Sign in with
Sign up | Sign in
Your question

The problem came back :(

Last response: in Memory
Share
January 14, 2012 8:17:07 PM

I thought my BSD problem was taken care of, for my comp didn't do it for over a week; but now it's back. cry

Thanks to some excellent advice I received here before, I installed and ran SIW. After checking a few categories, I found this under memory:

Memory Summary Reported by BIOS
Warning! Wrong values reported by BIOS
Maximum Capacity 32768 MBytes
Maximum Memory Module Size [unknown]
Memory Slots 4
Error Correction None
DRAM Frequency 665.2 MHz
Memory Timings 9-9-9-24 (CL-RCD-RP-RAS)

Any idea what all this means? Why the warning? How do I fix it? I am guessing that this may be related to my BSD problem.
Thanks in advance!

More about : problem back

a c 150 } Memory
January 14, 2012 8:42:40 PM

List your complete system specs. I never even heard of SIW before this post. I usually just use CPU-Z, Speccy, HW Monitor and PC Wizard 2010.

If you want to actually test parts use memtest for memory, Furmark for graphics cards and Prime 95 for overall system stability. Make sure you monitor temps when running Prime 95 and Furmark.
m
0
l
January 15, 2012 12:46:09 AM

Core i-7 2600k @ 3.4 GHz/ MSI Z68A-GD80 (gen3) mobo/ MSI GTX 560 graphics card/ Corsair Force 3 SSD 120 GB/ 23in. Samsung LED monitor/ G.SKILL Ripjaws Series 8GB (2 x 4GB) 240-Pin DDR3 1333 (PC3 10666) 9-9-9-24/ Windows 7 Professional upgrade 64-bit/ Antec 750 Earthwatts PSU. Anything else?
m
0
l
Related resources
a c 150 } Memory
January 15, 2012 4:06:46 AM

Ok a new build with quality parts.

Are you overclocking anything?

When does the computer crash? Randomly or doing certain tasks?

Have you run the tests I suggested above? Run memtest overnight or for 8 to 10 hours. That will rule out bad RAM. Furmark should be run for an hour or so to make sure the GPU is stable. Run Prime 95 to test overall system stability. Prime 95 should also be run 8 to 10 hours while monitoring temps.
m
0
l
January 15, 2012 5:30:16 AM

No overclocking. Will try to do the tests here quickly. And it happens when I put it to sleep. I'v also disabled auto restart, so hopefully I can pull some numbers off the screen.
m
0
l
January 15, 2012 5:33:44 AM

Oh, and I've run the standard memory test, but not one that would take all night. Where do i go to run that?
m
0
l
a b } Memory
January 15, 2012 7:58:41 AM

might try this

1) Looks like I may have found the solution: Changing PCI Link State Power Management to Moderate Power Settings. I discovered this after at least 50+ power cycles of adjusting one setting at a time, then once I started to narrow it down by re-enabling settings in the reverse order. Finally I can reproduce, and cure, the issue by toggling this setting between Moderate and Maximum. (Off also fixes it, but Moderate seems fine.)

www.forum-en.msi.com/index.php?topic=154280.0

if it only happens when you put it to sleep its most likely a motherboard issue. heard of several boards with sleep issues. try the latest BIOS and you may need to contact MSI. the other testing is great too, but it sounds like you.have a very specific issue related to one feature.

m
0
l
January 15, 2012 3:59:14 PM

Thanks to both of you for all the help.

Unksol, I get an invalid link to the MSI forums. Can you verify it?
m
0
l
January 15, 2012 4:06:14 PM

I also just turned of putting the computer to sleep or turning it off automatically (since I'm pretty good on doing it on my own anyway). It makes sense that it might be turning off after the preset amount of time has passed, even if it is asleep.

I'll run that mem test tonight as well. Thanks!
m
0
l
January 16, 2012 2:14:09 PM

This comp is driving me nuts. Now a new problem has appeared, and I don't know if it's a real problem or just an oddity.

When I start my comp (I turn it off at night because it's an SSD) there is the normal tone, but starting yesterday there is a tri-chirp after it. The first two chirps are really fast with a very small delay between the second and the third. Anyone ever heard of this just starting out of the blue?
m
0
l
a c 347 } Memory
January 16, 2012 3:27:55 PM

Look at the Event Manager, as shown below, and look for [+] Critical errors. Next MSI typically has a backwards DIMM slot arrangement; use the Blue DIMM slots.

I assume that you installed all of the latest drivers listed on MSI's website - http://www.msi.com/product/mb/Z68A-GD80--B3-.html#/?div... listed for your OS.

If you post the exact set of RAM then I can post the correct manual settings in the BIOS to see if that corrects the memory issues.

The next is to 'try' a BIOS update with M-Flash.

Event Manager:
m
0
l
January 16, 2012 10:02:15 PM

OK. Under Critical it says

Event ID: 41
Source: Kernel-Power
Log: System
3 in the last 7 days


I also have 76 errors and 24 warnings in the last 7 days. What does all of this mean?
m
0
l
a c 347 } Memory
January 16, 2012 11:31:03 PM

Critical is all you need to care about, stuff like some of Apple software throws all sorts of Errors and Warnings. Sure look at them, but they're not causing the problem(s) you posted.

I recommend SATA -> AHCI or if running RAID then obviously SATA -> RAID. If you changes the values in the BIOS then I recommend running 'Fix It' -> http://support.microsoft.com/kb/922976

Event ID: 41
Source: Kernel-Power

Without more details it means an unexpected shutdown in Windows typically not from a BSOD. Most BSOD's generate a detailed report.

IMO start with updating the BIOS that is if your PC is stable in the BIOS.
m
0
l
January 17, 2012 2:51:51 AM

My BIOS are up to date. Sure of it. Like I posted earlier I also turned off auto restart, so if it does it again, I'll be able to get more info.
m
0
l
a c 347 } Memory
January 17, 2012 11:02:29 AM

F3-10666CL7D-8GBRH DDR3-1333 7-7-7-21 @ 1.50v

Footnote: DDR3-1333 is fine but often CAS 7 is difficult for some MOBO's, often I see 'good' RAM that's has excellent Frequency to CAS cause problems. Below is the Rated, but if that fails then try 8-8-8-24.

DRAM Frequency -> DDR3-1333 ; optional but setting it locks it in
DRAM Timing Mode -> Link
Advanced DRAM Configuration /
tCL -> 7
tRCD -> 7
tRP -> 7
tRAS -> 21

DRAM Voltage -> 1.50v~1.55v ; IMO CAS 7 try 1.55v

Save & Exit = Yes

Good Luck! :) 
m
0
l
January 17, 2012 3:17:26 PM

Ok, I found the DRAM settings under the overclock section, and adjusted the to the given values (although there were a lot more than just the 7-7-7-21, but I left them alone because I was not sure) I set the DRAM voltage at 1.550. Guess we'll see how it goes.

As a side note, while I was looking for the DRAM settings, I found a setting that had the comp turn off when AC power was lost. Could that have anything to do with my initial problem? I set it to last power state by the way.

Once again, thanks for the help all!
m
0
l
a c 347 } Memory
January 17, 2012 3:27:36 PM

IF there's all sorts of brown/black-outs in your area or you often leave the PC for a longtime e.g. vacation then IMO have it set so the the PC remains off. I recommend an UPS, and I cannot stress how much longer life and less problematic your PC will be in the long run. In addition, I allow Hybrid Sleep (S3+S4 state) there's no sense in having a PC draw 100W~300W+ just sitting there if its not needed.

In my case I life in Florida with frequent brown/black-outs and lightning (FL is the lightning capital) during the warmer months.
m
0
l
January 17, 2012 3:50:21 PM

Ok, we don't usually have outage problems (I live in Phoenix).

I was also wondering about all the other DRAM values. Was I right in leaving them on auto, or do I need to adjust all of them as well?
m
0
l
a b } Memory
January 17, 2012 3:59:59 PM

Is your Corsair Force Series 3 120GB SSD running with the latest Firmware Update Revision 1.3.3 ?
m
0
l
a c 347 } Memory
January 17, 2012 4:20:21 PM

2kids said:
I was also wondering about all the other DRAM values. Was I right in leaving them on auto, or do I need to adjust all of them as well?

Leave them alone and as Auto, you only need to worry about (CAS-TRCD-TRP-TRAS) and if the RAM frequency is faster DDR3-1600 or faster then the Command Rate (2T).
m
0
l
January 18, 2012 4:08:29 PM

Thanks for the help everyone. I guess the thing to do now is wait and see what happens.
m
0
l
a c 347 } Memory
January 18, 2012 4:51:27 PM

As linked above, Memtest 4 passes or about a couple hours with 8GB. In addition, Prime95 + Blend test also a couple hours or preferably overnight.
m
0
l
January 19, 2012 10:58:43 PM

Ok cool. I actually got a blue screen today. Here are the numbers, because I have no idea what they mean.

***stop: 0x000000IE (0x0000000000000000, 0x0000000000000000,0x0000000000000000)
m
0
l
a c 150 } Memory
January 19, 2012 11:20:45 PM

I still do not see the results of the tests it was recommended you run.
m
0
l
a c 347 } Memory
January 20, 2012 12:15:26 AM

You'd need to look at Event Manager [+] Critical errors and post the results. Chances are a RAM issue, but I'd need to see the Event Manager or (.dmp) file to know.

If after running Memtest or Prime95 + Blend if you're getting errors then try:

DRAM Frequency -> DDR3-1333 ; optional but setting it locks it in
DRAM Timing Mode -> Link
Advanced DRAM Configuration /
tCL -> 8
tRCD -> 8
tRP -> 8
tRAS -> 24

DRAM Voltage -> 1.50v~1.55v

Save & Exit = Yes
m
0
l
January 20, 2012 1:51:29 AM

I am going to run Prime tonight (do I really need to run both it and Memtest?). Got the core temp going on, plus the gadget to constantly monitor. Usually in the low 30's (C), under load (MMH6 and Chrome ie at same time) the highest I ever saw on any of the cores was 59 (C). Usually in the the high 40's to low 50's (C) under that same load.

What's Furmark test again? Thanks! :D 
m
0
l
a c 347 } Memory
January 20, 2012 2:27:47 AM

If your Temps run too high that's a problem at it will end as a BSOD. Run Memtest first.
m
0
l
January 20, 2012 2:33:15 AM

So what is too high for a core-i7 2600k? I was under the impression that my temps were just fine.
m
0
l
January 20, 2012 3:14:51 AM

Found this link here in Tom's site.

http://www.tomshardware.com/forum/312569-28-core-2600k-...

According to it, I seem to be ok.

Now, I'm getting a little worried. Running Prime 95 and with the net going to type this, all my temps are between 69 and 73 C. According to the above thread, 72.6 is max safe range. ugh
m
0
l
January 20, 2012 3:21:23 AM

Yeah, so until I hear back from you (great) guys, I stopped the test. I definitely don't want to fry anything.
m
0
l
a c 150 } Memory
January 20, 2012 3:57:50 AM

That is mighty high for a stock 2600K. I idle in the upper 20s/low 30s and hit high 60s overclocked to 4.5Ghz. You are running stock speed/stock cooler right? What case and fan setup do you have?

If you are using stock cooling did you leave the stock thermal paste on the cooler when you installed it?
m
0
l
a c 347 } Memory
January 20, 2012 11:53:46 AM

If running 'stock' yep they're on the high side unless your ambient temps are really high.

BSOD/Post Failures during stress tests like Prime95 are common. If they were damaging then every build I did would have ended up badly. It's part of it.

If running stock you should be able to run Prime95 forever (lifespan).

Run Memtest, its more than likely a tell all. It stresses the RAM not the CPU. Prime95 + Blend stresses CPU + RAM.
m
0
l
January 20, 2012 2:32:25 PM

I don't know what's going on. I idle in the low 30's as well. But just a few min. of prime shot me up to 73. Yes stock cooler, w/ the paste already on the cooler. So I just stuck the cooler on. Case is NZXT Phantom. Do I need a better cooler?
m
0
l
a c 347 } Memory
January 20, 2012 2:52:09 PM

The stock HSF can only be installed once, and if you ever removed it then every time you also need to clean off the old thermal paste with e.g. Isopropyl alcohol w/lint free cloth, and replace the thermal paste. I recommend AC MX-4 - http://www.newegg.com/Product/Product.aspx?Item=N82E168...

Make sure there's more air going out than in; often folks pull in cool air but forget too much in keeps in the hot air trapping it inside the case and reducing 'cool air' airflow.

Test compare temps with the case's side on and off; if it's >2~3C then you've got an airflow problem.
m
0
l
January 20, 2012 2:53:32 PM

K. Will test. It'll be tonight or tomorrow though. Busy, busy, busy!
m
0
l
January 21, 2012 3:42:10 AM

Just back from a late night at work, and I have school in the morning. So the soonest I'll be able to do something is tomorrow afternoon. It's interesting, my wife has Netflix on pause while I checked a few things online, and the temperatures are spiked like when I was running Prime 95+ yesterday. Something's going on.
m
0
l
January 21, 2012 4:58:11 PM

Haha. I figured it out. Prime 95 started up automatically. Turned it off and the temps immediately dropped.
m
0
l
a c 347 } Memory
January 21, 2012 5:01:28 PM

Yeah sure Prime95 pegs the CPU to 100%. [ ] (unchecked) Options > Start at Bootup
m
0
l
January 21, 2012 8:47:32 PM

Yup, already done. Thanks though! Will be able to get the comp's side off tonight after the kids are in bed.
m
0
l
January 22, 2012 5:56:01 AM

Okay. So I got the main side off and let the comp run at idle before I took it off, and after. Not an airflow problem. The comp actually ran about 6 (C) higher out of it's cubby with the main side off, as opposed to inside it's cubby with the main side on. Note: the main side has 2 additional 120mm fans on it.

So, what next? :D 
m
0
l
a c 347 } Memory
January 22, 2012 12:34:04 PM

I assume the temperature problem was solved after you stopped Prime95 from running at startup. If you're still getting BSOD then I still haven't seen the results from Memtest?

If it's a RAM issue you can try slowing down the RAM even more:

DRAM Frequency -> DDR3-1333 ; optional but setting it locks it in
DRAM Timing Mode -> Link
Advanced DRAM Configuration /
tCL -> 9
tRCD -> 9
tRP -> 9
tRAS -> 24 -> 27

DRAM Voltage -> 1.50v~1.55v

Save & Exit = Yes
m
0
l
January 22, 2012 2:32:25 PM

I'll do my best to get memtest done tonight. I had forgotten about that one to be quite honest.

And yes, the earlier temp problem was an auto running prime 95.
m
0
l
January 23, 2012 3:48:02 AM

Unfortunately, I am having trouble with memtest. I have tried both the cd and flash drive, put my OS drive as the last on the list to boot, and I still keep having Windows start. Any suggestions? Thanks.

By the way, my BIOS is the new clickable type that works with a mouse.

This is where I went. Got the second one down and the fifth one.

Download - Pre-Compiled Bootable ISO (.zip)
Download - Auto-installer for USB Key (Win 9x/2k/xp/7) *NEW!*
m
0
l
a c 150 } Memory
January 23, 2012 4:42:24 AM

You have to mount the CD with something like ISOBurn to make it bootable. Download the Zip file, extract it and use ISOBurn to make a bootable CD. Then make the CD first to boot in BIOS.

http://sourceforge.net/projects/isoburn/
m
0
l
a c 347 } Memory
January 23, 2012 11:17:51 AM

How to CD/DVD Memtest and guide - http://www.geekstogo.com/forum/topic/246994-guide-to-us...

How to USB - http://forum.canardpc.com/threads/39168-Windows-HOWTO-B...

How to change Boot Priority - http://pcsupport.about.com/od/fixtheproblem/ss/bootorde...
Then in the BIOS select either the CD/DVD or USB Drive (native Intel/AMD USB 2.0 port) as the first priority in boot order. Otherwise if your HDD is in first priority it will ignore either the CD/DVD or USB drives.
m
0
l
January 23, 2012 2:38:45 PM

Cool, will make sure these things are happening, give it another go tonight for (hopefully) while I am sleeping.
m
0
l
January 24, 2012 11:39:40 PM

Oh, by the way, here is some information from when the comp restarts and I am first back in Windows. Don't know if it helps at all. Don't know why I have not posted it yet.

Problem signature:
Problem Event Name: BlueScreen
OS Version: 6.1.7601.2.1.0.256.48
Locale ID: 1033

Additional information about the problem:
BCCode: 1e
BCP1: 0000000000000000
BCP2: 0000000000000000
BCP3: 0000000000000000
BCP4: 0000000000000000
OS Version: 6_1_7601
Service Pack: 1_0
Product: 256_1

Files that help describe the problem:
C:\Windows\Minidump\011212-9609-01.dmp
C:\Users\Dylon\AppData\Local\Temp\WER-11122-0.sysdata.xml
m
0
l
a c 347 } Memory
January 25, 2012 10:31:04 AM

I hate to sound like a broken record here, but unless you 'can' run Memtest you're at a roadblock. The 'BCCode: 1e' error more than likely has it's roots in a RAM problem. You can try to boot with (1) stick see if the problem corrects itself and then swap out the stick with the other, and in both cases testing with (1) stick of RAM.
m
0
l
Related resources
!