Memory issues with 32GB on z87 and z97

[HapticBovine]

Hi all,

 

I am relatively new to OS X and hackintoshing. I have a peculiar problem with my kit of Corsair memory (4x8GB, 1866 MHz).

 

I had a system installed with <that which shall not be named>, and it started locking up. That was on my Gigabyte z87x-OC board, with a 4770k. I had weird glitches in Linux as well--firefox crashing, and kernel panics when I made a large ramdisk. This all occurred at stock speeds with my memory running at 1333 MHz.

 

I read about Gigabyte apparently having problems delivering power to all 4 modules under certain circumstances. I shelled out for a new board (not a total waste, because I'm trimming down the build platform to mATX), a maximus vii gene (z97).

With further research it seems I'm seeing the same behaviour. I can run memtest86+ with any combination of modules, it seems, and can't provoke any memory errors unless all 4 modules are installed at the same time. I have noted this both at 1333 MHz and the XMP profile at 1866 MHz. I have bumped my System Agent voltage (VCCSA) to ~1.3v but that hasn't seemed to help.
 

I am not trying to get support with the tonymac stuff per se, but I thought I would reach out here and see if I could learn more about these "0xdeadbeef" errors I keep getting (now I can get to password entry after install, then kernel panic).

From what I have read of other Unices, 0xdeadbeef is used to fill freed memory regions. When new memory is allocated, the kernel can check the values first; values filled with "0xdeadbeef" are safe to allocate, while other values are not; it means certain bugs such as the same memory being allocated twice may be occurring.

 

In the most general sense, the error indicates memory corruption. Of course, one common source of memory corruption is hardware failure.

 

The funny thing is another user on <you know where> also reported panics with the error "expected 0xdeadbeefdeadbeef but found 0xdeadbeef88098809". That leads me to believe the error is perhaps not due to faulty memory, but either overall system instability, or perhaps a hardware-dependent software (or configuration) bug.

 

What I am trying to ascertain is, how common are these memory errors? Do a lot of users have these problems with 32GB installed? How can I determine when I am dealing with a defective part, or when there is more work to do to stabilize my system?

 

[Mettiu]

I can't tell how common these problems are, although I had a similar experience with RAM related problems on a Gigabyte motherboard as well.

In my case the motherboard is a ga-z68mx-ud2h-b3, which RAM capacity I maxed out with four Patriot RAM sticks.

At first I tried to over volt the RAM, up to 1.64V, without results.

At the end, I "solved" the problem lowering the RAM timings, from the original 9-9-9-24 (for which those modules are certified), to 11-11-11-30.

 

I hope a way or the other can help you as well!