problem CPUID's CPU-Z program causes my desktop pc to "freeze".....

[apairofpcs]

Fellow pc lovers,

I'm having a problem that has literally "appeared from nowhere", regarding CPUID's CPU-Z program. It has suddenly caused my Windows Ultimate 32 bit desktop pc to freeze almost every time I open it. By "freeze", I mean the mouse pointer freezes, the keyboard freezes and I have to "hold instead of tap" the Reset button on the front of my case to do a Restart. I installed the last three versions of the program with the same result. When a version failed, I uninstalled it from Control Panel's Programs and Features applet, deleted the single reference I found in the registry and installed a different version. A restart is not required after an install. I was able to make the latest version work on the desktop in Safe Mode with Networking, only after a dialog box showing an error log file appeared. I attached the screen capture of the box to this post. The dialog box only appears in Safe Mode and the program always opens properly after the box appears, regardless of whether I display the contents or not. In GUI mode, the box never appears. The program may open once after the installation, but then it freezes the next time. I'll restart and it will open once, and then fail at my next attempt.

When the problem started on the desktop, I checked my two Windows 7 Home Premium Edition 64 bit laptops, installed the latest version and have had no problem with it. Neither laptop had the latest version when I checked, and each older version of the program worked fine.

This is the only program that has caused my desktop to freeze since it was built in April 2009. Does anybody have any idea why this program is creating aproblem?

 

[apairofpcs]

Sorry that I forgot to attach my screen capture file above. Here it is.....

 

[P5-133XL]

Doing a google search I noticed that that error is associated with RAM issues. Try running Memtest86+ on your machine

 

[apairofpcs]

Doing a google search I noticed that that error is associated with RAM issues. Try running Memtest86+ on your machine

Despite the fact that this is the ONLY program I'm using that has caused such a problem, and I ran Memtest86+ recently on the desktop, I'm going to run it all day long today.....because you suggested it. It won't take long per pass, because I only have 4 GB DDR2 RAM (2 x 2 GB) in the machine. I'll write back later.

 

[Stereodude]

If your PC works fine otherwise, don't use the program. It's a pointless program anyhow IMHO.

 

[apairofpcs]

If your PC works fine otherwise, don't use the program. It's a pointless program anyhow IMHO.

Indeed the pc works flawlessly without CPU-Z. I can always find another diagnostic program.

I started a Memtest86 + v.4.20 run 2 hours ago, so I may as well let it rip for 12 hours, a nice round number. So far, 2 passes have rendered no errors. A pass takes 42 1/2 min., so I'll have about 17 passes completed in 12 hours. I can spare the desktop for today, since I have 2 laptops to keep me busy.

In Dec. 2012, the desktop started freezing like crazy with almost every program I used. After 3 months of examinaion, I moved the 2 DIMMS to the 2nd pair of memory banks. I didn't have a freeze until now. In fact, a month after the switch I returned them to the original 2 banks, and the freezing resumed. They are currently in the 2 reliable banks. So it's not a waste of time to run Memtest today. Not knowing why the originally occupied banks failed, to this day, I'm wondering if/when the current banks will fail.

I will report the results of the testing tonight.

 

[Bozo]

Sounds like a motherboard issue if moving the RAM to different slots and they work fine.

 

[apairofpcs]

Sounds like a motherboard issue if moving the RAM to different slots and they work fine.

I agree. I'm happy that there are 4 memory banks on my motherboard, not 2. In my case, it doesn't matter that I wouldn't be able to upgrade my RAM to the motherboard's limit of 16 GB, because I'm running Win 7 32 bit for my OS. The 4 GB of RAM I have becomes 3.328 GB, as a result.

Progress report.....
10 passes with no errors.

 

[LunarMist]

It sounds like a good time to rebuilt that desktop system. Is the CPU an early i7 or something of the old C2D/C2Q variety?

 

[apairofpcs]

It sounds like a good time to rebuilt that desktop system. Is the CPU an early i7 or something of the old C2D/C2Q variety?

I have the same attitude as jtr1962 towards spending on a pc system before it's necessary. In my case, it's not necessary.

The CPU is an Intel Core 2 Duo E8400 "Wolfdale." Ever since the March 2012 memory bank swap event, it has been working flawlessly with the exception of the CPU-Z program failure which only started 4 days ago. The system is very stable, considering how often it's on in a day and how long it's been on active duty!

 

[apairofpcs]

Final status report:

At 10:15 PM last night, after 11 hours and 12 passes running Memtest86 + v. 4.20, the desktop pc had NO ERRORS. I'm satisfied with the integrity of my 4 GB of memory.

By the way, I removed every reference to CPUID CPU-Z from my hard drive and registry. I'm hoping this action will bring the end to my pc's freezes!

 

[Mercutio]

Memory tests of all kinds - even the ones that use a hardware tester - are notoriously unreliable. Passing Memtest86 really just means that RAM passes Memtest86, not that it's certifiably problem-free.

 

[Stereodude]

That may be true, but I've tracked down system instabilities to RAM before with Memtest 86+ & Goldmemory.

 

[ddrueding]

I think it might be more accurate to say that it can find problems, but not all problems. If RAM fails Memtest 86+ it is bad, if it passes it might be good.

 

[apairofpcs]

Memory tests of all kinds - even the ones that use a hardware tester - are notoriously unreliable. Passing Memtest86 really just means that RAM passes Memtest86, not that it's certifiably problem-free.

I'm happy to have a program that can verify that all of my memory is functioning.....at the time of the testing. Right now, I have 2 x 2 GB of memory that is working for me, especially since the only program to cause a freeze is gone. Do we as humans get a guarantee from a doctor that the health present during an office visit will continue indefinitely?

 

[apairofpcs]

That may be true, but I've tracked down system instabilities to RAM before with Memtest 86+ & Goldmemory.

I've ran Memtest86 on my desktop and 2 laptops many times, without being suspicious of the reliability of the memory. In each case, the result was No Errors. I've never had a pc where the memory failed. It was really no bother running Memtest for 11 hours "unattended."

If Memtest finds errors does it mean that the memory is defective, or can the result be a false positive? Is it logical to believe that a positive result is conclusive, and a negative result is inconclusive?

 

[apairofpcs]

I think it might be more accurate to say that it can find problems, but not all problems. If RAM fails Memtest 86+ it is bad, if it passes it might be good.

I couldn't have said it better myself!

 

[Mercutio]

If Memtest finds errors does it mean that the memory is defective, or can the result be a false positive? Is it logical to believe that a positive result is conclusive, and a negative result is inconclusive?

At the point that a memory test is finding errors, it's just time to replace the RAM. RAM is cheap.
I've seen quite a lot of RAM that can pass memory tests all day long yet was still defective, but I will also say that modern RAM is vastly more reliable than the stuff we were getting 10 or 15 years ago.

 

[apairofpcs]

At the point that a memory test is finding errors, it's just time to replace the RAM. RAM is cheap.
I've seen quite a lot of RAM that can pass memory tests all day long yet was still defective, but I will also say that modern RAM is vastly more reliable than the stuff we were getting 10 or 15 years ago.

I would rely on Memtest having doneit's job well, if it finds an error. No argument here. I think that we have to rely on the "scientific" method of carrying a rabbit's foot, keeping our fingers crossed and throwing salt over a shoulder, when there are no errors. Do you agree?

In our favor is the highly reliable manufacturing process that gives us such good products. I've benefited from this, with every pc I've owned never having bad memory.

 

[jtr1962]

At the point that a memory test is finding errors, it's just time to replace the RAM. RAM is cheap.
I've seen quite a lot of RAM that can pass memory tests all day long yet was still defective, but I will also say that modern RAM is vastly more reliable than the stuff we were getting 10 or 15 years ago.

For what it's worth, I've yet to see a bad stick of RAM in the 15 years I've been playing with PCs. Come to think of it, I haven't had a hard drive fail on me yet, either. Maybe I've just been exceeding lucky, or perhaps it's because I've taken certain precautions when dealing with anything electronic.

I'm sure though if my sample size was larger I would have encountered bad RAM by now.

What are the thoughts around here on using Prime95 and IntelBurnTest for stability testing? I've recently used both to uncover some stability issues on my new PC. The machine has basically been stable except when doing anything which stresses the GPU (one of my train sims pegs the GPU at 100% nearly all the time). Anyway, to make a long story short, I upped the north bridge voltage from 1.2875 to 1.3000, and also upped the GPU voltage from 1.20 to 1.25. The stability issues disappeared. As a bonus, I'm able to overclock the GPU (894 MHz instead of the stock 800 MHz). I also found out that I could reduce the CPU voltage slightly while still maintaining stablity. This helped with temperatures somewhat. Temps stay at 57°C or below no matter how hard I'm pushing the system. CPU multiplier is at 40 instead of the stock 38. Something psychological about having a 4 GHz machine. I probably could have pushed it to 43 or 44 without much trouble according to my reading but I'll opt for stressing the system less over slightly more speed which I'll seldom use. As it is CPU usage is usually under 10% most of the time.

BTW, IntelBurnTest is giving me 18.3 GFlops. Don't know how that compares to some other machines out there, but my old A7N8X-E with the XP3200 could barely manage 1 GFlop. IntelBurnTest doesn't test the GPU however, and AMD says it's capable of 672 GFlops (single precision). Precious few programs these days use the GPU though.

 

[Mercutio]

I very seldom bother with burn-in testing. Maybe once in a great while on a $1500+ desktop or on a server.

 

[Chewy509]

Stupid thought of the day, doesn't CPU-Z probe the chipset and hardware specifications in particular the memory bank setup? Maybe it's touching something that's causing the lockup? (eg it could be the chipset is faulty, but only seen when it's being probed by CPU-Z - as you did mention a memory bank swap).

PS. What motherboard/chipset? Being Wofdale, it's s775, so most likely a 9xx series or an early X series? (X38/X48)...

 

[ddrueding]

Stupid thought of the day, doesn't CPU-Z probe the chipset and hardware specifications in particular the memory bank setup? Maybe it's touching something that's causing the lockup? (eg it could be the chipset is faulty, but only seen when it's being probed by CPU-Z - as you did mention a memory bank swap).

PS. What motherboard/chipset? Being Wofdale, it's s775, so most likely a 9xx series or an early X series? (X38/X48 )...

Sounds like an excellent thought, actually.