Jump to content
Corsair Community

Bad Stick Twin2x2048-64004 G


Smithy152

Recommended Posts

Hi,

 

Hopefully this is just a formality, but I have a set of XMS 1 GB sticks, Twin2x2048-6400 G, and one of them produces errors in memtest at any settings I use. It is always at 003d4085c4 (980.5MB) on Auto 1.8v or manual EPP and SPD copied settings. Plus a whole range of values derived from my head, this forum, and our best friend Google. Some days it will run fine for an hour or too, others its a never ending stream at the same address. The other 'seems' to be fine, no errors ever, yet. I have already sent them back to the retailer, only to have them returned a week later. The report given to me on the phone before they were sent back to me was read to me aloud, word for word, by the nice (poor) man at the support desk and made no mention of the ram being tested at 2.1v and at SPD, EPP, 5-5-5-18 1.8v, 4-4-4-12 2.1v etc.. Or indeed any mention of bios settings used, or the duration of the memtest. The email I got 4 days after I complained to the poor, but nice support man (done that job so I was kind to him), stated that they had done the range of tests I had talked about on the phone... mmm... smells fishy. Either way I'm still with a pair, where one stick is bust, I believe. I'm not beyond the possibilty that its is the MB but the single stick tests indicate otherwise. In dual channel mode the errors are always at 1960MB (forget the exact value). MB is Asus P5K 0603 bios, have tried all bios's inc beta's and Google has been well and truely thrased on the subject. As of about 3 weeks ago I have tried everything that was in this forum on this brand of memory and the Asus P5 Family as a whole. Legacy USB off, slight changes in voltage, timings and so on.

 

The short and tall is, can I go ahead and fill in your RMA with 'the RAM Guys blessing' or is there any other information/test results you would like from me first. Or do you even have a possible solution I haven't found, I'm always learning.

 

Many thanks

 

Smithy.

 

P.S. Keep up the good work.

Link to comment
Share on other sites

Just thought about the power thing, but not having a spare supply rated for this system I endevoured to try some tests I hadent thought about before. So I whipped out the gts and put in an old Cirrius Logic PCI from days long gone. No errors, even at 4-4-4-12 2.1 V dual channel. But here's the trick. I put the GTS back in and ... no errors either. Its happened before, some days this thing runs like a horse on steriods (as it should) without any problems. And others the error just won't go away. It's always the same when trying to narrow in on one tiny little problem like this. I've taken a Multimeter to the motherboard, Vcore and Vdimm all report as perfect. I've also checked the rails under load, and everything looks good. I truely wish I had a complete set of known working spares to narrow the problem down, but this machine is new and my other kit is DDR1/AGP etc. I will wait untill it starts throwing errors up at me again, which shouldn't be long, and see if I can reproduce them with the Cirrus logic card. Motherboard DMI changes after switching the cards?

 

Oh last question, an important one perhaps, 28Amps on one rail ? isn't that 28 total. Here's my PSU

 

http://www.silverstonetek.com/products/p_spec.php?pno=st56f&area=usa

 

So it's actualy 2 x 20 AMPs, 40A/480W total. Everthing I've read says it should be fine for my system and a gts320. I'll look deeper into that issue, I'm learning everyday with this problem.

 

Again, many thanks for reading my ramblings.

 

Take care now.

 

Smithy.

Link to comment
Share on other sites

  • Corsair Employees
Please make sure that you have the latest BIOS for your motherboard and then load optimized defaults and set the Memory Voltage to 2.1 volts and then set the timings to 4-4-4-12 (CAS-tRCD-tRP-tRAS) and then test the modules one at a time with http://www.memtest.org! Test them both one at a time in the slot closest to the CPU and if one of them gives you errors and the other one fails, we would want to get them replaced for you. If both of them give you errors, then there may be some other issue.
Link to comment
Share on other sites

Well it wasn't long before the error came back. So I tested both sticks individualy and the same stick produces the same error at the same place over and over, as mentioned above. The second stick tests fine.

 

System is all auto defaults except for ram set at 4-4-4-12 and 2.1v and the slot next to the cpu was used in each testing cycle, as per instructions. The error will acctually show at auto everything (5-5-5-18) after long enough.

 

It seems that it only shows its head after being put through its paces for a while with something memory intensive. I would geuss that one of the chips is borderline, but faulty none the less.

 

Do you need anything else from me or can I RMA them?

 

Many Thanks

 

 

Smithy.

 

PS. Forgot to add. Legacy USB was off and i got exactly the same results with the old PCI Cirrus Logic Card as with the Geforce 8800.

Link to comment
Share on other sites

Archived

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

×
×
  • Create New...