Does anyone make good SSDs any more?

[Infraviolet]

I'm in the market for buying an SSD to replace an aging HDD. I've had enough of the mechanical limitations and vulnerability of HDDs. So I'm looking to buy an SSD (due to what it is going to replace: SATA, 2.5 inch, 1TB or close, must be able to use as boot drive for Linux OS on a UEFI system with secureboot and fastboot disabled), and yet the online reviews imply I'm in for nothing but a world of pain with just about every brand. For each brand I've looked at I'm shocked how many buyers report "this died in 3 months light usage", or, "medium usage, never more than quarter full, but at 2 years age SMART health rapidly declined and I'm getting file corruption". So rather than rely on anecdotes from single buyers alone I've tried to look upspecific bugs and models, the results are horifying, every brand seems to be producing utter junk.

Samsung, major scandals with the 870 EVO (big thread at https://www.techpowerup.com/forums/threads/samsung-870-evo-beware-certain-batches-prone-to-failure.291504/ )related to any with firmware which is not SVT02B6Q. Firmware will corrupt data and can brick the drive entirely. Samsung has also had trouble with 980 Pro and 990 series SSDs, suddenly turning read-only and absolutely racing through their write cycles life with only small amounts of data written. The older 840 line of SSDs suffered a bug where cells lost chage and data that wasn't read or written often enough would vanish, and such a firmware update was needed to make the drive automatically perform rolling rewrites across itself, which increased the wear somewhat and shotened the overall lifespan accordingly as, to prevent bits getting corrupted every bit had to be rewritten every few days.

WD and SanDisk, today the same company for SSD production. Currently being sued for having produced external SSDs under both brand names which suddenly bricked and wiped the data on them. Also a scandal with critical firmware updates being needed for the SA510 line of WD branded internal drives.

Crucial's MX500 line has been found to have defective firmware in some batches which are not using M3CR046 firmware, the issue appeared to be related to the controller writing in such a way as to cause excessive wear. And updating this apparently requires use of a Windows tool which even then can only perform a firmware update whilst the drive is being written, and Windows won't write to ext4 partitions as you'd have on a Linux system drive. It appears they also had a situation in recent years where they suddenly pulled some firmware update they were offering, for no explained reason. And there's speculation that some more recent driveshave moved to greater numbers of cell layers in their storage, for lower cost and lower endurance.

Kingston, particularly the A400 line, I've seen reports of very bad quality NAND memory chips bing used inside the drives, takes very few write cycles before dying. There were various forum threads where people opened them up and looked at chip markings on drives of different ages, less reliable drives after a certain date range suddenly had differently marked chips of what were assumed different brands. The A2000 also suffered from a bug related to an energy-saving standby mode. An error code "SATAFIRM S11" was often involved in troubles which some of Kingston's SSDs had.

Seagate, couldn't find criticism of any of their SATA SSDs as they don't seem to make SATA SSDs any more and seem to only do types with more modern connectors which aren't applicable to my situation.

An awful lot of sellers of cheap USB sticks, now seem to have SSD brnads, but one assmes they are either rebadged ones from the main manufactuers, as these companies do with USB sticks, or are made with very cheap internal parts.

All these failures seem to have been within the last 3 years, as if all the manufacturers are embracing shoddy production standards together.

When buying there is no way to know which firmware a drive is shipping with until you unbox, plug in and check with the likes of gsmartcontrol. On the outside drives which are very different internally have the same name and model number.

Furthermore all the manufacturers seem to be moving to QLC quad layer storage, which I understand is a lot more vulnerable to corruption than the original single layer SSD technology as each cell holds more bits and needs to stay at a more precise analogue level to be read out without corruption. All so they can save a bit of physical space in the drive, and make some cost savings which don't get reflected in the sale price. From what I can tell they're chasing bleeding edge technology changes for minor cost savings (fraction of percent of sale price) and making very unreliable devices because of it.

For all the brands it seems that detecting if the firmware is a bad version, and applying a firmware fix, is virtually impossible on non-dual-boot linux machines, as all the manufacturers produce junk Windows only software for SSD monitoring and don't make downloadable update files available online to be retrieved without said Windows software being a middle-man. I saw one brand, can't remember which, also offered a bootable iso for firmware updating, but this apparently was built on such an old DOS or UNIX core that when it booted it could only support PS2 type keyboards and not USB ones or the inbuilt ones on laptops, so the iso was unusable for virtually all modern PCs.

Are good SSDs, the main focus being more about long term reliability and not having to worry about PC being moved while running than speed performance,  just not manufatured any more?
Thanks

[kripton2035]

just for the records, I've installed 3 to 10 ssd's each week of work for more than 10 years now.
at most I would say 5 of all of them came back for defect ... and usually in the first month of usage.
just buy one of some reputable brand (I was selling samsung mainly), use the 3 to 5 years warranty if any bad luck,
AND MAKE AUTOMATIC REGULAR BACKUPS that is the key...
one of the first ssd I sold was a 500GB for ... EUR 1850 !!!! today these are less than EUR 50, and have far better specs.

[Benta]

I have five laptops where I installed Crucial MX500 250G and 500G SSDs. All been running daily the last three years without a single issue.
Maybe I'm just lucky.

[SiliconWizard]

Despite the Samsung hiccups - which were all fixed relatively quickly AFAIK - I have 4 Samsung SSDs (2 NVMe, 2 SSD) and had had absolutely zero issues with them.
The most recent one I have is a 980 Pro NVMe, the first firmware version had a problem but the one I bought had the second version which works fine.
As a general rule, not buying a product that just got released and waiting for a few months until initial problems are fixed is a good idea.
In my experience, Samsung has always fixed their issues rather promptly and as far as performance and reliability goes, this is still my go-to brand for SSDs.

[DavidAlfa]

My 500GB EVO 850 did well since 2018 until a month ago, zero issues.
I checked the SMART and had like 20K running hours (Edit : Why the hell did I write 400K? LOL)
I recently switched to a 2TB nvme SN850X just because it was time to upgrade.

[John B]

I'm using a Samsung 860, 2x 870s and a 970 and haven't had any issues. The latter ones are pretty new and it's hard to give solid feedback on them, but the disk reporting feature shows no errors or unusual readings. The 860 is maybe 2 years old and likewise hasn't caused any issues.

I have wanted to try some other brands, but I usually pick up the samsung drives when theyre on special. Last time I was going to pick up a western digital blue one, but it had a ton of feedback showing lots of failures.  Anecdotal of course so take it for what it's worth.

[thm_w]

just for the records, I've installed 3 to 10 ssd's each week of work for more than 10 years now.
at most I would say 5 of all of them came back for defect ... and usually in the first month of usage.
just buy one of some reputable brand (I was selling samsung mainly), use the 3 to 5 years warranty if any bad luck,
AND MAKE AUTOMATIC REGULAR BACKUPS that is the key...
one of the first ssd I sold was a 500GB for ... EUR 1850 !!!! today these are less than EUR 50, and have far better specs.

Yeah, get whatever drive is decent spec for your purpose with RAM (Teamgroup MP34 1TB 5yr warranty is $50), I wouldn't pay the samsung premium but you can if you want. Then back it up to a HDD.
End of discussion.

[tom66]

I've got a 5 year old Crucial MX500 (500GB, purchased for ~£120) still working fine as a main boot drive until very recently when it was replaced with an M.2 Samsung SSD.  Still in use as an aux drive.   I also have a BX500 (cheaper brother of the MX500, bad for peak write rate but great for reads) and a very old 120GB SSD from I think Patriot (?) but it's still alive.

The M.2 Samsung SSD (Evo 970) cost around £100 but has a 2TB capacity.  It claims a warranty of 1200TBW and 5 years, I'm pretty sure I'll exceed the years count before the TBW.

All drives are susceptible to failure.  Ensure you have a backup process in place, I use Backblaze myself for a few bucks a month but others are available.

I would never use a Kingston product.  They don't make their own flash and so have very variable quality.  I've had RAM and external USB drives from them fail.  And they're rarely any cheaper than the OEMs that do make flash.

[TERRA Operative]

I work for a multinational CDN provider, here our team oversees literally thousands of servers across multiple data centers and like 5 times that in SSD's

I haven't seen any specific brands standing out over others for failures.
We run the disks until they simply wear out, or they just randomly die as things sometimes do. The 'bathtub curve' is real, but skewed much towards the end-of-life end.

We use Samsung, Micron, and Intel SSD's mainly (WD, Samsung, HGST, and Toshiba for HDD's but they are phasing out for SSD's in all new machines).

For my personal laptops, I use Samsung nvme drives and was using a Samsung 500Gb SATA disk which was replaced with a Micron 1Tb disk simply because free upgrade. (Hooray for decomissioned servers sent to scrap!)

I like Samsung 'just because' and I've never had one die, but if you are really worried about failures, getting an enterprise rated drive from one of the brands I mentioned would be a good idea, if your budget allows.

[BradC]

I've got ~15 Samsung SATA SSDs starting with the 830. I did skip the 840 though. Unfortunately I recommend it to my brother and he lost a crapload of data to the 840 EVO.

I also have Intel SSDs, A couple of WD Green (slow as a wet week in winter but only used for bootloader), a Kingston or two, 2TB Silicon Power (big but _slooooow_) and a 2TB MX500 , and a few NVME of various makes (Samsung included).

My fav's would be the 980 Pro, older 970 Evo Plus and SK Hynix P31. The only one I regret buying is a Crucial P2 which was reviewed as a TLC then a bait and switch with QLC.

The only duds I've had were a PATA Kingspec back in 2008 and two OCZ Vertex Plus that died after ~10 years of 24/7 use.

Sometimes you win, sometimes you lose. RAID for availability and Backups on top of that.

I never buy at release time and let them settle in a year or so before making a move. That way it's almost certain the clangers have been worked out of the firmware.

[NiHaoMike]

Probably the worst has to be Sandforce back in the day, very high failure rate due to firmware bugs. It in fact made me question if SSDs were overcomplicated and should have been implemented as a dumb array of Flash, with wear leveling and bad block management done at the kernel level as some embedded systems do.

[wraper]

Almost any non DRAM-less from not some obscure Chinese company is quite decent. Of course you should avoid QLC NAND but they are not that bad either. Most of the problems you mentioned did not happen recently either. Firmware bugs that brick the drive are not nearly as common as in the past. And modern drives usually fail gracefully becoming stuck in read only mode. Rather than bricking themselves despite hardware being OK.

[Veteran68]

I have many Samsung and Crucial SSDs (both 2.5" and M.2) and have never had a single issue out of any of them. I also have several cheaper "off brands" like Silicon Power that have served me well. The last SSD I can remember failing on me in any way was a 80GB refurbed Corsair model about 10 years ago.

Remember that when legitimate defects happen, they often do not cause issues anywhere near proportional to the amount of media hype they receive.

[DimitriP]

"some batches" !!!
Some Ford engines pop out the spark plugs.
Some GM engines  have the head screws loosen and separate the cylinder head ( Helloooo Northstar engine!!)
Some Porsche crashes are fatal

At some point you gotta pick something!

Given a choice I prefer WD, Micron or Samsung. I'll use SanDisk on  "non-critical stuff"

And once installed ALWAYS check transfer speeds when doing 4K size transfers!!!
Sometimes the performance is abysmally bad compared to a mechanical drive.


 

[amyk]

Micron apparently started making SLC SSDs again: https://www.anandtech.com/show/18863/micron-updates-data-center-nvme-ssd-lineup-6500-ion-tlc-and-xtr-slc/2

True 100K endurance at a decent price.

[IanB]

I had the chance to compare a Samsung SATA SSD as an upgrade in an old machine, relative to NVMe/PCIe SSDs in new machines.

The short summary is that a SATA SSD is a bit faster then a SATA HDD as an upgrade, but there is absolutely no comparison with a modern machine that has an NVMe Gen 4 SSD as original equipment.

If you want to upgrade an old machine, I'd recommend to install a smaller capacity and inexpensive SATA SSD to boot the OS and install apps, but keep your files and data on a second high capacity HDD. If your SSD happens to fail, just reinstall the OS and apps and continue. I don't think you will see any speed boost from keeping your data on an SSD compared to an HDD.

However, upgrading old machines is not cost effective. The most effective approach is to retire the old machine and get new hardware with the latest CPU generation, fastest memory and fastest storage. You might find it goes 5-10x faster than the older machine. I didn't believe it until I experienced it.

[bdunham7]

For everyday cruising, a SATA SSD is HUGE improvement over a mechanical HDD.  For more demanding work and large transfers, yes NVME is better, but I'm working on my old 12-year old backup laptop right now and it is not slow at all.  A PC with the OS and programs on an HDD would be excruciating.

As far as SSDs go, it's Samsung for me.  Granted, the 840s were a problem and I never buy the latest and greatest so I don't have bleeding-edge problems.  I'm the slow second mouse that gets the cheese, I guess.  The 12-year old laptop I just mentioned was upgraded 10+ years ago with a Samsung 830 that currently has 64k hours and 54TB written with no hint of trouble.  Samsung is the only company that hasn't lost a single byte of my data, ever--and that includes numerous 2TB 2.5" SpinPoint HDDs.

[IanB]

Yes, a 12-year old laptop was my experience. It was a very expensive, workstation class machine, and with maxed out memory and an SSD it was much faster than with the original HDD. However, putting it side-by-side with a 2023 laptop shows the difference. One minute tasks take 5-10 seconds, and 10 second tasks take a second or two. All data paths between disk, CPU and memory are so much faster. A faster disk can only do so much for a slower motherboard and memory architecture.

[Dacian]

I have a 480GB Mushkin MKNSSDCR480GB that has about 50khours of usage (got it used on eBay and I used it for many years every day) still good.
Since I needed to update the OS I now use an INTEL SSDSC2BF240A4H again purchased used on eBay some years ago and this one is only 5khours so perfect health.
The SMART data on the Mushkin say 2 reallocate sectors but I think that was the case sine I got it many years ago.
I use Linux with no Swap partition or file and so there is almost no data written to disk.
I only do software and electronic design plus typical office work (email,web,office apps) so there is no stress. As backup I still use an old 1.5TB HDD 3.5" that is only powered few times per year when I need to save something and a 32GB SLC SSD.
Long therm storage data a HDD will be superior to SSD but for mostly read SSD is superior in all aspects.

[DimitriP]

However, upgrading old machines is not cost effective. The most effective approach is to retire the old machine and get new hardware with the latest CPU generation, fastest memory and fastest storage. You might find it goes 5-10x faster than the older machine. I didn't believe it until I experienced it.

"old machines" come in two flavors.
There are "old machines" that are fast (except when W10 starts thrasing the HDD in which case an SSD takes care of that problem)
and there are "old machines" that were not "speed deamons" to start with. They should have been returned/retired a long time ago !!

Example: an  i7-4790K running W10 with an SSD still flies.  Will it benchmark as good as a newer/faster CPU?. No, but it still flies!
for about $100 you can clone the HDD to and SSD and be back in business by the time you wake up after th e copying is done. AND you have a complete copy of all "your stuff" on the old/slow/constantly busy HDD.
Since you don't have to shell out a lot more than $100 to
a) buy a new machine and then
b) start reinstalling everything, that might just pass the cost-effective test.
Now if you have a machine with only "a few" applications installed,  maybe the re-installation is not as big of a deal...but the around $100 price tag may be

 

[TERRA Operative]

I have found in every instance bar none that an SSD upgrade is one of the most cost effective upgrades you can do to an old machine that has a SATA port, assuming you have at least a passable amount of RAM.
If not, upgrade both RAM and SSD, and it makes for a good experience.

An SSD and a few Gb of RAM is cheaper than a new PC..

[hans]

just for the records, I've installed 3 to 10 ssd's each week of work for more than 10 years now.
at most I would say 5 of all of them came back for defect ... and usually in the first month of usage.
just buy one of some reputable brand (I was selling samsung mainly), use the 3 to 5 years warranty if any bad luck,
AND MAKE AUTOMATIC REGULAR BACKUPS that is the key...
one of the first ssd I sold was a 500GB for ... EUR 1850 !!!! today these are less than EUR 50, and have far better specs.

I don't install so many per week, but other than that same experiences.
I got an OCZ Vertex 120GB when they were still 300+ euro. Best purchase ever.
Its still running, despite its atrocious reputation. It was running server duty for 2yrs until the firmware could take no more, and it would crash on heavy I/O first, then after some weeks of uptime.. then after just a few hours. I replaced it with a M.2 SSD and thats been rock solid as well. I moved the OCZ drive to my HTPC and it still runs mostly fine. No lock ups or uptime issues. Some boots into Ubuntu are slow for some reason, but when I restart the machine that issue is gone. The machine still feels response, despite the drive being slower than any harddrive of that era when it becomes to write speeds with randomized data (50-60MB/s on a half filed drive is about what you can expect). That drive is almost 14 years old now. All HDDs I had since then have been retired.

I've since then mostly bought Samsung drives. I've received a Samsung 830 as defective, but it failed on day 1, similar to a WD HDD also shipping with bad sectors. RMA'ed it, got it replaced in 2 days, and its been running rock solid.

I've had good experiences with Samsung and Crucial since then. Variety of 1TB and 2TB drives in my system, including 980Pro, no issues to report.
I'm even considering swapping some 4TB WD Red's in my NAS soon with SSDs if prices just drop a little further. The electricity costs on those kind of drives makes it worthwhile to make the jump. It would also give my NAS a big pool for VMs, docker containers and other live-work data.

I think despite SSDs having no moving parts, their overall failure rate is still on-par with regular HDD's. And we bought them for years without thinking about it. It looks like most SSD issues are related to firmware though, in contrast to HDDs which had more mechanical failure modes.
 Backups are always crucial. I'm typing this on a Macbook Pro with a soldered in SSD, which is a big shame on Apple, but I like the machine and OSX, and backuping data should be part of any good computer hygiene anyway.

[coppice]

The short summary is that a SATA SSD is a bit faster then a SATA HDD as an upgrade, but there is absolutely no comparison with a modern machine that has an NVMe Gen 4 SSD as original equipment.

That depends on what you are running. Try Linux with a SATA3 HDD drive, then move the same system to a SATA3 SSD. Things speed up, but its not a game changer. Try it with Windows 10, and its night and day. It feels like MS have gone completely in on the low seek time of an SSD, with Windows 10 and given up trying to mitigate the performance issues of accessing numerous small files. With Windows 7 you still had decent performance from an HDD system.

NVMe is in a different class, but that's not an option for trying to keep an old machine relevant.

[coppice]

SD cards are garbage. They fail in huge numbers. Used as a system disk, the lack of effective write spreading means they will rapidly fail as a swap space. Apart from the swap space wear type of issue, when I see an SD card have problems its not a bad block. The whole thing suddenly dies.

USB memory sticks are weak. They give a lot less trouble than SD cards, but they fail a lot more often than I'd like. Failures are like SD cards. The usually just die. If you get a bad block failure its usually in an application like swap space.

SATA SSD have served me very well. I have mostly used Crucial MX500s, and the prior Crucial models, and various Samsungs. This could be luck, but I've yet to have an issue with one, despite the scary tales of bugs.

Of course, all modern SSDs have storage lifetime issues when not powered up. Putting storage in a drawer to keep data safe for the long term still requires an HDD..... or the regular application of power for a considerable period, to refresh the data.

[Veteran68]

Don't forget that even old machines can benefit from M.2/NVMe if they have an available PCIe slot (x4, not sure that x1 bw would be enough). Obviously not talking about laptops here, but most all desktops except the tiny PCs would have expansion slots. You can buy a PCIe card with single or dual (maybe more) M.2 slots on them. I did so for an older desktop that had only one onboard M.2 when I wanted to add another.

[wraper]

Don't forget that even old machines can benefit from M.2/NVMe if they have an available PCIe slot (x4, not sure that x1 bw would be enough). Obviously not talking about laptops here, but most all desktops except the tiny PCs would have expansion slots. You can buy a PCIe card with single or dual (maybe more) M.2 slots on them. I did so for an older desktop that had only one onboard M.2 when I wanted to add another.

They won't be able to start booting from that drive though on machines that do not have NVMe support in BIOS. With PCI-E 1.x you need x4, and x2 for 2.x for it to be better than SATA.

[NiHaoMike]

They won't be able to start booting from that drive though on machines that do not have NVMe support in BIOS.

Get one of your old USB drives and use it for the /boot partition and bootloader, if the system has no other storage device.

[wraper]

They won't be able to start booting from that drive though on machines that do not have NVMe support in BIOS.

Get one of your old USB drives and use it for the /boot partition and bootloader, if the system has no other storage device.

Yes, that's why I wrote "start booting". However it a workaround requiring another drive and additional booting delay.

[bdunham7]

Example: an  i7-4790K running W10 with an SSD still flies.  Will it benchmark as good as a newer/faster CPU?. No, but it still flies!

Now if you have a machine with only "a few" applications installed, 

I have two old machines with no plans to replace.

One is my HTPC that I built with a i7-3770K.  It was (SATA) SSD from the beginning, but I upgraded to a 1TB Samsung and a 4K GPU when I was forced to downgrade to W10 from W7 and abandon Media Center.  The only thing it does that takes any amount of time is compressing/reencoding video files using Handbrake, and for that the latest, fastest versions would be faster, but not twice as fast--at least not for doing one file at a time.  Handbrake only uses about 3 cores at most per instance and per-core performance hasn't gone up as much as other parameters.  It easily plays videos while simultaneously encoding, so not much room for 'improvement'.

The other is my backup laptop, even older with an i7-2630QM and a Samsung 830 as the OS drive.  Among many other things it has a fully paid-up educational version of Adobe Photoshop (and other apps, the whole CS-5 suite) that cannot be reinstalled or licensed on any other machine.

Both of these machines would be intolerable with even a very fast spinning HDD, yet with a SATA SSD I'm not even considering replacing them.   

[David Hess]

I have had multiple data corruption problems with Samsung 870 EVO drives.

The only problem I have had with Crucial MX500 drives are 2TB models that had a hang bug when used with certain controllers (AMD Ryzen processors?), but there was a firmware update which fixed this.

It would be nice if firmware updates were available to run on any operating system, however it is not too taxing to temporarily install Windows on a system to update SSD firmware before putting the drives into service.

[BradC]

It would be nice if firmware updates were available to run on any operating system, however it is not too taxing to temporarily install Windows on a system to update SSD firmware before putting the drives into service.

I have a "Windows2Go" install of Win10 on a (fast) USB drive. It'll boot on pretty much anything eventually and can be used to do firmware updates on SSDs. The best part is having a clean backing file so I can replace the Windows config damage after it's done.

[David Hess]

It would be nice if firmware updates were available to run on any operating system, however it is not too taxing to temporarily install Windows on a system to update SSD firmware before putting the drives into service.

I have a "Windows2Go" install of Win10 on a (fast) USB drive. It'll boot on pretty much anything eventually and can be used to do firmware updates on SSDs. The best part is having a clean backing file so I can replace the Windows config damage after it's done.

I have a bunch of Windows 10 systems, including my laptop, but for my old workstation which is currently running FreeBSD, I have a Windows 10 boot SSD that I can swap in if needed.

[JoeRoy]

Samsung 970 Pro is the only MLC you can find (for a reasonable price).

[Infraviolet]

I don't think the 970 pro could work for me, I'm only looking at 2.5 inch SATA. This is for a laptop, I don't recall if it has fancy slots for non-SATA drives and can't tell without a slow and painful disassembly procedure. I'm therefore after a long-term-reliable 2.5 inch SATA SSD of a capacity the same as the HDDs I've gotten used to (1TB) to make a simple drop-in replacement*, I don't want to be trying to fidle with having multiple drives in the machine for different purposes, if multiple drives could fit at all.

*not sure if when doing so I'm better doing a fresh OS install on the SSD, then restoring from a timeshift external drive to get Mint back to my settings and programs, or doing some sort of dd of my existing drive to the SSD then putting it in and trying to boot straight to that.

[Haenk]

Not buying due to reports on the internet? That's BS.
Sure, at times, there is a bad batch, but that's rare.
Buy a quality brand one, I have mainly installed Samsung. Get a PRO edition, I would not bother spending a bit more for them - as they are really cheap now anyway.
Don't buy questionable chinese brands off Amazon. They are usually junk.

[Jeroen3]

If you have such high requirements just buy datacenter drives. And buy n+1 and put them in sw-raid.

I have 8 used (80-90%) Micron M600 1TB in a Windows Storage Space (4tb) and it works great. It's fast and it will remain fast the entire 4 TB.

[Infraviolet]

For SSDs does anyone know if Amazon's fulfilled by Amazon orders are safe for major brand SSDs (Samsung, Crucial...)? Or if there are loads of fakes on sale there? Guessing any which are sold as major brands but coming from a self-delivering seller with an unpronounceable name are probably fakes with the same internals as whatever no-name ones are sold under Chinese brands?

[David Hess]

I have had multiple data corruption problems with Samsung 870 EVO drives.

I finally gave up on the Samsung 870 EVO because of data corruption.  I am not sure what happened.  Maybe the 4TB 870 EVO draws too much power for my USB 3.0 enclosures, or maybe the 870 fakes synchronous writes to improve performance at the expense of reliability.  None of my 2TB Crucial MX500s have had a problem.

Continuous write speed of the Samsung 870 EVO was also not any better than the Crucial MX500s, and both were *slower* then the 14TB Ultrastar DC HC530 hard drives that I have started using.

[Veteran68]

Maybe the 4TB 870 EVO draws too much power for my USB 3.0 enclosures, or maybe the 870 fakes synchronous writes to improve performance at the expense of reliability. 

Both of our experiences are anecdotal, but while I can't speak to the 4TB 870 EVO, I've had a 4TB 870 QVO (cheaper) drive installed internally in my main desktop since Dec 2020 and it's been great. Some reviewers claimed it would be a slower & less reliable drive than the EVO, but that has not been my experience at all. It's the only 2.5" SATA SSD in this desktop, the other 3 are M.2 NVMe drives (two of which are Samsung gen4's, one a Silicon Power gen3).

[David Hess]

Maybe the 4TB 870 EVO draws too much power for my USB 3.0 enclosures, or maybe the 870 fakes synchronous writes to improve performance at the expense of reliability.

Both of our experiences are anecdotal, but while I can't speak to the 4TB 870 EVO, I've had a 4TB 870 QVO (cheaper) drive installed internally in my main desktop since Dec 2020 and it's been great. Some reviewers claimed it would be a slower & less reliable drive than the EVO, but that has not been my experience at all. It's the only 2.5" SATA SSD in this desktop, the other 3 are M.2 NVMe drives (two of which are Samsung gen4's, one a Silicon Power gen3).

I think the drive got corrupted after I moved it to a USB enclosure, but the nature of the corruption affecting the metadata makes me think it was a synchronous write problem where the drive reported a synchronous write completed, but it was not and more data was sent and then power was lost.

Maybe it is just a questionable drive but diagnostics reported no problems.

[Infraviolet]

Has anyone had corruption with the 1TB drives? Not sure how different the large capacity 4TB units might infact be internally as vs more standard 1TB units of supposedly the same model name. It wasn't that long ago after all that a 1TB HDD was the largest commonly found, so 4TB SSDs are quite a new thing by comparison.

[David Hess]

Has anyone had corruption with the 1TB drives? Not sure how different the large capacity 4TB units might infact be internally as vs more standard 1TB units of supposedly the same model name. It wasn't that long ago after all that a 1TB HDD was the largest commonly found, so 4TB SSDs are quite a new thing by comparison.

I only have 250GB and 2TB SSDs.  At this point I have collected a set of 4 x 250GB Crucial BX500s which were replaced with 250GB Crucial MX500s, so I think I will use them in RAID10 on my motherboard as a boot volume rather than let them sit around until useless.

[magic]

I think the drive got corrupted after I moved it to a USB enclosure

I have had bizarre incidents of data corruptions with USB enclosures, I simply don't trust those things as a matter of principle.

Even with the ones that seem to pass testing, I recently started to use BTRFS because it detects silent corruption.
Metadata redundancy (duplication) is also possible with it.

When I have to read an internal disk on a USB bridge, any important data I read twice and compare.

[thm_w]

For SSDs does anyone know if Amazon's fulfilled by Amazon orders are safe for major brand SSDs (Samsung, Crucial...)? Or if there are loads of fakes on sale there? Guessing any which are sold as major brands but coming from a self-delivering seller with an unpronounceable name are probably fakes with the same internals as whatever no-name ones are sold under Chinese brands?

If its shipped and sold by amazon, its generally safe. Though in Canada if you buy a samsung SSD from amazon it will not be under warranty (grey market). So you're better off buying elsewhere.
There are definitely many fakes available, just moreso on aliexpress and not in the top amazon results.

[tridac]

Slowly upgrading to an ssd future, now that enterprise quality ssd are becoming more affordable second user. Present server and development host is running FreeBSD 12,  has two 200Gb in a mirrored zfs root for the os, and 6 x 800Gb in a zfs raidz pool for data. Drives are Samsung, HGST or Toshiba with SAS interface. There was some risk to start, as no mention of wear index in the ads for the drives,  but using smart utils, (smartctl -a <device>) which extracts the drive info, the wear index is less than 5% for all of them, with some showing zero. It's a work in progress, as funds allow, but enterprise class sas drives from Samsung or HGST, for example, seem to be a good bet second user. Boot times and power consumption are much improved as well...

[Infraviolet]

The samsung 870 EVO is still a particular model to really avoid then? Or are things better with more recently manufactured ones?

[voltsandjolts]

You started this thread 8 months ago.
What did you buy in the end?
Or are you still waiting for "the one"?

[hans]

If you read back the OP, then the 870 EVO had "certain batches" that failed.

In other words, if you get a drive from that batch: upgrade firmware (if that helps) or return it. Otherwise consider it anecdotal evidence which says nothing about the general quality, as we don't know how what the failure rate is.


There is no "the one". Every manufacturer will screw up in one way or another, as that is what Murphy's law predicts. If all it takes is a small scandal to blacklist a manufacturer, then there is no mfgr left.

[Halcyon]

I stick with Crucial/Micron or Intel these days.

[tom66]

I just plugged in an old SSD from a system of mine.  It's 10 years old and everything seems fine with the drive.  120GB manufactured by PNY.  I also have a 6 year old MX500 drive in this system.  I think as long as these drives aren't write-abused, they should last a long time.

Meanwhile my Home Assistant instance died last night as the SD card in the Raspberry Pi crapped itself.  It went read only, which the Pi doesn't want to boot from, but I was fortunately able to recover data from it.  My guess is the bad block table has hit its limit.  At least they built the facility into it to allow data recovery before it is too late.  Also a good proof of my Docker backup strategy... NOT... everything fell apart... but at least I know now!

[Psi]

When it comes to SSDs, they are kind of unusual. 
With most things spending more gets you more reliability, but this is often not so with SSDs.
Spending more $$$ on a SSD gets you faster read/write speeds which comes with much higher temps and the bleeding edge FW version, so has a higher risk of failure or of FW bugs.

I'm not saying to buy cheap junk SSDs, you still want a known ok brand.


If you don't want to have to worry about it, get two SSDs and mirror them. Ideally different brands with similar specs.
But you do still need an offsite backup if you want to be safe.


+1 for Intel SSDs,  however i've not really looked into the brand in a while, my ones are still going strong.

[NiHaoMike]

It's worth considering used TLC and MLC SSDs. (Avoid used QLC unless it's insanely cheap.) Those often are cheaper than new QLC and have a lot more writes remaining than QLC ever had to begin with. The best deal I scored was $68 for a 2.6TB ioMemory PX600, MLC for less than the price of QLC. It's not bootable but that's no problem as the machine it's going in has other SSDs as well.

[Infraviolet]

I opened this thread a while ago when I last had HDD trouble and was considering an SSD replacement, now I have more HDD trouble and need a replacement urgently. And with my bad experiences with HDDs recently I think it's time to try SSD instead. So yes, I waited 8 months before this point.

Now I'm looking at the 1TB versions of:
Crucial BX500
Crucial MX500
WD Blue SA510
WD Green WDS100T3G0A

Any advice on whether any of those should be avoided?
I can't always see in the product descriptions which are MLC/TLC/QLC, any recommendations there?

EDIT: if anyone knows of an SSD brand that makes applying firmware fixes easy for pure linux users (no windows systems to hand at all) that might be a good choice, I keep seeing awful warnings about SSDs needing firmware fixes but the firmware updates being distributed only through convoluted Windows toolchains that won't work for drives with ext4 partitions on them.

Further edit: I can't seem to find any reports of a specific bug with the Crucial BX500 drives? They're apparently the low cost version, but appear to have the same endurance rating (360TB W) as the MX500? Is there a reason not to go with this sort? And I think the BX500 is TLC, not the dreaded QLC? The price difference of MX500 vs BX500 is a very small percentage, so in performance terms the BX500 may be a relative rip-off, but is it worthwhile in longevity terms, I see no reports of it having any firmware updates available, which suggests there have been no firmware mistakes in its design. Not having to apply any firmware updates would be good from a Linux perspective.

[BradC]

Crucial BX500
WD Green WDS100T3G0A

Avoid ^^^^.

Both are cacheless wonders. Can't comment on the WD Blue, but I have greens and they suck. I also have a couple of bigger MX500s and I hammer them as spool drives. Only broke one, once but nothing a secure erase didn't fix. I was trying something pathological and the drive called enough.

[Infraviolet]

"Both are cacheless wonders"
Does that just mean they're a bit slow? or does it affect lifespan and reliability?

To my understanding even a rather slow SSD is blazingly fast against an HDD?

If slowness, even down to near HDD levels, gets me longevity in excess of HDDs and a drive which won't get damaged during shipping or when moving the laptop... then I'm all for it.

[BradC]

"A bit slow" is subjective. If your write loads are small you may be ok. If you want to write significant data then you need to be prepared for write speeds to drop to 1990s hard drive speeds. I've seen 20-50MB/s from my WD Greens.

Aside from slow streaming writes, their performance is simply inconsistent. Dunno about where you are, but here the price difference between a BX and MX is less than 15% in most cases. Their performance difference under random writes is chalk and cheese.

[coppice]

Now I'm looking at the 1TB versions of:
Crucial BX500
Crucial MX500
WD Blue SA510
WD Green WDS100T3G0A

Assuming the Crucial MX500 is still the same product they've been making since 2017, and they haven't reworked it without change the name, that is a solid performer with a good track record.

[Infraviolet]

"Assuming the Crucial MX500 is still the same product they've been making since 2017"
Therein lies the problem, I don't think any SSD manufacturer has been making the same drive since 2017. it seems they al make revisions and changes, but keept the model numbers unchanged. The worry with the MX500 is the firmware bug which needs an update to M3CR046, but updating is virtually impossible on Linux. I don't know whether they've actually updated the firmware in the drives that are selling now, and there's no part number revision or anything to make it obvious what version of drive one is about to buy. That's why the BX500 got my interest, no mention of an equivalent firmware bug for it.

"I've seen 20-50MB/s from my WD Greens."
As performance comparisons go, the dying HDD I've been used to is a Toshiba MQ01ABD100, supposedly reading and writing at about 80MB/s? Although hdparm tells me:
sudo hdparm -Tt /dev/sda
/dev/sda:
 Timing cached reads:   10994 MB in  2.00 seconds = 5506.90 MB/sec
 Timing buffered disk reads: 260 MB in  3.01 seconds =  86.28 MB/sec

So I'd be looking for speeds no worse than this. The largest files I handle are m4v format videos of up to 4GB in size, but only occasionally do I move or copy them, mostly it's just reading them by watching via VLC. I have a tiny number of >10GB zip files around I occasionally extract hundred megabyte files from. Most of my usual work on the PC is web browsing, source code editing in text editors, and some relatively light-weight CAD (not much heavy rendering or computational solving involved).

[coppice]

"Assuming the Crucial MX500 is still the same product they've been making since 2017"
Therein lies the problem, I don't think any SSD manufacturer has been making the same drive since 2017. it seems they al make revisions and changes, but keept the model numbers unchanged. The worry with the MX500 is the firmware bug which needs an update to M3CR046, but updating is virtually impossible on Linux. I don't know whether they've actually updated the firmware in the drives that are selling now, and there's no part number revision or anything to make it obvious what version of drive one is about to buy. That's why the BX500 got my interest, no mention of an equivalent firmware bug for it.

The most recent firmware I can see on Crucial's site is M3CR045. I think updating from within Linux is not supported, but the bootable ISO images Crucial supply do not require an installed OS to update a drive. I have used MX500s extensively on Linux machines since they first appeared, and never had trouble.

[Monkeh]

M3CR046 never got published to their site, but it's easily downloaded and can be applied with hdparm. For older versions you can simply extract the firmware image and their own tool (which will do exactly the same thing as hdparm) from their bootable image.

They also offer both GUI and CLI Linux versions of the 'storage executive', which I've never felt the need to play with - but is the tool run by their own update images: https://www.micron.com/sales-support/downloads/software-drivers/storage-executive-software

[coppice]

M3CR046 never got published to their site, but it's easily downloaded and can be applied with hdparm. For older versions you can simply extract the firmware image and their own tool (which will do exactly the same thing as hdparm) from their bootable image.

They also offer both GUI and CLI Linux versions of the 'storage executive', which I've never felt the need to play with - but is the tool run by their own update images: https://www.micron.com/sales-support/downloads/software-drivers/storage-executive-software

Is there some key problem that is only fixed in M3CR046? I've never used an update that didn't come from Crucial's site, and I've never had a problem with the drives.

[Monkeh]

M3CR046 never got published to their site, but it's easily downloaded and can be applied with hdparm. For older versions you can simply extract the firmware image and their own tool (which will do exactly the same thing as hdparm) from their bootable image.

They also offer both GUI and CLI Linux versions of the 'storage executive', which I've never felt the need to play with - but is the tool run by their own update images: https://www.micron.com/sales-support/downloads/software-drivers/storage-executive-software

Is there some key problem that is only fixed in M3CR046? I've never used an update that didn't come from Crucial's site, and I've never had a problem with the drives.

There's some obscure bug which can cause the drive to hang under certain loads. None of my MX500s are a hardware revision which takes the 04x firmware series.

[Infraviolet]

"None of my MX500s are a hardware revision which takes the 04x firmware series."
The problem is the drives keep changing but keep the same model number. So I've no idea what will actually turn up when I order a drive, same situation for all manufacturers. Has anyone been buying MX500s, BX500s or EVO 870's recently?

[Monkeh]

You're overthinking it. If your data matters it doesn't reside on a single device - any loss of a device should cost you nothing more than the last few hours if it's a single drive system (backups cover everything else) and the time needed to swap it out.

Just buy something.

[coppice]

You're overthinking it. If your data matters it doesn't reside on a single device - any loss of a device should cost you nothing more than the last few hours if it's a single drive system (backups cover everything else) and the time needed to swap it out.

Just buy something.

better still buy two somethings, and have some redundancy. Even the best drives fail, and often suddenly and completely.

[Infraviolet]

"If your data matters it doesn't reside on a single device - any loss of a device should cost you nothing more than the last few hours"
Yes, I keep many copies of all my important data, but restoring my OS, programs, settings... even with some system images I've made to speed up the process, still costs me two days, and opening up the laptop to physically swap the drive out is unpleasant (it is one without a bottom cover, so you have to take everything apart, and had to remove the laptop's rubber feet to get at the screws when I last changed the HDD.

I've had enough of having to do this procedure multiple times in the last year, I want something that won't fail on me again so soon.

From what I'm reading the MX500's bug is perhaps less serious than the EVO 870's? The EVO 870 has an actual degradation of the physical flash chips I think, whereas the MX500 seems to be solely an issue in firmware. There look to be workarounds to maybe do MX500 firmware updating in Linux, or with bootable images, though I hear they need some tinkering wit before they'll work with keyboard input. So if I went for the MX500 I'd encounter any problems immediately, rather than later down the line?

Or there's the BX500, the one thing going for it is I don't see many reported errors with it, but would it infact give worse performance than an HDD, due to the lack of DRAM?

[Infraviolet]

Another thought, what about WD red drives? Are the higher reliability? But are the the sort which is unable to be used as a boot device?

I know it is similar to the SA500 Blue, but the Red is more of an enterprise specced version, SA500. Does this mean it will be a high reliability SSD?

[Someone]

"If your data matters it doesn't reside on a single device - any loss of a device should cost you nothing more than the last few hours"
Yes, I keep many copies of all my important data, but restoring my OS, programs, settings... even with some system images I've made to speed up the process, still costs me two days

So you need a better backup solution that does capture all that.

[David Hess]

From what I'm reading the MX500's bug is perhaps less serious than the EVO 870's?

I had the MX500 bug when I installed a 2TB one in my modern laptop.  It only resulted in the system hanging and no data was lost.  The EVO 870 bug silently corrupted data.

Or there's the BX500, the one thing going for it is I don't see many reported errors with it, but would it infact give worse performance than an HDD, due to the lack of DRAM?

Performance was practically identical, but I suspect the MX500 suffers from less write amplification because of the DRAM cache.

Something I recently noticed is that none of my SATA USB drives support TRIM through the USB interface, and apparently none do, at least with Windows, but my one NVMe USB enclosure does.  I am thinking of making a big USB disk array with my collection of 2TB SATA SSDs and replacing the two I regularly use with NVMe drives.

It's worth considering used TLC and MLC SSDs. (Avoid used QLC unless it's insanely cheap.) Those often are cheaper than new QLC and have a lot more writes remaining than QLC ever had to begin with.

I am glad that I avoided QLC drives.  My oldest SSDs are only a couple years old, and if they had been QLC drives, then they would have already have exceeded their write endurance.

As it is, I am going to exceed the write endurance on some of my TLC drives before reaching their 5 year warranty.

[Infraviolet]

I'm really getting tempted by the WD RED SA500 drive now. Looks like similar performnce to typical SSDs, but more enterprise grade, and to my understanding the design has barely changed, and no firmware updates released, since it was first available.

Can anyone give me a good reason not to select a WD RED SA500? (speed reasons are only "good" in my books if the speed of this would be actively worse than the spinning rust Toshiba MQ01ABD100 models I am used to)

I know the red is made for NAS usage, but that mainly just means its a bit tougher in its reliability ratings than a normal consumer SSD? And everything I'm reading online so far says WD Red drives can work fine as the sole drive, the boot drive, in a PC.

Thanks

[coppice]

I'm really getting tempted by the WD RED SA500 drive now. Looks like similar performnce to typical SSDs, but more enterprise grade, and to my understanding the design has barely changed, and no firmware updates released, since it was first available.

Can anyone give me a good reason not to select a WD RED SA500? (speed reasons are only "good" in my books if the speed of this would be actively worse than the spinning rust Toshiba MQ01ABD100 models I am used to)

I know the red is made for NAS usage, but that mainly just means its a bit tougher in its reliability ratings than a normal consumer SSD? And everything I'm reading online so far says WD Red drives can work fine as the sole drive, the boot drive, in a PC.

Thanks

Any SSD made in the last few years can keep a SATA bus saturated almost all the time. Its only the NVMe drives which show speed differences these days.

[IanB]

a) Some years ago (2019) I installed a Samsung SSD 860 PRO 1TB drive into an older laptop as an upgrade, and used the heck out of it as a daily workhorse for several years until I retired the PC. As of today, it reports good health on all the S.M.A.R.T. statistics, with 15.8 TB written out of a warranted 300 TB. Benchmark results show sequential read and sequential write near maximum, but random read and random write are not so good (however, this could be due to the ancient PC hardware).

b) The performance of a SATA interface SSD is woefully slow compared to an NVMe SSD.

Given (a), you can overthink things. Just find a good brand, reasonably priced SSD, like Samsung, install it, and forget it.

Given (b), you should give serious thought to replacing the PC. Upgrading components on older hardware is not really efficient.

[IanB]

Wow, is this thread so old? I just scrolled up and realized I've posted in it before. Here was me thinking it was a new thread.

[mariush]

Techpowerup has a nice SSD database here : https://www.techpowerup.com/ssd-specs/

IMHO WD Green SATA drives are not worth it. They usually have quite low endurance... It seems to me like they're using badly binned flash chips or chips that had a lot of  failed cells in them (like they're using 512 GB chips to make 240-256 GB functional drive)
They're fine for booting up an OS on an office machine but not storing long term.

The BLUE drives should be fine, and the reds should also work well but I'm not sure it's worth the extra money.

Other brands Samsung 860, 870 Evo ...  not sure who else I'd use in SATA format.  At work we bought Kingston A400 240 GB for office type machines, the documents and stuff is kept on NAS anyway.

Maybe Kingston KC600 would be OK as well, if you can find it.

Note that you can get  M.2 to SATA enclosures for around 10$, see for example : https://www.amazon.com/Sabrent-2-5-Inch-Aluminum-Enclosure-EC-M2SA/dp/B01N6PMZLW

Not many M.2 drives that are SATA based and good but there's a few: WD Red SA500 , WD Blue SA510, Silicon Power A55

[David Hess]

Any SSD made in the last few years can keep a SATA bus saturated almost all the time. Its only the NVMe drives which show speed differences these days.

SSD continuous write speeds are still slower than SATA3.  When I do my backups to an external USB enclosure, NVMe is not any faster unless the NVMe drive itself supports faster continuous writes.

Can anyone give me a good reason not to select a WD RED SA500? (speed reasons are only "good" in my books if the speed of this would be actively worse than the spinning rust Toshiba MQ01ABD100 models I am used to)

I have been burned by Western Digital too many times to trust them.  I was never able to confirm whether their SSDs support any form of power loss protection.

[paulca]

I learnt the hard way that not all SSDs are the same.

I bought crucial SSDs for a virtualisation server.  They ended up in a gaming PC, where they work fine.

"I am told":  Most, average, consumer SSDs use a limited amount of high speed flash and back it with a much larger slow (to write) flash.  The result is that for short, bursty writes like a normal OS or gaming PC, it's absolutely fine.  However, if you hit it with transfers larger than it's high speed cache, it's performance will collapse back to that of the slow write flash.  This can be an order of magnitude slower or more.

This is before you consider the "write leveling" and other dynamic sectoring which needs maintenance cycles for various filesystems or performance tanks in much the same way as an old HD that was fragmented would.

All of this results in bottle necks causing bottle necks causing bottle necks.  When the transaction queue in the kernel is overflowing with filesystem write requests and some of them start getting told "No".  Things get unstable.  I mean who actually checked the return code for half those functions? (J/k)

Examples of how to trash a cheap Crucial SSD?  Do a "docker pull" on a project with 6 images built from 12 dependencies and docker launches 18 threads all writing Gb image files to disk.  This along caused a server wide lag and even a few kernel lock errors.

I was also told to... check for full capacity write performance tests.  If you spend twice as much you will find a few that can sustain full write speed for their full capacity.  Usually, lately the Samsung Evo range.

If you want some confidence in what you are getting, get your wallet out and go enterprise.

[mariush]

Correct, most SSDs switch a part of the free space to pseudo-SLC mode, storing only 1 bit in each cell instead of 3 bits for TLC or 4 bits for QLC - so for example, if you have 100 GB of free disk space, you may get around 32 GB of SLC write cache on TLC drives, or around 25 GB of SLC cache in the case of QLC drives.

The speeds drop from 3-7 GB/s down to around 500 MB to 1.5 GB/s on TLCs, while QLC drives will drop to around 100-250 MB/s when the write cache is full.

Drives like WD SN570 using older controllers will use smaller write caches, up to 13 GB, and speeds drop to around 600 MB/s when this amount is full....  https://www.techpowerup.com/ssd-specs/western-digital-sn570-1-tb.d467

The newer SN770  can pretty much switch the whole drive to SLC mode, and if you manage to fill it up, write speeds will drop to around 560 MB/s  ...  https://www.techpowerup.com/ssd-specs/western-digital-black-sn770-2-tb.d647

Samsung 980 I own can do up to around 122 GB of cache (the 1 TB version goes up to around 160) ... From 3.1 GB/s reads , 2.6 GB/s writes will go down to around 400 MB/s when the write cache is full.

The Samsung 980 Pro and 990 Pro can go up to around 250 GB of write cache but manage to flush to TLC memory much faster, at up to 1.5 GB/s

[coppice]

I have been burned by Western Digital too many times to trust them.  I was never able to confirm whether their SSDs support any form of power loss protection.

Western Digital bought Sandisk, and I assume that is the heart of their SSD business now. I haven't used much Western Digital branded SSD hardware, but I had terrible experiences with Sandisk reliability, from their SD cards to their SSDs.

[tom66]

One thing for sure: do not buy Kingston devices.

Kingston do not make their own flash, and the quality of their drives is consequentially variable from batch and model to model.  I have had both Kingston RAM and Kingston SSDs go bad well within normal hardware lifespan, and the fiasco over the V300 SSDs where Kingston shipped review samples with faster flash and controllers and then changed the production units has put me off for good.

[David Hess]

Kingston do not make their own flash, and the quality of their drives is consequentially variable from batch and model to model.  I have had both Kingston RAM and Kingston SSDs go bad well within normal hardware lifespan, and the fiasco over the V300 SSDs where Kingston shipped review samples with faster flash and controllers and then changed the production units has put me off for good.

When I started considering SSDs a couple years ago, I evaluated many manufacturers but quickly narrowed my list to those who make their own Flash, so Crucial and Samsung.  Intel was a close third.

[terrance]

YMTC produces some nice stuff. But seemingly they don't export much.

A Chinese youtube guy did a lot of accelerated aging and other torture tests (on samples from China's market). Some are still on going.

in case you are interested

https://www.youtube.com/@XFJDIY

Basically:

samsung is mostly ok-ish, but power consumption/temperature is high, thus PMIC tent to blow. if controller is dead, data recovery almost impossible
intel/hynix: high leakage, not suitable for off-line storage.
toshiba/kioxia/sandisk/WD: ok. data recovery almost impossible
ymtc: above ok, data recovery almost impossible for now, due to LDPC, maybe possible later.
cheap, OEM and BS "gaming" brands: varies ok to bad, but for some controllers, data recovery is likely possible.

[mariush]

Kingston do not make their own flash, and the quality of their drives is consequentially variable from batch and model to model.  I have had both Kingston RAM and Kingston SSDs go bad well within normal hardware lifespan, and the fiasco over the V300 SSDs where Kingston shipped review samples with faster flash and controllers and then changed the production units has put me off for good.

When I started considering SSDs a couple years ago, I evaluated many manufacturers but quickly narrowed my list to those who make their own Flash, so Crucial and Samsung.  Intel was a close third.

Crucial is Micron's brand. So it will only ever use Micron flash memory : https://www.techpowerup.com/ssd-specs/filter/?mfgr=Crucial

Kingston buys wafers from Micron and at least for ram, they bought them untested and did the validation in-house to save money.  I doubt they're doing it for nand.

See https://www.techpowerup.com/ssd-specs/filter/?mfgr=Kingston , pretty much Micron and Toshiba for everything.


Western Digital has a partnership with Toshiba and invested heavily into the Toshiba / Kyoxia nand factories. So technically it can be considered manufacturer.

example

2019 https://www.westerndigital.com/en-ap/company/newsroom/press-releases/2019/2019-05-16-toshiba-memory-and-western-digital-to-jointly-invest-in-flash-manufacturing-facility-in-kitakami-japan

Toshiba Memory Corporation and Western Digital Corp. (NASDAQ: WDC) have finalized a formal agreement to jointly invest in the “K1” manufacturing facility that Toshiba Memory is currently constructing in Kitakami, Iwate Prefecture, Japan.

2022 https://www.westerndigital.com/en-ap/company/newsroom/press-releases/2022/2022-04-14-kioxia-and-western-digital-jointly-invest-in-new-flash-memory-manufacturing-facility

Kioxia Corporation and Western Digital Corp. (NASDAQ: WDC) have finalized a formal agreement to jointly invest in the first phase of the Fab7 (Y7) manufacturing facility at Kioxia's industry-leading Yokkaichi Plant in the Mie Prefecture of Japan. With construction of the first phase of Y7 completed, the joint-venture investment will enable initial production output beginning in the fall of this year. This marks another important milestone in the 20-year strategic joint-venture partnership between the two companies.

This joint-venture investment adds a sixth flash memory manufacturing facility to the Yokkaichi Plant, enhancing its position as the world's largest flash memory manufacturing site. The first phase of the Y7 facility will produce 3D flash memory including 112- and 162-layer and future nodes.


2024, quite recent news   https://www.trendforce.com/news/2024/02/26/news-nand-flash-manufacturers-kioxia-and-wd-reportedly-set-to-resume-merger-talks-in-late-april/

NAND Flash Manufacturers Kioxia and WD Reportedly Set to Resume Merger Talks in Late April


and some interesting bits from that press release ... at Q3 2023, market share:

31.4% Samsung  31.4%
20.2% SK Hynix + Solidigm
16.9% WDC
14.5% Kioxia
12.5% Micron
04.6% Others


Anyways... if you're crying for Intel flash memory,  well the flash memory division was sold to SK Hynix ... they're using the Solidigm brand to sell drives with QLC memory that's based on the old Intel flash memories.

Solidigm had some SSDs that were very optimized for low power consumption, great for use in tablets and laptops.

[voltsandjolts]

There is no "the one".

Yes, I know. That was "my point".
Just curious as to why the OP is still banging on about this 8 months later, instead of just buying a half decent drive and getting on with other actually interesting stuff.

[BennoG]

For me the most important factor for an SSD is the quality of the wear-leveling.
And I will stay far away from the QLC type of drives. The problem is most manufactures don't mention type type of cell anymore.
As you can image if you have 16 different voltage levels in a single cell then even if a couple of electrons leak you have a different value (this can possible be corrected by ECC etc)

P.S. this is my personal opinion.

source of the image   https://www.kingston.com/en/blog/pc-performance/difference-between-slc-mlc-tlc-3d-nand

Benno

[paulca]

I think it stands that if you expect "one size fits all", you will find yourself out of pocket pretty quickly.

If you want an SSD for a desktop PC which will not be processing multi-gigabyte disk IO frequently, then that small (ish) burst of high speed is all you will need.  So a bog standard 50 quid a terrabyte is all you want.

If you, like I, want a SSD to survive under a busy virtualised environment running several servers and multi-gig IO transfers are common (docker/kurbenetes et al as well as transfering entire VMs around)... a 50quid a terrabyte drive will absolutely NOT do.

The largest bulks storage I have, accounts for about 95% of storage by volume, is "infrequent write", "frequent read". 

In cases where speed is important on read and less important than reliability, I strip 2 or more cheap SSDs together.  (consider a Steam drive or a drive you load games from).

[madires]

WD and SanDisk, today the same company for SSD production. Currently being sued for having produced external SSDs under both brand names which suddenly bricked and wiped the data on them. Also a scandal with critical firmware updates being needed for the SA510 line of WD branded internal drives.

And this one:
SanDisk Extreme Pro Failures Result From Design and Manufacturing Flaws, Says Data Recovery Firm: https://www.tomshardware.com/news/sandisk-extreme-pro-failures-are-due-to-design-flaw

So far all manufacturers had/have some issues. It's quite similar to things we have seen for classic spinning disks. Some product families are great, some need a firmware update, and some are junk. If your data is important to you make backups!

BTW, when buying an SSD make sure to buy from a reputable seller to reduce the risk of getting a counterfeit.

[WatchfulEye]

Having gone through a couple of dozen SSDs in the last few years, I have had a couple of early incidents of data loss (blocks of sectors are unreadable) - rather than total drive failure.

One was found, annoyingly, because a backup was failing - it turned out that the backup software would panic when it received a "bad sector" notification, and abort the entire backup with an unhelpful error message. Overwriting the affected sectors triggered a sector reallocation event (deleting the affected files, and then filling the drive's free space with random bits), and the drive was restored to service, and has worked fine ever since. I suspect that the data loss occurred within a couple of weeks of installing the drive, but the affected sectors were infrequently accessed, and a change in backup rotation resulted in a re-read (a scheduled full backup, rather than an incremental backup).

The other was found proactively during a periodic test-reading of the drive, and the affected files were restored from backup.

I've been trying a "burn in" procedure on new SSDs, perfoming a set of 5 full drive write/verify cycles, before putting the drives into service. So far, however, it hasn't yielded any results other than wasting time.

[coppice]

As you can image if you have 16 different voltage levels in a single cell then even if a couple of electrons leak you have a different value (this can possible be corrected by ECC etc)

You do realise they don't even try to recover 16 distinct levels from the flash cells, right? ECC on these things is not handled at the digital level. They use coding, and soft bit data recovery from the medium. This is similar to how most modern comms channels, and hard disks work.

[amyk]

For me the most important factor for an SSD is the quality of the wear-leveling.
And I will stay far away from the QLC type of drives. The problem is most manufactures don't mention type type of cell anymore.
As you can image if you have 16 different voltage levels in a single cell then even if a couple of electrons leak you have a different value (this can possible be corrected by ECC etc)

P.S. this is my personal opinion.

source of the image   https://www.kingston.com/en/blog/pc-performance/difference-between-slc-mlc-tlc-3d-nand

Benno

Those numbers should really be considered false advertising now. I don't think 100K SLC is even made anymore, and what parts are available are tiny. For example Micron MT29F32G08ABAAA is only 60K cycles. Samsung K9L8G08U0M (MLC) is 1.5K cycles.  Look at the datasheets that have escaped into the public, or divide "TBW" by the capacity of an SSD to get the real endurance figures. They've gotten a lot more secretive with TLC and QLC. They've also been slowly decreasing the retention to make the endurance numbers look better, since you can program for more cycles and make the cells leakier if you don't need the charge to stay as long. Those numbers above are 60K/10 years and 1.5K/5 years. TLC is more realistically a few hundred cycles for 5 years, and QLC probably around 100 for 3 years.

[David Hess]

I don't think 100K SLC is even made anymore, and what parts are available are tiny.

Swissbit sells 128G SLC drives ... for more than $1000 each.

Look at the datasheets that have escaped into the public, or divide "TBW" by the capacity of an SSD to get the real endurance figures.

That is the first thing I calculate.

TLC is more realistically a few hundred cycles for 5 years, and QLC probably around 100 for 3 years.

400 to 600 for TLC, and 200 for QLC but I only have one entry for those.  There was an Intel TLC drive that I considered with a calculated endurance of 3750, which is more like how MLC performs, so I do not know what was going on there.

[Infraviolet]

I opted for the WD Red SA500 in the end, no firmware updates released since production egan in 2019 does imply they've made no mistales they needed to fix. I'm doing my reinstall on to it today, so will soon see what its like.

[PlainName]

Bit late to this thread but my personal experience of WD NVMe SSD might be useful.

I have several disks in my main PC and currently a 1TB NVMe SSD as the boot disk with the following volumes on:

C: - Windows system and boot
D: - General data
E: - Applications (much less writing that the other volumes)
F: - Temporary stuff. Typically not much used
G: - Chip SDKs, IDEs and associated tools

After a bit of use (of the previous drive to the WDC) I was seeing some performance issues, but whenever I ran a speed test it came out with good numbers, until it didn't and then I invested in the WDC replacement. Around then I invested in HDTune Pro and did a benchmark of the drive in November 2021 (screeny attached). You can see that the early part of the drive is a bit slow,with a huge dip around 180GB in.

[Note: HDTune hits the disks sectors directly so is unaffected by filesystem and similar high-level abstractions.]

The volume layout has space between volumes for expansion, so C: might be 90GB allocated then 25GB free for expansion, D: has 75GB then 10GB free, etc. The upper 420GB of the disk is unallocated (for expansion and/or wear levelling). The early dip corresponds to drive C:, which is quite heavily accessed in both read and write modes.

OK so far. Now wind forward to August 2023 and see the second HDTune benchmark. The space used by drive data is clearly visible from the massive performance drop, and the unallocated space is still as good as new.

This isn't unique to WDC but seems to affect any NVMe SSD I've tried, some quite a bit worse than the WDC. I guess SATA SSDs have a low enough bus speed that the performance degradation doesn't affect transfer speeds.

[thm_w]

Can anyone give me a good reason not to select a WD RED SA500? (speed reasons are only "good" in my books if the speed of this would be actively worse than the spinning rust Toshiba MQ01ABD100 models I am used to)

I have been burned by Western Digital too many times to trust them.  I was never able to confirm whether their SSDs support any form of power loss protection.

I doubt any consumer grade SSDs would have usable power loss protection.
If you care you'd buy an enterprise grade drive (which has expensive supercaps inside), or, use a UPS (probably the cheaper option).

[coppice]

Can anyone give me a good reason not to select a WD RED SA500? (speed reasons are only "good" in my books if the speed of this would be actively worse than the spinning rust Toshiba MQ01ABD100 models I am used to)

I have been burned by Western Digital too many times to trust them.  I was never able to confirm whether their SSDs support any form of power loss protection.

I doubt any consumer grade SSDs would have usable power loss protection.
If you care you'd buy an enterprise grade drive (which has expensive supercaps inside), or, use a UPS (probably the cheaper option).

Some of the early SATA SSDs claimed to complete work in progress as the power fell. I don't know about current SATA drives, but NVMe drives don't seem to have anything big enough to store some energy to do that.

[David Hess]

Enterprise drives not only complete writes, but protect data which is in transit, so nothing should ever be lost.

At a minimum drives need to be able to complete write operations to prevent corrupting data, including the flash translation data structures.  In the past not all drives achieved this, but I hope all modern drives manage it.  Consumer drives advertised with power loss protection should be achieving this.

[WatchfulEye]

Some of the early SATA SSDs claimed to complete work in progress as the power fell. I don't know about current SATA drives, but NVMe drives don't seem to have anything big enough to store some energy to do that.

In the absence of capacitors for energy storage, drives can achieve power loss protection by ensuring that the specifications for write command semantics are strictly adhered to (i.e. only signalling that a write command is complete when the data is flash and metadata has been appropriately updated), and maintaining internal metadata integrity with journalling.  This does not necessarily mean that all in-progress writes will be committed, but it does mean that applications and file systems can maintain internal consistency, and databases can maintain transactional integrity/atomicity.

Enterprise NVMe drives tend to feature a few mF of tantalum polymer capacitors for power loss protection, allowing them to finish all work in progress, when power is removed. This means the drives can legitimately signal that data has been committed immediately on receipt of the write command - which can be important for overall system performance in transactional workloads.

[David Hess]

In the absence of capacitors for energy storage, drives can achieve power loss protection by ensuring that the specifications for write command semantics are strictly adhered to (i.e. only signalling that a write command is complete when the data is flash and metadata has been appropriately updated), and maintaining internal metadata integrity with journalling.  This does not necessarily mean that all in-progress writes will be committed, but it does mean that applications and file systems can maintain internal consistency, and databases can maintain transactional integrity/atomicity.

There are two things which make that difficult:

The Flash write procedure is controlled by a state machine which may glitch as the supply voltage drops, causing the write to complete at the wrong address, destroying other data.  Early SSDs suffered from this, but maybe the Flash chips have been improved to avoid this problem.

I do not know why they did this, but Flash devices which store multiple bits per cell can perform multiple writes to the same page.  If the second write does not complete, then the data from the first write is destroyed.  This is solved by writing all of the data for a page at the same time, but doubles (or triples, etc.) the minimum write size.

[amyk]

TLC SSDs are around $0.10/GB now. Where are the $0.30/GB SLC SSDs, which would use exactly the same amount of die area but won't need the additional cost of a more complex FTL/ECC and controller, while also lasting many times longer?

[PlainName]

Presumably no-one wants to pay three times more for a quarter of the capacity, so there is no profit in them.

[NiHaoMike]

The manufacturers should provide a way to reformat TLC or QLC as MLC or SLC, for those who want to trade off capacity for more endurance.

[PlainName]

Doesn't wear leveling achieve that? Use half your disk and you've doubled the cells available for endurance.

[mariush]

It depends on the controller

Some can switch the whole free space to pseudo-SLC and only convert to TLC or QLC when nearly all buffer is full. 

Some controllers have a dynamic or static + dynamic portion, up to a percentage of drive's space ... for example some older WD drives like SN570 had a maximum of 12 GB pseudo-slc write cache. Some samsung drives have a guaranteed 6 GB fixed (always there) + up to something like 150-250 GB for every 1 TB or 2 TB of TLC memory.

[David Hess]

Doesn't wear leveling achieve that? Use half your disk and you've doubled the cells available for endurance.

It is not about using less of the drive; all of the cells are available for endurance.  Larger drives have greater total endurance because they have more cells available for wear leveling, and that is one way to get longer operating life, but of course it comes at the cost of buying a drive with more storage.

Drive endurance is proportional to size, but how many bits are stored per cell has an even greater effect, so a drive with the same number of cells but fewer bits per cell should have higher endurance despite lower capacity.

[PlainName]

but of course it comes at the cost of buying a drive with more storage.

Yes, exactly. The question was why don't they sell smaller drives with longer life (due to SLC vs QLC), and my contention was that you might achieve the same thing from just buying a bigger drive and not using some of it, even if it has to be a huge drive in comparison - same usable size, same lifetime, just a different way to get there.

Edit: Of course, that's predicated on unused space being usable for wear leveling (which apparently may not be the case) and wear leveling actually not losing data. But it's a mature technology and cheap as chips, now

[David Hess]

but of course it comes at the cost of buying a drive with more storage.

Yes, exactly. The question was why don't they sell smaller drives with longer life (due to SLC vs QLC), and my contention was that you might achieve the same thing from just buying a bigger drive and not using some of it, even if it has to be a huge drive in comparison - same usable size, same lifetime, just a different way to get there.

I suspect it just comes down to economics.  There is not enough demand for longer endurance storage to support mass production of SLC drives at a cost competitive with TLC drives, even though this should be more cost efficient where higher endurance is required.  Commercial and consumer grade customers can just buy larger drives to satisfy their longer endurance needs.  More critical customers can pay the high cost premium of true SLC drives, although some companies like Tesla have screwed this up.

Edit: Of course, that's predicated on unused space being usable for wear leveling (which apparently may not be the case) and wear leveling actually not losing data. But it's a mature technology and cheap as chips, now

A good SSD wear levels across the entire drive even if only part of the storage is in use, and should be performing scrub on read and idle time scrubbing to prevent data loss.

[aqarwaen]

i bit off topic here.but i got question let say i bought new ssd then write enough it reallocate sectors on flash.then swapped memories.
so my question is in this case would ssd/controller able to detect that i swapped memories.would it still keep or remove bad sectors if i did such thing?

[ejeffrey]

The manufacturers should provide a way to reformat TLC or QLC as MLC or SLC, for those who want to trade off capacity for more endurance.

My understanding is that MLC isn't a lot better than TLC.  So you would need to be going to pSLC to get a real benefit, and few people actually want to give up 66% of their storage.  So unfortunately we are stuck with the situation where only way to get SLC/pSLC drives is to get "industrial" SSDs that are insanely over priced and under performing.

MLC is basically in a weird untenable position for most use cases.  If you have MLC, it's generally better for performance to split the same total capacity into TLC and pSLC.  It's only where you need a specific high sustained write performance, and you don't benefit from often being able to burst faster than that.  There are applications like this, such as DVRs, data aquisition, and network capture systems.  However, they are exactly the applications where they also value high capacity.  In that case, just distributing the data across multiple TLC drives can get you the performance you need at a much lower cost/GB.

[amyk]

The manufacturers should provide a way to reformat TLC or QLC as MLC or SLC, for those who want to trade off capacity for more endurance.

They actually do, but are quite secretive about it. I just came across this:

i bit off topic here.but i got question let say i bought new ssd then write enough it reallocate sectors on flash.then swapped memories.
so my question is in this case would ssd/controller able to detect that i swapped memories.would it still keep or remove bad sectors if i did such thing?

That probably wouldn't work since they tend to store the firmware there too. But otherwise the FTL structure are stored in the flash, so if you reformat it using the manufacturer tool it'll rescan and establish new bad blocks.

[tridac]

I did that on a few older machines. One a 486dx that I keep for disk testing (suse 11.4), other utiliites and tools. The greatest gain is the bypassing disk rotational latency and seek times, both in milleseconds. Great improvement in boot times and general use, at very little expense. Otherwise, it's ex corporate server class sas 2.5 drives, which really do fly compared to 10k 2.5 types. Smart utils on all receipt, but many of the ex corporate sas drives show either zero or just a few percent wear index, and will last forever here. ZFS raidz for server and zfs mirrored for os system disks. Samsung preferably, but Micron and Toshiba sas pretty good as well....

[Geoff-AU]

Samsung .... The older 840 line of SSDs suffered a bug where cells lost chage and data that wasn't read or written often enough would vanish, and such a firmware update was needed to make the drive automatically perform rolling rewrites across itself, which increased the wear somewhat and shotened the overall lifespan accordingly as, to prevent bits getting corrupted every bit had to be rewritten every few days.

A bit hyperbolic.  The bug was that "old" data got slower and slower to read, because the drive tried many times to read the data until it succeeded.  I have an 840 with this issue, the firmware update did a full-drive refresh (which was slow, but I didn't lose any data), and the drive is still in daily use today so the overall lifespan was NOT adversely affected.  The data only has to be refreshed every 30+ days, not "every few days".  It's only a scandal because it became public.  If Samsung realised earlier that refreshing data was a thing, they would have just quietly done it and none of us would have been aware of it happening.

I wouldn't use NAND for archival storage though.  Magnetic disks are more stable (and my use-case is only for 3-5 years then I'm refreshing for bigger backup disks anyway)

[BrianHG]

The manufacturers should provide a way to reformat TLC or QLC as MLC or SLC, for those who want to trade off capacity for more endurance.

They actually do, but are quite secretive about it. I just came across this:

What a horrible esoteric hack tool and procedure used to patch the drive into SLC mode.  Talk about an error prone lengthy time pain in the ass... Though, once the drive is in SLC exclusive mode, the write performance uplift appears to function as I expected.

[David Hess]

Samsung .... The older 840 line of SSDs suffered a bug where cells lost chage and data that wasn't read or written often enough would vanish, and such a firmware update was needed to make the drive automatically perform rolling rewrites across itself, which increased the wear somewhat and shotened the overall lifespan accordingly as, to prevent bits getting corrupted every bit had to be rewritten every few days.

A bit hyperbolic.  The bug was that "old" data got slower and slower to read, because the drive tried many times to read the data until it succeeded.  I have an 840 with this issue, the firmware update did a full-drive refresh (which was slow, but I didn't lose any data), and the drive is still in daily use today so the overall lifespan was NOT adversely affected.  The data only has to be refreshed every 30+ days, not "every few days".  It's only a scandal because it became public.  If Samsung realised earlier that refreshing data was a thing, they would have just quietly done it and none of us would have been aware of it happening.

I wouldn't use NAND for archival storage though.  Magnetic disks are more stable (and my use-case is only for 3-5 years then I'm refreshing for bigger backup disks anyway)

Scrub-on-read and idle-time-scrubbing is suppose to handle that by rewriting blocks which have accumulated too many correctable errors, before they become uncorrectable errors.

I do not understand how Samsung could have been unaware of scrub-on-read and idle-time-scrubbing.

[Geoff-AU]

I do not understand how Samsung could have been unaware of scrub-on-read and idle-time-scrubbing.

I can only assume a Project Manager's deadline was threatened.

[David Hess]

I do not understand how Samsung could have been unaware of scrub-on-read and idle-time-scrubbing.

I can only assume a Project Manager's deadline was threatened.

Or they considered it a "premium" feature to be limited to data center products.  I only started using SSDs maybe 4 years ago, and at the time I evaluated various manufacturers.  Samsung discussed power loss protection only in connection with their data center products, and power loss protection is required to safely implement scrub-on-read and idle-time-scrubbing.

[tooki]

What I find unnerving is the completely misleading terminology of MLC and TLC: that suggests (and early explanations claimed) that MLC is two levels, and TLC three, which I think anyone who knows anything about electronics would take to mean voltages or charge levels or the like. That is, that excluding zero, SLC has one value, MLC two, and TLC three, giving 2, 3, and 4 possible values, respectively, including zero. But they're not. The "levels" are actually the exponent of a power of 2, so SLC has 2¹ = 2 possible values (incl. zero), MLC has 2² = 4 values, and TLC has 2³ = 8 possible values. This surely decreases S/N ratio appreciably (or whatever the correct term is in this situation; let me know!).

[David Hess]

What I find unnerving is the completely misleading terminology of MLC and TLC: that suggests (and early explanations claimed) that MLC is two levels, and TLC three, which I think anyone who knows anything about electronics would take to mean voltages or charge levels or the like. That is, that excluding zero, SLC has one value, MLC two, and TLC three, giving 2, 3, and 4 possible values, respectively, including zero. But they're not. The "levels" are actually the exponent of a power of 2, so SLC has 2¹ = 2 possible values (incl. zero), MLC has 2² = 4 values, and TLC has 2³ = 8 possible values. This surely decreases S/N ratio appreciably (or whatever the correct term is in this situation; let me know!).

You are right, and it is misleading.  I just always understood that it was a power of 2 because I do that sort of math all of the time ... to calculate things like signal to noise ratio, or how many "bits" is 40,000 counts? (15.3 bits)

2^x = 40,000
ln(2^x) = ln(40,000)
x * ln (2) - ln(40,000)
x = ln(40,000)/ln(2)
x = 15.3

Signal to noise is as good a term as any.

[amyk]

A bit hyperbolic.  The bug was that "old" data got slower and slower to read, because the drive tried many times to read the data until it succeeded.  I have an 840 with this issue, the firmware update did a full-drive refresh (which was slow, but I didn't lose any data), and the drive is still in daily use today so the overall lifespan was NOT adversely affected.  The data only has to be refreshed every 30+ days, not "every few days".  It's only a scandal because it became public.  If Samsung realised earlier that refreshing data was a thing, they would have just quietly done it and none of us would have been aware of it happening.

"in daily use" is the key word; that flash is so leaky that it can't hold readable charge for more than a month, and I bet that number would go down very quickly in a warmer environment as retention decreases exponentially with temperature.

I'm pretty sure people would've figured out sooner or later, as not everyone who has flash storage keeps it plugged in all the time. I have some old USB drives from SLC era which have gone through a lot of full-drive writes (hundreds, probably enough to wear out some modern SSDs) with no data loss after almost 2 decades. Here's a more recent case:

https://goughlui.com/2023/10/10/psa-ssds-with-ymtc-flash-prone-to-failure-check-your-ssds/

As for "TLC" "QLC" etc. it's definitely misleading, but they've used "levels" to refer to the "layers" of bits on the same page which are stored in the same cell. Programming pages happens one "level" at a time. It makes deep technical sense, but unfortunately is very useful for hiding the endurance and retention characteristics.

[tooki]

As for "TLC" "QLC" etc. it's definitely misleading, but they've used "levels" to refer to the "layers" of bits on the same page which are stored in the same cell. Programming pages happens one "level" at a time. It makes deep technical sense, but unfortunately is very useful for hiding the endurance and retention characteristics.

That sure sounds to me like it was selected very deliberately to be misleading. If it wasn’t deliberate, then it was a case of staggering incompetence.

[David Hess]

As for "TLC" "QLC" etc. it's definitely misleading, but they've used "levels" to refer to the "layers" of bits on the same page which are stored in the same cell. Programming pages happens one "level" at a time. It makes deep technical sense, but unfortunately is very useful for hiding the endurance and retention characteristics.

That sure sounds to me like it was selected very deliberately to be misleading. If it wasn’t deliberate, then it was a case of staggering incompetence.

What other terms could they have used?

Except for MLC, they unambiguously indicate how the capacity of the drive is multiplied, which is what matters to the customer.  They also give a good indication of how resilient the underlying storage is.

[tooki]

But the word “level” doesn’t normally indicate multiplication, never mind exponentiation. Parking garages often are labeled with “levels”, and if I see “parking level 3”,  I expect that to be either 2 or 3 floors off the ground (depending on whether one counts the ground level as #1). I certainly wouldn’t expect “level 3” to be 7 or 8 floors off the ground!!

I distinctly remember some early descriptions of MLC and TLC as explaining that SLC having one nonzero voltage level (let’s call the possible values 0 and 1), MLC having two nonzero voltage levels (values 0, 1, and 2), and TLC having three nonzero values (values 0, 1, 2, 3). That is clearly wrong, so clearly the naming had managed to confuse some people.

The numbering of single and triple accurately reflects bits per cell, and some name along those lines would have been more honest marketing.