Legit Reviews

Hello,

I got a rare problem with my system, where sometimes it boots normally and sometimes it doesn't. When it doesn't, there is no signal to the screen and the cpu is on, doing nothing. I got a MSI P6N Diamond motherboard, having this D-Bracket 2 connected to a JDB1 connector. This bracket has two USB ports and 4 Leds, which would turn on in a combination in such a way it "talks" to me what's going on. According to the manual, when Led 2 is green while the rest are red:

Memory Detection Test
Testing onboard memory size. The
D-LED will hang if the memory module
is damaged or not installed
properly.

So I thought, if the memory modules are damaged, the system would never boot correctly, right? But it boots fine sometimes. So I tried removing all the 4 memory modules (OCZ SLI Edition PC6400 1GB each), leaving only one of them installed. For simplicity, I also removed both GPUs (GTS G92 512mb) and the sound card (Sound Blaster Xtremegamer). I tried again and system did not boot. So I removed that memory module and installed another one, to no avail. Did the same with the rest, and system did boot with the 4th memory module. But when I installed the GPU to have screen signal, the system failed to boot again.. So I uninstalled the GPU again, to try to boot the way it just did before, and it did not boot... So I randomly tried by installing an additional memory module (having now 2 memory modules in total now), with the GPU again, and the system did boot..

And finally, the system was soaked in dust, making me think that maybe the dust was causing this. But I've already completely cleansed the system from dust. The temperature? I got the Gemini II CPU cooler, I've restored the BIOS setting to fail-safe defaults, so got no overclocking, and I got 9 fans blowing inside and outside.

What is going on here? I'm so lost, the system boots on its own whim. What can I do? Where should I look to find the issue which sometimes prevents my system from booting?

It could be the power supply. A Bad power supply voltage is not stable and when the computer is booting up if the power supply is not giving the correct voltages and or power then the system can lock up.

Also take a look at your capacitors on your motherboard and make sure none of them are bulging.

Well I got the Thermaltake W0131RU 850W Toughpower PSU, and I've heard how good a PSU that is. I did not find any capacitor bulging after checking all the motherboard completely.

kind of sounds like a PS problem to me too. It is a good PSU, but that doesnt mean that there is nothing wrong with it, have you checked the voltages? Also whatever memory you have in there you might want to run memtest on.

You're on the right track. When a computer doesn't turn on at all it usually points to the PSU or RAM. What voltage and timings are you currently running your RAM at in the BIOS? It might be as simple as loosening a timing or raising a voltage.

I haven't thought of running memtest, I will do so tonight and report back.

I will check the voltages. But how do I do that? I know almost nothing about electrical stuff.

And actually, there are serious electricity shortages here in Caracas, and perhaps maybe the electricity that the government sends us is poor quality? I'm just speculating here, but could this happen? Does poor quality electricity exists that could prevent my system from booting sometimes? I also have a voltage regulator. I don't think so since all computers in my office work fine, but are very low end systems with Pentium IV. Maybe an enthusiast gamer's system requires "better quality electricity"?

Major_A wrote:You're on the right track. When a computer doesn't turn on at all it usually points to the PSU or RAM. What voltage and timings are you currently running your RAM at in the BIOS? It might be as simple as loosening a timing or raising a voltage.

No, the computer does turn on, but it sometimes does not boot. I mean, everything is on, but there is no signal to the screen, the first welcome screen depicting motherboard version and all that does not even appear, nothing, screen is dead black. And sometimes it does boot correctly to windows and I can play Fallout 3 maxed up smoothly... what the heck?

In a successful boot, I loaded all to the fail-safe defaults, so I perhaps think that the timings should not be an issue. Anyways, before doing this, the RAM timings and voltages were set to exactly OCZ's recommended values.

When I said at all I meant the screen would not come on, not that nothing would happen. Could be an issue at the wall. Are you overclocking at all? When they say to check your voltages they are implying you use a multimeter or at least a piece of software (ala Everest or CPUID's HWMonitor).

My system was overclocked (Q6600 @3 ghz), prime95 stable. I was running fine for 2 years, and this problem begun recently. I forgot to mention that in one of the boots I got a screen saying that the previous overclocking failed and that the BIOS parameters had to be defaulted. So I went to the BIOS, saw that everything was at default, and could reboot fine. The next time I turned on the system, it did not boot, and I changed nothing. And strangely enough, this same screen appeared again in two more occasions, and my system was not overclocked anymore... I don't understand what's going on.

I'm concerned since now it almost never boots, I don't know if I will able to boot to try to run everest or any voltage measurer. The system has been booting each time less.

- There are no bulging capacitors
- The motherboard makes absolutely no beeps
- How can I test if my PSU is ok? I got no multimeter here
- My system is just not booting anymore.. I can't run everest nor any similar program, nor I can run memtest, the system just stays dead on, without absolutely any response
- There is no overclocking
- I only have one memory module connected (1GB OCZ SLI Ready Edition), the Q6600, its stock cooler, the cables from the PSU to power up the motherboard, and the aforementioned D-Bracket 2 to see if system boots fine. But nada, the green stays in LED 2 and the rest are red, meaning that it's dead stuck in the boot process.

Seriously, what can I do?

Sometimes, keep in mind sometimes, when it does say overclocking failed and you enter the BIOS it doesn't change anything. Reset your CMOS, after that pull the power plug out of the back of the PSU and hold down the power button on the tower for 10 seconds to "drain" the power from the capacitors on the motherboard. Report back.

Major, I did exactly as you told, and then plugged back the PSU cord, turned the system on, and still the issue persists.

Other options:
Reseat the CPU.

Is your 8-pin CPU power cable attached securely? Unplug all connections to the motherboard and reattach them.

There still is a possibility that the RAM you have is bad. After resetting the CMOS and trying to start it did you try any of your other RAM?

I've already tried all that, except for replacing the memory module after resetting the CMOS. So I just did this, and it did not work.

I don't think that one of the memory modules are defective, since if that was the case, then the system would just abruptly stop booting correctly forever, right? But that did not happen. When this issue began, the system almost always booted correctly. But as time passed, what begun to happen the most was failure to boot, until as of now the system never boots. So something was gradually deteriorating there.

Not necessarily. Somethings die over time, especially RAM. If it's requiring a lot of voltage to remain stable it can slowly kill it.

You need to find someone with a computer you can test components in. At this point it almost sounds like the motherboard might be bad.

I just could make the computer boot, I changed the surface from which the motherboard was lying, since it had a small metal part which it seemed it was shorting the system.

Now it always boots, but I have 2 problems now: it is always freezing randomly, and the system is only detecting 2 cores from the Q6600. In dxdiag menu, it shows that it is the Q6600 processor, with 2 CPUs.. and in CoreTemp it's showing that it's the E6600 (Conroe) processor.

bandieramonte wrote:I just could make the computer boot, I changed the surface from which the motherboard was lying, since it had a small metal part which it seemed it was shorting the system.

Now it always boots, but I have 2 problems now: it is always freezing randomly, and the system is only detecting 2 cores from the Q6600. In dxdiag menu, it shows that it is the Q6600 processor, with 2 CPUs.. and in CoreTemp it's showing that it's the E6600 (Conroe) processor.

search around in the bios advanced options for anything that would disable cores? most boards have an option for how many cores are used

I just did this, and there is nothing which would disable cores from the processor. I reseated the heatsink, and after rebooting, I found the windows began to detect the 4 cores again. But it's still so unstable, freezing randomly all the time.

have you played with the voltages(when i reset CMOS it defaults my DDR3 voltage too low which makes it very unstable, same can happen for CPU)

Okay so if it was shorting against the case then it might have been enough to damage the motherboard. At this point all you can really try to do is reset the CMOS, drain the caps, and start it back up.