VS 1GB Kit Cas2.5 issues

[McDoul]

Been having issues since upping from VS512DDR400 Kit to VS1GBDDR400 Kit.

 

BSOD's at random times.

 

Threads were active in Hexus, PCPer, XPForum, and god know where else leading to upping the voltage on mem a tenth - no good got worse - upping voltage on CPU - no good got worse.

 

If I took a stick out all was fine - problem is it applied to either stick. Each stick ran fine in each Dimm.

 

If I put the old memory back in (thanks to kind friend I sold it to! :laughing: ) it works fine too.

 

Memtest 86+ passes consistently - including repetitions of test 5 all except once when several errors showed up but not on subsequent hour of continued testing.

 

MS Mem Daig passes consistenly - including repetitions of test 6

 

CPU Burn-In runs just fine too passes without error.

 

Replaced Antec Smartpower PSU when PC Per article highlighted Asus A7N8Xe Deluxe issues with CPU drawing power from other rail -my volatges were quite bouncy and the BSOD messages wildly varied blaming everything from AGP to Sound to Kernel to Modem.

 

Slight improvement when memory reseated, but case and Dimms are clean - BSOD resumed soon after.

 

Temps are good.

 

All combo's of drivers tested with thorough cleaning.

 

Several Reinstalls (thank god for True Image and a clean install :sunglasse )

 

Installed Tagan TG480-U01 480W PSU in all its glory :D: - and certainly more stable but....

 

Still periodic BSOD

 

Will NOT last Prime 95 for more than 10 minutes... lorrd knows why tho.. not exactly explanatory is it? :confused:

 

Finally I discovered a gaming thread where a similar system to mine was suffering CTD's/BSOD's with event log message:

 

"The description for Event ID ( 0 ) in Source ( ) cannot be found. The local computer may not have the necessary registry information or message DLL files to display messages from a remote computer. You may be able to use the /AUXSOURCE= flag to retrieve this description; see Help and Support for details. The following information is part of the event: 6."

 

Said message pops up ALL over the web, in many, many, applications, and as yet I've not seen a single software/app developer find the cause in their apps.

 

I only get it when gaming :mad:

 

Can be FM2005, BF1942, Rome TW, Cricket 2004, Colin Mcrae 2005.... so not tied to an individual app..

 

...said thread contributor had same problem with same brand of memory as mine... he had it replaced and problem went away?

 

Is my memory shagged?

 

System:

 

Athlon XP 3200+ 'Barton'

Gigabyte Rocket Pro Cooler

Asus A7N8X-E Deluxe nForce2 DDR400

1GB Kit Corsair VS PC3200 DDR400

WD Caviar SE 120GB SATA150

Seagate Barracuda 80GB SATA150

Sapphire Radeon 9600

Vantec Silent 120mm Intake And Exhaust

Antec SLK3700BQE

Tagan TG480-U01 480W

 

BIOS at defaults in regard to CPU & Memory....

 

Help?

 

*Now sits cringing in the knowledge that if an RMA is required he has no flaming PC for weeks and Overclockers.co.uk are in danger of a really, really, cross letter :(: *

[RAM GUY]

Please test the modules one at a time with http://www.memtest.org and lets make sure one is not failing.

[McDoul]

Nope.

 

As before the tests were fine with each stick in each slot one at a time showing no errors.

 

This is with both Memtest & Windiag

 

Naturally I've had to do this several times now and it's only the once that it found a *****load of errors.

 

Oddly, the crashes go away when using a single stick, perhaps these don't like DC.... Scan Customer Care tried telling me they weren't comaptible with either my MoBo or DC :confused: .... odd 'cos you guys confirmed by email that they were. (Scan = Poo :p: )

 

Timings are 8-3-3-2.5 by the way...

 

Prime still fails.

[McDoul]

Also,

 

Having addressed the failing Prime95 issues with them, the mods there advised me to run the torture test with 1024k FFT to retest the RAM and see when it failed - 16 minutes.

 

Then to rule out other components I've been asked to try the small torture test as if that works then I can assume the memory is at fault.

 

The small torture test for Cache, Bus, and CPU stressing works just fine for well over 2 hours.

 

Incidentally, many of the people in so many forums seem to have Value Select PC3200 issues..... any thoughts? I'm just a little concerned that any RMA might still give me problems.

[RAM GUY]

Please try these settings:

With the A7N8X-E please set the FSB jumper to pins 2-3 with the modules in slots 2-3 and then reset or clear the bios and load setup defaults and set the Dim Voltage to 2.7 Volts and set the timings manually. Then please test the system with http://www.memtest.org.

CPU Freq: 200 MHz

Memory Frequency: 100%

Dim Voltage to 2.7 Volts

Resulting Frequency: 200 MHz

SDRAM memory timings : User Define

SDRAM CAS Latency: 2.5T

SDRAM RAS to CAS Delay (tRCD): 3T

SDRAM Row Precharge (tRP): 3T

SDRAM Active to Precharge Delay (tRAS): 8T

 

I have 4 of these exact MB's and this is what had to be done to solve problems with aggressive memory, no matters who's I tested. In addition, a few other's had some success disabling Spread Spectrum in the bios as well. I would take a look at http://www.asusforums.org as well in the AMD section.

[McDoul]

...and another BSOD this morning... nothing doing... just idle desktop and Outlook open in systray.

 

PC had been on for about 10 minutes.

 

No error message.

 

No Minidump

 

...and sod all in the event log.

 

Ran Prime and lasted 5 minutes on Large RAM Torture Test before error/halt, and has subsequently run Small CPU/BUS/Cache Torture test quite happily for the last 3 hours.

 

Is there anything else I can check?

[RAM GUY]

I posted right before you, please see my previous post and try those settings.

[McDoul]

....Oops we crossed over. [edit: Damn - Twice now - I'm gonna stop typing! :D: ]

 

I'll do as asked and feedback.

 

Thanks for the response... much appreciated.

[RAM GUY]

No worries, please let me know how you make out!

[McDoul]

Damn. :(:

 

Did as requested....

 

Jumper moved to pins 2-3

modules in slots 2-3 (The Blue DC ones)

BIOS Clear & Set To Default

DIMM @ 2.7 Volts

 

CPU Freq: 200 MHz

Memory Frequency: 100%

Resulting Frequency: 200 MHz

SDRAM memory timings : User Define

SDRAM CAS Latency: 2.5

SDRAM RAS to CAS Delay : 3

SDRAM Row Precharge : 3

SDRAM Active to Precharge Delay: 8

 

Memtest works fine for an hour.

 

Prime still fails on Large TT tho 2, 10, 17, and 40 minutes respectively, despite Small TT passing for over an hour again. (Twice)

 

I have a feeling the Blue Screening could be easier understood if someone had an explanation for this:

 

"The description for Event ID ( 0 ) in Source ( ) cannot be found. The local computer may not have the necessary registry information or message DLL files to display messages from a remote computer. You may be able to use the /AUXSOURCE= flag to retrieve this description; see Help and Support for details. The following information is part of the event: 6."

 

This event log entry never occurred under my old VS512MBKIT400 memory, which is now running well for a friend, just since installing the VS1GBKIT400... been going on since mid january now... dunno what eles to try.

 

It has improved massively from the frequency of BSOD/CTD since upping the PSU significantly in March but still continues to do so enough to rend the PC pretty lame.

[RAM GUY]

Well If the system is stable in all other respects I would not worry about it. But Prime 95 or the makers of that software will need to address that question. And anything from a driver to a weak PSU can cause that type of error.

[McDoul]

That would be the problem.. it's not stable in all other aspects... it keeps crashing to desktop or, as I say, less now than before, BSODing.

 

Currently preventing reliable use of Outlook, BitComet, Nero, WMP10, Pro Evo 4, Colin Mcrae, Rome TW, FM2005, CM0304, Cricket 2002, BF1942. :mad:

 

I'm gonna borrow back my old 512 VS Kit fro a mate and see if I can reproduce the problem.

 

both PC's have identical revision MoBo's, GPU's, CPU's, and at one point whilst troubleshooting even Drivers. We even use the same games.

 

I'll feed back when I've swapped the memory round.

 

I know it's not the Graphics Drivers - I've tried them all ( :mad: ) so I'm gonna look into the NVidia Drivers, that said, if it was the nForce2 drivers, my friend would have them too I'm sure, so I doubt I'll find much.

 

Thanks for your advice thus far.... If my memory crashes in his, and his is fine in mine can I assume its the memory?

 

There seem to be a lot of people who put little stock in the absolute reliabilty of Memtest or Windiag anyhow... so the passes in those tests mean little to me anyway. As said before, they've failed the same meory once, just not since.

[RAM GUY]

We can try replacing the modules if you like.

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

 

There seem to be a lot of people who put little stock in the absolute reliabilty of Memtest or Windiag anyhow... so the passes in those tests mean little to me anyway. As said before, they've failed the same memory once, just not since.

 

I would have to disagree, like with any software you have to understand or know how to interpret the results and diagnose the problems to fully understand this type of problem. Its not that cut and dry sometimes! And if you are passing memtest and Windows based memory tests but still having this type of problem then its most likely some other cause. Like for example; a weak or marginal PSU can cause this type of problem as well as a PCI card in the wrong slot to a conflicting driver.

[McDoul]

Ok.

 

I'll see what happens with swaps first.

 

My PSU is a 480W Tagan BTX compatible with ***-kicking rails... it was the first thing that got replaced whilst initially troubleshooting.

 

Cheers.

[RAM GUY]

No problem, please let me know!

[McDoul]

Long informative post!!! You have been warned. I apologise in advance, but its all about the facts.

 

So, its January 2005 and I'm running my own build flawlessly since November, the Spec being:

 

Athlon XP 3200+ 'Barton'

Gigabyte Rocket Pro

Asus A7N8X-E Deluxe

VS512MBKIT400 (2x256)

NVidia GeForce MX440 64MB

Seagate Barracuda 80GB

Antec SLK3700BQE & 350W Smartpower PSU

CDROM & CDRW/DVD Combo

120mm Front & Rear Vantec Silent Fans.

 

All is well.

 

I then build an identical PC for a mate with him taking the VS512MBKIT400 off my hands as I wanted to upgrade to a Gig.

 

This leaves me with the following build.

 

Athlon XP 3200+ 'Barton'

Gigabyte Rocket Pro

Asus A7N8X-E Deluxe

VS1GBKIT400 (2x512)

Sapphire Radeon 9600

Western Digital Caviar SE 120GB (In prior to rebuild as Seagate died)

Antec SLK3700BQE & 350W Smartpower PSU

CDROM & DVDRW

120mm Front & Rear Vantec Silent Fans.

 

So only changes from stable PC were the Mem, the DVDRW, and the GPU.

 

About a week after the build I start getting BSOD having reinstalled RoN to replay it on larger scale and settings. The only other things installed at this time are RTW and CM0304, and H&D.

 

I assume RoN is the problem as there appear to be errors throughout the install of the game, and it keeps crashing with the following error message:

 

"The description for Event ID ( 0 ) in Source ( ) cannot be found. The local computer may not have the necessary registry information or message DLL files to display messages from a remote computer. You may be able to use the /AUXSOURCE= flag to retrieve this description; see Help and Support for details. The following information is part of the event: 6."

 

I uninstall, and the BSOD's continue. Error messages ranging from wmaud, nvmcp, ati2mag, alc...modem thingy, and all manner of Memory, Graphics, and hardware related errors.

 

I swap the old NVidia card back in to see if the Radeon is causing the issue. Very carefully, full driverclean and all that, even a reinstall at one point.

 

Still BSOD'd.

 

Tested each and every NVidia nForce2 driver (My original stable 4.27 onward) with each and every RadCat 4.12 and up. ( :mad: ) Still not right.

 

Eliminated the nVidia 5.10 IDE lock up from the equation by leaving out the IDe Driver... no blame there.

 

Targetted the Memory. Does it crash with a stick removed? No. Swap the sticks, still fine. Swap slots. Still fine.

 

Test with WinMemDiag fine, Test with Memtest86+ fine too except one test which produced a shedload of errors. Oddly, the errors didn't show up again when I rebotted and left it for another hour. Prime wont pass the Large FFT torture test.... Prime forums indicate I should check for dust. A can of compressed air and I have an intimate hour in the crevasses of the MoBo... still BSOD, still silly unexpalined EL message, and still no Prime.

 

Next? PSU. Rails are weak, and altohugh fluctuation is not massive the rails are not up to an AMD chip on an NVidia board, and certainly not with a Radeon 9600 suckling on thy AGP teet.

 

So, in comes the Tagan TG480-U01 480W (with rails for a bullet train). I mean, HELL, this thing is getting some good reviews.

 

..and Lo, the BSOD's come to an end..... more or less.

 

Clearly there was more than one issue, the PSU needed upgrading, but that wasn't the only issue.

 

Still getting the odd blank one, and simply cannot play any games without, you've guessed it, the EL message that despite trawling over a thousand threads via Google, no-one has the answer to...except one lad who came back to a thread six months after it died to say he'd been getting EXACT same error message as me, was using VS1GBKIT400, RMA'd it, and never saw the issue again.

 

Odd thing is, others have claimed to have fixed it, but I think it's more PC specific than that. There are no common denominators in all the specs reviewed amongst the threads I've read.

 

I've used Prime on the small FFT test to stress CPU/BUS/Cache and it runs quite happily for the over 3 hours. It wont run large FFT to test the Memory tho for more than about 20 minutes.

 

I have a thread in the Corsair Mem forum here and have tried a couple of tricks recommended by RAM GUY to no avail.

 

An RMA has, in all fairness against the evidence of memtest been offered, and I'm waiting for my mate to return with my old memory to do a bit of swapping.

 

In the meantime, I'm sat watching disbelievingly as Prime runs a large FFT torture test on a single DIMM in slot 3. Was just a hunch you understand, and I'm obviously gonna need to run this on the other DIMM and the other DC slot with each DIMM but so far I'm intrigued.

 

Its working! :eek: Has been for an hour while I type this into a summary of the 50 or so posts in several forums begging for help in the first place.

 

My Temps are fine. My voltages are damn solid with a light fluctuation of insignificant proportion on the 5v.

 

My question is where do I go from here?

 

If both DIMM test fine in both slots individually what does it mean?

 

If they fail in 1 slot whats with my MoBo?

 

If one of them fails, is it faulty? Even tho Memtest & Windiag dont find nought? I think it must be. I've tried every driver combo under the sun, I've swapped PSU over, GPU out and in again, I've disabled and unplugged the DVDRW as it was the only other newcomer.

 

Prime is the only thing that stresses the memory to the level of a large game, and it keeps finding flaws.... until now, knock wood, crossy fingers, an' all that.

 

Any thoughts or related experiences appreciated....

 

(God the tension waiting for this test to fail on me... gonna nudge me knickers in a minute! :D: )

[Wired]

Testing both in slot 3 is a good idea. IIRC, try Dual Channel in slots 2 and 3. That mobo is wonky. Have you updated to the latest BIOS? Also, try bumping up the (Vcore I think, or whatever the agp voltage one si) from 1.5 to 1.6.

[McDoul]

I've got the DIMMS up from to 2.7 fro 2.6 default to no avail.

 

I've also upped the Rad this morning from 1.5 to 1.6 and still get the errors and crashes.

 

Mem has always been placed in the two blue slots which are what I perceive to be 2 & 3.

 

Asus manual labels them as A1 and B1, both Blue, with the solitary black slot labeled A2 closest to the CPU Socket and NorthBridge.

[McDoul]

Ok.

 

All done, and now any advice anyone has would most welcome.

 

2 DIMM's 676 & 677 or 1 and 2 respectively on the Asus A7N8X-E Deluxe.

 

DIMM 1 & 2 run fine in Slot 3 (Furthest from CPU/NB) for 2 hours plus. No Prime95 failures. No Game Crashes.

 

Both DIMM fail Prime in Slot 2 within 5 minutes.

 

Both DIMM fail Prime in Slot 1 within a minute. DIMM 2 invites a BSOD.

 

DIMM's in slots 2 & 3 Dual Channel, can game for short period before CTD or occasional BSOD. Fail Prime in 5 minutes. Notably system seems terribly sluggish with both DIMM in Dual Channel despite default and RAM GUY recommended set-ups.

 

DIMM's in Slots 1 & 3 cause BSOD within seconds of launching desktop. Tried again Prime failed in 1 minute or less. Not risking any gaming.

 

DIMM in Slot 1 & 2 No DC = BSOD and Prime fail in under a minute.

 

This means only current stable set-up is either DIMM in slot 3 single channel, suspiciously it still feels faster than both in DC.

 

So,

 

Any idea's whats wrong?

 

If it's the slots. Why? They were fine before the upgrade of memory and are free from dust.

 

Are there any test configurations I've missed?

 

...and finally, if my board needs replacing I'm screwed. Will Asus advance RMA? If not I'll need to buy a new board and then flog the RMA'd one on Ebay or the like. What's the best thing I can bget for my build? Originally when building the best reviews for DDR400 DC with AGP were the Abit NF7-?? but there were issues over exactly which version was best, and my current Asus Deluxe. Any recommendations would be fine.

 

Finally. Does this prove my RAM itself is Ok?

 

Thankyou to those of you with patience to get this far, it's much appreciated, I've almost reached the end of my tether over the last 4 months. :(:

[McDoul]

Wow. :bigeyes:

 

My old memory back from Mate (VS512MBKIT400) and swapped PC's.

 

Memtest, as said before, passes the memory in both machines, but since the memory is testing fine but not running fine I'm resorting to testing for stability with Prime95.

 

Now, the only way my memory works, having covered ALL combo's on default settings is with a single stick in Slot 3, and, oddly, the PC feels quicker with the half gig in Slot 3.

 

As unexpected, (beginning to believe my MoBo was at fault), my old memory passed with flying colours in my new PC, stable as a rock and feels quicker than the Gig did!

 

My Memory in his PC? Failed. Less than a minute. Crashes Apps/Games

 

His Memory in his PC? Passed. Over 2 hours. Fine.

 

I know 2 hours isn't exhaustive, but I know when it fails, and right now with failures taking place within 5 minutes anything lasting 2 hours is good enough to help spot the clear problem with what I now think is the memory again.

 

Sooo.

 

What do you guys think? My Mem works nowhere (Except singlky in Slot 3, my old Mem works everywhere and that must mean my Memory cannot run in Dual channel, Yes?)

[McDoul]

Wow! That explains it.

 

Got my old memory back from Mate (VS512MBKIT400) and swapped the DIMM's over.

 

Memtest, as said before, passes the memory in both machines, but since the memory is testing fine but not running under load fine I'm resorting to testing for stability with Prime95.

 

2 DIMM's 676 & 677 or 1 and 2 respectively on the Asus A7N8X-E Deluxe.

 

DIMM 1 & 2 run fine in Slot 3 (Furthest from CPU/NB) for 2 hours plus. No Prime95 failures. No Game Crashes.

 

Both DIMM fail Prime in Slot 2 within 5 minutes. both cuase CTD's.

 

Both DIMM fail Prime in Slot 1 within a minute. DIMM 2 invites a BSOD first time in but seems fina after altho wont pass Prime.

 

DIMM's in slots 2 & 3 Dual Channel, can game for short period before CTD or occasional BSOD. Fail Prime in 5 minutes. Notably system seems terribly sluggish with both DIMM in Dual Channel despite default and RAM GUY recommended set-ups.

 

DIMM's in Slots 1 & 3 cause BSOD within seconds of launching desktop. Tried again Prime failed in 1 minute or less. Not risking any gaming.

 

DIMM in Slot 1 & 2 No DC = BSOD and Prime fail in under a minute.

 

This means only current stable set-up is either DIMM, singly, in slot 3 single channel, suspiciously it still feels faster than both in DC.

 

That suggests to me that both DIMM's are fine and the MoBo is at fault, however....

 

As unexpected, beginning to believe my MoBo was at fault, my old memory (VS512MBKIT400) passed with flying colours in the default DC slots (2&3). Stable as a rock and feels quicker than the Gig did!

 

My Memory in his PC? Failed. Less than a minute.

 

His Memory in his PC? Passed. Over 2 hours and no crashes. He has has BSOD's in his event log in the past but know thats 'cos the SmartPower 350 5v rail isn't up to the combo of Asus and Athlon, let alone with an ATI card suckling on its AGP teet.

 

I know 2 hours isn't exhaustive, but I know when it fails, and right now with failures are taking place within 5 minutes anything lasting 2 hours is good enough to help spot the clear problem with what I now think is the memory again.

 

Sooo.

 

What do you guys think? My Mem works nowhere (Except singly in Slot 3), my old Mem works everywhere and that must mean my Memory cannot run in Dual channel, Yes?

 

But doesn't explain why although one stick is currently running well as tested in Slot 3, neither stick will run on its own in Slot 1 or 2.

 

Any advice, should I be looking to change both Mem & Mobo. Doesn't make sense too if my old memory runs like a dream in DC on my new PC, the fact that it runs fine in mine rules out the MoBo doesn't it?

[McDoul]

Oh, and now it appears that although corsair recommend the VS for ther MoBo.... it's not on the Asus list. Do I trust their listing or yours?

[RAM GUY]

McDoul

We do not submit our Value Select for Validation to any MB maker that is why it's not on ASUS's list. And if you are still having problems we can try replacing the modules for you, but it sounds like there maybe some other problem, like maybe a failing CPU. But lets try replacing your modules or if you can test your memory in another system if you have not done so that may help isolate the memory.

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

[McDoul]

Thanks.

 

The CPU is ok.... my memory fails in mates PC, and his works in mine.. thats the long and short of it after lots of testing.

 

I'll sort the RMA form and hopefully it'll sort it.

 

Thanks for your help.

[RAM GUY]

No problem, please let me know how you make out.