Tech Support Forum banner
Status
Not open for further replies.

Wanna Slam Ram

4K views 28 replies 6 participants last post by  Tyree 
#1 ·
Hello TSF,

Glad to see this site is still going strong,since my last conundrum as many years ago.

Hoping the wizards here can help solve yet another issue I'm dealing w/.

Recently built a PC w/ the main purpose of being a audio/vid./img. encoder/archiver.

Build was flawless,everything was like clockwork,smooth OS/driver install.

Which was a pleasant surprise considering my inexperience w/ SSDs, the fact i used some "budget" components.

So pleased was i, decided to forgo some of the burn-in test procedures memtest/prime95 etc.

Week or so goes by zero issues,and then dun! dun! dun! get my first program error..

No biggie or so i think,days pass then few more random program crashes/errors.

Off to the logs to see what's going on,find a few "common" errors mem/registry handle leaks.

Begin the process of investigating the problems driver/program compatibility etc.

Then..BAM! 1st BSOD get the not_less_than error,after a moment of Microsoft ranting i reset.

Boots clean,*sigh of relief* back to the logs, standard power kernel error but nothing "new".

No program errors just prior to the BSOD, still yet i go about cleaning up the OS.

Removed recent programs,cleaned up loose files,SFC command,ok maybe that fixed it?

30 minutes later..again BSOD different error this time..i forget what it was..

From there was all down hill, reliability floored..when it wasn't BSOD it was programs crashing.

Called a friend, looking for some SSD insight he say's "sounds like RAM to me".*facepalm*

Knowing i hadn't done a single hardware test, despite the fact my board came w. memtest integrated.(one of many reasons i chose it)

Sure enough errors start rolling by..out of 4 modules 1 made it past the first pass w/o errors.

Contact Patriot support,tech agrees it's bad RAM, issues RMA..

New RAM arrives..won't be so hasty this time, 4 passes each.(which takes foorever w/ 4GB modules)

Zero errors I'm happy ,all installed i boot it up..clear the logs and wait.

Beautiful no errors in 24 hours of normal use,time to do some encoding...

Like any good horror story, just when you think it's over..BAM! BSOD.

If i had any hair i would have pulled it out at that very moment,square one re-testing RAM.

Surprise surprise..insta-fail this time on all 4 modules,it's "budget"Ram but ***..

Contact Patriot again..this time the tech informs me "that RAM is untested on AM3 chipset" recommends i buy some new tested RAM.

OK, i know these modules wasn't on the boards approved list, nor was any at 4GB density.

The MB product page said it supported 16GB @ this RAM's Specs DDR3 1600 1.5v yada yada...

Honestly can't believe there is DDR3 floating around that wasn't tested on AM3 chipset,particular AM3 boards sure..but not entire AM3 line .

Whatever..the tech did issue yet another RMA , should be here in a few days..

My question is..what would cause RAM to degrade to failure so quickly?

Been around and around Google most say "cheap RAM" Heat or PSU.

Cheap RAM: sure at 200$ for 16GB..i was asking for it,but chances of 8 fail modules? Quality Control?

Heat: non issue, this rig ran like a ice cube I'm a stickler about temps, monitored w/software,bios,IR thermometer.

PSU: maaaybe..it's cheap also,i tested w/ multimeter w/o load today..all tested within 5% tolerances.(i think..1st time testing this way)

Thanks for reading my nightmare novel! feel free to tell me it was fluke and everything will be ok soon lol.

Build List.


MainBoard: BioStar TA870+ Ver.5.2

CPU: AMD PhenomIIx4 965 B.E.@ stock 3.4Ghz

RAM: Patriot Signature 240-pin DDR3 1600 16GB (4x4GB)@ Stock time/Voltage

SSD(OS): Crucial RealSSD C300 64GB SATA III

HDD(swap): WD Caviar Blue 250GB SATA III

VGA: HIS Radeon HD 4670 1GB PCI-E x16 @ Stock

PSU: Thermaltake TR2 500w 12v v.2.2 ATX

CPU Cooler: XIGMATEK LOKI H.D.T. SD963

Case: Thermaltake V3 Black Edition MidTower

Case Fans: 3x Scythe "slipstream" 120x120x25mm
2x Scythe 120x120x12mm

Fan Control: 1x Sunbeam Rheosmart3 30w
2x XIGMATEK Mono-Cool

OpticalDrive: 1x Asus combo SATA

OS: Win7 Home Prem.OEM
 
See less See more
#3 ·
Did you run MemTest on one stick of RAM at a time? I have found Patriot RAM to be very particular about compatibility and especially on lower tier Mobo's.
Your PSU is not the best quality and we suggest a minimum 550W good quality for a PCI-E GPU.
What are the Error Codes you see on the BSOD?
 
#4 ·
@KD5EEJ350
Yes, i am fairly certain RAM was running @ 1.5v.
The Bios load screen,Toverclocker utility,AMD software, all reported stock speed/voltage.
@Tyree
Yes,i did run MemTest one stick at a time, also in parity and as a whole.
Admittedly i did not run the 7-10 passes recommended as by the 2 pass there was hundreds to thousands of errors. Went as far substituting the VGA w/ another, fearing some of the errors maybe resulted from it's GPU.
Have had a lot of luck w/ BioStar MB's,tho always paired w/ G.Skill RAM.
This Board has alot of nice features,decent layout,even liked the color scheme.
Only complaints i have w/ them is the bios layout/function naming is anything but standard. "Particular" is a kind way to describe this RAM,never tested on the entire
AM3 line. Knew i was pushing it w/ the 500w Thermaltake,my reasoning was.
Tiny SSD handling the bulk of R/W,VGA that didn't require dedicated power,case fans all "leashed" on controllers,HDD only used intermittently.The only real concern i had was the pile of RAM plus high watt CPU.That said, will start shopping for a beefier PSU along w/ some G.Skill RAM. As for the BSODs the first was something like NOT_MORE_LESS error, honestly don't remember the many after that.Just after it happened i began researching the error,downloaded some BSOD/crash logging software from here or sevenforums,before i could so much as go over the readme for install..

Anyhow, thanks guys for the input, kinda knew it was a case of gambling w/ cheap RAM/PSU. Off to NewEgg to make my CC beg for mercy =/
 
#6 ·
@ wrench97
They where running @1600 stock speed.
Will absolutely give that a try when the replacements arrive,given I'm stuck w/ them regardless. Should i do that before i begin mem testing them? Will the Bios detect there stock speed and reset every time i pull and replace? Already decided to order some approved RAM,of course after i thoroughly harass both MB/RAM techs for there blessing.
If I'm able to adjust this RAM to work,I'll just save the new stuff for yet another build.
Thanks for the ray of hope 200$ is nearly two tanks of gas these days lol.
 
#8 ·
Thanks Wrench,
After reading your suggestions i looked around the net, as it turns out..
AM3 natively supports 1333,lot's of people having same issues with 1600.
Why the first RMA tech failed to mention this i do not know,why the second tech didn't i do know..Why give a simple fix for free,when you u can sell them another more expensive piece of hardware w/ the same problem.
Sadly, some guy on Patriot's forum w/ same kit/issue but i am unable to pass on this knowledge as there forums are as buggy as there techs.
Can only hope he finds TSF, where free tech support is far superior to the stuff you paid for..
 
#10 ·
Ok..So part two of luvv's long running story of the nightmare build..

In part one i had RMA'd the garbage Patriot RAM and was awaiting the shipment to arrive.
After getting some advice about my PSU being sub par,it was off to Newegg to purchase a PSU..CORSAIR|650W CMPSU-650TXV2 R to be exact.
While i was at it picked up 2X1GB G.Skill Model#F3-10600CL9D-2GBNQ RAM from the BioStar QVL.
And hey,what do i see but some other G.Skill RAM model #F#-10666CL8D-4GBHk
Yea, i got it, why not have more options..
So the Newegg shipment arrives ahead of the Patriot RMA shipment..big surprise there..
Off to the races, i go about swapping out the PSU, fits like a glove.
On goes the handy dandy static strap..first the 1x1xGB 10600 QVL RAM.
Clear the CMOS via the jumper method,gave the wiring a look over and power on.....NADA...NO Vid,NO Beep..nothing..
Oh,Wait ya the debug light is flashing, LED_1D while LED_2D is unlit..
Referencing the MB book it appears to be a RAM error,tho it doesn't mention flashing, only lit or unlit.
Whatever..i reseat, clear CMOS try again..again nothing but the debug light flashing..
OK, i'm not playing anymore ..moved the sticks to the secondary RAM slots A2/B2,clear CMOS..power on and..:upset: ahh a small change, the debug light LED_D1 is no longer flashing,just lit solid.
Time for clean benchtest,all set up reinstall RAM 1 stick.. 1st slot A_1..same thing flashing LED_1D/unlit LED_2D..
Swap,Clear,Swap,Clear..:sigh:
As it stands,any RAM 1 or 2 sticks in slots A_1/B_1 cause the flashing LED_D1/unlit LED_2D
Any RAM, 1 or 2 sticks in slots A_2/B_2 cause solid lit LED_1D/unlit LED_2D.
Tried the 10666 RAM,tho not officially supported had to try..same deal.
Have tried clearing the CMOS by removing the battery, thinking it maybe was a ram mismatch error of some sort..again no good..
Emailed BioStar,but in my experience with them its a 2 day response wait..
Now leaning towards shoddy MB as the cause of the original RAM deaths..
pretty soon will have RMA'd/Replaced the entire PC.....

To solve my issues please skip to PG. 84
To run away screaming please Skip to PG.96..

In all seriousness any ideas?
 
#12 ·
Thanks for the response Wrench,

At this point i wanna hop a plane to Cali,stand outside Patriot and shake my fist at them.
Depart LAX, on to Taiwan to do the same at BioStar...
Can a Main Board be considered carry on?

If anyone else has a unique knowledge of biostar's debug lights,or just say dude you should have got DELL..please feel free to add.
After all, the responses here will most likely come faster here then from biostar......................
 
#14 ·
GIGABYTE maybe...ASUS never...

Have a ASUS MB RMA nightmare story also,evolving a very expensive board with a bottom of the barrel SB fan.
On the third RMA the tech suggested i was letting something come in contact with the fan,instead of admitting it was a dime store fan as evident by the thousands of threads across the net.
In the end the solution was a $ 12 after market cooler,and negative one ASUS customer.
This was to be my 5th BioStar build,4 others so far as i know, are running flawlessly. This board, beyond price , had few negatives for me,a decent set of features,next to nill bad bad reviews.
Final selling point was seeing the benchtech guys push one over the wall o.c'n, and still remain stable.
That said,I'll give BioStar 3 swings same as ASUS, should they fail they will join ASUS MBs on the credit card blacklist.
 
#16 ·
You are correct, tho the reason i won't purchase there boards isn't so much the quality issue, it's the tech support insinuating it was somehow my fault.

My point also being, you can't always discern quality by brand name or price.
Just look at Toyota vs. Cadillac,a lot falls to plain old to luck of the draw.

A large % of my family worked at GM, they would all agree cars assembled on Friday where almost always less reliable then cars made on Tuesday..

All claims that one brand is better then another should have "should be" in them ie "ASUS should be better then BioStar"
 
#18 ·
Little update...got my BioStar response,

They more or less echo Wrench's response.

Memory Slot or CPU mem. controller,wanting me to test w/ a working AM3 CPU.

While i do have a several AM3 CPUs they are all installed in working rigs..and staying that way.
Going to try re mounting the CPU/HSF see if maybe its a issue of pin contact/debris.

If no change, will RMA both the CPU and MB...at this rate I'll be a honored guest at UPS/USPS.
Thanks again guys for the continued support
 
#20 ·
So are you ready for the possible end of this epic?

When we left off, i was going to attempt to remount the CPU and HSF.
And so i did,only during the process,discovered i was dangerously low on the AS5 supply..
Lucky for me i have tons of generic TIMs,so i start to remove the thumbscrews and notice there rather tight.
Lifting the HSF i realize i had applied the perfect application of AS5..don't happen often.
After a good cleaning, re applied, re mounted, this time not so firmly.
So as it turns out.. my most recent issue(boot failure)wasn't G.Skill's fault,nor Thermaltake's,nor BioStar's generic-ness.
The problem was between the chair and the keyboard...me.
My big dumb hands had over tightened the HSF mounting bolts,or i had somehow allowed a speck of dust in the socket at some point.

Memtest'd clean(2x1GB) Prime 95'd 1hr.- 50c ,little warmer then it ran before, but to be expected with junk TIM and looser HSF.

Either way stable for the time being,RMA RAM should be in Monday, will give it a go down-clocking it.
Side note: on my second tech support inquiry to Bio-Star, i mentioned RMA shipping charges where starting to give me hives.
To which he responded there was no need, he would accept a video of the symptoms of board failure, how flipp'n cool is that..go biostar =)
 
#22 ·
Honestly couldn't say what caused it, but the remount did in fact solve it.

This is the cooler Newegg.com - XIGMATEK LOKI SD963 92mm HYPRO Bearing CPU Cooler bracket included I7 i5 775 1155 AMD and dual fan push pull compatible

Picked it because it was ultra cheap,universal,fit my midtower,had decent reviews.
As a bonus newegg added a huge syringe of zigmatek TIM,love freebies even if they are nearly worthless.
After Memtest/Prime95, cued up a couple 4-6GB vid. files to convert and compress,flawless.
Although i did notice rather steep time increase w/ less and slower RAM,2GB don't cut it....
Maybe it was dust,tho i am aware of dust when i mount cpu so im quick about it.
Maybe i had it so tight it was distorting pins in the socket, or i had it so tight i was forcing a soldered point through the neoprene backing.cant say:4-dontkno
Going to order some AS5 and redo it when(if) my RMA ram gets lined out.
Maybe i will repeat the fault and see what's up.
Anyhow, thanks for hanging in with me,promise to stop typing out small sections of my life...just as soon as i get all 16Gigs of that junk ram working haha:grin:
 
#23 ·
the problem probably wasn't the quality of the asus board it was probably the fact that it was an nvidia chipset with an intel cpu in it. Those types of boards wether made by asus,gigabyte or msi etc were always prone to issues.
 
#24 · (Edited)
yes, you are correct about the chipset running crazy hot ,but it was a failing fan that was the issue i had.
It wasn't even a true HSF,just a small quarter sized fan inside of a tin cage,no aluminum, no copper,no kidding..had the words ASUS across the front of it.

But as i stated before..it wasn't so much the quality..i know these things happen after all.
It was a ASUS rep implying it had to be caused by me allowing objects to come in contact w/ it..3 times in a row are you kidding me?
He as a rep had to know those little fans where dropping like flies,all across the net was people complaining fan died..board fried
In the end i spent less to replace the HSF myself then i did on RMAs.
Brand to brand, model to model, version to version, heck day to day you don't know what you will end up w/until plug it in..up to that point it's all assumptions..
Oh ya wasn't a Intel CPU in my case was thousand dollar AMD FX60..still have it in the closet =p
 
#27 ·
Very possible,the neoprene is pretty thick, but not so thick it didn't concern me.

And the instructions where in quite possibly the worst broken engrish I've ever seen.

Coming from master of broken engrish such as myself,you gotta respect that.

When i seen how it mounted"that looks dangerous" did cross my mind,but if you knew how loud and annoying the stock AMD HSF was..

Had to risk it,my sanity was on the line....
 
#28 ·
RMA arrived yesterday,installed 8Gb of it today,set it to 1333 9-9-9-24 @ 1.5v..
Restart,started memtest, took my mother out for mums day, came back and.....ZERO ERRORS. *yay for me*

Not gonna install other 8Gb till i see how this works.
Maybe couple days after i get some AS5 and remount HSF and thoroughly inspect mounting back plate, I'll give the other 8 a spin.
soo far sooo good *crosses fingers*
 
Status
Not open for further replies.
You have insufficient privileges to reply here.
Top