Possible RAM Unstability?

[PCG520]

Hey everyone, I'm a new guy around here, so go easy on me. ;3

Well, here's my issue:

Recently I just built a new PC which has Windows Vista x64 Ultimate:

Intel Q9550@2.83 GHz (http://www.newegg.com/Product/Product.aspx?Item=N82E16819115041)
Zalman CNPS7500-ALCU (http://www.newegg.com/Product/Product.aspx?Item=N82E16835118035)
Gigabyte GA-EP45-DS3R (http://www.newegg.com/Product/Product.aspx?Item=N82E16813128344)
Crucial Ballistix Tracer 2x 2Gb DDR2 800 (http://www.newegg.com/Product/Product.aspx?Item=N82E16820148202)
MSI 4870x2 R4870X2-T2D2G-OC (http://www.newegg.com/Product/Product.aspx?Item=N82E16814127376)
Corsair CMPSU-750TX 750W (http://www.newegg.com/Product/Product.aspx?Item=N82E16817139006)
Western Digital WD5000AAKS (http://www.newegg.com/Product/Product.aspx?Item=N82E16822136073)
LG Black GH22NS30 (http://www.newegg.com/Product/Product.aspx?Item=N82E16827136152)
Linksys WMP300N (http://www.newegg.com/Product/Product.aspx?Item=N82E16833124069)


At first, it seemed everything was working, but now after two months, the PC has been getting numerous BSoDs (Blue Screens of Death) while doing anything from cruising the web to playing Left 4 Dead, Crysis, or Farcry 2. Also, I can keep reproducing BSoDs by trying to install World of Warcraft, as for once it starts installing a certain patch, it'll BSoD.

At first I thought it was the hard drive going because after trying to reinstall Windows, I couldn't reformat it, so I sent it back (I must have been on stupid pills because I just never deleted the partition and started over. Sorry Western Digital! :beer:) From there, everything seemed to work fine for about a week, then I continued to get this BSoDs. I checked the RAM by using MemTest86+, but got no errors after a 24 hour test. I checked on the Gigabyte website and found that my RAM wasn't on the supported list (Crap...). I know a couple people that have a P45 also and the exact same RAM, but their PCs run just fine. I investigated further into the speed of the RAM and the voltage, and I found a couple discrepancies (I've never really overclocked anything before, so I don't know much about it).

http://img179.imageshack.us/img179/9510/dsc01135fj0.jpg
http://img177.imageshack.us/img177/4562/dsc01136vl0.jpg
http://img523.imageshack.us/img523/6705/dsc01137se7.jpg
http://img218.imageshack.us/img218/1562/dsc01138jb0.jpg
http://img179.imageshack.us/img179/1105/dsc01139tp9.jpg
http://img186.imageshack.us/img186/6268/modramsx5.jpg

On Newegg.com, people have said that they have had to put their RAM up to 2.2v in order for it to properly work. What I'm wondering is why all the sudden I'm getting these errors, yet Memtest86+ didn't even find any :confused:. Should I just go into the BIOS and set the DRAM Voltage (pic 3) to 2.2v? Thing is, (pic 6) the RAM on CPU-Z has sections where it says 2.2V, so would I just be better off buying new RAM rather than possibly void my warranty by changing the voltage? Also, (pic 1) should I keep the RAM on 5-5-5-18 and on "Turbo" , or drop it to 4-4-4-12 and "Standard" like it is on the spec tab in Newegg's RAM specs? Right now I'm more focused on getting it stable than overclocking the RAM. Also, since Memtest86+ says there's no errors, yet it keeps crashing, could it be any other hardware problem that's causing this or could the RAM possibly be defective? Also I was considering just buying some pretty Corsair RAM (http://www.newegg.com/Product/Product.aspx?Item=N82E16820145184) instead of all these hassles, but would that even work with my P45 motherboard?

Also, (pic 6) this motherboard has built-in power-saving options which clocks the CPU (and maybe RAM?) down, so I'm wondering it that can be the problem. I can't find how to turn it off though in the BIOS.

If you need any more info, just post and I'll get right back. Right now this is my number one priority and I'll be checking on the thread almost 24/7. Thanks in advance for any kind of help.

[Priv]

Welcome to the forums!

If memtest does not find any errors there is no reason to believe that your memory is faulty. Did you check your error log in windows? And have you tried stress testing your CPU with prime95 yet?

Other than that you could try running the hard disk diagnostic tool of Western Digital. It is downloadable from their site.

If you can see the blue screen stop code, please tell us. If the blue screen provides any other information that could help us out alot more. Some BSODs are driver related and the blue screen will most of the times tell you what driver failed, something along the line of cdrom.sys or whatever driver it is.

[PCG520]

No, I haven't done a test with Prime95, but I'll get right on that.

Okay, I have a camera handy and am ready to snap a quick picture when it happens, which shouldn't be too long now.

[PCG520]

Okay, I just had a rare incident: I had two BSoDs, the first one different than the rest of them.

Usually, my BSoDs have been like this, (http://img246.imageshack.us/img246/8895/importants5aq7.jpg) but recently I just got this one. (http://img183.imageshack.us/img183/1870/importants3qf5.jpg) After I got that unique one, when I booted up Windows, it flooded me with things like this (http://img172.imageshack.us/img172/6338/importantsfx0.jpg) and this (http://img133.imageshack.us/img133/1738/importants2hc1.jpg).

I'll edit this post and upload more pics in a bit, just making sure I get this post up before I BSoD again.

EDIT: Okay, so I compiled a list of all the BSoDs I've gotten and put them into a .txt which can be found here. (http://www.megaupload.com/?d=YK69PRV7)

Note, Windows had me do a memory test, in which it said it failed, but Memtest86+ didn't. I'll Memtest86+ it again though and post the results on here.

[Trap05]

Try changing the sata cable to your hard drive. Run checkdisk or WD diagnostic tools after changing the cable. It could also be power related

[Priv]

I have a feeling that this is HDD or windows related. Unless memtest is bugged it should detect the errors if there are any. The first BSOD you got is what I always get when my computer is unstable and it crashes. This can be CPU, Northbridge or RAM related, so I still advise you to run prime95 blend test. And have you tried the WD diagnostic tools yet? We need to narrow down the problem, so please check your HDD.

Try disabling the Turbo setting in the bios, that could be the setting causing the problems. Even though I doubt it.

Another thing you could try is setting your RAM to 2.2V and have it run at its rated speed. DDR2-800 4-4-4-12.

[PCG520]

I'm really sorry I haven't gotten back to you guys yet, been really busy lately.

A bit more info that happened yesterday:

I was using this PC for about 30 minutes when it BSoD'd again. This time, the BSoD crashed (didn't even know that was possible...) and turned my PC off. I tried to turn it back on, but it wouldn't post. I panicked for about 4 minutes and tried to turn it back on. Thankfully, since I'm using the P45-DS3R, the backup BIOS chip recovered the main BIOS, then restarted and posted. From there, I haven't had any issues, but in the next couple hours, I'll change the RAM's clock speed and voltage, then run WD's HDD Analyzer thingamadoobob and Prime95.

[PCG520]

Okay, I just ran the WD test, and it came out with 0 errors. From there, I tried turning the voltage up on the RAM without dropping the timing, and that resulted in Windows BSoDing with a Memory error and then restarted. From there I just lowered the timing to 4-4-4-12 and now everything seems to be stable. No errors (excluding the "Your PC has shutdown unexpectantly"), but I've noticed one thing. Since yesterday, Mozilla isn't remembering to log me in automatically. Not sure if I'll reinstall Vista over it, but I might.

Running a Prime95 test now.

[PCG520]

Ran Prime95. When just running the regular scan, PC gets through with flying colors. When I do the blend test, 2 out of 4 of the workers fail, one immediately, and one after about 3 hours.

[Priv]

Try increasing the voltage on the northbridge, that is our last chance. If memtest86+ has no errors, and prime95 small fft has no errors, there is really only one option left, and that is the northbridge. How long did you run the small fft test?

[PCG520]

Well here's the weird thing: right after the BIOS reset itself, it seems the PC has been fairly stable. I've never BSoD'd yet, and the only one I could do was from incorrectly clocking RAM. Well, what kind of voltage increase should I do for the northbridge? I've never really overclocked at all. ^__^;

[Priv]

Is Prime95 blend test still giving you errors? If not, you shouldnt touch the NB voltage, if it is you can chance the voltage in the bios, find out what voltage it is at now with the auto setting and then work your way up from there. Trying one or two notches then test stability, if unstable go back to step one.

[PCG520]

Is Prime95 blend test still giving you errors? If not, you shouldnt touch the NB voltage, if it is you can chance the voltage in the bios, find out what voltage it is at now with the auto setting and then work your way up from there. Trying one or two notches then test stability, if unstable go back to step one.

Okay, thanks so much Priv, you're really saving my ass here, I really appreciate all your help, even to people that don't really know that much about overclocking. ;3

I'll post back with my results in an hour or so.

[PCG520]

Right before I was asbout to turn up the Northbridge, I BSoD'd, but this time there was no actual Bluescreen, it was just static. Booted up again and got this:

Product
Windows

Problem
Shut down unexpectedly

Date
1/16/2009 2:25 PM

Status
Report Sent

Problem signature
Problem Event Name: BlueScreen
OS Version: 6.0.6001.2.1.0.256.1
Locale ID: 1033

Files that help describe the problem (some files may no longer be available)
Mini011609-01.dmp
sysdata.xml
Version.txt

Extra information about the problem
BCCode: 4a
BCP1: 0000000074E63917
BCP2: 0000000000000002
BCP3: 0000000000000000
BCP4: FFFFFA60083A1CA0
OS Version: 6_0_6001
Service Pack: 1_0
Product: 256_1
Server information: 95d605e1-1435-471b-b265-0b74dc354ef1

Went to run Prime95, and the second I clicked on Torture Test, 3 out of 4 of the modules failed.

[Priv]

I'd still try increasing NB voltage. Otherwise it could be an OS problem. You'd most likely need to reinstall vista then.

[PCG520]

Well I increased the NB a couple times and that still resulted in instability.

Just out of curiosity I popped the Memtest86+ disc back in and I don't know how or why, but all the RAM errors just stockpiled. I don't know why they didn't even show up before, but now there's at least 350 or so in an hours worth of testing.

Time to go buy more RAM.

[Priv]

Your RAM is dead then I guess, the errors also turn op on 1.8V DDR2-800 5-5-5-15? If so, it's definitly dead.