MSI K8N Neo 2 Platinum + TwinX1024-3200XL

[unnamednewbie13]

System Stats:

 

Thermaltake Silent Purepower 560W

MSI K8N Neo 2 Platinum

*7025 v1.2 bios

AMD Athlon 64 3500+

Corsair TwinX1024-3200XL (CMX512-3200XL XMS) - 2x512MB Sticks

BFG 6800 Ultra O/C

*nVidia 66.93 and 67.02 drivers have been tested; 67.03 is available but only addresses an nVidia 6600 issue over 67.02.

DVD-ROM BDV212B

WDC WD740GD-00FLA0 (primary hard drive)

WDC WD2000JD-00HBB0 (secondary hard drive)

 

Issue: Random Software Crashes & Sytem Reboots

 

The TwinX's boot at CAS Latency 2T, tRCD 2, tRP 2 and tRAS 5. Memory voltage has been tested within a range from 2.5V to 2.7V. With them installed, this fresh installation of Windows is having plenty of glitches. When the video is stressed (during Doom 3 benchmarking, etc.), it will either crash to the desktop, or momentarily flash a blue screen of death before rebooting. During large segments of file transfers from one drive to the other (installations, backup restorations, etc.), the system will reboot without warning. Installed in a non-Dual Channel format (1-3 / green-green), they downgrade to DDR333 but are more stable, though quite slow. Dual Channel combo DIMMS 1/2 and 3/4 have both been tested.

 

I have swapped the memory sticks out for my old pair of CMX512-3200C2PT's, which work without any glitchiness, but are underperformers to the rest of the system's specs.

 

The 6800 Ultra video card is being fed power through two seperate lines, but occasionally a Power Indicator window pops up on the screen:

 

The NVIDIA System Sentinel is reporting that the NVIDIA-powered graphics card is not receiving sufficient power. To protect your hardware from potential damage or causing a potential system lockup, the graphics processor has lowered its performance to a level that allows continued safe operations.

 

This message has only popped up with the new pair of TwinX sticks installed.

 

==========================================================

 

Is there any known compatibility issues? Can this be fixed through any upgrades, or should I return the TwinX's and be satisfied with the low-performance 3200C2PT's?

[RAM GUY]

You will need to get the latest beta bios, I think it's 1.36 or newer especially if you have a 90nm 3500 CPU. And I would use slots 2-4 or 3-4 and set the Dim Voltage to 2.8 Volts.

[unnamednewbie13]

http://www.lejabeach.com/MSIK8N/k8nneo2.html lists these versions of the MS-7025 Award Bios:

 

"These BIOS's are the original issue as downloaded from the MSI Website. These BIOS's are the property of MSI. Each "zip" file contains the BIOS and a text file that describes the changes, and the flasher. The Beta BIOS's were downloaded from here The 7025v137mod and 7025v141mod was modified by "trats" of xtremesystems.org and un-hides many features. The BETA BIOS's 7025v141and 7025v151 were sent to a forum member buy MSI Customer Support. A Overclocking Review is here. The 64bit MSI Sector of Rebels Haven is here."

 

1.00

1.10

1.20

1.23

1.25

1.30

1.33

1.35

1.36

1.37

1.37mod

1.41

1.41mod

1.51

flasher AWFL833D.EXE

 

I'll try 1.51.

 

2.8V shows in the red zone, but I'll try it. Moving them back to DIMMS 3/4. As for slots 2/4, 1/3 are green, 2/4 are purple. MSI's manual discourages filling the purple slots while leaving the green slots empty. *shrug*

 

Do you suppose I'll have to boost AGP voltage too?

[unnamednewbie13]

Floppy drive not working. Currently trying to get Nero to burn a proper boot CD so I can install the darn bios update.

[unnamednewbie13]

Updated BIOS to 1.5B1. Moved memory to DIMMS 3 & 4. Set DIMM voltage to 2.8. Sytem became unstable and artifacts began to appear on 3D applications. Set voltage down to 2.6. Still testing.

 

Update 1: System still unstable. Set DIMM voltage back to 2.5.

 

Update 2: System still unstable. Swapped TWINX'S back out for the older RAM. Should I enable some other things like spread spectrum or whatnot? Awaiting input before I take further action.

[unnamednewbie13]

Still having problems. *bump*

[RAM GUY]

Min. suggested Voltage for these modules would be 2.75 Volts, please set the Dim Voltage to 2.8 Volts, or if you do not want to change the voltage I would suggest you use a JEDEC compliant module like CM64SD512-3200.

[unnamednewbie13]

I heard somewhere that the low latency memory runs optimally at a tRAS from 8 to 11, so I adjusted the tRAS from 5 to 8. Reset the voltage to 2.8V. It's seeming to run better now, so I'm left with troubleshooting the BFG 6800UOC video card.

 

Another question though...

 

MSI's manual states (strongly recommends, actually) that purple slots (DIMM 2 & DIMM 4) should not be used while green slots (DIMM 1 & DIMM 3) are empty. Unusual. However, Dual Channel memory only works on this motherboard in D1+D2 or D3+D4. What do you make of this? Should I pick up another pair of identical RAM sticks to fill all slots and eliminate this problem as it appears on paper, or is this an issue at all?

[unnamednewbie13]

Aside from the above comment/question, I'd like to note that system stability has just increased to where I haven't seen a single crash. This is with both the Corsair 3200XL's and the BFG 6800 Ultra installed. And how did this happen? I relaxed the CAS latency to 2.5-3-3-8 (or 10, for the last. forgot as of last boot).

 

As for the video card, I uninstalled all nVidia drivers, all ATI (temporary solution) drivers, rebooted into safe mode, ran Driver Cleaner to get rid of the traces, deleted the folders manually, emptied the recycle bin, rebooted, installed the downloaded motherboard drivers, rebooted, installed the downloaded Forceware drivers (both MB and video drivers gotten as official up-to-date versions from nVidia's site) and tested. Crashes were now mainly to the desktop, though there were a few instances of rebooting. Tried raising AGP frequency from 66 to 67. Instability. AGP voltage is now at 1.6, as per the BFG Tech staff's recommendation. Spread spectrum is off, HT is 5x (default), and no other overclock setting is on (besides the overvoltage on the memory and AGP). It was only when I relaxed memory latency that the problems stopped.

 

Heat shouldn't be a problem, because the Lian Li V1000's chassis' innards maintain room temperature even after a day of operation, due to the extra fans:

 

2x twin-fan hard drive coolers

1x triple-fan 5 1/4 cooler

1x PCI-slot turbine

 

Hopefully, I won't have to RMA the motherboard anytime soon, because I plan to drop heat sinks onto every large-and-hot microchip on it within reach.

 

While the system now works alright, relaxed memory seems a temporary solution at best, and I would dearly love to reduce it back down to 2-2-2-8 or even the factory default, 2-2-2-5. Again, any suggestions?

 

Note: Once I get this working at 100%, I'll post my BIOS settings here to help others out who are having the same sorts of problems.

[RAM GUY]

Lets try replacing your modules.

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

[unnamednewbie13]

Will do, Boss.

[RAM GUY]

Please let me know how you make out!

[dscheerder]

The MSI K8N Neo2 Platinum seems to be wicked picky about timings. Weird, considering it's touted as a savage overclocker. My TwinX1024-3200C2 only runs properly at 2.5-3-3-6 (at either 1T or 2T timings) and anything else won't POST or runs very poorly (almost 1/2 the 3DMarks or framerates in games). I've tested the RAM in another system using some diagnostic utilities and it doesn't report any errors whatsoever. For the power warning, I had the same with a BFG 6800 Ultra OC on a TTGI 450watt and the problem lessened when I used two independant leads from the power supply without anything else plugged into them. The problem went away when I RMA'd the video card. :laughing: I also now have an Enermax 550watt since I added two new HDDs, as I figured the 450watt was barely squeaking by before the two new drives and I didn't want to start a fire when I added the two new ones. I can't see AGP voltage being any kind of solution as you are already giving the card two leads to the PSU. I can't understand how the card can't draw enough power through two leads from the PSU and requires a bump from the AGP slot.

[RAM GUY]

What do you have the Dim Voltage set to? And for the AGP Voltage I would direct you back to the make of that Video Card for answers.

[unnamednewbie13]

Convinced it was the memory that was bad, I swapped it for some OCZ High Performance DDR400 2-3-2-5 1T's. Same reboot problems. As I already tested low-latency memory modes with an ATI 9600 Pro with the same memory settings, I don't think it's the BFG 6800UOC that's causing the problem.

 

Tomorrow I'll test the Thermaltake 560W power supply. Only the BFG, two SATA (one raptor) hard drives, a few fans and a DVD-ROM drive are being powered by that baby. Shouldn't be that strained...

 

If it turns out that the power supply is fine, I'll come to the conclusion that the (or at least this) MSI K8N Neo 2 Platinum Motherboard is a complete dud. Suprising, considering the positive tech reviews I've read on it prior to purchasing.

[slideshowbob]

I am having the same trouble with the same ram on the same MSI Neo2Plat Mobo. bios 1.4 (live update) and 5.3 drivers (live update)

 

I just got this board and ram on Friday from ZipZoomFly, maybe there is a bad batch of Ram, or maybe I too got a bad MSI board. I can't seem to get my Ram to stop showing errors and I do have the Memory voltage set at 2.8 V.

 

What to do?

[RAM GUY]

You know it might be your CPU as well.

[slideshowbob]

Good Idea Ram Guy, how do you test the CPU? AMD 64 3500+ 90nm (winchester)

 

I've just relaxed my Ram to 2.5 3 3 6 and am testing now. I should know in just a few more minutes if it helped me or not.

 

Thanks

[RAM GUY]

I would check for the latest beta bios for the MB with that CPU.

And you would need to test with known working hardware to be sure, or you might ask AMD if they have a utility ot test the CPU.

[slideshowbob]

Well I am a little confused I guess. I can't seem to ask the right question at AMDs site, to find a way to test the CPU. I did find newer beta bios, and I guess I can try that next, but it looks like it (1.5x bios) already was previously in the thread.

 

After 1.5 hours and 5 passes on Memtest-86 v3.2, I still have NO errors with the relaxed 2.5-3-3-6 settings.

 

Can anyone help me to ask the right question, so I can find out how to test the CPU?

PLEEZE?

 

Thanks again! :)

[RAM GUY]

Please go to http://www.amd.com and you should find a phone# and or email address or a link to there forum.

[unnamednewbie13]

Good call. It COULD be the CPU...

 

Or it could be the motherboard, or even the power supply. I'll have to test them one at a time.

 

One thing's for sure though, I'm going to exchange the OCZ's back for Corsairs. I'm crashing even more now.

[RAM GUY]

unnamednewbie13

Can you list your complete system spec and the bios version you have and tell me what is the exact problem you are getting? Like the system just reboots with or with out a BSOD and if you get a BSOD what is the message you get?

[unnamednewbie13]

Lian Li PCV-1000 chassis

AMD Athlon 64 3500+

1GB Corsair TWINX-3200XL /or/ 1GB OCZ

BFG 6800 Ultra OC (fan) (fed by two dedicated power cables)

SB Audigy 2 ZS Platinum

Thermaltake Silent PurePower 560W Power Supply

12X DVD

10K 70GB Raptor (main)

7.2K 200GB WD (storage)

USB2 250GB Lacey HD (storage)

 

Alright. I've tested the system with the Corsairs/OCZ's, and with the BFG6800UOC/9600PRO. The same problems result:

 

* Random application crashes.

* Random application lockups.

* Random reboots.

* POST lockups ~50% of the time, Windows startup lockups ~25% of the time.

 

I've tested the Corsairs, BFG and 9600 PRO combos with the included BIOS (1.2, I believe), and the Corsairs, OCZ's, BFG and 9600 PRO combos (Corsairs and OCZ's tested seperately, due to trade-in), with BIOS ver 1.5B1. If anything, more errors occur with the OCZ's, which I will soon trade back for the Corsairs. Relaxing the latencies didn't help very much, and I had to raise them to timings such as 3-4-4-10 2T. No overclocking features are enabled, though this motherboard is supposed to be very tolerant of such.

 

I've tried formatting and repartitioning the hard drives, and reinstalling Windows. I am going to try to swap the 3500+ processor today for a replacement to see if that helps. If that doesn't work, the motherboard (sigh) may need to be RMA'd.

 

Here's an attachment of information from DxDiag.

 

The warning "Sound Tab 1: The file ctaud2k.sys is not digitally signed, which means that it has not been tested by Microsoft's Windows Hardware Quality Labs (WHQL). You may be able to get a WHQL logo'd driver from the hardware manufacturer." should be irrelevant, due to the fact that these problems have persisted with the sound card drivers on CD, and with these new drivers downloaded from Creative's web site. I've also tried multiple ForceWare drivers to no avail.

DxDiag.txt

[RAM GUY]

There have been a few people reporting problems with the power supply that you have in your system, I would suggest you go back to our modules and you try another PSU!