Scott Mueller's Upgrading and Repairing PCs Forum

Greetings to all forum members.

I have committed an act of great foolishness, i.e. I used a vacuum cleaner to remove the dust from my PC. I know, you're probably thinking, “What kind of a dimwit would do that?” Well, you'd be justified in thinking that, but please don't rub it in, I am already in a very depressed state because of the consequences of my foolishness.

I have damaged some of my PC's hardware through electrostatic discharge from the vacuum cleaner. After I had vacuumed the dust out of my PC, the PC wouldn't even POST anymore. The screen was blank and it just kept restarting maybe a dozen times or more. Then after restarting after I don't know how many times, it would POST, but the POST screen showed 8 MB less RAM than is installed, i.e. the POST screen showed 8184 MB instead of 8192 MB. I was subsequently able to boot Windows 7 but the PC crashed relatively quickly.

Here are some relevant system details:

Motherboard: ASRock P55 Pro
RAM: four DDR3 modules, each one 2048 MB in size
CPU: first generation Intel Core i7 860 Nehalem
Video card: I think it has the AMD 5750 chip and 1 GB of RAM

After doing some online research I determined that the motherboard was defective. I purchased a used motherboard which is exactly the same as my original motherboard.

After installing the replacement motherboard, the PC now POSTs right away when the power is switched on. But I still have the problem with the POST screen showing 8 MB less RAM than is installed.

I have four RAM modules installed, each one 2048 MB in size. I tried installing just one module and the POST screen shows 2040 MB of RAM. I also installed the other three modules one after the other and the POST screen always showed 2040 MB of RAM for each module. When I install two modules, the POST screen shows 4088 MB of RAM instead of 4096 MB. When I install three modules, the POST screen shows 6136 MB of RAM instead of 6144 MB. And lastly, When I install four modules, the POST screen shows 8184 MB of RAM instead of 8192 MB. When I enter the BIOS setup, total RAM is always shown correctly. So when all four memory modules are installed, the BIOS shows a total RAM of 8192 MB and shows 2048 MB installed in each of the four slots. If the BIOS is simply reading the SPD chip of the memory modules to get the information regarding the amount of RAM installed, then that information isn't meaningful regarding the individual RAM module's state of health.

In addition to the problem with the RAM, I also get a “68” code on the two digit so called “Dr. Debug” display on the motherboard. This code appears approximately 1 to 1.5 seconds after I have powered on the computer, then remains displayed continuously for approximately 4 seconds. Also, when I enter the BIOS setup, code “87” is shown on the display. When I exit the BIOS setup, the two digit display goes dark again. Unfortunately ASRock's description for both of the aforementioned codes isn't very specific. to the description of the “Dr. Debug” error codes. Both codes fall into the number range whose description is, “Chipset initialization error. Please press reset or clear CMOS.” I first tried the reset option, that didn't help, and after that I cleared the CMOS using the button on the motherboard. That didn't change anything; the aforementioned error codes are still present and so is the problem with the RAM.

I am confused by the code system that displays on the two digit “Dr. Debug” display. Some (hexadecimal) numbers seem to represent specific steps in the PCs start up routine. These steps are documented on pages 32 – 34 of the manual for the ASRock P55 Pro motherboard (). So these steps are not errors. But then some of these “step code” numbers are also included in the number ranges which define error codes. For example, in the manual, code “75” means “Initialize Int-13 and prepare for IPL detection.” and code “78” means “Initializes IPL devices controlled by BIOS and option ROMs.” But 75 and 78 are also part of the number range 61 – 91 which is described as “Chipset initialization error. Please press reset or clear CMOS.” in the . I find this ambiguity confusing. Have I misunderstood something here?

I ran some memory tests using two of the programs on the Universal Boot CD and they didn't find any errors. However, I don't know if these tests are actually meaningful, because in URPCs 22nd ed. it says to disable the memory caching before running the tests. I was unable to find any options in the BIOS for disabling the cache memory. L1, L2, and L3 cache are all integrated into my Core i7 860 Nehalem processor so I thought that perhaps none of these caches can be disabled. Is this correct? If it is, are the memory tests I ran meaningless?

One of the two programs I used for memory testing was Memtest86+ 5.01. In the information block in the upper left hand corner of the screen it says, “Memory: 8183M 10187 MB/s”. There is also a line of information in the program that is labeled “Testing:”. At one point in time this line would display, “Testing: 6144M – 8192M 2048M of 8183M” and shortly after that “Testing: 8192M – 9124M 832M of 8183M”. I find this all quite confusing. How should I interpret the first part of the line (“8192M – 9124M”) in relation to the second part (“832M of 8183M”). Does the “M” in the first part stand for megabits and not for megabytes?

I haven't tried booting Windows since I'm afraid of corrupting my Windows installation.

I don't know what is causing the errors described above. I suspect it's either the motherboard – the used motherboard I purchased cannot be considered a known good replacement – or the processor. Or could all four of the RAM modules be faulty? I suppose this too is a possibility.

I'm not sure if my Core i7 860 Nehalem processor has an integrated memory controller and how it would affect the interpretation of the motherboard error codes if it does. When I read “Chipset initialization error.”, I thought that this must refer to a defect on the motherboard. But perhaps “chipset” could also refer to a fault in the processor? In my opinion these cryptic “Dr. Debug” codes aren't really much help in pinpointing the cause of any faults.

I tried removing the video card and powering up the system. The “68” error code still shows up as described. I also get an additional “2E” code. According to the motherboard manual, this code means “Initializes all the output devices.”; so this code is to be expected since a necessary output device is missing from the system. I can't verify the “87” code, since without video, I can't access the BIOS setup.

I've also tried removing all of the RAM, leaving the video card installed, and then powering up the PC. Then all I get is an “E8” code on the two digit “Dr. Debug” display, there are no codes shown before or after this “E8” code. The code is accompanied by three beeps in succession which repeat continuously. The PC won't POST or run without RAM installed. It goes straight to the “E8” code and there it remains until it is switched off. I suppose this is to be expected.

I should also mention that once or twice during all of my troubleshooting endeavors I powered on the PC and “00” appeared on the “Dr. Debug” display, no codes were displayed before or after the “00” code. The PC wouldn't POST. When I hit the reset button on the motherboard, the PC would start up again and POST. In the “Dr. Debug” error code list “00” means, “Please check if the CPU is installed correctly and then clear CMOS.” I didn't do either, since I am positive that the CPU is installed correctly.

I should probably mention that during all of the hardware installation procedures I performed during the motherboard swap and subsequent troubleshooting I was very careful to use all of the necessary ESD protection measures. So I used an antistatic mat and connected myself and the motherboards to a common connection point on the mat. Once the motherboard was installed, the PC case was connected to the common connection point. I even wore ESD protection gloves.

I know I've screwed up really badly here. I'm in such a depressed state and I would really appreciate any help you, Scott, and any other forum members can give me.

I don't have access to any known good spare parts and just purchasing used replacement parts because I suspect a part is defective, but am not certain that it is, isn't cost effective. I really don't know what to do now. I'm hoping you folks can point me in the right direction.

Sorry for this complicated post, but since this ESD disaster happened over a week ago I have been in a mixed up state of mind.


Kind regards,

Daniel

I suspect there may not be anything wrong. You should try to boot into Windows or do a clean install and see how things go. You could always do a clean install on a spare hard drive you have, run it for a week or so just to see how things go. If everything is fine, either continue to use that install or switch back to whatever drive you were using before.

Other than that, I'd try re-seating the cpu and clearing the cmos following page 24 in your manual. Just because you said have not tried that yet. I'd follow the directions on page 24 directly after re-seating the cpu.

That said, I'm not exactly sure and I'd wait for Scott to reply as I'm also interested to hear what he has to say about this.

It sounds like everything is normal after the motherboard replacement. The CPU and chipset appear to reserve 8MB of the memory for internal use, and the "diagnostics" you are running are either completely ineffective or misleading. What happens after you install a new drive and reload Windows or boot to your existing copy? Scott.

So, Daniel, How did this turn out?