reaonable time, and some will fail early. So regardless of whether drive is new or old, regular image and data backups are the best protection against drive failures. One should always take the view i.e. it is not "when will the drive will fail" but "the drive will fail at any point in time".

So in the end, no amount of trying to predict when a drive will fail really helps. Assume the drive will fail and protect against a failure.

Having said that, statistically older drives are more likely to fail than a newer drive. I actually keep a spare nvme drive for use as backup drive, knowing I can put it into service temporarily as main drive if my laptop internal nvme drive fails.

However, I admit I have not always made backups to exterenal drives as frequently as I should and nearly got bit up my posterier recently but was able to fully recover after some messing around (the partition holding my latest backup had been deleted somehow but I was able to restore the partition and then restore latest backup made that morning).

retiredat44 · Sep 14, 2025

Hard Disk Sentinel Giveaway Download & Key Official I found this seems legit I am running it. Free! If I broke some rule I am sorry.

cereberus · Sep 14, 2025

MisterEd said:
I think you are misunderstanding what the health means. The drive is not going to stop working when the health reaches 0%. The health is only a probability of its lifespan. I assume it is based on average lifespans for SSDs. However, like any drive it will eventually fail. The older it is and the more it is used the probability that it will fail increases. In any case you should be making regular backups and be prudent and replace it before it fails.

It is not even average lifespan - it is a conservatively low vendor figure. Tests on various ssds suggest the vendor nominal life is only around 40% of actual life. See my earlier post.

gunrunnerjohn · Sep 14, 2025

I try not to run a disk until it drops, that always seems to come at an awkward time and never when it's convenient to deal with it.

gunrunnerjohn · Sep 14, 2025

cereberus said:
It is not even average lifespan - it is a conservatively low vendor figure. Tests on various ssds suggest the vendor nominal life is only around 40% of actual life. See my earlier post.

That would make sense since it's beneficial to the manufacturer for you to replace the drive as soon as possible.

hdmi · Sep 14, 2025

S.M.A.R.T. attributes for NVMe

Byte#0 Critical Warning
This field indicates critical warnings for the state of the controller. 1) If Available Spare is lower than the Available Spare Threshold, 2) If Temperature is not within the normal temperature range (above an over temperature threshold or below an under temperature threshold), 3) If NAND reliability is degraded due to internal issues, 4) If entered read only mode
Byte#2:1 Temperature
Indicates the temperature of the overall device (controller and NAND included) in units of Kelvin.
Byte#3 Available Spare
Indicates a normalized percentage (0 to 100%) of the remaining spare capacity available. If the Available Spare is lower than the Available Spare Threshold, the status is bad.
Byte#4 Available Spare Threshold
Value is indicated as a normalized percentage (0 to 100%).
Byte#5 Percentage Used
Indicates a vendor specific estimate of the percentage of device life used based on the actual device usage and the manufacturer's prediction of device life. A value of 100 indicates that the estimated endurance of the device has been consumed, but may not indicate a device failure. This value is allowed to exceed 100. Percentage values 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). Refer to the JEDEC JESD218 standard for SSD device life and endurance measurement techniques.
Byte#47:32 Data Units Read
Indicates the number of 512-byte data units the host has read from the controller; this value does not include metadata. This value is reported in thousands (i.e., a value of 1 corresponds to 1000 units of 512 bytes read) and is rounded up. When the LBA size is a value other than 512 bytes, the controller shall convert the amount of data read to 512-byte units.
Byte#63:48 Data Units Written
Indicates the number of 512-byte data units the host has written to the controller; this value does not include metadata. This value is reported in thousands (i.e., a value of 1 corresponds to 1000 units of 512 bytes written) and is rounded up.
Byte#79:64 Host Read Commands
Indicates the number of read commands completed by the controller.
Byte#95:80 Host Write Commands
Indicates the number of write commands completed by the controller.
Byte#111:96 Controller Busy Time
Indicates the amount of time the controller is busy with I/O commands.
Byte#127:112 Power Cycles
Indicates the number of power cycles.
Byte#143:128 Power On Hours
Indicates the number of power-on hours. This does not include the time that the controller was powered and in a low power state condition.
Byte#159:144 Unsafe Shutdowns
Indicates the number of unsafe shutdowns. This count is incremented when a shutdown notification is not received prior to loss of power.
Byte#175:160 Media Errors
Indicates the number of occurrences where the controller detected an unrecovered data integrity error. Errors such as uncorrectable ECC, CRC checksum failure or LBA tag mismatch are included in this field.
Byte#191:176 Number of Error Information Log Entries
Indicates the number of Error Information log entries over the life of the controller.
Byte#195:192 Warning Composite Temperature Time
This field indicates the minimum Composite Temperature field value that indicates an overheating condition during which controller operation continues. Immediate remediation is recommended (e.g., additional cooling or workload reduction). The platform should strive to maintain a composite temperature less than this value.
Byte#199:196 Critical Composite Temperature Time
Contains the amount of time in minutes that the controller is operational and the Composite Temperature is greater than or equal to the Critical Composite Temperature Threshold (CCTEMP) field in the Identify Controller data structure.

Manufacturer-defined S.M.A.R.T. attributes

(The ID number and whether the attribute exists can vary with the make and model of the SSD.)

ID#5 Reallocated Sector Count
Count of reallocated sectors. When the drive finds a read/write/verification error, it marks that sector reallocated and transfers the data to a new sector from a special reserved area.
ID#9 Power-On Hours
The number of hours the drive has been in the “on” state. This value simply shows user behavior and is not directly related to the health of the drive. The raw value of this attribute shows the total count of hours (or minutes, or seconds, depending on manufacturer) the drive has spent in the power-on state. When the machine is in Hibernation Mode, the Power-On Hours value does not increment.
ID#12 Power-On Count
The raw value of this attribute reports the cumulative number of power on/off cycles. This includes both sudden power off and normal power off cases.
ID#177 Wear Leveling Count
This attribute represents the number of media program and erase operations (the number of times a block has been erased). This value is directly related to the lifetime of the SSD. The raw value of this attribute shows the total count of P/E Cycles.
ID#179 Used Reserved Block Count
This attribute represents the number of reserved blocks that have been used as aresult of a read, program or erase failure. This value is related to attribute 5 (Reallocated Sector Count) and will vary based on SSD density.
ID#181 Program Fail Count
This attribute represents a total count of the number of failed program requests (failed writes).
ID#182 Erase Fail Count
This attribute represents a total count of the number of failed erase requests.
ID#183 Runtime Bad Count
Equal to the sum of the Program Fail count (attribute 181), the Program Erase Fail count (attribute 182), and the Read Fail count. This summary value represents the total count of all read/program/erase failures.
ID#187 Uncorrectable Error Count
The total number of errors that could not be recovered using ECC.
ID#190 Air Flow Temperature
The current temperature of the area surrounding the NAND chips inside of the SSD.
ID#195 ECC Error Rate
Count of correctable errors. Number of errors corrected by the internal error correcting mechanism.
ID#199 CRC Error Count
The number of Cyclic Redundancy Check (CRC) errors. If there is a problem between the host and the DRAM or NAND flash, the CRC engine will tally the error and store it in this attribute.
ID#235 Power Recovery Count
A count of the number of sudden power off cases. If there is a sudden power off, the firmware must recover all of the mapping and user data during the next power on. This is a count of the number of times this has happened.
ID#241 Total LBAs Written
Represents the total size of all LBAs required for all of the write requests sent to the SSD from the Host. To calculate the total size (in bytes), multiply the raw value of this attribute by 512 Bytes.

Scannerman · Sep 14, 2025

I try to keep things as simple as possible for my aging and infantile mind. Besides, not everyone reading these forums is a computer guru and some actually visit so they can learn the hows and whys of PCs. Simply put "spinners" are mechanical hard drives that store data magnetically by means of a spinning disk. Not everyone knows this but it is common knowledge. SSDs, on the other hand, store data electrically by means of tiny cells in the silicon. This knowledge is less common. These tiny cells have finite life spans and can only be over written X number of times before they are completely spent. In this regard, some mechanical drives actually boast much longer life spans because spinners can be over-written and re-written more times than a silicon cell can. Yes, there are some exceptions to this but the exceptions only prove the rule.

This is partly why mechanical hard drives are still popular today despite all the problems associated with moving parts, speed, carbon foot prints, mobility, and plain old convenience. They are also considerably less expensive than SSDs boasting the same storage capacity. Generally you can get more re-writes on a mechanical drive than you can on an SSD. When SSDs first came out there were all sorts of issues with these cells filling up and 'burning out' fast. Programs that used to rewrite data to other sectors on the drive that ran in the background (often unbeknownst to the user) were a menace to SSDs. Today, much of this has already been dealt with but it still does happen on occasion. For example, most people know that you do NOT defrag an SSD as it will lower the life expectancy of the drive.

Today we use TRIM on SSDs much the same way we used to de fragment mechanical drives, but TRIM can only do so much. When a cell is finished its life cycle it is done. When spinners are built to last they will out last most silicon despite having moving parts. I have 14 year old hard drives still running and functioning correctly in my PC because they were designed for commercial use. Since then I've had a number of silicon drives cap out on me and a couple brick on me. They barely lasted 5 years. This is not to dis SSDs. Silicon drives certainly have their benefits but they are not without their own caveats. Much like a high performance vehicle a PC built for speed requires more attention and care. The need for speed can get expensive.

There is a good possibility that the humble "Klunk Drive" will still be around for another decade serving common requirements of utility and affordability. In a humorous way, I can personally relate. I might be old, and I might be slow, but I can still swap out a battery in a vehicle and I can still drive.

IF your SSD is fairly new and the health of that drive is diminishing quickly I'd look into how much that drive is being rewritten and how often. That's all I got.

billmcct · Sep 14, 2025

Scannerman said:
SSDs, on the other hand, store data electrically by means of tiny cells in the silicon. This knowledge is less common. These tiny cells have finite life spans and can only be over written X number of times before they are completely spent. In this regard, some mechanical drives actually boast much longer life spans because spinners can be over-written and re-written more times than a silicon cell can. Yes, there are some exceptions to this but the exceptions only prove the rule.

This is why it is NOT recommended to use SSDs for long term storage. If the SSD isn't powereed for a long period of time the cells degrade

gunrunnerjohn · Sep 14, 2025

billmcct said:
This is why it is NOT recommended to use SSDs for long term storage. If the SSD isn't powereed for a long period of time the cells degrade

You have to do more than just power them on. You have to at a minimum read each cell, that will allow the drive to notice any data degradation and write the affected cells before the errors exceed the ability of the error correction to correct it. It's actually the write that recharges the charge in the cell, not just powering the drive on. The reading allows the drive to notice that some blocks don't pass the parity checks and triggers an error correction and rewrite of the block.

SSD drives have varying amounts of background processes that are doing wear leveling, etc. but those aren't really insuring data integrity over the entire data set.

How much longer should I keep my C drive? (pics)

Well-known member

My Computer

Syzygy

My Computers

Well-known member

My Computer

Well-known member

My Computer

Well-known member

My Computer

Well-known member

My Computer

Well-known member

My Computers

Well-known member

My Computers

Jack of all trades – master of none

S.M.A.R.T. attributes for NVMe​

Manufacturer-defined S.M.A.R.T. attributes​

My Computers

Well-known member

My Computer

Active member

My Computer

Well-known member

My Computers

Similar threads

S.M.A.R.T. attributes for NVMe

Manufacturer-defined S.M.A.R.T. attributes