Sign in with
Sign up | Sign in
Your question

Memory Controller Issues On The BioStar M7NCG 400!

Tags:
Last response: in Memory
Share
July 24, 2004 8:55:50 AM

Problem Summary:
================
Windows(NT/2000 Kernel) BSOD.

STOP Exceptions Encountered:
============================
1. DRIVER_IRQL_NOT_LESS_OR_EQUAL(nv4_disp.dll)
2a. KMODE_EXCEPTION_NOT_HANDLED
2b. PAGE_FAULT_IN_NON_PAGED_AREA
3a. BAD_SYSTEM_CONFIG_INFO (During Bootup)
3b. BAD_POOL_DATA (During XP Clean Install)

1 and 2 happens while the system is running.

My System Config:
=================
AMD Athlon XP 2000+ (1.67GHz/FSB266)
BioStar M7NCG 400 (nForce2 400)
Hynix 512MB DDR333 SDRAM DIMM(8-4-4-3)
Creative SB Live!
Seagate Barracuda 80GB PATA
Quantum Fireball lct15 20GB PATA

Details:
========
I tried updating my nVidia GeForce MX IGP drivers to
the latest available from nVidia. Windows stopped saying that the trouble maker was the nVidia IGP driver. The problem persisted.
Then I thought this was a memory timing related issue. I did a look-up of the memory timing values from the Hynix website for my HY5DU56822AT based DIMM and found that VDD/VDDQ should be 2.5V and CL-tRCD-tRP should be 2.5-3-3. I set the memory timings and voltages manually to these exact values in the BIOS config.
I have checked every component of my system and found nothing wrong (whatever I could test with software. eg:Memtest86+). No virus, trojan etc. I tried another Hynix 512MB RAM module but Memtest did not indicate any errors with the module!
Could this be a memory controller issue? Considering that the M7NCG400 is a low-end mainboard, I'm not sure whether I should beleive that BioStar designs crappy mainboards!

I'd really appreciate if somebody would help me solve this "mystery". :-(

Thanks Guys.
July 24, 2004 5:29:06 PM

I'm not sure whether I should beleive that BioStar designs crappy mainboards!
\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\

Yup, believe it! Before you upgrade the board though, try some different ram in it. Some motherboards just hate certain ram. It may solve the problem. Of course, if you want to put an end to all issues, order an ABIT NF7-S and keep on trucking! :smile:


Abit IS7 - 2.8C @ 3.4 - Mushkin PC4000 (2 X 512) - Sapphire 9800Pro - TT 420 watt Pure Power
Samsung 120gb ATA-100 - Maxtor 40gb ATA - 100
Sony DRU-510A - THAT'S MORE LIKE IT!
July 24, 2004 10:49:23 PM

Cybercraig, I ain't lookin' for an alternative. Can we tahe a look at solving this problem rather than finding "other" ways? I ran Memtest+ v1.20. No errors detected!
Related resources
July 25, 2004 12:56:02 AM

Leave the timings set to auto. Pull your SB Live card and disconnect the slave drive you are not using. Leave the on-board audio disabled and try a boot here with the bare minimum. Check device manager for conflicts. Worse comes to worse borrow an AGP card and try it from there. On-board Video is always a hassle because you have to share your ram with everything else that needs it. You should have either 32mb or 64mb allocated in Bios. Try eliminating problems one at a time.

Abit IS7 - 2.8C @ 3.4 - Mushkin PC4000 (2 X 512) - Sapphire 9800Pro - TT 420 watt Pure Power
Samsung 120gb ATA-100 - Maxtor 40gb ATA - 100
Sony DRU-510A - THAT'S MORE LIKE IT!
a b } Memory
July 25, 2004 3:32:51 AM

Try underclocking the RAM to DDR266 (PC2100 speed), to match the bus speed of your processor.

<font color=blue>Only a place as big as the internet could be home to a hero as big as Crashman!</font color=blue>
<font color=red>Only a place as big as the internet could be home to an ego as large as Crashman's!</font color=red>
July 25, 2004 12:55:40 PM

It seems to me to be a memory controller issue with the Hynix HY5DU56422, HY5DU56822 and HY5DU561622 series 268, 435 and 456 bit DDR SDRAM chips @ 166MHz. Mine is the J grade chip with CL2.5. The timings provided on the datasheet from Hynix is 7-3-3-2.5!

I'm led to beleive so, since initially, I had thought that there could be a problem with my DIMM. I swapped my DIMM at the dealer for a brand new DIMM (of the same specs) just to check! The new module again gave me the same old problems...
KMODE_EXCEPTION_NOT_HANDLED
PAGE_FAULT_IN_NON_PAGED_AREA

I had also heard that the nForce2/Crush 18G memory controller had a few "minor" problems with "some" dual-sided memory sticks. Mine is a 512MB double sided module. Oh! What a coincidence!

Though it sounded ridiculous at that time, I had reasons to beleive that "theory". This is because, when I tried my friend's DDR400 256MB x 2 sticks (also Hynix), I could play Hidden & Dangerous 2 as well as NFS UG for a "long" while. It was a single sided module and was running at 400MHz, 8-4-4-3.

Then...
I inserted my DIMM and reset the memory timings/speed in the BIOS settings to Optimal/SPD.... Hmmmm...

Down the drain again!

Talking about falling into an ocean of sh**... I did a backup of my HDD, checked it with SeaTools, zero-filled it and repartitioned it. Now even WinXP won't install! During the file copy, setup halts with a STOP: 0x00000B8; A wait operation, attach process, or yeild was attempted from a DPC routine.

It also seems that my system vendor is clueless about this "particular" "issue"! Somebody gimme a gun!
July 25, 2004 1:05:53 PM

I had tried down-clocking with my earlier DIMM. There were very "few" problems/BSoDs after that. I tried Memtest86+ with the earlier DIMM and it showed up a sh** load of errors! Unfortunately, I exchanged that for a brand new DIMM and now I'm screwed more than ever! :-(

Now even WinXP installation halts with a STOP:0x00000B8!

The really weird part is that I've been running Memtest86+ v1.2 for 3 days continuous now! Not one error! Not ONE error!!!

Man! This is driving me up the wall!
July 26, 2004 12:10:58 AM

I ran into a similar problem w/ n/force 2 IGP board
check in bios if there is a "memory precharge" option
disable it if there is
July 31, 2004 8:40:19 AM

I tried setting all memory settings to Auto and not by SPD.
The board now runs the DDR333 RAM at DDR266 with imings set toa weird 6-3-3-3! Everything's fine now! XP and Suse 9.1 are working fine now!

So who's to blame?

The memory controller or the DDR RAM?

A positive attitude may not solve all your problems, but it will annoy enough people to make it worth the effort.
Herm Albright (1876 - 1944)
July 31, 2004 3:44:40 PM

I've seen that happen before! Definitely the board, but, if it runs good what the heck!

Abit IS7 - 2.8C @ 3.4 - Mushkin PC4000 (2 X 512) - Sapphire 9800Pro - TT 420 watt Pure Power
Samsung 120gb ATA-100 - Maxtor 40gb ATA - 100
Sony DRU-510A - THAT'S MORE LIKE IT!
August 4, 2004 7:10:00 AM

So the question here is...

1. Is this behaviour a problem with the way BioStar has designed their mainboard around the nForce 2 400 chipset?

OR

2. Is it a problem with the Hynix RAM Module/Module Series?

It is pointless to run your memory at a higher clock speed than your FSB. Nevertheless, if the God damn RAM module states FSB333 then it God damn work at FSB333 or you flush it down the toilet or shove it up the dealer's <wherever>!

A positive attitude may not solve all your problems, but it will annoy enough people to make it worth the effort.
Herm Albright (1876 - 1944)
August 6, 2004 1:32:45 AM

Mushkin PC4000 uses Hynix D-5 modules which I have. I could not get them to run in a MSI-875Neo board at anything other than DDR-333. I could clock them up 5:4 to 193mhz, but that was it. One stick ran perfectly, but if I put in two, back to 333mhz. End result, no more MSI board. Nobody ever could figure out why.

Abit IS7 - 2.8C @ 3.4 - Mushkin PC4000 (2 X 512) - Sapphire 9800Pro - TT 420 watt Pure Power
Samsung 120gb ATA-100 - Maxtor 40gb ATA - 100
Sony DRU-510A - THAT'S MORE LIKE IT!
!