Jump to content
Corsair Community

MSI K8N Neo2 Platinum, TWINX1024-3200XLPRO & Random Reboots


thoradin

Recommended Posts

Ok, I've been reading through here since I got all my parts in November and was experiencing random reboots in XP Pro SP2. I've re-installed XP Pro and left it at SP1A, tried changing timings and voltages to what you all suggested. I re-installed SP2 from CD again with no changes. I took out my sound card and i'm just using everything that is onboard to try and narrow down a possible hardware failure. I've also tried un-hooking my DVD and CDRW drives thinking it was a power issue with my Antec TruePower 430W supply. I've upgraded to a bigger PSU "Ultra ULT31562 X-Connect 500W" it has 34A on the +12v rail which should be well enough. I took out my PowerColor Radeon 9700 Pro and put in an old Creative Geforce2 with still the same random reboots. Tested the memory one at a time using Memtest86 with lots of errors on bank one "1st Green" but switched to the 3rd Bank "2nd Green" with no problems on both modules. Then I tested both modules together on the last two banks "Green & Purple". After 12 hours of testing there was only 5 errors found which I think is ok *shrugs*. I'm at my wits end on what to do next. Since it's been since November and I haven't been able to solve my problem I can't return anything back to where I purchased them. :( Oh yeah, I even tried patching the BIOS to Beta 1.5 with no changes, I'm back to using the offical 1.4 release now. Hope all this info can help you with my problem. I even replaced all my IDE and floppy cables. I don't have another board that supports this memory or cpu so I can't test that option.

 

I have the following installed right now. (bare minimum) I don't need to OC my system.

 

AMD Athlon64 3500+ (Winchester) Boxed Retail

MSI K8N NEO2 Platinum with 1.40 BIOS revision

Corsair Twinx1024-3200XLPRO at auto settings 2-2-2-5-1T

PowerColor Radeon 9700 Pro (R97-C3G)

2 Western Digital 7200 RPM 8mb cache

JustLink 24x10x40 CDRW

4 80mm Aux Fans in a Lian-Li PC-65 case

Windows XP Pro SP2

Ultra ULT31562 X-Connect 500W (+3.3v 28A, +5v 30A & +12V 34A)

All connected to an APC BackUPS Pro 650 so i know it's not a power condition issue.

 

I know the NEO2 board likes to under volt everything so i've adjusted everthing in BIOS to compensate so that each device should be getting enough power under load.

 

My Athlon64 3500+ Winchester needs 1.4v (1.5v for a Newcastle) minimum as per AMD's site and my core center says that it's at 1.47-1.48v having it set to 1.55v in BIOS

AGP voltage is set to 1.60v in bios *has no idea how to check this value in windows*

Memory voltage is set to 2.75v right now to be safe...I've tried 2.70, 2.80 and 2.85 with no changes. Not sure how to check the actual value in windows.

 

My MSI Core Center tells me this right now:

vCore 1.47-1.48v

+3.3v is at 3.31v steady

+5.0v is at 5.05-5.08v

+12v is at 12.20-12.26v

 

CPU Temp under heavy load 49ºC

CPU Temp at idle 38ºC

System Temp under heavy load 33ºC

System Temp at idle 32ºC

 

BIOS settings for CPU are set to manual

with spread spectrum enabled and Cool n' Quiet disabled

200Mhz x 11 = 2210Mhz

 

Memory is set to auto detect on all. I have tried manual settings

 

I hope all this info I provided will help me get up and running soon.

Link to comment
Share on other sites

have you tried setting cas to 2T or 2.5T

 

Yep, tried all that with no change. The only thing out of the ordinary is if I leave the computer idle with one application running like AIM, it will usually stay on for at least 12-24 hours with out rebooting. If I run memory intensive apps like SETI@home, 3DMark03 or 3DMark05 I can make the reboot happen within the 1st hour. I guess the programs are trying to write in an address that either my system or memory do not like. That's why I'm thinking it might be a memory related problem rather than hardware.

Link to comment
Share on other sites

The Dim Voltage should be set to 2.8 Volts with XMS3200XL and I would try and run them one module at a time and they should be installed in slots 1-2 as well.

 

I was reading around and I've noticed that slots 3-4 seem to be more stable than 1 & 2 for some unknown reason. I'll switch them back to slot 1-2 and I'll bump my memory voltage back up to 2.8v again and test them separately.

 

I also purchased Hot CPU Tester Pro 4 tonight so we'll see if that program can narrow the problem down.

Link to comment
Share on other sites

  • Corsair Employees
I would check the user manual for the correct slots. It does depend on how the MB maker does the MB layout as to which slots to use. AMD suggests skipping the first slot when onky 2 modules are installed. But I would check with the MB maker.
Link to comment
Share on other sites

OK, this is wierd. I've tested these modules before separately with memtest86 and they did not error out but I used slot 3 on the motherboard to test them. Now just using slot 1, only one of my modules is giving me errors now. I'm running on the good one now and no locks ups so far.

 

As far as the test results with Hot CPU Tester Pro 4, all my tests passed on the good DIMM after doing a 6 hour test. With the bad DIMM I only get errors on the Sorting Algorithms module trying to run a 6 hour test "usually locks up after 15 minutes". Here are a couple of error messages I got from the programs log file after it crashes on the algorithm test.

 

Error:CPU 0: MFC QuickSort: Checksums do not match; a[135131] = 11114, a[135131] = 11115

 

Error:CPU 0: MFC QuickSort: Checksums do not match; a[212561] = 17449, a[212561] = 17448

 

Error:CPU 0: Memory cells do not match; cMem1[1194208] = -32, cMem2[1194208] = -31

 

Error:CPU 0: MFC QuickSort: Checksums do not match; a[44959] = 3688, a[44959] = 3689

 

Error:CPU 0: MFC QuickSort: Checksums do not match; a[16177] = 1327, a[16177] = 1326

 

At their user forums it says that Sorting Algorithms can reflect either a bad CPU, memory or chipset. Since my chipset test had no failures and the rest of my CPU tests passed except for the sorting algorithms with the bad DIMM installed and now it passes everything with the good DIMM. I'm pretty certain I have a bad DIMM in my matched set. What do I do now?

 

If any of you are interested in this diagnostic program and wonder if AMD supports it, look at this article at their site. Try the lite version and see if it helps any of you out.

 

http://www.amd.com/us-en/Processors/ProductInformation/0,,30_118_9485_9488%5E9611~75112,00.html

Link to comment
Share on other sites

  • 3 weeks later...

I just want to let everyone know that I received my new modules the other day and I've been running SETI, Hot CPU Tester Pro 4, 3DMark03 & 05 with no errors or reboots in dual channel mode. ^____^

 

I'm using slots 1&2 (Green/Purple)

In DRAM Configuration, the memory timings are all set to Auto and the mobo correctly detects the timings at 2-2-2-5-1T

High Performance Mode = Manual

Aggressive timings = Enabled

Dynamic Overclocking = Disabled

HT Frequency = 5x

Cool'n Quiet = Disabled

Adjust CPU Ratio = 11

Adjust CPU FSB Frequency = 200

Adjust AGP Frequency = 66

CPU Voltage = 1.45v (stays at 1.41-1.42v under load)

Memory Voltage = 2.80v

AGP Voltage = 1.60v

 

CPU Temp under load = 48ºC

Sys Temp under load = 37ºC

Current Room Temp = 24ºC

 

It's been running under load for 10 hours now with no hiccups, I'm pretty sure everything is stable/fixed.

 

Thanks so much for your help and the quick turn around for getting my new modules. :biggrin: I'll be doing business with you all again.

Link to comment
Share on other sites

Archived

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

×
×
  • Create New...