|Topic Review (Newest First)|
|04-09-2012 07:10 PM|
Thanks, it's still running the Prime95 test, I used it throughout the day, but the memory change made Prime95 ate a very noticable part of the machine. I did fire up a VM, but only one. I grabbed Memtest86 today and burned it to CD, I guess if it's still running in the morning, I'll toss that in toward the end of the day. (I sort of need to use the machine)
Good KB, it would seem if it applies to Vista and 2K8 it might apply to 7 too. I'm updating my DirectX, and checking out that DriverScanner, I've not seen it before and need a tool like that. Thanks for the forum links, I will jump over there after testing this thing about. One thing I have is plenty of BSOD files they like
Thanks again for the help, I'm quite happy to have those voltages turned down, I was at the deer in the headlights, what now I thought. After everything I learned / purchased, once I get this rig stable, then I'm going to try to OC it. that ought to be interesting.
|04-09-2012 06:51 PM|
I've been doing some internet searching and I'll link to things that have worked for some people. stop x101 errors are tough when it isn't a voltage problem (this really does not seem to be a voltage problem), since there are so many different things that can cause it.
This can happen from bad RAM, which is why I recommended that Prime95 test. If that runs fine without any issues, then move on to the memtest86 I linked above (you download it, burn it to a cd, boot from the cd, and it runs the test). If that doesn't bring up any errors, we can basically rule out the RAM.
I have heard of other people getting this from doing a lot of virtualization as you are, here is a fix from microsoft, although I don't think it applies to Windows 7. There may be something equivalent though.
Sort of related, people have gotten this error from problems with directx, the latest patches fix that. Really, it can be any number of bad drivers. Since this only happens when you do a specific thing (running virtual machines), that doesn't seem very likely. (useful utility for checking drivers if you are curious)
I have heard that it can be a problem with the BIOS itself - you could flash it - but I really don't recommend that.
All that stuff aside, if it is a hardware problem, I would expect a problem to show up with the RAM. If that turns out to be good, I think it is a Windows problem (probably not some obscure driver problem like what I just listed). I am not going to say re-install Windows, because that is a pain. But I would recommend going over to sevenforums.com and asking there since those people know about windows things, or posting in answers.microsoft.com if you want.
One last test that is quick and easy to run: Open command prompt as an administrator (search for it, right click and select run as administrator), and type "sfc.exe /scannow" (without the quotes).
Hope some of that stuff helps!
|04-09-2012 08:49 AM|
Thanks, you are more an expert than I at this.
I set my VCore to 1.2000, and memory to 1.500 and am currently running Prime95, showing 11.2 gig in use. I also learned that my temps were incorrect, I had TJ Max set to 100 in RealTemp, in BIOS it was set to 90. Now I like my temps better, that was the cheapest 10° C I've ever spent.
The Memory showed a Profile DDR Voltage at 1.65, the spec sheet seemed to say it would run at 1.5 as long as the clock rate was 1333. (or lower I presume)
So far I've been running the custom test for about 90 minutes, while also doing a little other stuff on the pc. It is set to a 5 minute test cycle.
I'll post back when something happens.
|04-08-2012 11:26 PM|
I am not an expert, but I know some people here are, so do defer to their judgement. But, I'll give you my thoughts:
I don't think this is an issue with the processor being generally unstable, and I am certain that more vcore is not going to help. My best guess is something to do with the RAM. I saw that you ran Prime95, which is great. It only tests 1600 MB of RAM though by default. Go ahead and lower your vcore back to normal, and DRAM voltage to whatever the stock value is for your ram, probably 1.5 V. Open up Prime95, set it to custom, and put in something like 10000 MB for ram so that it has to test all three sticks, and see what happens (I'm going to try that out on my system in a minute to make sure that is ok lol). I am thinking that you will see a crash, since it crashed when loading up many virtual machines.
Another test to try if this doesn't show anything useful is memtest86.
I hope that is helpful; this really is an odd problem to be having.
Edit: Yep, set Prime95 to 10000 MB of ram, that gets about 11 GB loaded up for me. Also, apparently you can get a more thorough test in less hours by setting the time thing to 5 or 10 minutes rather than the default 15.
|04-08-2012 03:09 PM|
I am having problems getting my rig to run even at stock speeds. The main issue is Stop 101 errors, the last crash was a solid freeze which also happen maybe 30% of the time. I am beyond my expertiese. I really don't care about OC as much as rock solid stable operations. My other machine, a Q6600 runs rock solid, I built this one about a year ago, I've never got it stable.
My machines run 24/7, on this one I run multiple virtual pc's, do development in Visual Studio 2010, and run typical LOB applications. (Quickbooks, Word, Excel, etc...) I game a few times a year, X-Plane or an Old Nascar game. I'd run Prime95 for 14-24 hours without a crash. Then start 5 Virtual PC's and within a short time stop 101. The VPC's were not doing much, all xp or win7 clients connected to remote networks via VPN and sitting there for the most part. Temps during crashes monitored with RealTemp are low. I'm seeing about 8 Gig of ram in use with those VPC's running.
I've replaced a ton of stuff in the past year, Motherboard, Ram, Powersupply, CPU in that order. The most recent addition was the CPU cooler because I could not raise vCore without getting too hot, where I called 90 pucker tight. Shortly before that Intel sent me a replacement CPU, which threw a 101 within 36 hours.
I filled out my Rig, and am running the BIOS settings below. Lately I've been bumping vCore, now that I'm over 4, that pucker tight feeling has started again, and I thought it well time to ask those that know. So what am I doing wrongly?
Thanks for thinking about my issue, it's been a long road.
Here's my BIOS settings
CPU Clock Rate 21
QPI Clock Ratio Auto
Uncore Clock Ratio x16
BCLK Frequency 133
PCI Express MHZ 100
cpu Clock Drive 700mv
cpu Click Skew 0ps
Load-Line Calibration Standard
QPI/VTT Voltage 1.395
CPU PLL 1.88
PCIE 1.5 Auto
QPI PLL 1.2
IOH CORE 1.2
ICH I/O 1.5 Auto
ICH Core 1.2
Dram Voltage 1.66
Dram Termination .75 Auto
Ch-A Data Vref .75 Auto
Ch-B Data Vref .75 Auto
Ch-C Data Vref .75 Auto
Ch-A Address Vref .75 Auto
Ch-B Address Vref .75 Auto
Ch-C Address Vref .75 Auto
System Memory Multiplier 8
Performance Enchance Standard
Advanced and Misc Timing Auto
Temps in BIOS 37,35,37,35
Intel Turbo Boost Tech Disabled
CPU Enhanced Halt C1E Disabled
C3/C6/C7 State Support Disabled
CPU ESIT Function Disabled
CPU Multi-Threading Enabled
HPET Support Enabled
HPET Mode 64-Bit Mode
Here's a short verion of my rig
Gigabyte X58A-UD3R(R-2.0) X58 chip is Rev 13, Bios version- FG
i7 - 930 CPU, CPUID= 00106A5, Update Rev=00000011
CPU Cooling - Thermalright HR-02 Macho
12 Gig, 3x4GB Kingston HyperX, KHX1600C9D3K3/12GX
2 Radeon HD5770 - (running 4 monitors)
4 - Seagate 1 TB Drives in RAID10 Array ST31000333AS
2 - Seagate Momentus 500Gig Hybrid drives Mirrored Boot ST95005620AS
OCZ ZX Series 1000 Watt Power Supply (power draw 467 watts running Prime95 torture test, with a RAID rebuild going on)
Enzotech SLF-1 heatsink for ICH10R (note as stock doesn't fit this board)
2 - Noiseblocker NB-BlackSilentFan X2 in case
1 - DVD iHAS524
I've replaced the MB, Ram, Powersuply, CPU (rma) and added cooling in my quest.
Some of this was due to either Stop 3b, 7x, and 9f Errors
MSI X58 PRO-E
6 - ocz3P1333LV6GK
700 Watt CoolerMaster 3 Rail HDD
When I preview this, the CRLF's go away.