EVGA X79 Dark + XMP Profile

I used memory part number CML32GX3M4A1600C10 successfully on an EVGA X79 SLI motherboard at 1600 with no problems. I have upgraded to an EVGA X79 Dark so when I use the XMP profile and save the changes, boot the PC and then shutdown I can start up the PC but I get nothing on the monitor, no BIOs beep and it doesn't boot. If I use the 2nd XMP profile which seems to set everything to automatic except for memory speed the same problem occurs (if I set the speed to 1600). I have to reset the BIOs back to default in order to boot which seems to set the memory speed at 1333. Here is a screenshot of the XMP profile number 1. If anyone sees anything wrong just let me know:

 

http://i516.photobucket.com/albums/u321/robert3892/130924202920_zpsd5dc2a77.jpg

Well Corsair tech support is not answering my query and nothing has been posted here. Any ideas anyone?

Hi Robert, I would suggest calling CS by phone if you can. That would be the easiest way to get a hold of them.

 

All I can say is sometimes you get components that just don't like each other for some reason. There is no real reason why they shouldn't run on that board. I'm at a loss.

 

If you want you can have them replaced and see if a second kit would work. Just use the link on the left to request an RMA

Well first I want to see what Corsair and EVGA have to say. I don't have the time or money to pull my kit waiting for a replacement. I don't think the kit is defect as this same kit ran perfectly with the Corsair recommended settings on my previous X79 EVGA SLI. I think the kit is simply incompatible with the X79 Dark or the X79 Dark needs a BIOs update for the memory

I wasn't saying it is defective, just that sometimes components do clash, even though they should be 100% compatible. A BIOS update from EVGA could very well fix this but how long are you willing to wait for that to happen if it even does.

 

You don't have to be with out your memory to have it replaced. Advanced replacements are available so you will only be down the time it takes to change the memory.

 

If that is something you want to do,m then just use the link on the left to request your RMA.

 

Have you by any chance tried clearing the CMOS before using the memory on the new board? If not it may be worth trying.

If I make any memory change in the BIOs the PC will not boot so I have to reset the BIOs to default in order for the PC to boot.

By the way calling CS from Europe isn't cheap

I'm Experiencing the exact same issue now on my new system :(

I'm Experiencing the exact same issue now on my new system :(

 

Do you have the same memory and you're using the X79 dark?

Do you have the same memory and you're using the X79 dark?

Yes,

i got it fixed by setting the dimm voltage at 1.56 instead of the standard value, and the last bios. I know it's not maybe the best solution and a fix should be find for that problem, but for now it works, at least for me. Now it works perfectly also from a cold boot.

Robert, are you still troubleshooting this? I was curious to see what happened so far.

I think I will try forum member's Xeonn's suggestion

What version of XMP is on your modules and what version is your MOBO compatible with?

In the OP's case YB, Both boards are XMP 1.3 ready and the modules show as XMP v1.3 in the pic of the BIOS in the OP. What confuses me and why I didn't suggest just upping the DIMM voltage was the lack of a video signal. That usually doesn't happen. Especially with a CPU with native 1600mhz support.

 

I would have to think it's a combination of needing a BIOS update along with poor voltage regulation on the board as per Xeonns posts that in his case setting DIMM voltage to 1.56 cued his issuue. But he had a video signal beefore hand.

 

Xeonn posted:

i got it fixed by setting the dimm voltage at 1.56 instead of the standard value, and the last bios. I know it's not maybe the best solution and a fix should be find for that problem,

That is actually totally the best bethod. Some boards no matter the quality just don't regulate voltage correctly or show correct values in the BIOS. Even though you set the voltage to 1.56v in the BIOS i would bet money that it is actually closer to 1.5v when tested with a DMM. As long as you system is stable you should be good to go. Just need to figure out why roberts having issues.:)

Can you verify VDIMM with a meter?

Unfortunately I don't have a meter to test this but I'd like everyone to know that setting the voltage to 1.56 as per Xeonn's advice seemed to have fixed this issue. I relayed this info to Jacob at EVGA. I can only assume the voltage reading at XMP auto is not good enough for these modules

I can only assume the voltage reading at XMP auto is not good enough for these modules

No, thats not the case. The only thing that changed was your MB. And as I ststed earlier some boards either don't regulate the voltage correctly, or the BIOS reading are wrong. Which is ALWAYS a possibility. I'm sure if you were able to test VDIMM with a DMM you would find out that the actual volatge is closer to the default values.

 

The other thing you need to consider is that your applying a slight overclock. And again, depending on your MB having to add a bit of voltage is not uncommon to compensate for that overclock, despite what the modules are rated for. .

 

While XMP is supposed to be a one click overclocking setting, sometimes it just doesn't work out because of the other factors I mentioned above.

 

In other words your results are completely within what would be called "normal".

I struggled with the same issue. This board is really finicky or perhaps the bios is not mature enough. The problem was voltage. My memory is spec'd to run at 1866 1.5v CL9.

However, I kept getting B7 errors on post (memory training) Although I could boot with each stick individually and with 2 sticks in the first two slots, the error would pop up with three or mor3 sticks. 1.6 v for the memory didnt even do it for me. At 1.65 though I am at 1866 mhz and 10,10,10,27. CR 2T using all 4 sticks. I havent played around much because this board spikes voltage to 1.6 vcore on boot before it settles down to 1.3. At that voltage I am Intel Burn test stable at 4.5 ghz at 70 Degrees. 4.6 ghz is impossible though.

 

So, What are the results people have seen with this memory? Will 1.65 kill it?

Actually your results aren't uncommon. The sticks are tested @ 1.5v , but that doesn't mean they won't require additional tweaks depending on the system. Your results of having to add voltage as you add sticks is just compensating for the extra load on the memory controller. And this very well could be your board limiting them too.

 

Your running a fairly high OC with a mild memory oc ...your volatges are on par for sure.

 

And No, 1.65v won't kill them. They are good to 2.0v before you could potentially damage them although over 2.0v voids your warranty.

 

As long as you are stable i wouldn't worry about it.

upping the vccio to 1.1v allowed me to go to CL 9. With a 10% increase in mem read/write speeds as well as a drop in latency from ~100 to 70.

seems im a bit too late to say that such aggressive timings and such a low voltage to the memory would be my first guess

 

 

im glad top see that others are able to point out the same thing :)