RM-VS256MC(DDR400) in DDR266

[maxlaw]

I have just brought a 256MB DDR 400 CL 2.5 value select memory module

(RM-VS256MC) to replace my broken 256MB DDR 266(Its 266, I am using Althon

1600+, 266FSB at 1400MHz internal freq) memory module. My motherborad is

Epox EP-8RDA3I v3.x (nForce 2 Ultra 400), which support FSB up to DDR 400.

 

What I experience is, my system is unstable in all following settings (I

didn't overclock, only try to "underclock" my CPU, which retain in 266FSB or

200FSB without changing frequency multiplier):

RAM: DDR200, CPU FSB 100MHz, 2-3-3-6

RAM: DDR200, CPU FSB 100MHz, 2.5-3-3-5

RAM: DDR200, CPU FSB 100MHz, 2.5-3-3-5

 

RAM: DDR266, CPU FSB 133MHz, 2-3-3-6

RAM: DDR266, CPU FSB 133MHz, 2.5-3-3-6

RAM: DDR266, CPU FSB 133MHz, 2.5-3-3-8

RAM: DDR266, CPU FSB 133MHz, 3-3-3-8

 

(Asynchronus)

RAM: DDR333, CPU FSB 133MHz, 2-3-3-6

RAM: DDR333, CPU FSB 133MHz, 2.5-3-3-7

RAM: DDR333, CPU FSB 133MHz, 2.5-3-3-8

 

(Asynchronus)

RAM: DDR400, CPU FSB 133MHz, 2.5-3-3-8 (Default)

RAM: DDR400, CPU FSB 133MHz, 3-3-3-8

 

The term "Unstable" means, my machine would restart within 2 hours

after a sccessful boot on Windows Xp. It keeps restarting after that reboot

once Windows boot up logo appears until I cool power-off then re-power-on the computer. If I press restart button instead of power-off, the machine would hang(Monitor shows no signal) and the Debug-LED shows C1 error (L2 cache or memory module), which can

also be solved by cool-power off then re-power-on the computer. But the same

situation happens within 2-hours after this re-power-on~~~

 

I have performed MemTest for the module, it is good to pass all

tests. But if I keep iterate all tests again, start at the 3rd or 4th iteration of tests, errors start to pump out in lots of addresses which spread entire memory addresses. In addition, the MemTest will prompt me an unexpected interupt occur and halt after around 30000 errors has been detected(it halts in no particular individual test). Similar problems also occur in the tester "Stress Memory Test".

 

I have tested my MB with a King-ston DDR333 CLL2.5 module, it's fine, it

passes all tests and stable in Windows XP. This King-ston module and my MB

also work in Dual channel mode with a Magic Pro DDR266 memory.

 

I have tried to rise the Vdimm to 2.75v or Voltage of nForce chipset to

1.8v, it doesn't help. The most mysterious is, when I offer an good cooling

condition for my computer (Temp CPU < 16C at full load), it prolongs the duration of stable time(the machine won't reboot) to around 4 ot 6 hours from 1 or 2 hours.

 

My Power supply is 480W (Theorictical peak 480W, pratical output 400W, used 1 month, all voltage tolerence within 5%), I have 2 7200 rpm ATA Harddisk, DVD RW, CD RW, Geforce2 MX 400 display (yes, it's an antique, you know my system work fine over a year except I have changed a higher output powersupply recently), plus my 8RDA3I MB and Athlon 1600+, it should not be power supply problem. All the above tests including tests of King-ston and Magic-Pro RAM are all in the same environment.

 

My CPU never exceed 50C (normally below 45C at room temp 25C, I forget the model of my CPU cooling device, it is a coolermaster product). And I have obtained the most update BIOS (basically, the newest version of BIOS only solve compatibility issue of King MAX chips)

 

What should I do? Does a DDR400 dot not support DDR 266? Because I couldn't find the internal timing settings inside the RAM module(using CPU-Z) in 266MHz mode.

[maxlaw]

Does anybody has suggestion? :[pouts:

[RAM GUY]

Please make sure that you have the latest bios and then load setup defaults and set these settings and test the modules one at a time with http://www.memtest.org.

CPU Freq: 133 MHz

Memory Frequency: 100%

Dim Voltage to 2.7 Volts

Resulting Frequency: 133MHz

SDRAM CAS Latency: 2.5T

SDRAM RAS to CAS Delay (tRCD): 3T

SDRAM Row Precharge (tRP): 3T

SDRAM Active to Precharge Delay (tRAS): 7T