Problems with computer. No display. (DRAM debug light on)

LorenaRose

Commendable
May 6, 2016
9
0
1,510
Hello everyone.

I recently put together a custom computer but I could use some help.

I recently got it up and running and everything was going fine and I was playing a game and all of a sudden the screen froze and I was unable to get a response from the mouse or keyboard and had to shut it down. I turned it back on to try and see why it had crashed and it turned on for a few minutes and then did the same thing, only now everything turns on and it sounds like everything is fine, it no longer displays anything.

The DRAM light is solid and will sometimes flicker to the CPU light as well. I've tried using only one RAM stick and I still got nothing. I have already tested the screen and it works fine. Does anyone know what could be the problem?

FULL SPEC LIST:
Intel Core i7-6700 3.4GHz Quad-Core Processor
MSI B150 Gaming M3 ATX LGA1151 Motherboard
Corsair Vengeance LPX 16GB (2 x 8GB) DDR4-2133 Memory
Seagate Barracude 1TB 3.5" 7200RPM Internal Hard Drive
MSI GeForce GTX 970 4GB Twin Frozr V Video Card
Corsair CX 600W 80+ Bronze Certified Semi-Modular ATX Power Supply

This is my first build so please, have mercy. :)
 
Solution
You don't have to purchase the PSU I linked; it was just a suggestion to show you the difference between PSU specs. Ideally, though, having at least two +12V rails will help, as one rail will be dedicated to the GPU, while the other will be used for the rest of your system.

This tip is by no means a requirement for stability, considering that ultimately, as long as the +12V rail has sufficient amount of amps needed, your system would be fine. However, having the multiple rails improves the cleanliness of the voltage, which improves stability (less electrical traffic equals less noise, which in turn equals better flow, which results in better performance).

MSI has tested the Corsair CX600 and it has been proven successful; however...
First suspect is RAM. This isn't indicative of the problem, but I don't see your RAM on the compatibility list. Generally, you should purchase parts listed on the compatibility list. This is good practice for at least a couple of reasons: 1. Manufacturer can't hide behind incompatibility when RMA or tech support is requested; and 2. The compatible RAM has been tested and known to work.

If you got another DDR4 DIMM that is known to be working, try it in your mobo (motherboard). Alternatively, test your Vengeance sticks in another PC.

The next suspect is PSU. Though the brand is known, that PSU isn't very good, in my opinion. Your PSU has only one +12V Rail, which is what is taking the load of all your connected parts. Discrete GPUs are very power hungry, so you might want to consider a better quality PSU. This PSU could handle your build better - it has 2 x +12V 30A rails for better load balance.

Third suspect would be motherboard. Though your mobo is a good quality board, it is still possible to have defects, but unfortunately there isn't much you can do to test a mobo, so try RAM and/or GPU tests first (test them in another PC would be ideal).

Lastly, your CPU. The CPU is the least common failing part, but it is still possible. Ensure your temps are within tolerable ranges. Software like Prime95 is useful at stress testing your CPU. You can monitor your temps with CPU-Z.
Good luck
 

LorenaRose

Commendable
May 6, 2016
9
0
1,510


Thank you for the reply. I will be returning a few of my items soon to get some of MSI's confirmed compatibility list. http:// This is the link to the PSU list from their site. Would you recommend any of them? Or should I stick with the one you linked?
 
You don't have to purchase the PSU I linked; it was just a suggestion to show you the difference between PSU specs. Ideally, though, having at least two +12V rails will help, as one rail will be dedicated to the GPU, while the other will be used for the rest of your system.

This tip is by no means a requirement for stability, considering that ultimately, as long as the +12V rail has sufficient amount of amps needed, your system would be fine. However, having the multiple rails improves the cleanliness of the voltage, which improves stability (less electrical traffic equals less noise, which in turn equals better flow, which results in better performance).

MSI has tested the Corsair CX600 and it has been proven successful; however, you don't know what they used to make the build that had the CX600. It is that reasoning that I always buy a PSU with at least two +12V rails (w/ min 20A each), to ensure that I have more than the requirement of the GPU.
 
Solution

LorenaRose

Commendable
May 6, 2016
9
0
1,510


Oh ok. And yeah I replaced my ram with compatible ones and my computer turned on and worked for a few minutes, froze, and then I had to shut it down again. Again there is no display. I tried running it without the graphics card and still nothing. Only now none of the debug lights are on.
 
Try this:

1. Disconnect all cables, except for the CPU fan
2. Remove motherboard from the case
3. Place motherboard on the motherboard box, or any non-conductive surface (newspaper, cardboard, wood, etc.)
4. Insert ONE stick of RAM into the closest slot to the CPU
5. Connect the PSU - ensure you connect both the P1 (24-pin) and P4 (8-pin) connectors

At this point, all you should have connected to your motherboard is RAM and PSU. Connect a monitor to either the HDMI or the DVI port. To turn on your motherboard, use a screwdriver (or other metal object) to jump the two PWR_SW pins (do the same thing to turn off). If you get a display, turn off and add another DIMM, and then turn on again. If you get display again, turn off and then try adding the GPU, and then turn on again.

If you get a display with the GPU, turn off, and then add your HDD. Keep repeating this pattern of adding device and testing for display until:
A. You've successfully gotten a display with everything connected to the motherboard, while it is outside of the case (at which point the likely problem is an electrical short caused by the motherboard making contact with the case); or
B. You've found the part that will not allow your computer to boot with display (in which case there's an open circuit within the device, thus making the electrical current flowing through unable to reach ground, and ultimately causing an incomplete circuit).

Edit: While removing your motherboard from the case, take a good look at the standoffs, as there may be one installed unnecessarily. If this is true, remove the standoff and before adding parts back into the case (and after performing the above steps).
 


I'm sorry if you have already done this, but i want to be clear, so i'll ask specifically...
1. Have you tested known-to-be-working RAM on your motherboard?
2. Have you tested a known-to-be-working GPU on your motherboard?
3. Have you tested a known-to-be-working PSU on your motherboard?

If the answer is yes to all of the above and the results are the same, then the likely culprit is a bad motherboard.

If the answer is no to any of the above, you may want to perform the tests.

Here are my observations thus far:

1. Initially, you built the computer, successfully, i might add, but then the computer froze. Freezing is typically the result of faulty RAM, but the problem may also be HDD, Windows, or CPU (in that order of likelihood)
2. Upon a forced reboot (you had no choice), the problem happened again. A second time of the initial problem in such short span of time suggests hardware over software. Time to start looking at your parts a little closer.

3. When trying to turn it on a third time after the aforementioned freezing, you powered on, but had no display. The absence of display is typically caused by faulty RAM; PSU; GPU; mobo; or CPU (in that order of likelihood). In this instance, you had the motherboard LEDs lighting up, but the DRAM LED would flicker (this suggests a PSU or mobo problem, as the LED is supposed to illuminate consistently, but depending on POST, the pattern will differ)

4. As per my recommendation, you replaced your RAM with something from the compatibility list; tried again to boot, but found that the freezing continued (sounding like a PSU or HDD problem)

5. You have not, or least have not mentioned if you did, replaced your PSU, but have performed the steps I outlined to build your system outside of the case (to rule out case shorts), but you have not specifically mentioned what the outcome was (Did the system boot properly with display, but with the lines across the screen, as mentioned? or did the system boot and freeze, with the lines mentioned?)

If the computer does in fact boot w/o GPU; does not freeze, but has the problematic lines, this suggests PSU or mobo as the culprit.

On the other hand, if the computer boots, but still freezes, this would suggest PSU.

So now, the burning question is: Did you replace your PSU (or test a known-to-be-working one in your system)?

 

LorenaRose

Commendable
May 6, 2016
9
0
1,510


I have not tested a new power supply.

When I rebuilt out of the case I got to the point where everything was in, but after about 10 minutes of it being on it would still freeze with hatched lines.

As far as the flickering goes, I think blinking might have been a better term. The DRAM and CPU would blink back and forth, or it would have a solid DRAM led.

So at this point, the motherboard is back in the case. When I power it on, the DRAM and CPU lights blick back and forth, stop, then one solid DRAM light shines for a few seconds, stops, then the VGA light is solid, stops, and then the computer boots with display. After the computer boots up for a few minutes, it freezes, gets colored hatched lines across the screen, and I am forced to shut it down. All this is without my GTX 970 attached.
 
OK. Thank you for the great details - this helps. Right now, it looks like the motherboard is the culprit, but because of the consistent ability to POST w/o error, and boot Windows w/o error, there is also the possibility of CPU. Typically, a CPU will shut down when the temperature gets too high. This is done to prevent damage from overheating. However, with a problematic motherboard or PSU, this process can be interfered with, thus causing freezing.

The PSU is still a possibility here because it may not be providing enough power to keep the system up, or the PSU may have 'dirty' power flowing through, causing instability.

Since there isn't much you can do to rule out motherboard (the only way i know of is to put known working parts on it, and then fire it up), I suggest we start looking at other things.

Try this:

1. Turn on your computer, and go into your BIOS/UEFI (usually F2 or Delete at motherboard/BIOS splash screen)
2. Let your PC sit here for 10-20 minutes
3. Check to see if it froze by moving the mouse (UEFI) or using the arrow keys (BIOS)

If computer freezes here, you definitely have a hardware problem. If it does not, your problem may be in Windows, so try this:

1. Turn on your computer
2. Immediately after the motherboard/bios logo, repeatedly press F8
3. Once in Advanced Boot Options, use the down arrow and choose to start in safe mode. Keep in mind that safe mode starts with minimal drivers and services, so you won't have network and you'll be limited to a low resolution.
4. Once you're in Windows, press/click the start/windows key/logo, and type "Event Viewer"
5. Open the Event Viewer, save the log, and then post it here for analysis.

At this point, the true test will be to let your computer sit (or do what you were doing when computer froze previously) and see if the freezing happens again. If your PC freezes again, then it may be worth trying a new PSU. If you don't want to try this, consider cleaning your CPU and its heatsink to remove all thermal paste. Then reapply thermal paste. I have complete faith and trust in Artic Silver 5, but have also used some unknown brand (and I can't recall the name either) from Best Buy with success. Arctic Silver 5 Application Guide (PDF)
 

LorenaRose

Commendable
May 6, 2016
9
0
1,510


It's a hardware problem (froze up on the BIOS screen). Could it be the CPU? I'll look into that PSU as well.
 
I believe your problem is with the CPU. If I'm not mistaken, that CPU is/was code named "Skylake", which has had problems, but the problem was with high performance/heavy load. See article here. MSI has updated BIOS patches that should address your concern. See here
If you're uncertain about flashing the BIOS, you may want to reach out to MSI and ask them if either the two recent updates will address the freezing problem you're having, for your problem isn't the same as the one noticed by the guys at GIMPS (the makers of Prime95).
 

LorenaRose

Commendable
May 6, 2016
9
0
1,510


So all I have to do is download this driver to see if it's the problem?
 
The BIOS update is not a driver, rather think of it as the complete programming for your motherboard. BIOS flashing has gotten increasingly less difficult over the years, but must still be done in a proper manner. Improper BIOS flashing can lead to a dead motherboard. I'm not an MSI tech, so I can't tell you conclusively that the BIOS update will fix your problem, but if your CPU is one that has the codename Skylake, it has been known, acknowledged, and documented by Intel.

Chances are that if you were to start a case with MSI, you'd have to provide them with the make and model of all the parts of your build. With the Skylake problem widely known, MSI would likely point the finger at Intel and tell you to update your BIOS to address the freezing concern. The BIOS update will have a different revision or version number than currently loaded on your motherboard.

Actually, now that I think about it, maybe you should check your BIOS version first, and then compare it to the latest update. If your BIOS version matches the update, then you don't have to do this and you should then make contact with MSI to see about getting a replacement motherboard.

So, to sum it all up:

1. Your problem could be the CPU or the mobo
2. Check your BIOS version and compare it with the latest update
3. If your BIOS version and the update's version match, contact MSI tech support
4. If the BIOS versions do not match, update your BIOS with the latest version, and then see if the freezing continues
5. If freezing continues, contact MSI and Intel for RMA (Return Merchandise Authorization) - maybe even contact the vendor which sold you the parts, as they may have a solution for you, too.
 

LorenaRose

Commendable
May 6, 2016
9
0
1,510


So I ended up buying a new CPU and that seems to have fixed my crashing problems. (Yay!) However, when I try and reinstall my GPU, I get no display. I'm also stuck in an infinite Checking for updates loop for windows. It's like I fix one thing and then another problem always another..
 
I'm thinking the PSU is the culprit, but that's assuming you've tested your GPU in another system. Remember, MSI tested the PSU you bought, but their compatibility list doesn't disclose what other parts were used with the PSU. So, for all we know, they tested that PSU w/o GPU.

I certainly understand your frustration; I've been down that road several times with my builds, but after you figure out the problem and get it fixed, you learn from the problem. Soon you'll be diagnosing computers here.
 

LorenaRose

Commendable
May 6, 2016
9
0
1,510


I'm just an idiot and forgot to change the display to the GPU instead of the Mobo. Now I just need to figure out why I'm stuck in the never ending checking for Windows update