Samsung 16GBit A-die / M-die (32GB unbuffered DIMM, NOT double-height DIMM) - Overclock.net - An Overclocking Community

Forum Jump: 

Samsung 16GBit A-die / M-die (32GB unbuffered DIMM, NOT double-height DIMM)

 
Thread Tools
post #1 of 4 (permalink) Old 06-17-2019, 04:17 PM - Thread Starter
New to Overclock.net
 
ssateneth's Avatar
 
Join Date: Feb 2012
Posts: 888
Rep: 43 (Unique: 38)
Samsung 16GBit A-die / M-die (32GB unbuffered DIMM, NOT double-height DIMM)

There are now 32GB DIMMs available that use 16GBit Samsung IC's. I'm not talking about those weird double-stacked DIMM's that only work in 2 motherboards (ASUS ROG MAXIMUS XI APEX to name one). They're available for as low as $145.99 per DIMM with a JEDEC rating of 2666MHz CAS19.

https://www.superbiiz.com/detail.php?name=D432G266S1
https://www.avadirect.com/32GB-M378A...oduct/12763278

Has anyone tried these yet?

Intel Core i9-9900k (Silicon Lottery 5.1GHz @ 5.3GHz 1.36 VCore)
ASUS ROG MAXIMUS XI APEX
G.SKILL F4-4000C19D-32GTZSW (2x16GiB 4000MHz 19-19-19-39 2T, tuned to 4133MHz 14-17-17-31 2T)
EVGA NVIDIA GeForce GTX 1080 Ti FTW3
Seasonic PRIME 1300W 80+ Platinum

Last edited by ssateneth; 06-25-2019 at 07:58 PM.
ssateneth is offline  
Sponsored Links
Advertisement
 
post #2 of 4 (permalink) Old 06-18-2019, 04:47 PM
Collector
 
emissary42's Avatar
 
Join Date: Aug 2013
Location: Nehr'esham
Posts: 81
Rep: 12 (Unique: 10)
These are Samsung 16GBit M-Die and while you can overclock them, you won't get anywhere near the performance of your F4-4000C19D-32GTZSW @ DDR4-4133 14-17-17 out of them.

Team HARDWARELUXX | B-Die Memory List @ Hardwareluxx
emissary42 is offline  
post #3 of 4 (permalink) Old 06-25-2019, 07:39 PM - Thread Starter
New to Overclock.net
 
ssateneth's Avatar
 
Join Date: Feb 2012
Posts: 888
Rep: 43 (Unique: 38)
Extremely preliminary results shows these can run 3600MHz. But they are also not binned like what your usual RAM assembler would (gskill, corsair, etc) so your results could vary wildly. I got a pair and so did Buildzoid. I'm running them on an Asrock Z370 Taichi, which has a daisy chain arrangement and not quite the best physical arrangement. I also have a ASUS Z390 ROG MAXIMUS XI APEX but I'm reluctant to shove these DIMMs in there since I use it for day to day tasks (I know, seems like a waste, right?)

M-Die seems to respond well to voltage. 3600 CAS14 does not POST @ 1.5 volts. It does POST @ 1.65 volts with occasional errors in memtest. It runs fine @ 1.7 volts, but then I run into another wall. These IC's seem to be temperature sensitive. After about 5 minutes, many frequent memory errors. These DIMMs do not have an integrated temperature sensor, so I just used my finger. Yup, these puppies get hot with no heatspreader @ 1.7 volts. Very hot to the touch, but not enough to burn finger. My 16GB x 2 B-die also run 1.7 volts day to day, and will not exceed 45C. Will work on 3600 CAS15. 3700 does not seem to be stable (yet). Will see.

3600 CAS14 1.5V = no POST
3600 CAS14 1.65V = Runs memtest with occasional error (1 per minute)
3600 CAS14 1.7V = Runs memtest OK, but temperature limited (frequent errors after 5 minutes)

3600 CAS15 1.4V = no POST
3600 CAS15 1.45V = Almost passes POST sequence
3600 CAS15 1.475V = Runs memtest with frequent errors
3600 CAS15 1.5V = Runs memtest with occasional error
3600 CAS15 1.525V = No errors after 13 hours memtest

TRCD and TRP seem to need to be loose. Current running 22 on TRCP/TRP. Any lower throws errors.
TRFC also needs to be pretty high. Currently set to 560. Any lower throws infrequent errors. 65535 TREFI is OK.
Can set CAS Write Latency pretty low. Runs at 8 @ 3600MHz OK. Lower = no post.

Intel Core i9-9900k (Silicon Lottery 5.1GHz @ 5.3GHz 1.36 VCore)
ASUS ROG MAXIMUS XI APEX
G.SKILL F4-4000C19D-32GTZSW (2x16GiB 4000MHz 19-19-19-39 2T, tuned to 4133MHz 14-17-17-31 2T)
EVGA NVIDIA GeForce GTX 1080 Ti FTW3
Seasonic PRIME 1300W 80+ Platinum

Last edited by ssateneth; 06-26-2019 at 03:56 PM.
ssateneth is offline  
Sponsored Links
Advertisement
 
post #4 of 4 (permalink) Old 09-17-2019, 12:36 PM
New to Overclock.net
 
Join Date: Sep 2019
Posts: 1
Rep: 0
I had the opposite experience with M378A4G43MB1-CTD. I had a pair on MSI B450 Tomohawk for 3700X, and in order to reach 3600 (with atrocious CL20) I had to lower the Vdimm from default 1.35V to 1.19V to get semi-stable (still error out in memtest86 test #6 in pass 1). Here is a list of Vdimm and frequency that will POST:

Vdimm 1.35V: under load 1.360V reported, 3200 CAS22
Vdimm 1.34V: under load 1.360V reported, 3400 CAS22
Vdimm 1.33V: under load 1.344V reported, 3400 CAS22
Vdimm 1.32V: under load 1.344V reported, 3466 CAS22
Vdimm 1.31V: under load 1.328V reported, 3466 CAS22
Vdimm 1.30V: under load 1.312V reported, 3533 CAS22 (WIN10 blk screen)
Vdimm 1.29V: under load 1.296V reported, 3533 CAS22
Vdimm 1.28V: under load 1.296V reported, 3600 CAS22 (geekbench 5 crash)
Vdimm 1.27V: under load 1.280V reported, 3600 CAS22 (geekbench 5 crash)
Vdimm 1.26V: under load 1.264V reported, 3600 CAS22 (geekbench 5 pass, AIDA stability test crash)
Vdimm 1.25V: under load V reported, 3600 CAS20 (geekbench 5 pass, AIDA stability test crash)
Vdimm 1.24V: under load 1.248V reported, 3600 CAS20 (geekbench 5 pass, AIDA stability test crash)
Vdimm 1.23V: under load 1.248V reported, 3600 CAS20 (geekbench 5 pass, AIDA stability test crash)
Vdimm 1.22V: under load 1.232V reported, 3600 CAS20 (geekbench 5 pass, AIDA stability test crash)
Vdimm 1.21V: under load 1.232V reported, 3600 CAS20 (geekbench 5 pass, AIDA stability test pass for 1 min)
Vdimm 1.20V: under load 1.216V reported, 3600 CAS20 (geekbench 5 pass, AIDA stability test pass for 2 min)
Vdimm 1.19V: under load 1.200V reported, 3600 CAS20 (geekbench 5 pass, AIDA stability test pass for 2 min)
Vdimm 1.18V: under load 1.184V reported, 3600 CAS20 (geekbench 5 pass, AIDA stability test crash)

The DRAMs did not feel hot at all at ~1.20V, so I doubt it is high temp that rendered them unstable. Setting Vdimm to above 1.45V the system simply refused to boot.

Anyway, it is a bit disappointing that these sticks can not get through the first pass in memtest86 test #6 without error even at 3600 CL20. I had high hope after watching Buildzoid's video. If someone has some advice on OCing these DRAMs I am all ears.


Quote: Originally Posted by ssateneth View Post
Extremely preliminary results shows these can run 3600MHz. But they are also not binned like what your usual RAM assembler would (gskill, corsair, etc) so your results could vary wildly. I got a pair and so did Buildzoid. I'm running them on an Asrock Z370 Taichi, which has a daisy chain arrangement and not quite the best physical arrangement. I also have a ASUS Z390 ROG MAXIMUS XI APEX but I'm reluctant to shove these DIMMs in there since I use it for day to day tasks (I know, seems like a waste, right?)

M-Die seems to respond well to voltage. 3600 CAS14 does not POST @ 1.5 volts. It does POST @ 1.65 volts with occasional errors in memtest. It runs fine @ 1.7 volts, but then I run into another wall. These IC's seem to be temperature sensitive. After about 5 minutes, many frequent memory errors. These DIMMs do not have an integrated temperature sensor, so I just used my finger. Yup, these puppies get hot with no heatspreader @ 1.7 volts. Very hot to the touch, but not enough to burn finger. My 16GB x 2 B-die also run 1.7 volts day to day, and will not exceed 45C. Will work on 3600 CAS15. 3700 does not seem to be stable (yet). Will see.

3600 CAS14 1.5V = no POST
3600 CAS14 1.65V = Runs memtest with occasional error (1 per minute)
3600 CAS14 1.7V = Runs memtest OK, but temperature limited (frequent errors after 5 minutes)

3600 CAS15 1.4V = no POST
3600 CAS15 1.45V = Almost passes POST sequence
3600 CAS15 1.475V = Runs memtest with frequent errors
3600 CAS15 1.5V = Runs memtest with occasional error
3600 CAS15 1.525V = No errors after 13 hours memtest

TRCD and TRP seem to need to be loose. Current running 22 on TRCP/TRP. Any lower throws errors.
TRFC also needs to be pretty high. Currently set to 560. Any lower throws infrequent errors. 65535 TREFI is OK.
Can set CAS Write Latency pretty low. Runs at 8 @ 3600MHz OK. Lower = no post.
choumore is offline  
Reply

Quick Reply
Message:
Options

Register Now

In order to be able to post messages on the Overclock.net - An Overclocking Community forums, you must first register.
Please enter your desired user name, your email address and other required details in the form below.
User Name:
If you do not want to register, fill this field only and the name will be used as user name for your post.
Password
Please enter a password for your user account. Note that passwords are case-sensitive.
Password:
Confirm Password:
Email Address
Please enter a valid email address for yourself.
Email Address:

Log-in



Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)
 
Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may post new threads
You may post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off