Jump to content
Corsair Community

Mestest errors with TR3X6G1600C9


lm2b

Recommended Posts

Hi,

 

Afte 1.5 years running fine, my PC has started to crash a month ago.

 

I've found errors running Memtest86+ (even after clean contacts), and now I'm tryng to get right configurations to avoid this errors, but no success.

 

My original configuration was 2 modules in AUTO, by SPD (1333MHz, 1.5v, 9-9-9-24), dual channel ungaged.

Searching in Corsair foruns and specs, I've set to manual 1600MHz, 1.65v, 9-9-9-24, but no success also.

 

I know, this modules are not recommended to my system (AMD) - I didn't see the compatibility table at buy time, just ordered with my vendor "a pair of corsair DDR3 1600 great memory =)". I think thats is not a trouble, I'm right? It worked fine for 1.5 years.

 

Any suggestions?

Tks!

Link to comment
Share on other sites

  • Corsair Employees
Your system will not support DDR1600 with the CPU that you have, I would suggest setting the memory Frequency at DDR1333 and set the memory Voltage to what is printed on the modules label and and set Command Rate to 2t and disable legacy USB then test the modules one at a time with http://www.memtest.org.
Link to comment
Share on other sites

With one stick at time got no errors at all.

Even with 2 sticks sometimes I got no errors (an if so, I need to wait one day to get errors again).

 

I'll give a try with RAMGuysettings. By the way, even if processor specs say DDR1333, mobo specs says that can handle until ddr1800 under OC.

 

NOTE: I cannot find the 2T sessing on my mobo, but can set a lot of parameters:

 

DRAM Timing Mode: [Auto] [both] [DCT 0] [DCT 1]

TCL: [Auto] [4 CLK] ~ [12 CLK]

TRCD: [Auto] [5 CLK] ~ [12 CLK]

TRP: [Auto] [5 CLK] ~ [12 CLK]

tRTP: [Auto] [4 CLK] ~ [7 CLK]

TRAS: [Auto] [15 CLK] ~ [30 CLK]

TRC: [Auto] [11 CLK] ~ [41 CLK]

tWR: [Auto] [5 CLK] ~ [12 CLK]

TRRD: [Auto] [4 CLK] ~ [7 CLK]

tRWTTO: [Auto] [3 CLK] ~ [17 CLK]

tWRRD: [Auto] [2 CLK] ~ [10 CLK]

tWTR: [Auto] [4 CLK] ~ [7 CLK]

tWRWR: [Auto] [3 CLK] ~ [10 CLK]

tRDRD: [Auto] [3 CLK] ~ [10 CLK]

tRFC0/1/2/3: [Auto] [90ns] [110ns] [160ns] [300ns] [350ns]

Link to comment
Share on other sites

I'll give a try with RAMGuysettings. By the way, even if processor specs say DDR1333, mobo specs says that can handle until ddr1800 under OC.

Correct some motherboards do. But just because the MB supports it, it is still CPU dependent. And your CPU only supports a max frequencies of 1333mhz. So anything above that would require overclocking.

Anyway heres the spec for your MB taken from ASUS site. Even according to this you are still needed to overclock at 1600mhz!

4 x DIMM, Max. 16 GB, DDR3 1800(O.C.)*/1600(O.C.)/1333/1066/800 ECC,Non-ECC,Un-buffered Memory

Also this information is on page 1-13 in your manual. It lists all 1600mhz modules as O.C.

 

NOTE: I cannot find the 2T sessing on my mobo, but can set a lot of parameters:

On some ASUS boards it's called 1n or 2n instead of 1t or 2t. But it is still the same.

Link to comment
Share on other sites

Tks, Peanutz, you're right!

 

But even with no overclock (just as the start case - everithing in AUTO and, so, DDR1333), now the errors occurs. Remember that memory become unstable after a long time of non-trouble day-by-day use.

 

The most strange thing is that when I power up the machine I can see errors on memtest (variable count of errors, between 400 and 2000, but at the same area - near the end of the 4Gb) and errors persist until I remove one stick (anyone of them - so, no errors in single module configuration). Removed a stick, I got no errors, and if I put it again, magic!, no more errors. Until next day, and start it again...

 

This ASUS MB appear not have the configuration. Google for it an you'll se a lot of people angry:

http://www.google.com/search?q=Asus+M4A785TD-V+EVO+command+rate

 

Maybe I can test with another "fail-safe" configuration? Any sugestions?

Link to comment
Share on other sites

Well, I've made a lot of tests every day, when power up my PC. And I've got errors on memtest:

- with "legacy USB" enabled or disabled (I've also discarded false positives because of OS hangs coincide with memtest errors);

- with memory settings totally on AUTO (DDR1333, 1.5V, 9-9-9-24) and with manual here recommended settings (like DDR1333, 1.65V, 9-9-9-24, CPU/NB overvoltage [+.1 and +.2V])

- with both memory modules on dual channel, sigle channel, ganged or unganged mode.

- with just one module on different banks/slots.

- with and without processor overclock (auto MB overclock 10%).

 

BUT after some time and sort of tests, specially after remove and place modules again - or after overclock and underclock sequentially, memtest stop to show errors (and my system go stable). But on the next day, there are the errors again. And if I open my OS and just wait a few minutes to get warm (supposing that could be some kind of cold start problem), still getting errors. All contacts are clean and visually OK, on both: modules and MB (all cleaned with isopropyl alcohol, with system off until get dry).

 

When I didn't get more errors, my system can run (as a test) at DDR1600 9-9-9-24 1.5V and 10% of auto-overclock (or the equivalent showed by POST DDR-1760) without any problem.

 

The case is clean and have small temperatures and good ventilation (12x12 fans on front/back doing a wind tunnel, 14x14 fan under the power supply, and a big 12x12 fan on a big copper cooler on processor).

 

As I can I'll test my system with another-brand common DDR1333 modules to be sure that is an memory problem. Can be an incompatibility by modules and AMD system? Even after worked >1.5 year?

 

Hugs,

Luiz

Link to comment
Share on other sites

Can be an incompatibility by modules and AMD system? Even after worked >1.5 year?

 

BUT after some time and sort of tests, specially after remove and place modules again - or after overclock and underclock sequentially, memtest stop to show errors (and my system go stable). But on the next day, there are the errors again.

Not really. But your MB can go bad. And that's what i am betting on. But as you said you would need another set to test out on your MB to be sure.

Link to comment
Share on other sites

  • Corsair Employees
We can try and replace the modules but from what you have posted I would suspect some other issue and if you can I would test the modules on another system to be sure. But if you want to try and replace them please use the link on the left and request an RMA.
Link to comment
Share on other sites

  • 2 weeks later...

Tks Peanutz and Ramguy.

 

I'm running my system about 10 days with new other-brand modules, WITHOUT any problem.

I'm using it's default config (2x2Gb, dual channel unganged, DDR1333, 1.5V, 9-9-9-24), just like original corsair AUTO mode (remember that even in auto mode corsair modules fails).

 

I cannot test my corsair modules in another machine (I cant find anyone that can test it for me), but after 10 days stable with another memory, the problem seems to be my corsair modules.

 

I'll request RMA. Thanks!

Luiz

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...