One single error in test #13 "hammer test"

[Smooph]

Hi there,

yesterday I upgraded my (new) mid-2012 MacBook Pro non Retina with two CT2K8G3S160BM Crucial 8GB DDR3L 1600Mhz Sticks.

To be sure that they work properly, I ran memtest86 6.0 free - one pass all tests.

All tests ran flawlessly - except test #13 "hammer test".

There was one single one-bit-error.

Should I be concerned now? Should I run several (2?, 4?) passes?

I've also contacted Crucial to ask about their experience with the Rowhammer-Issue.

Thank you very much for your assistance!


Best regards


Simon

[David (PassMark)]

See this page for details.
http://www.memtest86.com/troubleshooting.htm

There is a special section on row hammer errors.

[Smooph]

Hi David,

thank you for the link.

I understand now what's happening during the "Row Hammer Test".

What I don't understand is, why it's only one single bit? If the physical layout causing these errors, there would still be more than one?

Or am I completely wrong?

In another Slideshow from Yoongu Kim there is talk of millions of errors, which also seems logic to me.

Sorry for the stupid question, but maybe I have a error in reasoning...

[David (PassMark)]

The test isn't perfect for all CPU types. The efficiency of finding faults depends somewhat on the memory layout (for example the use of dual or quad channel, or not), clock speeds and timings.

So it is possible the test isn't optimal for your setup, but there is also the possible there is just one weak cell in your RAM.