Boot problems, Memory failure? HELP!

[Anzy]

Hey guys, having an odd problem with my rig last few days and I'm at a stand still right now. Basically I was getting stop errors (0x00000050, 7E, etc) and I tried to reformat,no go. So I was testing things and after I pulled my TV tuner card out I was able to finish the Vista setup but then the problems came back again. I ran each stick separately under memtest, each module failed miserably in less than a minute. I'm thinking thinking that either the memory is truly bad (Though this setup ran flawlessly since August), the motherboard DIMM slots died, the power supply is dying or the memory isn't getting enough voltage. I'm by no means an expert and these conclusions are only based off my limited knowledge.

Voltages:
VcoreA: 1.30 V
Vcore2: 2.03 V
+3.3V: 3.38 V
+5V: 5.89 V
+12V: 4.61 V
DDR Voltage: 0.13 V
+5V Standby Voltage: 5.12 V

These readings are from a third party boot disk that has a bunch of utilities on it as I was unable to locate the section in the BIOS that has these readings. Do those voltages look good? At the moment I'm quite lost so, please, any help would be greatly appreciated! Thanks!

System Specs:
mushkin 4GB(2 x 2GB) 240-Pin DDR2 SDRAM DDR2 800 (PC2 6400) Dual Channel Kit Desktop Memory Model 996564
GIGABYTE GA-G33M-DS2R LGA 775 Intel G33 Micro ATX Intel Motherboard
Western Digital Raptor WD1500ADFD 150GB 10,000 RPM Serial ATA150 Hard Drive
EVGA 640-P2-N825-AR GeForce 8800GTS Superclocked 640MB 320-bit GDDR3 PCI Express x16 HDCP Ready SLI Supported Video Card
Intel Core 2 Duo E6750 Conroe 2.66GHz LGA 775 65W Dual-Core Processor Model BX80557E6750
CORSAIR CMPSU-620HX ATX12V v2.2 and EPS12V 2.91 620W Power Supply

Thanks again!

[Anzy]

Update: I tested a different power supply and still had the memory fail, not sure what else to do now... It's either the motherboard or both memory modules... which do you guys think it is? Thanks!

[jjdog711]

did you try resetting the bios defaults, just pull the battery for a min or two if you can't get into bios. i had i similar problem and that fixed it.

[redduc900]

There may not necessarily be a problem w/ your RAM, DIMM slots, or PSU... but w/ the memory timings (or insufficient Vdimm, like you mentioned). If you can, write down all of your RAM timings and voltages in the BIOS, and post them in this thread. Also, some of the voltages you have listed don't jibe... VcoreA looks OK at 1.30v, DDR Voltage isn't right at .13v, +5v probably isn't right either at 5.86v (it's probably the 4.61v shown next to +12v; although that's lower than the +/- 5% tolerance), or the 5.12v shown next to +5v standby voltage, etc.)

Another thing you want to try, if you haven't already, is either resetting CMOS (like jjdog711 suggested) or loading Optimized defaults in the BIOS. After loading Optimized Defaults, only change those settings required to POST and load the OS without errors. Before trying to install Vista another time though, make sure your RAM passes Memtest first.

[Anzy]

Thanks for the replies! jjdog, I did reset the CMOS earlier today, pulled the battery for a good 30 minutes. redduc, I hope to God you are right, I was thinking it was something along the lines of problems with the voltages... the thing that confuses me though is that I had been running the system without many problems since August and all the sudden this stuff is happening. I just completely rewired my system thinking that something may have been loose, set BIOS to optiimized settings and in about 15 seconds through memtest I have a ton of errors,
Errors per Memory Slot: 84208028

I tried looking for the RAM timing and voltages but I'm unable to find where they are, I looked under the Health settings and the Overclocking menu but everything is set to auto. Redduc any help you could provide would be greatly appreciated! Thanks!

[redduc900]

Here's some screenshots of your BIOS settings (this and the next two pages of the article)...

http://www.ocworkbench.com/2007/gigabyte/GA-G33M-DS2R/g4.htm

[Anzy]

redduc, now I'm confused because most of those settings there are not even on my screen, some of them are but not many, I'm going to see if there is a reason why my settings aren't showing up.

Nevermind, found you have to hit CTRL + F1 at main screen to see advanced settings... brb with those numbers

[Anzy]

CPU Clock Ratio: 8x
CPU Freq: 2.66 Ghz (333x8)
CPU Host Freq: 333
Memory Freq: 800
Timing: 5-5-5-18
ACT to ACT Delay: 3
Rank Write to READ Delay: 3
Write to Precharge Delay: 6
Refresh to ACT Delay: 52
Read to Precharge Delay: 3
Static tRead value: 7
Static tRead Phase Adjust: 29

Then it says System Voltage Optimized and lists overvoltage and undervoltages controls, all set to Normal
CPU GTLREF1 Voltage Ratio: .636

Under the Health menu, it lists Vcore, DDR18V, +3.3V and +12V all "OK"

I don't see anything saying exactly what the voltage is at...

[El<(')>Maxi]

If the system ran perfectly for a period of time with the same components and BIOS settings, I'd be leaning towards parts going bad. Check for badcaps and test the suspect ram in another system to clarify the problem if you can.

[redduc900]

That RAM is rated at 5-5-5-18 w/ 1.8Vdimm, so your timings look OK. The Vdimm shown of 1.8v is in line w/ the rated voltage, although since you're running 2x2GB sticks I think I might try increasing the Vdimm by .1v-.2v

You can use the "DDR2 Overvoltage Control" setting in the BIOS to first set it to +0.1V, which will adjust the Vdimm to 1.9v Again run Memtest and check for errors. You might even try the +0.2 setting to increase it to 2.0v, although that shouldn't be necessary. If you still have errors in Memtest, again run one stick at a time like you did previously, but with the increased Vdimm of +0.1v You might also want to try increasing the MCH voltage ("MCH OverVoltage Control" setting) in the BIOS... increase it by 0.1v

[Anzy]

Maxi, I'm not exactly sure how to test for badcaps, I assume that physically checking the motherboard? The only other system I have at my disposal right now is this notebook, so I'm unable to test the memory at the moment.
redduc, How do we know that the vdimm is running at 1.8V? Also, earlier I had tried doing exactly as you had mentioned with increasing it through Overvoltage control and took it step by step all the way up to .4v, all had the same results, except the error took slightly longer to come up when at .4v overvoltage. I wish that worked... this is something of an urgent thing as I need this system up ASAP.

I suppose if I knew how to check for badcap I would but I'm not exactly sure how, at this point it's most likely the RAM or motherboard at fault? ... this sucks lol

[El<(')>Maxi]

Visual inspection is the quickest, look for leaking or bulging caps on the board.

Here is where you set the ram voltage.

http://www.ocworkbench.com/2007/gigabyte/GA-G33M-DS2R/Gigabyte%20G33%20008%20copy.jpg

[Anzy]

Maxi, thanks again for the reply, I'll definitely take a look at the board, probably right now, what do you suggest if I don't spot anything wrong on the board? Buy new RAM? Also I'm curious is it just extreme coincidence that both of these sticks died at the same time?

[redduc900]

redduc, How do we know that the vdimm is running at 1.8V?
In post #8 above, you noted this...
Under the Health menu, it lists Vcore, DDR18V, +3.3V and +12V all "OK"
.. so I assumed the Vdimm (DDR Voltage) was 1.8v
As a side note, are you testing your RAM w/ Memtest86 or Memtest86+? If using Memtest86 3.4(a), instead give Memtest86+ 1.70 a try...

http://www.memtest.org/

... but like I suggested above, test w/ only one stick at a time, increasing pertinent voltages by 0.1v and taking notes as to the number of errors in Memtest and at what point they occur. Another good memory diagnostic utility is MS Memory Diagnostic...

http://oca.microsoft.com/en/windiag.asp

Edit: What slots do you have your RAM installed to?

[El<(')>Maxi]

Maxi, thanks again for the reply, I'll definitely take a look at the board, probably right now, what do you suggest if I don't spot anything wrong on the board? Buy new RAM? Also I'm curious is it just extreme coincidence that both of these sticks died at the same time?

No I would test this ram in another system first before making that conclusion. Have you tried any of what redduc is suggesting?

[Anzy]

The RAM is installed to the two yellow slots, but I have switched them with the same results. I have tried two different versions of Memtest with the same result, I'll have to find out exactly what versions they are. I have my computer dismantled and will remain that way until my new RAM comes in. Ordered these. (http://www.newegg.com/Product/Product.asp?Item=N82E16820231145) Someone from the [H] forum told me that he thinks that it is improbable it is just the RAM? This is troubling to say the least. His other options were Bad BIOS settings (can't be since I have reset them several times with the same result) or bad motherboard or even bad CPU. If either of these had a problem wouldn't I at least get a beep error? Right now there are no beep errors so I remain confused. I realize buying new RAM may not be the fix, but at this point I don't believe it can be settings since the same settings worked for how many months and all of the sudden this problem happened. Unless, the settings were a problem to start with and only killed my memory slowly over time. I don't forsee gaining access to a computer where I may test my RAM, not in the near future at least.

[Anzy]

Well, the order came in and I replaced the old memory with the new memory and everything is working fine again... I'm just curious as to what killed the last set? Is this next set going to die too by the same cause? Thanks for all the help guys!