Weird motherboard XMP issue

Joined
Nov 5, 2018
Messages
20
So, I just built a new machine using the z390 aorus master. So far, everything is excellent. I haven't gotten into testing overclocking yet, but I plan to soon.

I've now noticed a quirk x2 though, which makes me think it might not go away, so I wanted to check in with you all who know more than me. Twice now, when booting from cold, I get a post code of C4 (which doesn't actually have anything associated with it) and the machine won't boot. I cleared CMOS and no help. The first time this happened, it was shortly after enabling XMP, so I removed one stick of ram, and the machine booted. Put the stick back in, and the machine booted. Enabled XMP, and the machine booted. Ran Prime95 blender for 1-2 hours, no issues. Gamed for a few hours, no issues. Turned my computer off, went to bed, this morning, same thing, with the same solution.

The fact that once I boot, everything goes well, including with XMP, it makes me think this is related to the board moreso than the RAM. The only additional observation I have, and I apologize if this shows ignorance, is that the ram I have is listed at:
DDR4 3200 (PC4 25600)
Timing 16-18-18-38
CAS Latency 16
Voltage 1.35V

Currently, it's running those timings, but it looks like the voltage is 1.2V. Could this be the problem?
 
Yes, most certainly.
If the memory spec is 1.35v, then bump it up to that voltage.

Also check for BIOS updates that may help with memory issues.

.
 
Nice that it hasn't failed again, but you can absolutely count on more future failures until that memory voltage is corrected.

.
 
I'm surprised that the XMP doesn't set the voltage as well as the timings.

Occasionally, I've had to set memory voltage to something like 1.37V to get memory stable at rated speeds (more so with AMD systems with older AGESA code). If it doesn't run at 1.35V, RMA the memory.
 
more voltage needed isn't a reason to rma, the replacement set will most like do the same thing. if bumping the voltage gets it stable, leave it at that. ive been running corsair vengeance ddr3 with extra voltage for years without an issue.
 
more voltage needed isn't a reason to rma, the replacement set will most like do the same thing. if bumping the voltage gets it stable, leave it at that. ive been running corsair vengeance ddr3 with extra voltage for years without an issue.

I guess it depends on how much extra voltage is needed. In my example of 1.37V...I probably wouldn't RMA. If I needed 1.4V+ I probably would.
 
Had a similar problem which turned out to be one or more of the DIMMS was not inserted completely despite all apearances of being so and both retaining clips being locked. Some can be really tight and require more pressure than normal to make full contact.
 
The ops problem is most likely the memory voltage being too low as already stated.

But I had some terrible issues with Corsair LPX 3000 CL15 1.35V 4x4GB that turned out to be down to the motherboards.
This is sold as a strong overclocking ram yet I couldnt get it to run at 3000MHz on my first 2 motherboards, it was only fully stable @ 2500MHz.
They were cheaper gaming motherboards.
As soon as it was put it in a higher quality board it clocked like a demon.
Its been running @ 3733MHz CL17 1.44V for well over a year.
 
Back
Top