Question My Computer won't post with any kind of RAM overclock.

Aug 29, 2019
26
3
45
I built my first gaming PC in April this year and for the first few months everything was running fine. I turned on the XMP profile for my RAM to run at 3200mhz instead of 2133mhz and "Game Boost" for an automatic CPU overclock. I then had an issue with the RAM and had to get it replaced by Corsair. I got the replacement and everything was fine again, using XMP and Game boost. Recently, I've been having issues where my PC will occasionally not post. I can use it for a long gaming session without issue, turn it off and come back to it an hour later for it not to post. I am not 100% sure of the root cause of this issue, but since I plugged the PSU directly into the wall socket and don't have any overclock it has booted more consistently. I decided to try activating the XMP profile again for my RAM and it didn't post afterwards. I also tried the "Memroy try It!" function to get it to run at 3200 (no go) and then 3000 (also didn't work).

My motherboard has a second CPU power socket that, according to the manual, is for "exrtreme overlockers" so I never used it, I don't even think I have a 4-pin cable for it. I'm thinking that maybe it is a power issue using just the one CPU cable?

I am curious if anyone else has had this issue, and if so, what can be done about it? I'm not a heavy user, but I bought 3200mhz RAM and would like to be able to run it at that speed.

My System:
MB: MSI MPG z390 gaming pro carbon
CPU: i5 9600k
RAM: Corsair Vengeance RGB Pro 3200mhz
PSU: Corsair RM1000i
GPU: MSI 1660ti Armor OC 6G

Thanks in advance for the help!
 
Aug 29, 2019
26
3
45
Drop the automatic cpu overclocking
Try the cpu at stock and just set the ram to xmp
I am 99% sure the root of your problem is the automatic overclocking
Thanks for the reply. I'll try it out tonight!
Do you have any idea what would cause it to stop working like that though, after months of using them both at the same with no apparent issue? Or could that potentially be what killed off my first set of RAM?
 

germanium

Distinguished
Jul 8, 2006
93
9
18,565
If you remove power from the computer or turn the power supply switch off in the back of the computer the RAM will need to be retained & this will in some cases of a high overclock on ram to not post initially & may reboot multiple times before it is able to post. lesson here is not to unplug your computer of turn the power supply off in the back if at all possible.

Try waiting through that process. Power may have went out overnight that causing it to have to retrain the RAM. All it takes is a few seconds of power loss to cause this but once it posts once it will continue to do so until you get another power outage.

Getting yourself an uninterruptable power backup will likely completely avoid this issue except for long power outages that is.
 
Last edited:
Aug 29, 2019
26
3
45
If you remove power from the computer or turn the power supply switch off in the back of the computer the RAM will need to be retained & this will in some cases of a high overclock on ram to not post initially & may reboot multiple times before it is able to post. lesson here is not to unplug your computer of turn the power supply off in the back if at all possible.

Try waiting through that process. Power may have went out overnight that causing it to have to retrain the RAM. All it takes is a few seconds of power loss to cause this but once it posts once it will continue to do so until you get another power outage.

Getting yourself an uninterruptable power backup will likely completely avoid this issue.
Interesting, I didn't know that... So you're saying that if, for whatever reason, my computer won't post (i.e. spins fans and lights up for a few seconds before shutting off and repeating this process) I should just wait it out until it POSTs? I'd gotten into a habit of having to reset the CMOS every time it did this...
 

Karadjgne

Titan
Ambassador
Ddr4 has no issues at 1.5v.
You have 2 issues.
First is unstable ram OC. It's only 99% stable, and that last 1% is somewhere other than what's normally used, so never gets noticed. Except on boot.
Second is the use of Game boost. That doesn't just change your cpu speeds, it messes with a lot of settings you don't even see in bios, including the usb, pcie and ram transmission speeds and timings and latencies. It's not a program written specifically for your specific motherboard and cpu, but a general program that cannot think or make decisions on the fly. It sets specific target changes and these are not always beneficial. Every cpu, gpu, ram has slightly different needs, that program is written to cover the basics, but not specifics. It's a seriously bad idea all around and nothing much more than a sales gimmick.

Like in a new car lot, having a tan car sit next to a red car, and the red car has a sticker on the windshield "Fast!" and the tan car has "Great gas milage!" (but really both cars are identical except the paint)

Game boost, OC Genie, Turbo V, all the same, different names, garbage.

Turning off Game Boost isn't easy as a switch. You'll need to reset the bios to optimal factory default settings, turn the pc off, reset cmos. Otherwise some of those 'additional changes' won't get reset, just continued and you'll still have instability at boot until bios reverts to its 'last stable settings', boots, loads the remnants and orphans of game boost, and starts all over with the hassles.
 
Last edited:

germanium

Distinguished
Jul 8, 2006
93
9
18,565
DDR4 may not have issues with 1.5 volts but I have read that Intel's memory controller may have issues with 1.5 volts as some people have claimed ruined memory controllers on Core I9 9900K processors from running memory at 1.5 volts
 

Karadjgne

Titan
Ambassador
? Ram voltage shouldn't affect the memory controller. Ram voltage is the voltage actually going to the physical slot.

Now if they bumped the vccio/system agent voltage (VDDQ) upto 1.5v, then yes, I'd expect issues since the memory controller shouldn't see anything above @ 1.1v - 1.2v

Legit Reviews contacted Intel about the safe voltage range on DDR4 memory and we received this response.

“1.5v is the absolute max we allow for XMP certifications. However, good DDR4 memory will run at 1.35v up to 3200. Technically, no “safe” (guaranteed) OC over-voltage but 1.35v or lower is best.” – Intel

Jedec states DDR4 voltage specs are 1.2v to 1.4v, however that doesn't take user OC into consideration. Those voltages are to safeguard the ram and slot and whatever is powering them as far as mosfets go, but the only thing the memory controller gets is data in channels, it doesn't also get voltage from the ram slot. It gets its voltage through the cpu/motherboard.
 
Last edited:

germanium

Distinguished
Jul 8, 2006
93
9
18,565
? Ram voltage shouldn't affect the memory controller. Ram voltage is the voltage actually going to the physical slot.

Now if they bumped the vccio/system agent voltage (VDDQ) upto 1.5v, then yes, I'd expect issues since the memory controller shouldn't see anything above @ 1.1v - 1.2v



Jedec states DDR4 voltage specs are 1.2v to 1.4v, however that doesn't take user OC into consideration. Those voltages are to safeguard the ram and slot and whatever is powering them as far as mosfets go, but the only thing the memory controller gets is data in channels, it doesn't also get voltage from the ram slot. It gets its voltage through the cpu/motherboard.

Ram voltages can in fact cause damage to CPU if excessive. All Nehalem processor compatible motherboards came with such a warning not to go over 1.65volts on the RAM. This was with DDR3 ram which ran at higher voltages than DDR4.