2500k not stable with RAM at 1600

trents · Apr 13, 2014

Just got my 2500k and having instability when trying to run the GSKill Ripjaws 1600 at 1600 with 11 11 11 28 1T timings which are recommended by the JEDEC tables. And these are the non XMT timings. Only completes two lines of the IntelBurnTest wihen on 1600 (high stress setting).

Is there a component in Intel architecture that is the equivalent of the CPU/NB in AMD that allows you to add more juice just to the IMC?

Or does it really make any real performance difference with Intel CPUs what frequency the memory is running when you aren't concerned about the IGP performance? I read conflicting things about this.

ATMINSIDE · Apr 13, 2014

Is this pushing 2 or 4 DIMMs?

What motherboard?

trents · Apr 13, 2014

Two DIMMS. With Latencies in the 11 range at 1600 it might be better from a performance standpoint to go with 1333 and lower latencies anyway.

trents · Apr 13, 2014

Motherboard is Asus P8Z68-VLX.

redduc900 · Apr 13, 2014

What model are the modules, and does Memtest86+ pass at their rated timings and voltage?

ATMINSIDE · Apr 13, 2014

I'm assuming these are 1.5V modules?
Try running 2T instead of 1T. If that doesn't work try ~0.05V higher than rated.

trents · Apr 13, 2014

The modules ran a long time at 1600+ in my AMD system without any issues at 1.55v and XMP-1600 profile timings. Dual channel kit. F3-12800CL9D-8GBRL. I haven't run memtest on them but I was recently using them in another system w/o any issues. Haven't tried 2T at 1600 and CL11 yet. I did try upping the voltage on the present Intel system to 1.585 on the RAM to see if it would help and it didn't. Seems to be an ICM issue. Right now I am running them at 1333 8 8 8 22 1T in the intel system at 1.55 volts and they are stable.

ShrimpBrime · Apr 13, 2014

Those sticks are recommended 2T with Cas 9 latency and 1.5v. You shouldn't have any issues with this.

DDR3 1600 (PC3 12800)
Timing 9-9-9-24-2N
Cas Latency 9
Voltage 1.5V

If you do, the IMC just doesn't like that kit of ram. it happens.

I'd try for 8-8-8-24-32-1/2T see what happens.

ShrimpBrime · Apr 13, 2014

Oh wanted to mention I have a hard time with OCZ 1600mhz with the i7. Dang thing just isn't stable. No blue screens, it black screen restarts, even with one stick. Weird as heck. Tested each stick quickly on the AMD rig with no issues...

trents · Apr 13, 2014

ShrimpBrime said:
Those sticks are recommended 2T with Cas 9 latency and 1.5v. You shouldn't have any issues with this.

My sentiments exactly!

Is there any way to supplement the IMC voltage?

trents · Apr 13, 2014

ShrimpBrime said:
I'd try for 8-8-8-24-32-1/2T see what happens.

No dice.

ShrimpBrime · Apr 13, 2014

trents said:
My sentiments exactly!

Is there any way to supplement the IMC voltage?

From what I read this isn't always safe. Neither is OVing the ram either.

If it was mine, I'd get something that it'll agree with. Cas9 is cool, but Cas 11 should have you netting some 1100mhz effective imo.

trents · Apr 13, 2014

Well, I've got some Team Vulcan 2400 ram in my AMD machine. Maybe I'll see if it will run at 1600 in the Intel.

trents · Apr 13, 2014

Overvolting the RAM is not safe? You mean for the RAM?

ShrimpBrime · Apr 13, 2014

trents said:
Overvolting the RAM is not safe? You mean for the RAM?

For the IMC and the Ram.

ShrimpBrime · Apr 13, 2014

Hmm. lemme put it this way.

Intel is pretty much set it and forget it.

You buy a quad stick kit designed for Intel and run 1200mhz effective, either the IMC can handle it or it won't.

There's no FSB adjustments so you pretty much get what you get from the list.

If you think the IMC can handle 2666 memory, that's what you purchase.

I've yet to find any average AMD/INTEL chip to run much more than 1100mhz effective daily and that includes an overvolt, slightly.

The fast kits get expensive....

trents · Apr 13, 2014

But the voltage to the RAM isn't going through the IMC is it? Right now I'm running the RAM at 1333 with 8 8 8i 22 1T timings which is a little more aggressive than the JEDEC would suggest and that's why I'm giving it 1.575v instead of 1.5v. Just can't see how that could damage the IMC.

EarthDog · Apr 13, 2014

If you have yet to see it, shrimp, you haven't been looking. From sandybridge on, I habent had one not do 2200+...2400+ really.

Trents, 1.575v won't remotely hurt a thing. What he read was voltage over 1.65 and still plenty of people run over that 24/7 and especially for benching.

So, just setting the xmp profile isn't working?

ATMINSIDE · Apr 13, 2014

Try 1600 9-9-9-24-2T 1.5V

I think its an issue with the 1T command rate.

trents · Apr 13, 2014

ATMINSIDE said:
Try 1600 9-9-9-24-2T 1.5V

I think its an issue with the 1T command rate.

Nope. Won't even boot all the way into Windows. Now Windows wants to do a repair.

2500k not stable with RAM at 1600

Senior Member

Sim Racing Aficionado Co-Owner

Senior Member

Senior Member

Inactive Moderator

Sim Racing Aficionado Co-Owner

Senior Member

~MadHatDeLidder~

~MadHatDeLidder~

Senior Member

Senior Member

~MadHatDeLidder~

Senior Member

Senior Member

~MadHatDeLidder~

~MadHatDeLidder~

Senior Member

Gulper Nozzle Co-Owner

Sim Racing Aficionado Co-Owner

Senior Member

Similar threads