Are issues with overclocking RAM even beneficial?
Are issues with overclocking RAM even beneficial?
I have a Gigabyte Aorus mobo with 64 GB DDR4 4000Mhz RAM. It came with 32GB, and I added another 32GB of the exact same mfg, make, timing (I know this is the first question I'll be asked, but I checked and double checked - its the exact same 4 pieces of RAM).
The mobo defaults to 2400MHz. Since I have 4000MHz RAM, I thought to enable the XMP profile. There is only one profile in BIOS when I enable XMP, called Profile 1. It lists the RAM as 4000MHz. But after enabling it, I would get random BSOD (clock watchdog timeout). I am going to be fair and say that I did encounter BSOD before installing the RAM, but it seemed to go away with a BIOS update. I cannot be sure of this as the computer is relatively new and BSOD is tricky to reproduce. If I just leave the computer on, it can go hours (days) without BSOD. DisplayCal seems to cause it to BSOD more often, but I cannot be sure.
I read somewhere that I could try lowering the clock speed of the RAM. At last count, 3600 still caused BSOD or hard crash without BSOD (particularly with DisplayCal).
I have just lowered it further to 3400.
But my question here is: is it even worth it? I mean, I really would love to get the performance I paid for with the expensive 4000MHz RAM, but if I keep getting BSOD, I would stick with disabling XMP and just living with the reduced speed. Is it even that much reduced for that matter, how much would I notice the difference? OTOH I will surely notice the difference when a BSOD occurs!
If it is worth it, then there seems to be a lot to do, with timings and whatnot, and all the while the only way to know if it worked is to wait for a BSOD to maybe or maybe not eventually happen. Seems like a pain, so I'd do it only if its worth the gain.
Any help or suggestions would be appreciated, thanks!
System specifications are available, including the PSU make and model. Have you verified the installation of the motherboard drivers? What attempts have been made to run memtest?
The information retrieved from the System Information in Windows is as follows:
OS Name: Microsoft Windows 11 Pro
Version: 10.0.22000 Build 22000
Other OS Description: Not Available
OS Manufacturer: Microsoft Corporation
System Name: Z590 AORUS PRO AX
System Manufacturer: Gigabyte Technology Co., Ltd.
System Model: Z590 AORUS PRO AX
System Type: x64-based PC
System SKU: Default string
Processor: 11th Gen Intel® Core™ i9-11900KF @ 3.50GHz, 3504 Mhz, 8 Core(s), 16 Logical Processor(s)
BIOS Version/Date: American Megatrends International, LLC. F9a, 1/24/2022
SMBIOS Version: 3.3
Embedded Controller Version: 255.255
BIOS Mode: UEFI
BaseBoard Manufacturer: Gigabyte Technology Co., Ltd.
BaseBoard Product: Z590 AORUS PRO AX
BaseBoard Version: Default string
Platform Role: Desktop
Secure Boot State: Off
PCR7 Configuration: Elevation Required to View
Windows Directory: C:\WINDOWS
System Directory: C:\WINDOWS\system32
Boot Device: \Device\HarddiskVolume5
Locale: United States
Hardware Abstraction Layer: Version = "10.0.22000.778"
User Name
Time Zone: Pacific Daylight Time
Installed Physical Memory (RAM): 64.0 GB
Total Physical Memory: 63.9 GB
Available Physical Memory: 56.2 GB
Total Virtual Memory: 73.4 GB
Available Virtual Memory: 62.6 GB
Page File Space: 9.50 GB
Page File Path: C:\pagefile.sys
Kernel DMA Protection: Off
Virtualization-based security: Running
Required Security Properties: Base Virtualization Support, UEFI Code Readonly, SMM Security Mitigations 1.0, Mode Based Execution Control, APIC Virtualization
Available Security Properties: Hypervisor enforced Code Integrity
Hypervisor Enforcement: Enforced
User Mode Policy: Windows Defender Application Control policy Enforced
Application Control Policies:
Device Encryption Support: Elevation Required to View
A hypervisor is present. Required features for Hyper-V will not be shown.
I would first attempt to stabilize the system using one kit of RAM, only adding another if necessary. Even though the new kit matches the original one you’re mixing, this can still lead to issues, which is why RAM is sold in matched sets. ee odd man out section here. https://forums. With two kits installed, it’s unclear if the problem stems from mixing kits or another factor. On the motherboard compatibility list, does it mention any 4 dimm kits at 4000mhz? Running four dimms might reduce the maximum stable speed.
I previously shared a discussion about overclocking, but now I suspect it's not the main concern, so this thread feels more suitable.
I've attached a detailed timeline of events since purchasing the PC. I bought the new system from NZXT last December. It was intended to replace a failing machine (which is now non-functional), but because the other unit performed better than expected, it mostly remained idle. I ran some light tasks to confirm it worked, though never for more than about an hour. It triggered a BSOD around a week after delivery, prompting me to contact support, who recommended updating the BIOS. That seemed to temporarily fix the issue.
However, I remained uncertain since the PC was mostly inactive.
Later, I discovered the RAM wasn't operating at its full speed—its rated speed was 4000MHz, while the default was set at 2400MHz. I activated the XMP profile, which then caused another BSOD. On forums, others advised testing the memory and adjusting timings. I tried lowering the timings by increasing the system memory multiplier (e.g., from 4000 to 3800), but the problem persisted.
Then a memory sale at NewEgg prompted me to add another 32GB of RAM. The new RAM matched the manufacturer's specifications, and it brought more BSODs. Still, due to infrequent use, I wasn't sure if this was a recurrence or a new issue from the installation. The advice was to test the RAM again, which I did without finding any errors.
I then reset the BIOS to "optimized defaults" to clear previous overclocking settings. Unfortunately, the CWT BSOD continued. Following that suggestion, I increased the Vcore to 1.4 (a conservative safe setting), but the issue remained.
I became increasingly frustrated because some of my assumptions—like resetting everything as a last resort—didn't resolve the problem.
Another suggestion someone made has since caused me some regret: updating the motherboard drivers. I updated all drivers listed on my specific motherboard from the manufacturer's site. However, this didn't fix the issue and I'm wondering if it might have introduced new problems.
I also experience CWT, WHEA Uncorrectable, and Kernal Security BSODs occasionally, though most are silent. Occasionally, system crashes or reboots happen without a BSOD.
The timing of these crashes is quite erratic. I suspect copying many files is the main trigger for the CWT BSOD—almost every time I copy files from an external HDD or network server, the PC freezes after a few minutes. I can play games without issues, but copying them guarantees a crash within 30 minutes. Initially, I thought it was related to the HDD or USB, but switching to an internal SSD didn't help. The crash happens regardless of whether I'm copying from a backup HDD or network drive, and whether the destination is external or internal storage.
Other factors that seem to cause crashes include running DisplayCal for monitor calibration, watching YouTube videos (possibly coincidental), and sometimes background programs running silently.
Playing games doesn't seem to affect stability. I've tested multiple memory diagnostic tools, but none have crashed the system or reported errors. If memory is the root cause, it might be a missing or misconfigured test.
Below is my current system configuration:
OS Name: Microsoft Windows 11 Pro
Version: 10.0.22000 Build 22000
Other OS Info: Not available
OS Manufacturer: Microsoft Corporation
System Name: MARKS-NZXT
System Manufacturer: Gigabyte Technology Co., Ltd.
System Model: Z590 AORUS PRO AX
System Type: x64-based PC
System SKU: Default string
Processor: 11th Gen Intel® Core™ i9-11900KF @ 3.50GHz, 3504 Mhz, 8 Core(s), 16 Logical Processor(s)
BIOS Version/Date: American Megatrends International, LLC. F9a, 1/24/2022
SMBIOS Version: 3.3
Embedded Controller Version: 255.255
BIOS Mode: UEFI
BaseBoard Manufacturer: Gigabyte Technology Co., Ltd.
BaseBoard Product: Z590 AORUS PRO AX
BaseBoard Version: Default string
Platform Role: Desktop
Secure Boot State: Off
PCR7 Configuration Elevation Required: To View
Windows Directory: C:\WINDOWS
System Directory: C:\WINDOWS\system32
Boot Device: \Device\HarddiskVolume5
Locale: United States
Hardware Abstraction Layer Version: 10.0.22000.778
User Name: MARKS-NZXT\markw
Time Zone: Pacific Daylight Time
Installed Physical Memory (RAM): 64.0 GB
Total Physical Memory: 63.9 GB
Available Physical Memory: 55.8 GB
Total Virtual Memory: 73.4 GB
Available Virtual Memory: 62.4 GB
Page File Space: 9.50 GB
Page File Location: C:\pagefile.sys
Kernel DMA Protection: Off
Virtualization-based security: Not enabled
Device Encryption Support: Elevation Required to View
Hyper-V - VM Monitor Mode Extensions: Yes
Hyper-V - Second Level Address Translation Extensions: Yes
Hyper-V - Virtualization Enabled in Firmware: Yes
Hyper-V - Data Execution Protection: Yes
I'm really hoping for any guidance, even a small detail that could help. Please let me know if anything I've shared can be useful! Thank you!
And that's why you encounter problems. Your understanding is off. The RAM underneath the heatsink contains PCBs with silicon chips attached. In kit #1, all those chips originated from a single silicon sheet. In kit #2, each chip was made from a different sheet. The RAM exterior remains consistent, but internal specifications vary due to differences in impurity levels and composition across each silicon sheet, affecting the Secondary and Tertiary timings.
The sole shared feature is the Primary timings, model number, speed/size, and heatsink color. Under typical conditions, this CPU should run smoothly with four sticks. However, unusual situations may arise. You might need to enable XMP, adjust the DRAM voltage manually to around 1.38v±0.02v, and possibly tweak VCCIO slightly—perhaps 1.2v for VCCIO and 1.3v with ±0.05v for VCCSA.
I'm starting to question if it's actually the RAM and it could be a wrong assumption. My reasons are:
using certain apps like games that usually need a lot of RAM rarely (almost never) cause BSOD.
running memory tests I've done multiple times hasn't detected any errors or BSOD.
I've already reset the BIOS to default, so overclocking shouldn't be the problem (xmp is now off).
the only thing that definitely causes the BSOD is copying a large number of files from one hard drive to another. I have a backup drive that I'm trying to copy to my internal drive, and this process always leads to BSOD.
Or would you still consider RAM as a possible cause for another reason?
What is the bsod? Are there any significant issues in the Event Viewer besides a sudden power outage? Doing this frequently increases the likelihood of data corruption. Try running CMD with admin rights.
Type : sfc /scannow
Type : dism /online /cleanup-image /restorehealth
Check the results.
I encounter three distinct BSOD scenarios. The most frequent is CLOCK_WATCHDOG_TIMEOUT. In other cases, I may face WHEA_UNCORRECTABLE_ERROR or KERNAL_SECURITY_CHECK_FAILURE. Please let me know if minidumps would be useful—I've begun saving them and sharing them online to aid in troubleshooting. Regarding the sfc and dism checks, they are now showing normal results (I've resolved past issues), though the BSODs persist without any detected problems from those checks.