• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

100% of my QLC drives are now dead.

Joined
Mar 21, 2021
Messages
4,607 (3.89/day)
Location
Colorado, U.S.A.
System Name CyberPowerPC ET8070
Processor Intel Core i5-10400F
Motherboard Gigabyte B460M DS3H AC-Y1
Memory 2 x Crucial Ballistix 8GB DDR4-3000
Video Card(s) MSI Nvidia GeForce GTX 1660 Super
Storage 500GB SN550 WD Blue NVMe M.2
Display(s) Dell P2416D (2560 x 1440)
Power Supply Apevia 600W Gold 80 Plus
Software Windows 11 Home
All the more reason to run my boot SSDs mostly empty so they run in SLC mode (for endurance, not speed).
 
Joined
May 9, 2012
Messages
8,441 (1.91/day)
Location
Ovronnaz, Wallis, Switzerland
System Name main/SFFHTPCARGH!(tm)/Xiaomi Mi TV Stick/Samsung Galaxy S23/Ally
Processor Ryzen 7 5800X3D/i7-3770/S905X/Snapdragon 8 Gen 2/Ryzen Z1 Extreme
Motherboard MSI MAG B550 Tomahawk/HP SFF Q77 Express/uh?/uh?/Asus
Cooling Enermax ETS-T50 Axe aRGB /basic HP HSF /errr.../oh! liqui..wait, no:sizable vapor chamber/a nice one
Memory 64gb Corsair Vengeance Pro 3600mhz DDR4/8gb DDR3 1600/2gb LPDDR3/8gb LPDDR5x 4200/16gb LPDDR5
Video Card(s) Hellhound Spectral White RX 7900 XTX 24gb/GT 730/Mali 450MP5/Adreno 740/RDNA3 768 core
Storage 250gb870EVO/500gb860EVO/2tbSandisk/NVMe2tb+1tb/4tbextreme V2/1TB Arion/500gb/8gb/256gb/2tb SN770M
Display(s) X58222 32" 2880x1620/32"FHDTV/273E3LHSB 27" 1920x1080/6.67"/AMOLED 2X panel FHD+120hz/FHD 120hz
Case Cougar Panzer Max/Elite 8300 SFF/None/back/back-front Gorilla Glass Victus 2+ UAG Monarch Carbon
Audio Device(s) Logi Z333/SB Audigy RX/HDMI/HDMI/Dolby Atmos/KZ x HBB PR2/Edifier STAX Spirit S3 & SamsungxAKG beans
Power Supply Chieftec Proton BDF-1000C /HP 240w/12v 1.5A/4Smart Voltplug PD 30W/Asus USB-C 65W
Mouse Speedlink Sovos Vertical-Asus ROG Spatha-Logi Ergo M575/Xiaomi XMRM-006/touch/touch
Keyboard Endorfy Thock 75% <3/none/touch/virtual
VR HMD Medion Erazer
Software Win10 64/Win8.1 64/Android TV 8.1/Android 13/Win11 64
Benchmark Scores bench...mark? i do leave mark on bench sometime, to remember which one is the most comfortable. :o
This thread reminds me: 100% of my OCZ drives are dead, and I owned like every model they made.
as i mentioned previously ... my Vertex 3 is still alive and kicking after 9yrs ... although enjoying a nice retirement in his packaging at 97% health remaining ...
tho i only owned this model and got informed of the "issues" OCZ had with their SSD only 5 yrs after i bought it :laugh:


i guess it make a statement that per case issue can't be generic, for QLC or OCZ :laugh:
 
Joined
Jul 30, 2019
Messages
2,608 (1.46/day)
System Name Not a thread ripper but pretty good.
Processor Ryzen 9 5950x
Motherboard ASRock X570 Taichi (revision 1.06, BIOS/UEFI version P5.50)
Cooling EK-Quantum Velocity, EK-Quantum Reflection PC-O11, EK-CoolStream PE 360, XSPC TX360
Memory Micron DDR4-3200 ECC Unbuffered Memory (4 sticks, 128GB, 18ASF4G72AZ-3G2F1)
Video Card(s) XFX Radeon RX 5700 & EK-Quantum Vector Radeon RX 5700 +XT & Backplate
Storage Samsung 2TB 980 PRO 2TB Gen4x4 NVMe, 2 x Samsung 2TB 970 EVO Plus Gen3x4 NVMe, AMD Radeon RAMDisk
Display(s) 2 x 4K LG 27UL600-W (and HUANUO Dual Monitor Mount)
Case Lian Li PC-O11 Dynamic Black (original model)
Power Supply Corsair RM750x
Mouse Logitech M575
Keyboard Corsair Strafe RGB MK.2
Software Windows 10 Professional (64bit)
Benchmark Scores Typical for non-overclocked CPU.
I was seriously thinking running 4 x 8TB QVO in a raid 6 would be pretty fast and affordable for a redundant array if it wasn't for the prospect of the rebuild time getting nuked from exhausting the cache.
 
Joined
Jul 5, 2013
Messages
25,778 (6.44/day)
Not to say they were ever a GOOD purchase as an OS drive but complete failure of all of them just seems too much to me.
There are always going to be anecdotal stories on all three sides of this issue. The important thing to remember is that they are all anecdotal. That's all they can be.
Perhaps. However, a 23% field failure rate is unacceptable in ANY industry.

No, just the standard old "QLC is evil WAAAH" nonsense, as usual from the QLC refuseniks.
Physics is the only thing required to understand the failure potential for QLC. IF you understand the process by which data is stored and the chemistry by with the memory operates, it doesn't take much to conclude that QLC is crap.

It's gotten better, but at the end of the day, no matter how much polish you use, a turd is still a turd no matter how shiny you get it.
 
Last edited:
Joined
Feb 20, 2020
Messages
9,340 (5.92/day)
Location
Louisiana
System Name Ghetto Rigs z490|x99|Acer 17 Nitro 7840hs/ 5600c40-2x16/ 4060/ 1tb acer stock m.2/ 4tb sn850x
Processor 10900k w/Optimus Foundation | 5930k w/Black Noctua D15
Motherboard z490 Maximus XII Apex | x99 Sabertooth
Cooling oCool D5 res-combo/280 GTX/ Optimus Foundation/ gpu water block | Blk D15
Memory Trident-Z Royal 4000c16 2x16gb | Trident-Z 3200c14 4x8gb
Video Card(s) Titan Xp-water | evga 980ti gaming-w/ air
Storage 970evo+500gb & sn850x 4tb | 860 pro 256gb | Acer m.2 1tb/ sn850x 4tb| Many2.5" sata's ssd 3.5hdd's
Display(s) 1-AOC G2460PG 24"G-Sync 144Hz/ 2nd 1-ASUS VG248QE 24"/ 3rd LG 43" series
Case D450 | Cherry Entertainment center on Test bench
Audio Device(s) Built in Realtek x2 with 2-Insignia 2.0 sound bars & 1-LG sound bar
Power Supply EVGA 1000P2 with APC AX1500 | 850P2 with CyberPower-GX1325U
Mouse Redragon 901 Perdition x3
Keyboard G710+x3
Software Win-7 pro x3 and win-10 & 11pro x3
Benchmark Scores Are in the benchmark section
The 970EVO and 970EVO plus also has proven to be very very reliable SSD's. Cheap but they at least don't use QLC Nand. I know they are trying to make QLC reliable also, but at the moment there's to much trash drives with QLC on the market. I would not use it anyway for data storage, then give me an good old HDD, they have at least proven themself.
Hi,
Yeah just attempting to dust my 500gb evo plus atm
Win-11 is proving relentless boot wise with it :laugh:

Yep you can put a tuxedo on a goat but it's still a goat :slap:
 
Joined
Jun 2, 2022
Messages
349 (0.47/day)
System Name HP EliteBook 725 G3
Processor AMD PRO A10-8700B (1.8 GHz CMT dual module with 3.2 GHz boost)
Motherboard HP proprietary
Cooling pretty good
Memory 8 GB SK Hynix DDR3 SODIMM
Video Card(s) Radeon R6 (Carrizo/GCNv3)
Storage internal Kioxia XG6 1 TB NVMe SSD (aftermarket)
Display(s) HP P22h G4 21.5" 1080p (& 768p internal LCD)
Case HP proprietary metal case
Audio Device(s) built-in Conexant CX20724 HDA chipset -> Roland RH-200S
Power Supply HP-branded AC adapter
Mouse Steelseries Rival 310
Keyboard Cherry G84-5200
Software Alma Linux 9.1
Benchmark Scores Broadcom BCM94356 11ac M.2 WiFi card (aftermarket)
Crucial BX has always been QLC afaik.
Crucial BX500 used to be TLC (it has always been DRAMless, however). I have an old 120 GB model from before they switched to QLC. I imagine that the 300 and 100(?) were certainly also TLC then.
 
Last edited:
Joined
Jan 10, 2011
Messages
1,347 (0.27/day)
Location
[Formerly] Khartoum, Sudan.
System Name 192.168.1.1~192.168.1.100
Processor AMD Ryzen5 5600G.
Motherboard Gigabyte B550m DS3H.
Cooling AMD Wraith Stealth.
Memory 16GB Crucial DDR4.
Video Card(s) Gigabyte GTX 1080 OC (Underclocked, underpowered).
Storage Samsung 980 NVME 500GB && Assortment of SSDs.
Display(s) ViewSonic VA2406-MH 75Hz
Case Bitfenix Nova Midi
Audio Device(s) On-Board.
Power Supply SeaSonic CORE GM-650.
Mouse Logitech G300s
Keyboard Kingston HyperX Alloy FPS.
VR HMD A pair of OP spectacles.
Software Ubuntu 24.04 LTS.
Benchmark Scores Me no know English. What bench mean? Bench like one sit on?
Physics is the only thing required to understand the failure potential for QLC. IF you understand the process by which data is stored and the chemistry by with the memory operates, it doesn't take much to conclude that QLC is crap.
A simplistic physics interpretation would also balk at the extremely low glide elevations of your typical spinning rust. Yet in practice they still ran fine, except for a minority of cases we invented RMAs for.

We cannot do a meaningful risk analysis or gauge the reliability of any working system based solely on a conceptual understanding of it. I wish that was true. But unfortunately, given our extremely limited ability to process this world and how little we understand it, we're stuck making do with the teachings of damned eugenicists...
 
Joined
Jul 5, 2013
Messages
25,778 (6.44/day)
A simplistic physics interpretation would also balk at the extremely low glide elevations of your typical spinning rust.
That would be a misunderstanding of the science involved in "spinning rust". Magnetic fields are, when properly attenuated, extremely reliable. That was a comparison that did not lend itself to a good argument.

A rechargable battery would be a better comparison and more similar in function. A battery works by applying a voltage to change the chemistry of the compounds within a battery for the purpose of holding a charge. When that charge is used, the chemistry reverts back to a lower charge state until recharged again. However, the chemical compounds within that battery can only withstand a certain number of charge discharge cycles before rendering them unusable.

NAND cells are very similar. When voltage is applied, the charge state for the cell changes. Change the voltage in a different way and the cell resets to it's default state. That is an SLC cycle. 2 Voltage cycles are needed for change an SLC block to default(NAND is always written to in blocks) and 1 more to set the state of each NAND cell within that block.

An MLC write cycle works as follows: A voltage state is applied to reset the Block being written too. Then another is applied the set the new data being written to the block. The voltage is then altered and applied again to make a third state. This requires a slight increase of voltage when compared to the first voltage application cycle. A total of three voltage cycles need to be applied to the NAND to alter MLC cells.

A TLC write state works the same way, but has a fourth voltage application cycle.

A QLC write cycle adds a fifth cycle.

With each write cycle, an additional voltage is applied to the cell and at a slightly increased voltage each time. With each of these write cycles, the chemistry of the NAND cell degrades by a small but not insignificant amount.

With SLC, there is 2 voltage cycles applied for every change made. With MLC, 3. TLC, 4. QLC, 5. PLC, 6.

It should be very clear to anyone who understands this process that with each voltage cycle, at an increased voltage with each, that a NAND cell will quickly and easily degrade. TLC is on the bleeding edge of what can be considered an acceptable level of NAND cell wear. QLC, pushes beyond that acceptable level and effectively relegates it to the areas of incidental storage. QLC is is fine if you only plan to use it lightly a for a year or two. Beyond that, it becomes increasingly unfavorable and unreliable. And that is ONLY if the NAND cells do not experience a cascade voltage failure, which is a known problem within the NAND industry.

QLC is untrustworthy after a certain period of time, no matter what the manufacturers claim.
 

hat

Enthusiast
Joined
Nov 20, 2006
Messages
21,734 (3.39/day)
Location
Ohio
System Name Starlifter :: Dragonfly
Processor i7 2600k 4.4GHz :: i5 10400
Motherboard ASUS P8P67 Pro :: ASUS Prime H570-Plus
Cooling Cryorig M9 :: Stock
Memory 4x4GB DDR3 2133 :: 2x8GB DDR4 2400
Video Card(s) PNY GTX1070 :: Integrated UHD 630
Storage Crucial MX500 1TB, 2x1TB Seagate RAID 0 :: Mushkin Enhanced 60GB SSD, 3x4TB Seagate HDD RAID5
Display(s) Onn 165hz 1080p :: Acer 1080p
Case Antec SOHO 1030B :: Old White Full Tower
Audio Device(s) Creative X-Fi Titanium Fatal1ty Pro - Bose Companion 2 Series III :: None
Power Supply FSP Hydro GE 550w :: EVGA Supernova 550
Software Windows 10 Pro - Plex Server on Dragonfly
Benchmark Scores >9000
use hard drive for game storage.
That's what I've been doing... two 1TB drives in RAID 0. I feel like it offers a fair middle ground between the performance of an SSD and the cost effectiveness of a hard drive, especially when I've never been blown away by an SSD as far as games go. It made a huge difference for Windows, but games, not so much.
 
Joined
Jan 10, 2011
Messages
1,347 (0.27/day)
Location
[Formerly] Khartoum, Sudan.
System Name 192.168.1.1~192.168.1.100
Processor AMD Ryzen5 5600G.
Motherboard Gigabyte B550m DS3H.
Cooling AMD Wraith Stealth.
Memory 16GB Crucial DDR4.
Video Card(s) Gigabyte GTX 1080 OC (Underclocked, underpowered).
Storage Samsung 980 NVME 500GB && Assortment of SSDs.
Display(s) ViewSonic VA2406-MH 75Hz
Case Bitfenix Nova Midi
Audio Device(s) On-Board.
Power Supply SeaSonic CORE GM-650.
Mouse Logitech G300s
Keyboard Kingston HyperX Alloy FPS.
VR HMD A pair of OP spectacles.
Software Ubuntu 24.04 LTS.
Benchmark Scores Me no know English. What bench mean? Bench like one sit on?
That would be a misunderstanding of the science involved in "spinning rust". Magnetic fields are, when properly attenuated, extremely reliable. That was a comparison that did not lend itself to a good argument.
I was actually going for the extremely delicate fluid mechanics involved in keeping the head from going all zen garden on the platters...

What I'm saying, put bluntly, is that numbers from an empirical test matters, personal fears and opinions don't. As far as I'm aware, only manufacturer's own endurance rating come close to serving the former. Anyone claiming otherwise has got to show us some figures...
 
Joined
Aug 12, 2020
Messages
1,198 (0.85/day)
This thread reminds me: 100% of my OCZ drives are dead, and I owned like every model they made.
RIP OCZ. I still have lanyard with their logo back from I tested CoreXStream model.

But more on-topic, their first JMicron SSDs SURE had issues lol. Revodrive was a marvel though, a trailblazer if you will.
Personally, I now run 1TB Kioxia Exceria + Transcend SSD370 for 1.5TB combined storage - enough for me. I already moved the goalpost of "I will keep looking for MLC", as former is 3D TLC. It still keeps just fine after almost 2 years though, so looks like that goalpost was fine to move. And keeping drive mostly free sure does help for controller.

But most importantly, no spinner = FAR less noise!
 
Last edited:
Joined
Jul 5, 2013
Messages
25,778 (6.44/day)
I was actually going for the extremely delicate fluid mechanics involved in keeping the head from going all zen garden on the platters...

What I'm saying, put bluntly, is that numbers from an empirical test matters, personal fears and opinions don't. As far as I'm aware, only manufacturer's own endurance rating come close to serving the former. Anyone claiming otherwise has got to show us some figures...
Way to move the goal post. :rolleyes:
And no, us users(and in my case retailer) do NOT have to show figures for our reports to be considered reasonable.

News flash for you there: Manufacturers LIE! They exaggerate certain facts to make things look good, obscure other facts to hide things that make them look bad and last but not least they flat out lie about things they can legally get away with.

Why do you think places like Tech Power Up exist? Hmm? It's to review the products on offer to verify claims and show just what a product can ACTUALLY do. If manufacturers were honest, review sites wouldn't be needed because consumers could trust what the manufacturers say. But manufacturers are NOT honest. They are LIARS and thus TPU and others are needed so people can get properly accurate information. Forums like this exist in part so that the real public can chime in and offer up the experience THEY have. And that brings us to this thread. TAA DAA!

The OP and others shared their experiences because we're trying to HELP other users figure out that QLC is, in fact, crap and should be avoided if long-term use is what you want from your storage. This is not because we have an vendetta against any one company, but because we have had experiences that show a result and we're trying to show that this technology is not acceptable. QLC is planned obsolesce at it's most exquisite.

You nay-sayers can mock and whatnot, but all you are doing is making yourselves look like sheep who can't think for themselves. Kinda sad. What's even more sad is that it is a bit funny.
 
Last edited:
Joined
Jan 10, 2011
Messages
1,347 (0.27/day)
Location
[Formerly] Khartoum, Sudan.
System Name 192.168.1.1~192.168.1.100
Processor AMD Ryzen5 5600G.
Motherboard Gigabyte B550m DS3H.
Cooling AMD Wraith Stealth.
Memory 16GB Crucial DDR4.
Video Card(s) Gigabyte GTX 1080 OC (Underclocked, underpowered).
Storage Samsung 980 NVME 500GB && Assortment of SSDs.
Display(s) ViewSonic VA2406-MH 75Hz
Case Bitfenix Nova Midi
Audio Device(s) On-Board.
Power Supply SeaSonic CORE GM-650.
Mouse Logitech G300s
Keyboard Kingston HyperX Alloy FPS.
VR HMD A pair of OP spectacles.
Software Ubuntu 24.04 LTS.
Benchmark Scores Me no know English. What bench mean? Bench like one sit on?
Way to move the goal post. :rolleyes:
And no, us users(and in my case retailer) do NOT have to show figures for our reports to be considered reasonable.
I didn't move anything.
The issue has always been reliability of QLC flash. To make a valid statement about said reliability requires proper sampling and probability analysis. This is a given in all of science and engineering.

but because we have had experiences that show a result
And all I'm asking is you to quantify this experience and show me that it's significant compared to others' that tell a different story.
Being unable to do so doesn't remove it as a requirement for a valid conclusion on reliability, it only reduces your choices for said conclusions to "We don't know."
 
Last edited:
Joined
Mar 21, 2021
Messages
4,607 (3.89/day)
Location
Colorado, U.S.A.
System Name CyberPowerPC ET8070
Processor Intel Core i5-10400F
Motherboard Gigabyte B460M DS3H AC-Y1
Memory 2 x Crucial Ballistix 8GB DDR4-3000
Video Card(s) MSI Nvidia GeForce GTX 1660 Super
Storage 500GB SN550 WD Blue NVMe M.2
Display(s) Dell P2416D (2560 x 1440)
Power Supply Apevia 600W Gold 80 Plus
Software Windows 11 Home
Way to move the goal post. :rolleyes:
And no, us users(and in my case retailer) do NOT have to show figures for our reports to be considered reasonable.

News flash for you there: Manufacturers LIE! They exaggerate certain facts to make things look good, obscure other facts to hide things that make them look bad and last but not least they flat out lie about things they can legally get away with.

Why do you think places like Tech Power Up exist? Hmm? It's to review the products on offer to verify claims and show just what a product can ACTUALLY do. If manufacturers were honest, review sites wouldn't be needed because consumers could trust what the manufacturers say. But manufacturers are NOT honest. They are LIARS and thus TPU and others are needed so people can get properly accurate information. Forums like this exist in part so that the real public can chime in and offer up the experience THEY have. And that brings us to this thread. TAA DAA!

The OP and others shared their experiences because we're trying to HELP other users figure out that QLC is, in fact, crap and should be avoided if long-term use is what you want from your storage. This is not because we have an vendetta against any one company, but because we have had experiences that show a result and we're trying to show that this technology is not acceptable. QLC is planned obsolesce at it's most exquisite.

You nay-sayers can mock and whatnot, but all you are doing is making yourselves look like sheep who can't think for themselves. Kinda sad. What's even more sad is that it is a bit funny.

But it says 9900mAh :D and is sold on Amazon.com

Amazon.com : 9900 mAh
9900mAh.jpg
 
Last edited:
Joined
Jul 5, 2013
Messages
25,778 (6.44/day)
The issue has always been reliability of QLC flash. To make a valid statement about said reliability requires sampling and probability analysis. This is a given in all of science and engineering.
It doesn't take a genius to conclude that more voltage applied more times equals greater wear more quickly.

And all I'm asking is you to quantify this experience and show me that it's significant compared to others' that tell a different story.
Those who have a different experience aren't going to say anything, are they.

Being unable to do so doesn't remove it as a requirement for a valid conclusion on reliabilty, it only reduces your choices for said conclusions to "We don't know."
You're missing the point. Context is important.
 
Joined
May 8, 2020
Messages
578 (0.39/day)
System Name Mini efficient rig.
Processor R9 3900, @4ghz -0.05v offset. 110W peak.
Motherboard Gigabyte B450M DS3H, bios f41 pcie 4.0 unlocked.
Cooling some server blower @1500rpm
Memory 2x16GB oem Samsung D-Die. 3200MHz
Video Card(s) RX 6600 Pulse w/conductonaut @65C hotspot
Storage 1x 128gb nvme Samsung 950 Pro - 4x 1tb sata Hitachi 2.5" hdds
Display(s) Samsung C24RG50FQI
Case Jonsbo C2 (almost itx sized)
Audio Device(s) integrated Realtek crap
Power Supply Seasonic SSR-750FX
Mouse Logitech G502
Keyboard Redragon K539 brown switches
Software Windows 7 Ultimate SP1 + Windows 10 21H2 LTSC (patched).
Benchmark Scores Cinebench: R15 3050 pts, R20 7000 pts, R23 17800 pts, r2024 1050 pts.
Funny how ppl nowadays say that tlc is reliable... A tlc with dram - yea, kinda. But i still prefer true mlc such as 970 Pro, or you can use 3dxpoint drives as well (64gb optane is about similar to a 1tb tlc drive in TBW, but $ per gig is ofc higher). Still its always a good idea to make backups.
 
Joined
Mar 21, 2021
Messages
4,607 (3.89/day)
Location
Colorado, U.S.A.
System Name CyberPowerPC ET8070
Processor Intel Core i5-10400F
Motherboard Gigabyte B460M DS3H AC-Y1
Memory 2 x Crucial Ballistix 8GB DDR4-3000
Video Card(s) MSI Nvidia GeForce GTX 1660 Super
Storage 500GB SN550 WD Blue NVMe M.2
Display(s) Dell P2416D (2560 x 1440)
Power Supply Apevia 600W Gold 80 Plus
Software Windows 11 Home
How does one test/assess reliability given it will run in SLC mode when near empty and thrash when near full?
 
Joined
Jul 5, 2013
Messages
25,778 (6.44/day)
How does one test/assess reliability given it will run in SLC mode when near empty and thrash when near full?
That is not exactly how it works. Not all drives configured with SLC caching modes behave the same way. The limits of SLC cache are different for each manufacturer and again different for each model of drive.
 
Joined
Jan 10, 2011
Messages
1,347 (0.27/day)
Location
[Formerly] Khartoum, Sudan.
System Name 192.168.1.1~192.168.1.100
Processor AMD Ryzen5 5600G.
Motherboard Gigabyte B550m DS3H.
Cooling AMD Wraith Stealth.
Memory 16GB Crucial DDR4.
Video Card(s) Gigabyte GTX 1080 OC (Underclocked, underpowered).
Storage Samsung 980 NVME 500GB && Assortment of SSDs.
Display(s) ViewSonic VA2406-MH 75Hz
Case Bitfenix Nova Midi
Audio Device(s) On-Board.
Power Supply SeaSonic CORE GM-650.
Mouse Logitech G300s
Keyboard Kingston HyperX Alloy FPS.
VR HMD A pair of OP spectacles.
Software Ubuntu 24.04 LTS.
Benchmark Scores Me no know English. What bench mean? Bench like one sit on?
It doesn't take a genius to conclude that more voltage applied more times equals greater wear more quickly.
No one is contesting the fact that QLCs are less reliable than coarser cell tech. Even those lying manufacturers admit this. My position is less reliable =\= useless or "crap," and that this alleged tendency for premature failure is just bad luck.

Those who have a different experience aren't going to say anything, are they.
Indeed they don't. That's why all the comments and posts about the quality of QLC are inherently biased, ergo unusable for a valid conclusion.
 
Joined
Jan 14, 2019
Messages
10,316 (5.21/day)
Location
Midlands, UK
System Name Holiday Season Budget Computer (HSBC)
Processor AMD Ryzen 7 7700X
Motherboard MSi PRO B650M-A WiFi
Cooling be quiet! Dark Rock 4
Memory 2x 16 GB Corsair Vengeance EXPO DDR5-6000
Video Card(s) Sapphire Pulse Radeon RX 6500 XT 4 GB
Storage 2 TB Corsair MP600 GS, 2 TB Corsair MP600 R2, 4 + 8 TB Seagate Barracuda 3.5"
Display(s) Dell S3422DWG, 7" Waveshare touchscreen
Case Kolink Citadel Mesh black
Audio Device(s) Logitech Z333 2.1 speakers, AKG Y50 headphones
Power Supply Seasonic Prime GX-750
Mouse Logitech MX Master 2S
Keyboard Logitech G413 SE
Software Windows 10 Pro
I didn't move anything.
The issue has always been reliability of QLC flash. To make a valid statement about said reliability requires proper sampling and probability analysis. This is a given in all of science and engineering.
We'll probably never have that, so the second best we have is personal anecdotes, which have some validity, in my opinion. We're all subjective human beings living a subjective experience on this planet. On that analogy, it's fair for me to conclude in another thread, based on my sample size of 1, that DLSS at 1080p is crap. It's equally fair for OP to conclude that QLC drives are unreliable. Without large-scale data (what is "large-scale" is a different question worth discussing), you're fair to conclude that you disagree.
 
Joined
Sep 27, 2008
Messages
1,055 (0.18/day)
I find it hard to believe that all the big SSD manufacturers (Micron, Samsung, Kioxia, ex-Intel, WD...etc.) are all in cahoots to sell drives that are not fit for purpose. There has to have been some pretty thorough validation of QLC drives such that the companies are reasonably certain that customers won't flood them with complaints and lawsuits.
 
Joined
Jul 30, 2019
Messages
2,608 (1.46/day)
System Name Not a thread ripper but pretty good.
Processor Ryzen 9 5950x
Motherboard ASRock X570 Taichi (revision 1.06, BIOS/UEFI version P5.50)
Cooling EK-Quantum Velocity, EK-Quantum Reflection PC-O11, EK-CoolStream PE 360, XSPC TX360
Memory Micron DDR4-3200 ECC Unbuffered Memory (4 sticks, 128GB, 18ASF4G72AZ-3G2F1)
Video Card(s) XFX Radeon RX 5700 & EK-Quantum Vector Radeon RX 5700 +XT & Backplate
Storage Samsung 2TB 980 PRO 2TB Gen4x4 NVMe, 2 x Samsung 2TB 970 EVO Plus Gen3x4 NVMe, AMD Radeon RAMDisk
Display(s) 2 x 4K LG 27UL600-W (and HUANUO Dual Monitor Mount)
Case Lian Li PC-O11 Dynamic Black (original model)
Power Supply Corsair RM750x
Mouse Logitech M575
Keyboard Corsair Strafe RGB MK.2
Software Windows 10 Professional (64bit)
Benchmark Scores Typical for non-overclocked CPU.
I find it hard to believe that all the big SSD manufacturers (Micron, Samsung, Kioxia, ex-Intel, WD...etc.) are all in cahoots to sell drives that are not fit for purpose. There has to have been some pretty thorough validation of QLC drives such that the companies are reasonably certain that customers won't flood them with complaints and lawsuits.
My big takeaway with QLC
  1. performance sucks for huge file copies to disk that exhaust cache
  2. less overall lifespan due to limited writes
  3. arguably poorer quality/endurance due to nand density (but may still have a reasonable warranty period)
  4. uses a bit less energy ( https://www.reneelab.com/difference-slc-mlc-tlc.html )
If your needs fit in the criteria above QLC is ok, probably even fine. Just don't burn the drive to the ground with writes, have a good backup plan for your data, and buy from a reputable brand a model not known to be problematic. (also don't supercool it, superheat it, or smash it with a hammer)
 
Last edited:
Joined
Oct 6, 2021
Messages
1,567 (1.59/day)
QLC is only good for SSD manufacturers' profit margins. It's bad for the consumer and the environment (it becomes trash faster), a waste of sand and energy. Governments are creating laws to facilitate the repairability of smartphones to supposedly extend their useful life, they should do the same with PCs by banning the use of QLC.
 
Joined
Jul 5, 2013
Messages
25,778 (6.44/day)
I find it hard to believe that all the big SSD manufacturers (Micron, Samsung, Kioxia, ex-Intel, WD...etc.) are all in cahoots to sell drives that are not fit for purpose. There has to have been some pretty thorough validation of QLC drives such that the companies are reasonably certain that customers won't flood them with complaints and lawsuits.
It's all about profit. Companies will do whatever they feel they need to do in order to make money.

My big takeaway with QLC
  1. performance sucks for huge file copies to disk that exhaust cache
  2. less overall lifespan due to limited writes
  3. arguably poorer quality/endurance due to nand density (but may still have a reasonable warranty period)
If your needs fit in the criteria above QLC is ok, probably even fine. Just don't burn the drive to the ground with writes, have a good backup plan for your data, and buy from a reputable brand a model not known to be problematic. (also don't supercool it, superheat it, or smash it with a hammer)
I would not disagree with your statement. QLC based drives from a company that strives for quality will be a reasonable experience as an OS drive as long as you don't expect the drive to last long. Even some of the lesser known brands offer reasonable drive experiences, again as long as you're ok with the drive not lasting more than a couple years and degrading in performance as it ages.
 
Last edited:
Top