Failed DIMM detected by MEMTEST86+

[isatis]

Hello,

 

I got my computer with 2 TwinX1024-3200C2PT modules in a ABIT IC7-MAX3 mobo, P4C 2,6GHz@3,25 RATIO 5:4 to keep good timings on RAM, AGP/PCI Fixed to 66/33MHz

 

My DIMMs were set like this :

 

DIMM1 : Kit1 TwinX CMX512-3200C2PT v1.2

DIMM2 : Kit2 TwinX CMX512-3200C2PT v3.1

DIMM3 : Kit1 TwinX CMX512-3200C2PT v1.2 (this was later detected faulty)

DIMM4 : Kit2 TwinX CMX512-3200C2PT v3.1

 

BIOS Setup : 2,7v 200MHz 2-3-3-7.

 

I started wondering when Windows XP started to crash unexpectedly a week ago

 

Since the last 2 days there was specific BOSD : STOP 0x00000050 "PAGE_FAULT_IN_NON_PAGED_AREA", this meant RAM modules.

 

I did run MEMTEST86+ (http://www.memtest.org) and got errors with both kits installed. I got errors on address 0x7fb49de8 (i stopped the test after 5% done as there were 100+ errors).

I made this test 4 times, swapping kit1 with kit2 and pairs within the kits, and still got errors (i stopped the test after 5% done as there were 100+ errors).

 

I did tests with the kit2 only, and got no error (1 full test done ).

UPDATE : 5 tests done with this kit. I stop there, since i need that PC =)

 

I did tests with the kit1 only, and got errors on address 0x2fbcaef0 (i stopped the test after 5% done as there were 100+ errors)

 

I did tests in DIMM1 with the first DIMM of the kit1, and got no errors (1 full test done ).

 

I did tests in DIMM1 with the second DIMM of the kit1, and got errors on address 0x1fb0eef0 (i stopped the test after 5% done as there were 100+ errors)

 

I did that last test 4 times.

 

I made test with all 3 others modules, and got no error (1 full test done )

 

I called the vendor but i purchased the kit on october, 25th 2003 so the do not longer support the DIMMs.

 

Thank you for you response.

 

--

Remy C.

[RAM GUY]

Please make sure that you have the latest BIOS for your MB and load Setup/optimized default settings, and try the following BIOS settings;

TwinX1024 or TwinX512-3200C2

Softmenu Setup

CPU Freq: 250

DRAM Clock: 200 MHz

CPU:DRAM ratio: 5:4

Memory or Dim Reference voltage: 2.8/2.85 Volts

AGP Voltage: Default *unless you have ATI (9200 or 9600) or NVIDIA (5200 or 5700) then 1.6 Volts suggested!*

Advanced/Chipset

Configure Dram timing by SPD: Disabled/User define

SDRAM CAS Latency: 2.5T

SDRAM RAS to CAS Delay (tRCD): 3T

SDRAM Row Precharge (tRP): 3T

SDRAM Active to Precharge Delay (tRAS): 7T

SDRAM Burst Length: 8

All other settings should be set to default settings!

Then please test them one at a time with http://www.memtest.org and let’s make sure it's not some other issue! I would run the test for at least 2-3 passes to be sure!

[isatis]

Hello,

 

Same HW config but :

 

DIMM1 : Kit2 TwinX CMX512-3200C2PT v3.1

DIMM2 : Kit1 TwinX CMX512-3200C2PT v1.2

DIMM3 : Kit2 TwinX CMX512-3200C2PT v3.1

DIMM4 : Kit1 TwinX CMX512-3200C2PT v1.2

 

And graphic card changed with an Asus V9999 6800GT

 

I did reset BIOS to optimized defaults.

 

At 2,8V, memtest crashed doing 7% of sub test (0% overall)

At 2,9V, system hang up on scanning the RAID drives

 

I try with only one couple of dimms and i'll post back

[RAM GUY]

Can you tell me the bios settings you have set for both CPU and memory and any performance settings that you may have set?

[isatis]

Using the couple supposed to be good, the system crashed at 2,8V

 

I put back the CPU to 2,6GHz 1,6V and tested again with all four modules at 2,8V, i.e. :

 

DIMM1 : Kit2 TwinX CMX512-3200C2PT v3.1

DIMM2 : Kit1 TwinX CMX512-3200C2PT v1.2

DIMM3 : Kit2 TwinX CMX512-3200C2PT v3.1

DIMM4 : Kit1 TwinX CMX512-3200C2PT v1.2

 

Still failing at address 0x7fb49de0

 

I inverted module 2 and 4 to see

 

Hum.. still failing at address 0x7fb49de8...

 

i try inverting generation types (i.e. 1 with 2 and 3 with 4)

 

This time failing 0x7fa49de8

 

I tried this :

 

DIMM1 : Kit1 TwinX CMX512-3200C2PT v1.2

DIMM2 : Kit1 TwinX CMX512-3200C2PT v1.2

DIMM3 : Kit2 TwinX CMX512-3200C2PT v3.1

DIMM4 : Kit2 TwinX CMX512-3200C2PT v3.1

 

Failing at 0x0fbcaef0

 

Then this :

 

DIMM1 : Kit2 TwinX CMX512-3200C2PT v3.1

DIMM2 : Kit2 TwinX CMX512-3200C2PT v3.1

DIMM3 : Kit1 TwinX CMX512-3200C2PT v1.2

DIMM4 : Kit1 TwinX CMX512-3200C2PT v1.2

 

Failing at 0x4fbcaef0

 

To find out a mobo issue i'll try 8 tests with 1 couple in slots 1-3 then 2-4 and again swapping them.

All settings default, but modules at 2,8V, AGP 1,6V, CPU 1,6V

 

This time i did think to write numbers on modules...

 

1/

DIMM1 : Kit1 TwinX CMX512-3200C2PT v1.2 (1)

DIMM3 : Kit1 TwinX CMX512-3200C2PT v1.2 (2)

Failing at 0x3fb49de0

 

2/

DIMM2 : Kit1 TwinX CMX512-3200C2PT v1.2 (1)

DIMM4 : Kit1 TwinX CMX512-3200C2PT v1.2 (2)

Failing at 0x3fb49de0

 

3/

DIMM1 : Kit1 TwinX CMX512-3200C2PT v1.2 (2)

DIMM3 : Kit1 TwinX CMX512-3200C2PT v1.2 (1)

Failing at 0x3fb49de8

 

4/

DIMM2 : Kit1 TwinX CMX512-3200C2PT v1.2 (2)

DIMM4 : Kit1 TwinX CMX512-3200C2PT v1.2 (1)

Failing at 0x3fb49de8

 

At this point i can say that module 1 on Kit1 is broken.

 

**following next post**

[RAM GUY]

That tells me there is some other problem, because you have it failing at the same address no matter what module you have installed in what slot. At least that's what it appears from what you have posted. I would suggest testing the modules one set at a time on another system to be sure. In addition, I am sorry but we only guarantee them with one set at the tested settings. In addition you are over clocking your CPU, I would suggest testing with the memory and CPU not over clocked to help isolate the problem. So please set the CPU too it’s default and then test them if you do not have another system. If they run stable one set at a time I would try and set the memory frequency manually at DDR400 1-1 with your CPU and set the timings manually to Cass 2.5-3-3-7 and see if the system will pass http://www.memtest.org. I have no problem replacing your modules but I am not sure that will solve the issue.

[isatis]

[Note : All performance settings were set to auto, since there seems not be a way to disable them, CPU to defaults but 1,6V, AGP to 1,6V, RAM to 2,5-3-3-7 200MHz 2,8V]

 

I think you missread the address i got errors on :

 

On the first part, i test both 2 kits, and there the differences bolted :

 

0x7fb49de0

0x7fb49de8

0x7fa49de8

0x0fbcaef0

 

Those are for the 4 dimms tests.

In the beginning of the post i did the test with the CPU back to 2,6GHz (was 3,2 with overclocking)

 

So my post was already at your previously told specification with no overcloking this time.

 

Furthermore, i did begin a 8 pass test protocol with only 1 kit at a time, anticipating your demand.

 

The kit 1 has its tests already posted :

 

Tests 1/ & 2/ are to isolate bank problem but the system sees the modules as they did not move.

 

Same for test couple 3/ & 4/, 5/ & 6/ and 7/ & 8/

 

The 5/ & 6/ (run 37 times all night long) and 7/ & 8/ (3 times) did not return any error.

 

Here are the errors reported by couple test 1/ and 2/ compared to 3/ & 4/ :

 

1/ & 2/ 0x3fb49de0

3/ & 4/ 0x3fb49de8

 

Note the last bit difference !

 

It is due to dual channel that strip memory accesses to be one byte on one module and the following on the second.

 

I do tests with two modules as there a TWINX set and you change both if one damaged.

 

I do now a test with kit 1 modules, but one module at a time, still on the standard specification, with your modified timings and voltage.

 

**results in next post**

[isatis]

Here are the results for module 1 identified on kit1 :

100+ errors on address 0x1fb0eef0

*Some more addresses failed once :

[some more errors could place here but error addresses are too close so the screen scrolled without let me time to read]

0x1fb0f490

0x1fb0f4d0

0x1fb0f510

0x1fb0f550

0x1fb0f590

0x1fb0f5d0

0x1fb0f610

0x1fb0f650

0x1fb0f690

0x1fb0f6d0

and 2 mores on 0x1fb0eef0

419 errors total in the end of the first full test sequence.

609 errors total in the end of the second full test sequence.

 

And there for the module 2 :

No errors found on 3 tests

[RAM GUY]

Please follow the link in my signature “I think I have a bad part!” and we will be happy to replace them or it!