Question Thoughts on this nemix ram?

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Soulkeeper

Diamond Member
Nov 23, 2001
6,712
142
106
I noticed this on the nemix website after seeing the 2933 on newegg.
Apparently it has hynix mem chips in a 9 (cause of ecc) on each side configuration.
It claims to be 2Rx8 just like my 16GB with hynix chips and the same 9 chips per side.

The odd thing is the chips in the picture clearly say "AFR", I'm guessing it's the identical chips and manufacturing process ?
Maybe each chip has 2 dies in them, and just a higher speed binning ?
At first I thought maybe the picture was wrong, and it would have 19 chips per side...
This appears to be it, just no mention of "AFR" ... D-die ?

This is the first time i've seen 32GB dimms in this configuration for ecc udimm that are readily available.
These would be perfect for a ryzen build, and cost 1/2 the price of the hard to find samsung dimms.
Still 21nm hynix stuff ?

Here are some PDFs i've collected on the subject
 
Last edited:

Soulkeeper

Diamond Member
Nov 23, 2001
6,712
142
106
I'd contact the motherboard maker. Maybe they didn't add support for ecc.
Maybe boot linux and run edac-utils -vr

Code:
edac-util -vr
mc0: 0 Uncorrected Errors with no DIMM info
mc0: 0 Corrected Errors with no DIMM info
mc0: csrow0: 0 Uncorrected Errors
mc0: csrow0: mc#0csrow#0channel#0: 0 Corrected Errors
mc0: csrow0: mc#0csrow#0channel#1: 0 Corrected Errors
mc0: csrow1: 0 Uncorrected Errors
mc0: csrow1: mc#0csrow#1channel#0: 0 Corrected Errors
mc0: csrow1: mc#0csrow#1channel#1: 0 Corrected Errors
mc0: csrow2: 0 Uncorrected Errors
mc0: csrow2: mc#0csrow#2channel#0: 0 Corrected Errors
mc0: csrow2: mc#0csrow#2channel#1: 0 Corrected Errors
mc0: csrow3: 0 Uncorrected Errors
mc0: csrow3: mc#0csrow#3channel#0: 0 Corrected Errors
mc0: csrow3: mc#0csrow#3channel#1: 0 Corrected Errors
edac-util: No errors to report.

You'd also see something like this in dmesg:
EDAC amd64: Node 0: DRAM ECC enabled.
 
Last edited:

Ryan Koller

Junior Member
Apr 26, 2017
7
3
51
The live linux distro I used didn't have edac-utils but dmesg did show the "EDAC amd64: Node 0: DRAM ECC enabled." I ran through all my bios settings again and the only ECC options I have are on. I decided to run memtest86 again to see what could be found there. The front info page still shows ECC as "no". When I inspect the individual modules, memtest does show them as ECC.

Everything I've done confirms that ECC is indeed on and working. Maybe I do need memtest86 pro paid version to get it to test. I've done over three passes and no errors and no ECC correction events have been shown.

I'm going to put proxmox on this machine as the base operating system. I'll get edac-util and check for errors as well as try to check the logs. I'll look for a memory stress test in linux to try to ferret out of the ram is operating without any base errors that error correction could be hiding.

So far I think the memory is good, but after your experience I want to know for sure.
 

Vincent

Platinum Member
Oct 9, 1999
2,030
2
81
I ordered four of the shopblt.com Crucial DDR4-3200 EUDIMMs found by SoulKeeper (thanks!). Two arrived a week ago and the remaining two are arriving next week. They seem to be working fine (ASRock X470 Taichi, Ryzen 5 3600) as far as I can tell.

Code:
$ dmesg | grep EDAC
[    0.813496] EDAC MC: Ver: 3.0.0
[    9.978560] EDAC amd64: Node 0: DRAM ECC enabled.
[    9.978561] EDAC amd64: F17h_M70h detected (node 0).
[    9.978591] EDAC MC: UMC0 chip selects:
[    9.978591] EDAC amd64: MC: 0:     0MB 1:     0MB
[    9.978592] EDAC amd64: MC: 2: 16384MB 3: 16384MB
[    9.978594] EDAC MC: UMC1 chip selects:
[    9.978594] EDAC amd64: MC: 0:     0MB 1:     0MB
[    9.978595] EDAC amd64: MC: 2: 16384MB 3: 16384MB
[    9.978595] EDAC amd64: using x16 syndromes.
[    9.978595] EDAC amd64: MCT channel count: 2
[    9.978624] EDAC MC0: Giving out device to module amd64_edac controller F17h_M70h: DEV 0000:00:18.3 (INTERRUPT)
[    9.978629] EDAC PCI0: Giving out device to module amd64_edac controller EDAC PCI controller: DEV 0000:00:18.0 (POLLED)
[    9.978630] AMD64 EDAC driver v3.5.0

$ sudo dmidecode -t memory
# dmidecode 3.1
Getting SMBIOS data from sysfs.
SMBIOS 3.2.1 present.
# SMBIOS implementations newer than version 3.1.1 are not
# fully supported by this version of dmidecode.

Handle 0x000E, DMI type 16, 23 bytes
Physical Memory Array
    Location: System Board Or Motherboard
    Use: System Memory
    Error Correction Type: Multi-bit ECC
    Maximum Capacity: 128 GB
    Error Information Handle: 0x000D
    Number Of Devices: 4

Handle 0x0016, DMI type 17, 40 bytes
Memory Device
...

Handle 0x0018, DMI type 17, 40 bytes
Memory Device
    Array Handle: 0x000E
    Error Information Handle: 0x0017
    Total Width: 128 bits
    Data Width: 64 bits
    Size: 32 GB
    Form Factor: DIMM
    Set: None
    Locator: DIMM 1
    Bank Locator: P0 CHANNEL A
    Type: DDR4
    Type Detail: Synchronous Unbuffered (Unregistered)
    Speed: 3200 MT/s
    Manufacturer: Micron Technology
    Serial Number: XXXXX
    Asset Tag: Not Specified
    Part Number: 18ASF4G72AZ-3G2B1  
    Rank: 2
    Configured Clock Speed: 3200 MT/s
    Minimum Voltage: 1.2 V
    Maximum Voltage: 1.2 V
    Configured Voltage: 1.2 V

Handle 0x001B, DMI type 17, 40 bytes
Memory Device
...

Handle 0x001D, DMI type 17, 40 bytes
Memory Device
    Array Handle: 0x000E
    Error Information Handle: 0x001C
    Total Width: 128 bits
    Data Width: 64 bits
    Size: 32 GB
    Form Factor: DIMM
    Set: None
    Locator: DIMM 1
    Bank Locator: P0 CHANNEL B
    Type: DDR4
    Type Detail: Synchronous Unbuffered (Unregistered)
    Speed: 3200 MT/s
    Manufacturer: Micron Technology
    Serial Number: XXXXXX
    Asset Tag: Not Specified
    Part Number: 18ASF4G72AZ-3G2B1  
    Rank: 2
    Configured Clock Speed: 3200 MT/s
    Minimum Voltage: 1.2 V
    Maximum Voltage: 1.2 V
    Configured Voltage: 1.2 V

$ edac-util -v
mc0: 0 Uncorrected Errors with no DIMM info
mc0: 0 Corrected Errors with no DIMM info
edac-util: No errors to report.
 

Soulkeeper

Diamond Member
Nov 23, 2001
6,712
142
106

These seem to be the best settings i'm gonna get without going insane on the voltage.
Slightly different subtimings than the nemix, but generally similar.

Interesting note that might help someone. The AddrCmdSetup under CAD bus setup timings needed to be manually set or the system wouldn't post. 1 was needed to post at 3200Mhz 2 for 3400MHz.
 
Last edited:

Soulkeeper

Diamond Member
Nov 23, 2001
6,712
142
106
Just a heads up to anyone considering the nemix or crucial.
I realized that the chip markings were different.

Crucial/Micron ones have Rev.B, the Nemix had Rev.E
D9XPF vs. D9ZFV chip markings

Which seems odd to me because the Crucial ones were "newer" direct from manufacturer ordered, newer date on the SPD.
So micron is making both Rev.B and Rev.E chips of the same size at the same time ? strange
 

Soulkeeper

Diamond Member
Nov 23, 2001
6,712
142
106
@Soulkeeper

Crucial is essentially Micron's storefront brand. Wouldn't surprise me if they give the best/newest ICs to Crucial.

Yes, but the products have Rev.B not the Rev.E.
So I dunno.
I suspect Rev.B and Rev.E are pretty much identical tbh. Maybe the same dies just with a slight process tech mod.
The few days I had with the Rev.E nemix suggest overclocking and timings are essentially identical.

running 3466MHz 16-21-20-20-44-64 timings now at 1.38V (few corrected ecc errors here and there).
trcd_rd and rp don't scale at all with voltage.
cl scales, rfc(edit: no) and trcd_wr scale a little bit.
I tried for a long time to get 3533MHz stable, I couldn't manage it. I tried up to 1.46v, but i'm not willing to go further. It would get atleast halfway thru memtest.
 
Last edited:

DrMrLordX

Lifer
Apr 27, 2000
21,582
10,785
136
@Soulkeeper

If Micron is anything like Samsung, they have different runs of ICs for different purposes. Among the older 10x-fabbed RAM ICs, Samsung had B, D, and E-die with B-die having the best performance and density (E-die was almost as good, but was featured in lower-density arrangements than most modern B-die). Rev. E may not necessarily be better than Rev. B, and it may not even be newer (?).
 

Soulkeeper

Diamond Member
Nov 23, 2001
6,712
142
106
I just noticed the samsung 32GB modules of ecc udimms 3200 are now available (M391A4G43AB1-CWE).
They just started showing up at etailers this week for about $1000, so about 40% more expensive than the micron options.

I'm tempted to test them out, but can't really afford/justify it.
 
Last edited: