Question Windows 10 became slow after removing a corrupted memory module

Aug 23, 2020
5
0
10
Hi
I have 4x8GB DDR3 RAM modules that have been working perfectly for several years. I started receiving random BSOD and the investigations led to a single RAM module being the defective one. After removing it, the BSODs stopped.
However, Windows 10 (up to date) became too laggy with the CPU/RAM not being fully utilized, event after a fresh restart. I remove another RAM module so that the system runs using a 2/4 memory bank combination, and am still facing a very laggy experience which was not happening before. SFC/scannow does not yield a problem, neither does windows' memory diagnostic tool upon starting the computer.
I am using a i7 4790k with Corsair Vengeance Pro 4x8 2400 running at 1600MHz as per CPU's maximum RAM frequency support (was running 32Gb perfectly fine at 2400 according to their label and motherboard, but then changed it to 1600 MHz after facing this problem and removing the 8GB module), with a Gigabyte Z97X-UD5H-BK motherboard with an up to date BIOS. XMP is disabled and wasn't enable before.
Any help would be appreciated.
Thank you
 
Last edited:
Hi
I have 4x8GB DDR3 RAM modules that have been working perfectly for several years. I started receiving random BSOD and the investigations led to a single RAM module being the defective one. After removing it, the BSODs stopped.
However, Windows 10 (up to date) became too laggy with the CPU/RAM not being fully utilized, event after a fresh restart. I remove another RAM module so that the system runs using a 2/4 memory bank combination, and am still facing a very laggy experience which was not happening before. SFC/scannow does not yield a problem, neither does windows' memory diagnostic tool upon starting the computer.
I am using a i7 4790k with Corsair Vengeance Pro 4x8 2400 running at 1600MHz as per CPU's maximum RAM frequency support (was running RAM perfectly fine at 2400 according to their label and motherboard, but then changed it to 1600 MHz after facing this problem), with a Gigabyte Z97X-UD5H-BK motherboard with an up to date BIOS. RAM settings are set to Auto in BIOS, XMP is disabled.
Any help would be appreciated.
Thank you
I don't fully understand. You decreased your RAM speed from 2400 to 1600 Mhz by disabling XMP ?
 
Aug 23, 2020
5
0
10
I don't fully understand. You decreased your RAM speed from 2400 to 1600 Mhz by disabling XMP ?
Hi Egda23
I never enable XMP before when I had the 32GB modules installed, but change the frequency multiplier in BIOS so that the speed became 2400 MHz. I changed the multiplier after removing the 8 and 16GB modules in case it was affecting the CPU's performance. It is now working as per the CPU's limits.
I updated my original post so that it is clearer now.
Thanks
 
Hi Egda23
I never enable XMP before when I had the 32GB modules installed, but change the frequency multiplier in BIOS so that the speed became 2400 MHz. I changed the multiplier after removing the 8 and 16GB modules in case it was affecting the CPU's performance. It is now working as per the CPU's limits.
I updated my original post so that it is clearer now.
Thanks
This is even more confusing. You are saying you had been running 4x8Gb at 2400MT/s, but that one of the modules was faulty. So you removed modules until you were at 2x8GB and also dropped the sped to 1600, but you also removed a 16GB module? Did you mean removing 16GB total from two modules?

What exactly does laggy experience mean? I don't get why you would think it wouldn't be slower if you dropped the speed to 1600. If you lowered the speed to 1600 and didn't decrease the cas latency, it will be slower and possibly make the system slower as well. If the modules are capable of 2400MT/s, you can likey run 1600 at CL9 or possibly even CL8.

Also, unless you tested your modules at 1066 instead of 2400, it's possible the faulty module is fine. Running 4x8GB at 2400MT/s, which would likely require more than 1.65v, would probably become a problem over time.
 
Aug 23, 2020
5
0
10
This is even more confusing. You are saying you had been running 4x8Gb at 2400MT/s, but that one of the modules was faulty. So you removed modules until you were at 2x8GB and also dropped the sped to 1600, but you also removed a 16GB module? Did you mean removing 16GB total from two modules?

What exactly does laggy experience mean? I don't get why you would think it wouldn't be slower if you dropped the speed to 1600. If you lowered the speed to 1600 and didn't decrease the cas latency, it will be slower and possibly make the system slower as well. If the modules are capable of 2400MT/s, you can likey run 1600 at CL9 or possibly even CL8.

Also, unless you tested your modules at 1066 instead of 2400, it's possible the faulty module is fine. Running 4x8GB at 2400MT/s, which would likely require more than 1.65v, would probably become a problem over time.

I have been running 4x8Gb at 2400 for 5 years with perfect speeds. Now one of the 8GB became faulty and I removed it and kept the speed at 2400. After removing it, the performance became extremely slow. I thought it might have to do with using 3 modules in dual channel mode, hence I removed another 8GB module just in case. Performance did not improve. I then reduced the speed from 2400 to 1600 MHz if that was causing the issue.

By laggy I mean opening a folder would take 8-9 seconds. Moving a window makes its content appear tearing. Boot up time has increased by 2-3 folds (I am using an M.2).

I am aware that reducing RAM speed from 2400 to 1600 will have an impact, but to the extent not in standard usage (it will be dramatic in demanding apps).

Will test the faulty module at 1066 and update this thread.
 
KhaldoonSaid

Post the rest of your system specs. Also, I forgot to ask what program you used to test the memory to determine which module was faulty. I usually use memtest86, because it's usually the most accurate or accurate enough. Don't run memory tests in Windows to determine if a module is faulty.

The problem you describe sounds like an issue with the SSD and possibly corrupted data which could have been caused by the faulty module. You mentioned you already did a file check, but it's possible that there are still corrupted system files as well as the GPU driver and files.

It's also possible something is wrong with the motherboard. Test the SSD with the manufacture's diagnostic software if it's available or a third=party software like CrystalDiskInfo or SSD Life.
 
Aug 23, 2020
5
0
10
I used "Windows Memory Diagnostic Tool" upon booting and right before Windows starts. I tested it with the faulty module alone after removing the other modules, and it failed in every memory slot, event when testing it at 1066 MHz. I removed it and used the other modules and it passed.
100% agree on not running memory tests in Windows.

The problem might be corrupted data as I had many BSODs in the past week while using the faulty module along with the others, which might have had an impact on system files. What made me think otherwise is that SFC/Scannow did not yield any file corruption. Any recommendations on how to further test system files integrity?

the M.2 tool (Western Digital SSD Dashboard 2.5.0.0) says the drive is in 100% good condition, and CyrstialDiskInfo gives me a "Good 100%" result. Drive's firmware is up to date. The S.M.A.R.T. test completed with no detected problems.

System specs are:
  • Intel Core i7 4790K @4.0 GHz
  • Gigabyte Z97X-UD5H-BK motherboard with Intel HD Graphics 4600 (not used)
  • 2x8GB RAM modules - Corsair Vengeance Pro DDR3 running at 1600 MHz (11-13-13-31)
  • Gigabyte GTX 1080 Founder's Edition 8GB
  • 1 NVMe M.2 Western Digital 500 GB (WDS500G3X0C-00SJG0) used only for Operating System (Windows 10 Pro 64-bit - Version 2004 build 19041.450)
  • 5 HDD (2x8 TB, 2x6TB, 1x4TB), SATA
  • Corsair TX850M power supply - 850W
  • Wired LAN connection for internet
  • DVI connection to monitor
  • Bluetooth USB adapter for sound connectivity
  • Bluetooth USB adapter for keyboard + mouse connectivity