91888 errors - erm....are these genuine?

[drsp123]

So I ran memtest on a 2x32 GB kit and it seems every single address is bad - which seems odd. Everything seems OK when running the PC, but both Memtest and Windows memory diagnostic seem to think there is a HW issue. Report below:

I don't really understand the detail - but it looks like the failures are related to the actual results including a prefix for the memory address - am i doing something wrong in the config?

Summary

Report Date	2021-01-01 15:50:38
Generated by	MemTest86 V8.4 Free (64-bit)
Result	FAIL

System Information

EFI Specifications	2.70
System
Manufacturer	ASUS
Product Name	System Product Name
Version	System Version
Serial Number	System Serial Number
BIOS
Vendor	American Megatrends Inc.
Version	1401
Release Date	12/03/2020
Baseboard
Manufacturer	ASUSTeK COMPUTER INC.
Product Name	ROG STRIX B550-E GAMING
Version	Rev X.0x
Serial Number	200873128502886
CPU Type	AMD Ryzen 9 5900X 12-Core
CPU Clock	3693 MHz
# Logical Processors	24 (12 enabled for testing)
L1 Cache	24 x 64K (288952 MB/s)
L2 Cache	24 x 512K (118924 MB/s)
L3 Cache	1 x 65536K (40978 MB/s)
Memory	65465M (37229 MB/s)
DIMM Slot #0	32GB DDR4 XMP PC4-28800
	G Skill Intl / F4-3600C16-32GTZN
	16-22-22-42 / 3602 MHz / 1.450V
DIMM Slot #1	32GB DDR4 XMP PC4-28800
	G Skill Intl / F4-3600C16-32GTZN
	16-22-22-42 / 3602 MHz / 1.450V

Result summary

Test Start Time	2021-01-01 15:44:13
Elapsed Time	0:05:53
Memory Range Tested	0x0 - 1030000000 (66304MB)
CPU Selection Mode	Parallel (All CPUs)
ECC Polling	Enabled
# Tests Passed	1/1 (100%)

Lowest Error Address	0x50028100 (1280MB)
Highest Error Address	0x5300FEB8 (1328MB)
Bits in Error Mask	000000FF00000000
Bits in Error	8
Max Contiguous Errors	2

Test	# Tests Passed	Errors
Test 0 [Address test, walking ones, 1 CPU]	1/1 (100%)	0
Test 1 [Address test, own address, 1 CPU]	0/0 (0%)	91888
Test 2 [Address test, own address]	0/0 (0%)	0
Test 3 [Moving inversions, ones & zeroes]	0/0 (0%)	0
Test 4 [Moving inversions, 8-bit pattern]	0/0 (0%)	0
Test 5 [Moving inversions, random pattern]	0/0 (0%)	0
Test 6 [Block move, 64-byte blocks]	0/0 (0%)	0
Test 7 [Moving inversions, 32-bit pattern]	0/0 (0%)	0
Test 8 [Random number sequence]	0/0 (0%)	0
Test 9 [Modulo 20, ones & zeros]	0/0 (0%)	0
Test 10 [Bit fade test, 2 patterns, 1 CPU]	0/0 (0%)	0
Test 13 [Hammer test]	0/0 (0%)	0

Last 10 Errors

2021-01-01 15:50:06 - [Data Error] Test: 1, CPU: 0, Address: 5300FEB8, Expected: 000000005300FEB8, Actual: 000000235300FEB8

2021-01-01 15:50:06 - [Data Error] Test: 1, CPU: 0, Address: 5300FEB0, Expected: 000000005300FEB0, Actual: 000000865300FEB0

2021-01-01 15:50:06 - [Data Error] Test: 1, CPU: 0, Address: 5300FEA8, Expected: 000000005300FEA8, Actual: 000000E85300FEA8

2021-01-01 15:50:06 - [Data Error] Test: 1, CPU: 0, Address: 5300FEA0, Expected: 000000005300FEA0, Actual: 0000000D5300FEA0

2021-01-01 15:50:06 - [Data Error] Test: 1, CPU: 0, Address: 5300FE98, Expected: 000000005300FE98, Actual: 000000C75300FE98

2021-01-01 15:50:06 - [Data Error] Test: 1, CPU: 0, Address: 5300FE90, Expected: 000000005300FE90, Actual: 000000625300FE90

2021-01-01 15:50:06 - [Data Error] Test: 1, CPU: 0, Address: 5300FE88, Expected: 000000005300FE88, Actual: 000000355300FE88

2021-01-01 15:50:06 - [Data Error] Test: 1, CPU: 0, Address: 5300FE80, Expected: 000000005300FE80, Actual: 000000A05300FE80

2021-01-01 15:50:06 - [Data Error] Test: 1, CPU: 0, Address: 5300FE78, Expected: 000000005300FE78, Actual: 0000006F5300FE78

2021-01-01 15:50:06 - [Data Error] Test: 1, CPU: 0, Address: 5300FE70, Expected: 000000005300FE70, Actual: 000000025300FE70

Many thanks!

[David (PassMark)]

Something is wrong. But it isn't a typical memory failure. It is strange to have so many errors in test 1 and so many bits in error (0xFF, which is 8 bits).
Might be interesting to test 1 memory stick at a time.

So maybe it is a BIOS bug. Or maybe some poorly setup hardware device (e.g. PCIe network card, etc..) that is using some memory addresses, but BIOS doesn't notice they are in use. So the same memory addresses get used twice, leading to MemTest86 and the hardware device corrupting each other's data.

[drsp123]

Thanks - I ran for the modules individually using the same slot - the first DIMM ran fine with no errors, though the second module showed the same error behaviour as before (though generated twice the number of errors of around 183k errors)......The only other devices I have connected are one graphics card (RTX3080 to CPU) and a NVME SSD (chipset) plus some SATA drives and USB devices. Any ideas on where to go from here?

[David (PassMark)]

I assume when testing individual stocks you were testing them in the same slot?

So maybe the RAM is bad and it is an atypical failure.

[drsp123]

Thanks, yes using the same slot. I guess it's time to start the RMA process. Thanks for your help, much appreciated.