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

What causes driver corruption?

Darkroman

New Member
Joined
May 8, 2022
Messages
2 (0.10/day)
After a while or just randomly driver issues might pop up that'll cause your computer to BSOD. What causes this? Is it just the constant overwriting of old drivers that at some point it just gets corrupted? Wouldn't/shouldn't an overwrite of the corrupted install fix that? Also why does using DDU to essentially nuke your graphics driver have a better success rate than just uninstalling the driver and reinstalling?
As a computer techie this is one of those questions I've been pondering for a while as I have used DDU (when appropriate) just based on symptoms found and short of a hardware failure, seems to fix graphics issues. I don't know WHY though as in the nitty gritty of it all. I can only speculate but I wanted a more in-depth explanation as to how this happens.
Definite causes I already know of are malware, computer shutting down during the install/update, and possibly bad sector on the drive that the driver files are written to before sector itself became corrupted.
 

eidairaman1

The Exiled Airman
Joined
Jul 2, 2007
Messages
35,926 (6.60/day)
Location
Republic of Texas (True Patriot)
System Name PCGOD
Processor AMD FX 8350@ 5.0GHz
Motherboard Asus TUF 990FX Sabertooth R2 2901 Bios
Cooling Scythe Ashura, 2×BitFenix 230mm Spectre Pro LED (Blue,Green), 2x BitFenix 140mm Spectre Pro LED
Memory 16 GB Gskill Ripjaws X 2133 (2400 OC, 10-10-12-20-20, 1T, 1.65V)
Video Card(s) AMD Radeon 290 Sapphire Vapor-X
Storage Samsung 840 Pro 256GB, WD Velociraptor 1TB
Display(s) NEC Multisync LCD 1700V (Display Port Adapter)
Case AeroCool Xpredator Evil Blue Edition
Audio Device(s) Creative Labs Sound Blaster ZxR
Power Supply Seasonic 1250 XM2 Series (XP3)
Mouse Roccat Kone XTD
Keyboard Roccat Ryos MK Pro
Software Windows 7 Pro 64
windows updates, overclocking which affects the os.
 
Joined
Mar 21, 2019
Messages
297 (0.25/day)
Location
Texas
System Name nogame
Processor AMD R9 5950x
Motherboard MSI MAG B550 Tomahawk
Cooling corsair h115i hydro series 280mm
Memory corsair vengeance rgb pro 32gb
Storage WD SN850 2tb + 3 WD and 1 Samsung ssd
Case Corsair icue 220t rgb
Power Supply corsair RM850 modular
Mouse logitech
Keyboard logitech
Software w10,
multiple Blue screen's of death will corrupt a system
 
Joined
Mar 10, 2010
Messages
10,069 (2.26/day)
Location
Manchester uk
System Name RyzenGtEvo/ Asus strix scar II/Trig
Processor Amd R5 5600G/ Intel 8750H/3800X
Motherboard Crosshair hero8 impact/Asus/crosshair hero 7
Cooling 360EK extreme rad+ 360$EK slim all push, cpu ek suprim Gpu full cover all EK
Memory Corsair Vengeance Rgb pro 3600cas14 16Gb in four sticks./16Gb/16GB
Video Card(s) Sapphire refference Rx vega 64 EK waterblocked/Rtx 2060/GTX 1060
Storage Silicon power 1TB nvme/8Tb external/1Tb samsung Evo nvme 2Tb sata ssd/1Tb nvme
Display(s) Samsung UAE28"850R 4k freesync.dellshiter
Case Lianli p0-11 dynamic/strix scar2/aero cool shiter
Audio Device(s) Xfi creative 7.1 on board ,Yamaha dts av setup, corsair void pro headset
Power Supply corsair 1200Hxi/Asus stock /850 watt ?
Mouse Roccat Kova/ Logitech G wireless
Keyboard Roccat Aimo 120
VR HMD Oculus rift
Software Win 10 Pro
Benchmark Scores 8726 vega 3dmark timespy/ laptop Timespy 6506
Poor memory or IMC tolerance to heat or frequency.
 
Joined
Jul 25, 2006
Messages
9,643 (1.67/day)
Location
Nebraska, USA
System Name Brightworks Systems BWS-6 E-IV
Processor Intel Core i5-6600 @ 3.9GHz
Motherboard Gigabyte GA-Z170-HD3 Rev 1.0
Cooling Quality case, 2 x Fractal Design 140mm fans, stock CPU HSF
Memory 32GB (4 x 8GB) DDR4 3000 Corsair Vengeance
Video Card(s) EVGA GEForce GTX 1050Ti 4Gb GDDR5
Storage Samsung 850 Pro 256GB SSD, Samsung 860 Evo 500GB SSD
Display(s) Samsung S24E650BW LED x 2
Case Fractal Design Define R4
Power Supply EVGA Supernova 550W G2 Gold
Mouse Logitech M190
Keyboard Microsoft Wireless Comfort 5050
Software W10 Pro 64-bit
Improper shut down of computer is a common cause for file corruption - and drivers are files too. So always "gracefully" shut down Windows, then power off the computer. Or, just let the computer go to sleep.

Of course, a graceful shutdown is unavoidable if you suddenly lose power. If your grid is unstable, consider getting a good UPS with AVR. In fact, IMO, every computer should be on a "good" UPS with AVR.

That said, a faulty power supply can cause sudden shutdowns, reboots or lockups too - and those can corrupt files.

Any number of other faulty components that result in system shutdowns, reboots, or lockups can do this too.
 
Joined
Feb 22, 2022
Messages
55 (0.57/day)
Processor AMD Ryzen 9 5950X
Motherboard Asus Crosshair VIII Dark Hero
Cooling Custom Watercooling
Memory G.Skill Trident Z Royal 2x16GB
Video Card(s) MSi RTX 3080ti Suprim X
Storage 2TB Corsair MP600 PRO Hydro X
Display(s) Samsung G7 27" x2
Audio Device(s) Sound Blaster ZxR
Power Supply Be Quiet! Dark Power Pro 12 1500W
Mouse Logitech G903
Keyboard Steelseries Apex Pro
If you are very, VERY unlucky a cosmic ray can cause a bit flip and corrupt a file, as it passes through your computer o_O
 
Last edited:
Joined
Mar 21, 2021
Messages
2,033 (4.68/day)
Location
Colorado, U.S.A.
System Name HP Compaq 8000 Elite CMT
Processor Intel Core 2 Quad Q9550
Motherboard Hewlett-Packard 3647h
Memory 16GB DDR3
Video Card(s) NVIDIA GeForce GT 1030 GDDR5 (fan-less)
Storage 2TB Seagate Firecuda 3.5"
Display(s) Dell P2416D (2560 x 1440)
Power Supply 12V HP proprietary
Software Windows 11 Pro 64-bit
It is great that error correction is part of DDR5
 
Joined
Feb 20, 2022
Messages
175 (1.79/day)
System Name Custom Watercooled
Processor 10900k 5.1GHz SSE 5.0GHz AVX
Motherboard Asus Maximus XIII hero z590
Cooling XSPC Raystorm Pro, XSPC D5 Vario, EK Water Blocks EK-CoolStream XE 360 (Triple Fan) Radiator
Memory Team Group 8Pack RIPPED Edition TDPPD416G3600HC14CDC01 @ DDR4-4000 CL15 Dual Rank 4x8GB (32GB)
Video Card(s) KFA2 GeForce RTX 3080 Ti SG 1-Click OC 12GB LHR GDDR6X PCI-Express Graphics Card
Storage WD Blue SN550 1TB NVME M.2 2280 PCIe Gen3 Solid State Drive (WDS100T2B0C)
Display(s) LG 3D TV 32LW450U-ZB and Samsung U28D590
Case Full Tower Case
Audio Device(s) ROG SupremeFX 7.1 Surround Sound High Definition Audio CODEC ALC4082, ESS SABRE9018Q2C DAC/AMP
Power Supply Corsair AX1000 Titanium 80 Plus Titanium Power Supply
Mouse Logitech G502SE
Keyboard Logitech Y-BP62a
Software Windows 11 Pro
Benchmark Scores https://valid.x86.fr/2rdbdl https://www.3dmark.com/spy/27927340 https://ibb.co/YjQFw5t
After a while or just randomly driver issues might pop up that'll cause your computer to BSOD. What causes this? Is it just the constant overwriting of old drivers that at some point it just gets corrupted? Wouldn't/shouldn't an overwrite of the corrupted install fix that? Also why does using DDU to essentially nuke your graphics driver have a better success rate than just uninstalling the driver and reinstalling?
As a computer techie this is one of those questions I've been pondering for a while as I have used DDU (when appropriate) just based on symptoms found and short of a hardware failure, seems to fix graphics issues. I don't know WHY though as in the nitty gritty of it all. I can only speculate but I wanted a more in-depth explanation as to how this happens.
Definite causes I already know of are malware, computer shutting down during the install/update, and possibly bad sector on the drive that the driver files are written to before sector itself became corrupted.
Run sfc /scannow in the cmd.exe prompt with admin and see if you have corrupted system files. You can also have hardware stability issues of faulty storage. A hard disk with a bad/damaged read head is a great one of random corruption that can lead to BSOD. RAM overclocking is the fastest way to corrupt everything.
 
Joined
Aug 20, 2007
Messages
17,712 (3.28/day)
System Name Pioneer
Processor Ryzen R9 5950X
Motherboard GIGABYTE X570 Aorus Elite
Cooling Noctua NH-D15 + A whole lotta Sunon and Corsair Maglev blower fans...
Memory Crucial Ballistix 64GB (4 x 16GB) @ DDR4-3600 (Micron E-Die, dual rank sticks)
Video Card(s) EVGA GeForce RTX 3090 Ti FTW3
Storage 2x Crucial P5 Plus 2TB PCIe 4.0 NVMe SSDs
Display(s) 55" LG 55" B9 OLED 4K Display
Case Thermaltake Core X31
Audio Device(s) TOSLINK->Schiit Modi MB->Asgard 2 DAC Amp->AKG Pro K712 Headphones or HDMI->B9 OLED
Power Supply EVGA SuperNova T2 Titanium 850W
Mouse Razer Deathadder v2
Keyboard WASD CODE Mechanical KB w/ Cherry MX Green switches
Software Windows 11 Enterprise (yes, it's legit)
Joined
Dec 11, 2019
Messages
286 (0.32/day)
Processor Xeon E-2124
Motherboard Fujitsu D3644-B1
Cooling Akasa AK-959CU
Memory 32GB Kingston DDR4 (4x KSM26ES8/8ME)
Video Card(s) Sapphire Pulse 550 2G 64-bit
Storage Samsung 870 Evo, Samsung 980
Display(s) Fujitsu B22-8 TS Pro
Case Akasa Raptor
Power Supply Antec HCG-620
Mouse Logitech MX500
It is great that error correction is part of DDR5

It was added to make poorer quality memory (and more profit), not to protect your data.
 
Joined
Jul 25, 2006
Messages
9,643 (1.67/day)
Location
Nebraska, USA
System Name Brightworks Systems BWS-6 E-IV
Processor Intel Core i5-6600 @ 3.9GHz
Motherboard Gigabyte GA-Z170-HD3 Rev 1.0
Cooling Quality case, 2 x Fractal Design 140mm fans, stock CPU HSF
Memory 32GB (4 x 8GB) DDR4 3000 Corsair Vengeance
Video Card(s) EVGA GEForce GTX 1050Ti 4Gb GDDR5
Storage Samsung 850 Pro 256GB SSD, Samsung 860 Evo 500GB SSD
Display(s) Samsung S24E650BW LED x 2
Case Fractal Design Define R4
Power Supply EVGA Supernova 550W G2 Gold
Mouse Logitech M190
Keyboard Microsoft Wireless Comfort 5050
Software W10 Pro 64-bit
It was added to make poorer quality memory (and more profit), not to protect your data.
:( That, of course, is nonsense.

Error correction has been around for decades. It is there to protect the data/avoid data corruption (what other purpose would it serve?) in specific scenarios. It was commonly used in servers and "mission critical" systems.

If the goal was to make poorer quality RAM, you would not see the least expensive DDR5 on Newegg still come with a lifetime warranty.

If this is a function of anything, it is deeper densities and faster speeds allowing for even less room for error. In other words, a good thing.
 
Joined
Dec 11, 2019
Messages
286 (0.32/day)
Processor Xeon E-2124
Motherboard Fujitsu D3644-B1
Cooling Akasa AK-959CU
Memory 32GB Kingston DDR4 (4x KSM26ES8/8ME)
Video Card(s) Sapphire Pulse 550 2G 64-bit
Storage Samsung 870 Evo, Samsung 980
Display(s) Fujitsu B22-8 TS Pro
Case Akasa Raptor
Power Supply Antec HCG-620
Mouse Logitech MX500
:( That, of course, is nonsense.

Error correction has been around for decades. It is there to protect the data/avoid data corruption (what other purpose would it serve?) in specific scenarios. It was commonly used in servers and "mission critical" systems.

If the goal was to make poorer quality RAM, you would not see the least expensive DDR5 on Newegg still come with a lifetime warranty.

If this is a function of anything, it is deeper densities and faster speeds allowing for even less room for error. In other words, a good thing.

Sigh. On-die ECC in consumer DDR5 is not equivalent to ECC in servers and "mission critical" systems. The lifetime warranty is irrelevant to the selection of memory chips during manufacture.

"On-die error correction code (ECC)3 and error check and scrub (ECS), which were first to be adopted in DDR5, also allow for more reliable technology node scaling by correcting single bit errors internally. Therefore, it is expected to contribute to further cost reduction in the future."

 
Joined
Jan 17, 2010
Messages
12,113 (2.68/day)
Location
Oregon
System Name Juliette // HTPC
Processor Intel i7 9700K // AMD Ryzen 5 5600G
Motherboard ASUS Prime Z390X-A // ASRock B550 ITX-AC
Cooling Noctua NH-U12 Black // Stock
Memory Corsair DDR4 3600 32gb //G.SKILL Trident Z Royal Series 16GB (2 x 8GB) 3600
Video Card(s) EVGA GeForce RTX 2080 XC ULTRA GAMING// APU
Storage Samsung 970 EVO NVMe 1Tb, Intel 665p Series M.2 2280 1TB // Samsung 1Tb SSD
Display(s) ASUS VP348QGL 34" Quad HD 3440 x 1440 // 55" LG 4K SK8000 Series
Case Seasonic SYNCRO Q7// Silverstone Granada GD05
Audio Device(s) Focusrite Scarlett 4i4 // HDMI to Samsung HW-R650 sound bar
Power Supply Seasonic SYNCRO 750 W // CORSAIR Vengeance 650M
Mouse Cooler Master MM710 53G
Keyboard Logitech 920-009300 G512 SE
Software Windows 10 Pro // Windows 10 Pro
Joined
Jul 25, 2006
Messages
9,643 (1.67/day)
Location
Nebraska, USA
System Name Brightworks Systems BWS-6 E-IV
Processor Intel Core i5-6600 @ 3.9GHz
Motherboard Gigabyte GA-Z170-HD3 Rev 1.0
Cooling Quality case, 2 x Fractal Design 140mm fans, stock CPU HSF
Memory 32GB (4 x 8GB) DDR4 3000 Corsair Vengeance
Video Card(s) EVGA GEForce GTX 1050Ti 4Gb GDDR5
Storage Samsung 850 Pro 256GB SSD, Samsung 860 Evo 500GB SSD
Display(s) Samsung S24E650BW LED x 2
Case Fractal Design Define R4
Power Supply EVGA Supernova 550W G2 Gold
Mouse Logitech M190
Keyboard Microsoft Wireless Comfort 5050
Software W10 Pro 64-bit
Sigh. On-die ECC in consumer DDR5 is not equivalent to ECC
Double sigh.

I never said it was equivalent.

I said it was nonsense to suggest error correction was integrated into DDR5 so, as you contend, the makers could produce "poorer quality" RAM.

And of course, error correction is there to protect the data. I ask again - what other function would it have? Just because it does not function exactly the same way as error correction used in server RAM, that does not mean it is not there to correct errors and help prevent data corruption.

The lifetime warranty is irrelevant to the selection of memory chips during manufacture.
Of course it is. Perhaps not directly since the chip maker often is not the RAM stick maker. But ultimately, of course it is relevant. The stick makers are not going to buy chips from chip makers if their chips have a high failure rate. And they surely are not going to buy chips from those makers if they have to keep honoring free replacements to the end users.

Cost reduction does NOT imply "poorer quality".
 
Joined
Feb 22, 2022
Messages
55 (0.57/day)
Processor AMD Ryzen 9 5950X
Motherboard Asus Crosshair VIII Dark Hero
Cooling Custom Watercooling
Memory G.Skill Trident Z Royal 2x16GB
Video Card(s) MSi RTX 3080ti Suprim X
Storage 2TB Corsair MP600 PRO Hydro X
Display(s) Samsung G7 27" x2
Audio Device(s) Sound Blaster ZxR
Power Supply Be Quiet! Dark Power Pro 12 1500W
Mouse Logitech G903
Keyboard Steelseries Apex Pro
Cost reduction does NOT imply "poorer quality".
This! So much this!

Stop with the conspiracy theories that lower cost = lower quality.
 

qubit

Overclocked quantum bit
Joined
Dec 6, 2007
Messages
17,211 (3.25/day)
Location
Quantum Well UK
System Name Quantumville™
Processor Intel Core i7-2700K @ 4GHz
Motherboard Asus P8Z68-V PRO/GEN3
Cooling Noctua NH-D14
Memory 16GB (2 x 8GB Corsair Vengeance Black DDR3 PC3-12800 C9 1600MHz)
Video Card(s) MSI RTX 2080 SUPER Gaming X Trio
Storage Samsung 850 Pro 256GB | WD Black 4TB | WD Blue 6TB
Display(s) BenQ XL2720Z (144Hz, 3D Vision 2, 1080p) | Asus MG28UQ (4K, 60Hz, FreeSync compatible)
Case Cooler Master HAF 922
Audio Device(s) Creative Sound Blaster X-Fi Fatal1ty PCIe
Power Supply Corsair HX 850W v1
Mouse Microsoft Intellimouse Pro - Black Shadow
Keyboard Yes
Software Windows 10 Pro 64-bit
If you are very, VERY unlucky a cosmic ray can cause a bit flip and corrupt a file as it passes through your computer o_O
Oh yeah, aircraft computers have a surprising amount of bit flips due to cosmic rays due to less atmospheric protection at high altitude.

The designers try to protect against and mitigate these effects as much as possible.
 
Joined
Dec 11, 2019
Messages
286 (0.32/day)
Processor Xeon E-2124
Motherboard Fujitsu D3644-B1
Cooling Akasa AK-959CU
Memory 32GB Kingston DDR4 (4x KSM26ES8/8ME)
Video Card(s) Sapphire Pulse 550 2G 64-bit
Storage Samsung 870 Evo, Samsung 980
Display(s) Fujitsu B22-8 TS Pro
Case Akasa Raptor
Power Supply Antec HCG-620
Mouse Logitech MX500
Double sigh.

I never said it was equivalent.

I said it was nonsense to suggest error correction was integrated into DDR5 so, as you contend, the makers could produce "poorer quality" RAM.

And of course, error correction is there to protect the data. I ask again - what other function would it have? Just because it does not function exactly the same way as error correction used in server RAM, that does not mean it is not there to correct errors and help prevent data corruption.


Of course it is. Perhaps not directly since the chip maker often is not the RAM stick maker. But ultimately, of course it is relevant. The stick makers are not going to buy chips from chip makers if their chips have a high failure rate. And they surely are not going to buy chips from those makers if they have to keep honoring free replacements to the end users.

Cost reduction does NOT imply "poorer quality".

Memory that is so unreliable it needs on-die ECC is poor quality, you can say "it is progress man, we need it for density and moar speed", I can say "poor quality", whatever.

If they were really interested in protecting your data, they'd implement proper ECC support in the cpu, dimms and motherboard. But, they didn't.

Suggesting that the ECC in consumer DDR5 is there to protect our data, while perhaps technically accurate (in terms of what it does), is misleading as to the reason it was implemented and the consequence for consumers. I would not replace DDR4 with DDR5, with the assumption I'm now protected like a server (if my data were that important to me), by the presence of on-die ECC. This is what we're being led to believe (for marketing purposes) is the case.

What it actually means is (during manufacture and selection) more memory errors will be considered acceptable than they were before, but that doesn't sound so appealing to buy, does it?

Just cos it is broken doesn't mean it will fail prematurely. For example: all flash memory is broken (yeah, yeah, 'broken' is a relative term), some is just a little less broken and SSDs still come with decent warranties regardless of their status out of the box.

"Cost reduction does NOT imply "poor quality". - Sure, sure, keep the faith.
 

Aquinus

Resident Wat-man
Joined
Jan 28, 2012
Messages
12,683 (3.36/day)
Location
Concord, NH
System Name Apollo
Processor Intel Core i9 9880H
Motherboard Some proprietary Apple thing.
Memory 64GB DDR4-2667
Video Card(s) AMD Radeon Pro 5600M, 8GB HBM2
Storage 1TB Apple NVMe, 4TB External
Display(s) Laptop @ 3072x1920 + 2x LG 5k Ultrafine TB3 displays
Case MacBook Pro (16", 2019)
Audio Device(s) AirPods Pro, Sennheiser HD 380s w/ FIIO Alpen 2, or Logitech 2.1 Speakers
Power Supply 96w Power Adapter
Mouse Logitech MX Master 3
Keyboard Logitech G915, GL Clicky
Software MacOS 12.1
Improper shut down of computer is a common cause for file corruption - and drivers are files too.
Actually I think a lot of filesystems will not overwrite the existing file, but will write the data to a free block(s) and then write it to the journal, leaving the space for the "old file" open for reclamation. From that perspective, computers are actually pretty resilient to unexpected shutdowns because an incomplete write is just treated as free space after reboot and the existing file is still there.

Memory that is so unreliable it needs on-die ECC is poor quality, you can say "it is progress man, we need it for density and moar speed", I can say "poor quality", whatever.

If they were really interested in protecting your data, they'd implement proper ECC support in the cpu, dimms and motherboard. But, they didn't.

Suggesting that the ECC in consumer DDR5 is there to protect our data, while perhaps technically accurate (in terms of what it does), is misleading as to the reason it was implemented and the consequence for consumers. I would not replace DDR4 with DDR5, with the assumption I'm now protected like a server (if my data were that important to me), by the presence of on-die ECC. This is what we're being led to believe (for marketing purposes) is the case.

What it actually means is (during manufacture and selection) more memory errors will be considered acceptable than they were before, but that doesn't sound so appealing to buy, does it?

Just cos it is broken doesn't mean it will fail prematurely. For example: all flash memory is broken (yeah, yeah, 'broken' is a relative term), some is just a little less broken and SSDs still come with decent warranties regardless of their status out of the box.

"Cost reduction does NOT imply "poor quality". - Sure, sure, keep the faith.
If non-ECC memory is so bad, maybe you could enlighten us as to how often bits flip on machine running standard JEDEC speeds and not with an overclocked (sp. out of spec,) CPU. I suspect it's not very often and you're probably not doing something "mission critical" in the truest sense of the term.

The reality is that overclocking and running mismatched DIMMs is the biggest source of data corruption, at least from my perspective. Fight me. :p
 
Joined
Apr 18, 2019
Messages
491 (0.43/day)
Location
The New England region of the United States
System Name Daily Driver(gaming, browsing)
Processor Ryzen 7 3800X
Motherboard Gigabyte X570 Aurus Pro Wifi
Cooling Noctua NH-D15 chromax.black
Memory 32GB(2x16GB) Patriot Viper DDR4-3200C16
Video Card(s) EVGA RTX 3060 Ti
Storage Samsung 970 EVO Plus 1TB (Boot/OS)|Samsung 980 1TB (Games)|2 x Samsung 860 EVO 1TB & 500GB (Misc)
Display(s) Gigabyte G27F
Case Corsair Graphite 600T w/mesh side
Audio Device(s) Logitech Z625 2.1 | cheapo gaming headset when mic is needed
Power Supply Corsair HX850i
Mouse Redragon M808-KS Storm Pro (Great Value)
Keyboard Redragon K512 Shiva replaced a Corsair K70 Lux - Blue on Black
VR HMD Nope
Software Windows 10 Pro x64
Benchmark Scores Nope
Memory that is so unreliable it needs on-die ECC is poor quality, you can say "it is progress man, we need it for density and moar speed", I can say "poor quality", whatever.

If they were really interested in protecting your data, they'd implement proper ECC support in the cpu, dimms and motherboard. But, they didn't.

Suggesting that the ECC in consumer DDR5 is there to protect our data, while perhaps technically accurate (in terms of what it does), is misleading as to the reason it was implemented and the consequence for consumers. I would not replace DDR4 with DDR5, with the assumption I'm now protected like a server (if my data were that important to me), by the presence of on-die ECC. This is what we're being led to believe (for marketing purposes) is the case.

What it actually means is (during manufacture and selection) more memory errors will be considered acceptable than they were before, but that doesn't sound so appealing to buy, does it?

Just cos it is broken doesn't mean it will fail prematurely. For example: all flash memory is broken (yeah, yeah, 'broken' is a relative term), some is just a little less broken and SSDs still come with decent warranties regardless of their status out of the box.

"Cost reduction does NOT imply "poor quality". - Sure, sure, keep the faith.
Wow, so cynical... I'm sure you work for a DRAM manufacturer or JEDEC and know all the reasons why ECC was added internally for DDR5. AFAIK, with larger and larger amounts RAM being installed in PCs, the chances of a flipped bit in data at rest has increased substantially. This new feature is meant to prevent that. Yeah, it isn't full-blown end to end ECC like we see in workstations and servers but do most people really need that anyway? Most people who have bad RAM don't know it in the first place, they think the software is junk or that it must be a virus.
 
Joined
Mar 21, 2021
Messages
2,033 (4.68/day)
Location
Colorado, U.S.A.
System Name HP Compaq 8000 Elite CMT
Processor Intel Core 2 Quad Q9550
Motherboard Hewlett-Packard 3647h
Memory 16GB DDR3
Video Card(s) NVIDIA GeForce GT 1030 GDDR5 (fan-less)
Storage 2TB Seagate Firecuda 3.5"
Display(s) Dell P2416D (2560 x 1440)
Power Supply 12V HP proprietary
Software Windows 11 Pro 64-bit
Last edited:
Joined
Dec 11, 2019
Messages
286 (0.32/day)
Processor Xeon E-2124
Motherboard Fujitsu D3644-B1
Cooling Akasa AK-959CU
Memory 32GB Kingston DDR4 (4x KSM26ES8/8ME)
Video Card(s) Sapphire Pulse 550 2G 64-bit
Storage Samsung 870 Evo, Samsung 980
Display(s) Fujitsu B22-8 TS Pro
Case Akasa Raptor
Power Supply Antec HCG-620
Mouse Logitech MX500
Wow, so cynical... I'm sure you work for a DRAM manufacturer or JEDEC and know all the reasons why ECC was added internally for DDR5. AFAIK, with larger and larger amounts RAM being installed in PCs, the chances of a flipped bit in data at rest has increased substantially. This new feature is meant to prevent that. Yeah, it isn't full-blown end to end ECC like we see in workstations and servers but do most people really need that anyway? Most people who have bad RAM don't know it in the first place, they think the software is junk or that it must be a virus.

You don't have to believe me man, just read the non-public facing documentation from people like Hynix or Micron, or look at the more informed/technical articles on the net. I'm not the only one who has addressed the misconception about the nature of DDR5's ECC *. Is my interpretation of the facts cynical AF? Probably, but like I said, this stuff is in the public domain. If you ask someone from Micron directly, obviously they're not going to describe it to you the way I just did. They'll tell you about the shrinking nodes, the density and the speed, just like Bill did in his reply.

If non-ECC memory is so bad, maybe you could enlighten us as to how often bits flip on machine running standard JEDEC speeds and not with an overclocked (sp. out of spec,) CPU. I suspect it's not very often and you're probably not doing something "mission critical" in the truest sense of the term.

The reality is that overclocking and running mismatched DIMMs is the biggest source of data corruption, at least from my perspective. Fight me. :p

It wasn't my point to say if non-ecc memory is bad, or not. I was referring to the memory chips themselves and the differences between ddr4/ddr5 and how they're made. Believe it or not, I actually have a fairly high trust in the non-ecc DDR4 I own. I can only answer your question privately and I've never had a corrected error recorded in the logs on this system (run for several years 24/7), or the other ECC system I run that was built a few years prior.

* The salient part is after 6:11 (this video is by Ian Cutress, senior editor of Anandtech for 11 years, if you don't believe him either, idk what to tell you).


For those who dislike watching videos:

What on-die ecc support means think of it this way, on-die ecc allows memory manufacturers to go denser on the process to get higher density memory and more of it comes out the factory and lowers the cost, what it does not do is protect your data, it enables more scaling down to denser process nodes.

So the takeaway here is that on die ecc is not an ecc for you or me it is simply a device that helps make the memory cheaper and better yielding. If you do need a proper end-to-end ecc solution where your data is fully protected from cpu to memory or memory to accelerator then you still need to invest in an ecc based platform, it's going to be fun to see how these memory manufacturers deal with customers who say you said i had ecc when in actual fact you really don't.

on die ecc is a way of managing those bit flips so more cells at the production stage pass the validation method, it's simply to say well normally some cells won't work because you get defects in the manufacturing process, but with this on die ecc you can actually make sure that more of those cells reach the required jedec specification and you can sell that memory
 
Last edited:

qubit

Overclocked quantum bit
Joined
Dec 6, 2007
Messages
17,211 (3.25/day)
Location
Quantum Well UK
System Name Quantumville™
Processor Intel Core i7-2700K @ 4GHz
Motherboard Asus P8Z68-V PRO/GEN3
Cooling Noctua NH-D14
Memory 16GB (2 x 8GB Corsair Vengeance Black DDR3 PC3-12800 C9 1600MHz)
Video Card(s) MSI RTX 2080 SUPER Gaming X Trio
Storage Samsung 850 Pro 256GB | WD Black 4TB | WD Blue 6TB
Display(s) BenQ XL2720Z (144Hz, 3D Vision 2, 1080p) | Asus MG28UQ (4K, 60Hz, FreeSync compatible)
Case Cooler Master HAF 922
Audio Device(s) Creative Sound Blaster X-Fi Fatal1ty PCIe
Power Supply Corsair HX 850W v1
Mouse Microsoft Intellimouse Pro - Black Shadow
Keyboard Yes
Software Windows 10 Pro 64-bit
Yeah, it isn't full-blown end to end ECC like we see in workstations and servers but do most people really need that anyway? Most people who have bad RAM don't know it in the first place, they think the software is junk or that it must be a virus.
My PC (see specs) originally had 16GB DDR3 as a 4x4 arrangement. For years it was intermittently unstable, ie BSOD and other niggly errors. Really intermittent, so much so that troubleshooting just couldn't pinpoint it, including removing the memory modules. It often looked like software was doing it, but I didn't think so.

A couple of years ago, the instability started to get noticeably worse where it would BSOD every few minutes which allowed me to finally track it down to one of the memory modules with 100% certainty. I did notice that with 4 modules installed, they were all running hot, with the one closest to the CPU very hot, even though the PC isn't overclocked. I suspect that this heat was enough to make the module unstable and eventually to damage it, or maybe it was slightly faulty from new, I'll never know. I don't know if that faulty module was actually in that slot though.

I then replaced all the modules with new 2x8 Corsair modules of the same type and model range and put them in slots 1 & 3 so that the closest slot isn't occupied and the PC has run fine ever since - a blessed relief after all that time not running quite right. The two modules run a lot cooler, too. I've actually got another 2x8 modules used bought from a friend as part of a spare mobo / CPU / RAM / cooler bundle. I tried them and while it was great seeing 32GB in this old Sandy Bridge rig, the modules ran hot again, so I've removed them. I don't need 32GB and it's just not worth the headache. I'd have to add an extra fan there to fix the heat problem and that's not good for a PC built to be very quiet.
 
Joined
Mar 21, 2021
Messages
2,033 (4.68/day)
Location
Colorado, U.S.A.
System Name HP Compaq 8000 Elite CMT
Processor Intel Core 2 Quad Q9550
Motherboard Hewlett-Packard 3647h
Memory 16GB DDR3
Video Card(s) NVIDIA GeForce GT 1030 GDDR5 (fan-less)
Storage 2TB Seagate Firecuda 3.5"
Display(s) Dell P2416D (2560 x 1440)
Power Supply 12V HP proprietary
Software Windows 11 Pro 64-bit
Strange, my DDR3 sticks are not even warm to the touch.
 

qubit

Overclocked quantum bit
Joined
Dec 6, 2007
Messages
17,211 (3.25/day)
Location
Quantum Well UK
System Name Quantumville™
Processor Intel Core i7-2700K @ 4GHz
Motherboard Asus P8Z68-V PRO/GEN3
Cooling Noctua NH-D14
Memory 16GB (2 x 8GB Corsair Vengeance Black DDR3 PC3-12800 C9 1600MHz)
Video Card(s) MSI RTX 2080 SUPER Gaming X Trio
Storage Samsung 850 Pro 256GB | WD Black 4TB | WD Blue 6TB
Display(s) BenQ XL2720Z (144Hz, 3D Vision 2, 1080p) | Asus MG28UQ (4K, 60Hz, FreeSync compatible)
Case Cooler Master HAF 922
Audio Device(s) Creative Sound Blaster X-Fi Fatal1ty PCIe
Power Supply Corsair HX 850W v1
Mouse Microsoft Intellimouse Pro - Black Shadow
Keyboard Yes
Software Windows 10 Pro 64-bit
Strange, my DDR3 sticks are not even warm to the touch.
I think it's the ventilation around that area is poor.
 
Top