Test 7 / CPU 6 / 0x159c4xxxx failure with 2x 8GB stops with 1x 8GB

[jamesetaylor]

I have a failure in test 7 on cpu 6 that happens consistently in the same bit in the same address range.

Here are the tests:

1. Original config (8Gb in DIMMA2 (stick A), 8Gb in DIMMB2 (stick B). Error.
2. Limit address range of test. No error.
3. Limit test to CPU 6. No error.
4. Switch stick B into DIMMA2; remove stick A. No error.
5. Add stick A into DIMMB2; Error present with same addresses, CPU etc (despite the sticks being in different slots now).

I'll be trying a few more options (I have an identical machine I can swap components with) but does what I'm seeing point at the CPU? I am surprised that switching sticks in their sockets would yield errors in the same addresses and I would expect a memory issue at a particular address to happen across CPUs.

System details:

Intel Core i5 10600
MSI MAG B460M MORTAR WIFI
16GB (2x8GB) Corsair DDR4 Vengeance LPX Black, PC4-21300 (2666), Non-ECC Unbuffered, CAS 16-18-18-35, XMP 2.0, 1.2V

Any ideas would be gratefully received!

Cheers,
James

[David (PassMark)]

The mapping of RAM addresses to slots / sticks is complex.
There are hardware features such as multiple channels, interleaving, scrambling, Channel Hashing, NUMA & XOR schemes make it nearly impossible (for an arbitrary CPU & RAM stick) to know which memory addresses correspond to which rows in the RAM

It is likely bad RAM.
See,
https://www.memtest86.com/troubleshooting.htm