Sensor Test Crash

[xenogear74]

Thanks for the quick reply!

I'm not overclocked and am running at stock voltage. I'm going to try doing 1 stick of ram at a time next and go from there.

I was just surprised last night when i turned off EIST, etc... and then ran the test and had it work! Leads me to believe it may be something with the board? I think it's the same one the others had as well.

I'll let you know from there!

 

[unclewebb]

I think I found your problem.

I'm not overclocked and am running at stock voltage.

Don't let anyone else find out that you are not overclocking or they might kick you off of TechPowerUp.

Usually when people contact me they are running their systems on the verge of self destruction. That test should work for you. If you are using mostly AUTO or default voltage settings in the bios then the CPU or some other component might need a slight tweak in voltage. Try watching the CPU voltage with CPU-Z during idle and during that test. One of the power saving features might be causing too big of a drop in voltage.

 

[xenogear74]

Didn't even think of that! Great idea.

And I'll try to at least overclock it a little just to make sure i'm still part of the club

 

[xenogear74]

I tried upping the voltage and no go. I did however leave C1E on but turned of EISE (speedstep) and low and behold it worked fine.

CPUZ did see a dip in voltage, but only by .10 and I don't know if that's enough to really consider a problem.

I'm tempted to try this on a dual core i have lying around to see if this is something the mobo doesn't like or may show a defective processor?

 

[xenogear74]

Well, I wanted to check back in with findings.

Even increasing my voltage I couldn't get the test to work, but I was able to find 2 things that did. (just note I am running Windows 7 x64)

1) turning of EIST in the bios. for some reason when this was off the processor was fine and the test went through swimmingly.

2) Leave EIST on, and change the windows 7 power management features to not throttle down the CPU (turn it on High Performance). At this point the program still throttled the multi like it should, and the test passed fine even with EIST on. I guess it might have been something to do with the windows power management and the EIST not playing nice.

Either way, I thought I'd share.

Thanks for all your help!

 

[guitarfreaknation]

I am running Windows 7 Professional (64 bit) and my computer crashes when I try to run the sensor test.
Details:
E8400 (not overclocked)
My Idle temps are supposedly around 45-50c

When I click sensor test it opens the box, I select Prime95 and let the test run. Everything seems to be going fine until the idle test when it says 'stopping Prime95'. At this point, my entire computer freezes, nothing on the display changes at all. Numbers stop changing, even the second hand on the windows clock gadget on the desktop stop, no a single pixel on my display will change. I have to use the hardware reset button on my computer and then everything boots normally.

Why is this happening, and what can I do to fix it.
I tried the 32 and 64 bit versions of Prime95, is there a 64 bit realtemp?

P.S. I've heard around the internet that Tjmax for an E8400 is 100c, is this correct?

I have the exact same problem BUT on Windows Vista Ultimate 64-bit

 

[unclewebb]

guitarfreaknation: Post some screen shots and some details so I can have a look. There is nothing evil about the test. It just uses features of your CPU that most other software doesn't. A stable CPU should be able to pass the RealTemp sensor test. I've never tested a single CPU that failed this test and locked up and froze the computer towards the end of the test.

RealTemp 3.58.1
http://www.fileden.com/files/2008/3/3/1794507/RealTempBeta.zip

 

[burebista]

Kevin, guitarfreaknation is that guy with Q9550 unrecognized.
Here is his problem.

 

[guitarfreaknation]

guitarfreaknation: Post some screen shots and some details so I can have a look. There is nothing evil about the test. It just uses features of your CPU that most other software doesn't. A stable CPU should be able to pass the RealTemp sensor test. I've never tested a single CPU that failed this test and locked up and froze the computer towards the end of the test.

RealTemp 3.58.1
http://www.fileden.com/files/2008/3/3/1794507/RealTempBeta.zip

I am aware that this test isnt "evil"
I just wanted to get some input on how to resolve this issue.

 

[unclewebb]

It's possible that the voltage drops too low during the transition from load to idle. What does CPU-Z show for your voltage when at full load and when at idle?

Is your computer otherwise rock solid stable? Can it run Prime95 and LinX and anything else you throw at it?

Is your CPU voltage set manually in the bios or is it set to AUTO? The last person that had troubles was able to fix it with a little more voltage so maybe that will help you too.

If you are tired of having to sit through the whole Sensor Test again and again only to see it crash then try running Prime95 and go into the RealTemp Settings window and while P95 is running, adjust the Clock modulation up and down from 100% to 12.5% and see how that goes. With it down at 12.5% if it hasn't locked up yet then try stopping Prime95 and see what happens. This might allow you to test your setup a little quicker.

 

[guitarfreaknation]

It's possible that the voltage drops too low during the transition from load to idle. What does CPU-Z show for your voltage when at full load and when at idle?

Is your computer otherwise rock solid stable? Can it run Prime95 and LinX and anything else you throw at it?

Is your CPU voltage set manually in the bios or is it set to AUTO? The last person that had troubles was able to fix it with a little more voltage so maybe that will help you too.

If you are tired of having to sit through the whole Sensor Test again and again only to see it crash then try running Prime95 and go into the RealTemp Settings window and while P95 is running, adjust the Clock modulation up and down from 100% to 12.5% and see how that goes. With it down at 12.5% if it hasn't locked up yet then try stopping Prime95 and see what happens. This might allow you to test your setup a little quicker.

Yes, the system is solid. It gets a bit hot in my cheap case and stock CPU fan but other than that runs great. I did run p95 when took these screenshots. First is the idle CPU-Z and load. Second one is on load only.

 

[unclewebb]

If your computer is 100% stable except for the RealTemp sensor test then I would suggest not running that test.

1.152 volts at full load is very low. Did you set that manually or is your bios set to AUTO? Did you try raising the voltage manually in the bios to see if you can pass the RealTemp sensor test? Maybe it will pass at 1.175 or 1.200 volts.

 

[95Viper]

If your computer is 100% stable except for the RealTemp sensor test then I would suggest not running that test.

1.152 volts at full load is very low. Did you set that manually or is your bios set to AUTO? Did you try raising the voltage manually in the bios to see if you can pass the RealTemp sensor test? Maybe it will pass at 1.175 or 1.200 volts.

Just wondering about the comment? As, my Q9550 is quite happy at these voltages on my GA-x48-DS4 for over a year.

Just my two cents;I do believe(disclaimer:been known to be wrong once or twice), if, he(guitarfreaknation) is not overclocking, he needs to lower the CPU voltage a tad to help lower the CPU temps, and maybe bump his NB a little and check his memory voltages. Gigabyte boards are finicky about memory.
Idle: Load:

 

[unclewebb]

95Viper: You're right and I'm wrong.

I'm used to overclocking everything and anything so I tend to use more voltage than that. If you can run reliably at those speeds with that much voltage then that's great.

Have you ever completed the full 10 minute RealTemp Sensor Test with those voltages? Some users have trouble completing this test even though they are 101% stable in Prime95, LinX, etc.

When you have 10 minutes to kill can you run that test and post the results so other users can see that it is possible. It could be something like NB voltage that is the problem for guitarfreaknation or perhaps his CPU simply needs a little more voltage compared to yours.

His CPU is an E0 compared to your C1 so that usually means it should be happy with less voltage compared to yours. I'm not sure why Intel didn't include the model number in his CPU. The CPU-Z specification box shows the information that Intel writes into the chip and your CPU clearly shows Q9550 while his does not.

 

[95Viper]

I know something is screwy with his, some suggested it might be a ES. You can get some weird results off of them.

Yep, completed that test a few times and run prime 95 always after replacing tim or maintenance, to make sure I ain't fubared something. Am replacing tim this week if Dow Corning can get me some of their TC-5688, if not I will be using(testing) Zalman Stg2 (am using Stg1 now).

 

[unclewebb]

ES CPUs will always have (ES) in the CPU-Z Specification box. guitarfreaknation's CPU doesn't have that and it has the correct CPUID 0x1067A which all of the E0 processors have. I've seen a few CPUs where Intel forgot to write in the correct model number. CPU-Z uses a look up table and compares the CPUID value with the GHz amount and the amount of cache, etc. and determines that it really is a Q9550. It might be an early model or some other unknown reason why Intel did this. I've seen some where Intel writes the correct model number into only 1 of the 2 cores. I've learned that anything is possible. I'll probably add this unique CPU to the RealTemp look up table in the near future.

 

[95Viper]

Is this what you requested?

 

[unclewebb]

That's what I like to see. It drives me nuts because I know the Sensor Test doesn't do anything too wild and crazy but some extremely stable computers choke on this test. I'll never know why. Thanks for showing that it's possible to complete this test on a Q9550.

 

[guitarfreaknation]

If your computer is 100% stable except for the RealTemp sensor test then I would suggest not running that test.

1.152 volts at full load is very low. Did you set that manually or is your bios set to AUTO? Did you try raising the voltage manually in the bios to see if you can pass the RealTemp sensor test? Maybe it will pass at 1.175 or 1.200 volts.

Thanks. I like the simplicity.

 

[baldrick1001]

I'm suffering with the same freeze problem when prime is being stopped.. I believe that this issue may have something to do with the Gigabyte p45 chipset motherboards - as everyone on this thread that has experienced this issue has one....

Spec

Q9650 (At stock)
EP45-UD3R
4gig Corsair RAM
80 gig Intel SSD
2 X Samsung 1t HDD
Corsair HX620 PSU

 

[guitarfreaknation]

I'm suffering with the same freeze problem when prime is being stopped.. I believe that this issue may have something to do with the Gigabyte p45 chipset motherboards - as everyone on this thread that has experienced this issue has one....

Spec

Q9650 (At stock)
EP45-UD3R
4gig Corsair RAM
80 gig Intel SSD
2 X Samsung 1t HDD
Corsair HX620 PSU

My problem fixed itself or maybe because I oc'ed from 2.8 to 3.0Ghz... Dont know how or why but I do the sensor test just fine now..

 

[unclewebb]

baldrick1001: Maybe it has something to do with the default chipset voltage being used on the Gigabyte boards. I like hearing any explanation that points the finger at something besides RealTemp.

My problem fixed itself or maybe because I oc'ed from 2.8 to 3.0Ghz..

Did you change the core voltage when you increased your overclock? On my old Asus board when things are set to AUTO and I start overclocking, various voltages start going up and up.

 

[baldrick1001]

No overclock applied (don't tell anyone, but I'm a bit of a coward with O/C really!!). Everything is at stock and I've never really felt the need to overclock as its pretty fast as it is (that's the party line statement) - the SSD makes a huge difference to general usability.

So in answer to your question, everything is set to auto - someone early in this thread stated that turning off “eist” remedied their system hang – I’ll give that a go later and feed- back. Just seems really strange that turning off Prime 95 would cause the hang .


Many thanks

Baldrick1001

 

[guitarfreaknation]

No overclock applied (don't tell anyone, but I'm a bit of a coward with O/C really!!). Everything is at stock and I've never really felt the need to overclock as its pretty fast as it is (that's the party line statement) - the SSD makes a huge difference to general usability.

So in answer to your question, everything is set to auto - someone early in this thread stated that turning off “eist” remedied their system hang – I’ll give that a go later and feed- back. Just seems really strange that turning off Prime 95 would cause the hang .


Many thanks

Baldrick1001

Well perhaps if you just set in to manual and bump the FSB just by 10 or so points it would fix this issue. You should try it, its really simple and take a few minutes..

Good luck anyway

 

[MaZz]

Sorry for the 4 year bump!

Not sure if this is at all relevant anymore.

I tried this test on my old Gigabyte EP45T-UD3LR motherboard and E8400 cpu and I got the system lock up and had to hard reset three times.

I read the entire thread and the one thing everyone had in common is the Gigabyte EP45 series motherboard, so it must just have something to do with this particular motherboard.