Black Screen When Running FurMark and Almost All Games

[iCsav]

Hey guys. So, my first post here. I got a custom build about 1.5 months ago. It has got a GTX 950. The mistake I made was not testing some games on the damn thing sooner. Approx. 20 days ago, I installed Dying Light and on starting it, it played out the intro, got to the checkpoint screen and blacked out. I could hear weird noises. Not the game sounds, but buzzing, whirring and beeping until the monitor switched off after a minute or two. The only solution was to hard reset using case button. I thought it might have something to do with settings and left it alone. Due to real life issues, I couldn't do anything till a day ago. Optimistic, I tried it again. Same as before. Tried reinstalling and everything on forums about DL. I tried other games. This time from Steam. No luck. Then, I even tried a very new game - The Lost Mythologies. The opening screen comes on and the the same thing. This was suspicious. I tried FurMark (at 720 preset). I got the same results. Black screen, no sounds this time and monitor's message: "Power saving mode", which is basically no output signal from the GPU. I can't even monitor GPU temps or anything while running these things. The only game running (out of Windows 7 provided games) is Osu!. I found this thread. Seems pretty similar to mine. Would you guys suggest anything from it? Any other solutions? I can provide any other info you might want. I don't have any overclocking applied.

 

[ICIT2LOL]

Hello and welcome iCsav mate I am assuming the machine runs normally otherwise?? If so please run speccy and let us know what the temps are at "normal" pace.
Speccy - Free Download - Piriform.

You might find a bit more in depth details using this ditty of mine
Using HW Info
You can test the volts on the PSU with HW Info HWiNFO, HWiNFO32/64 - Download < download the right bit version and close the right hand window select Sensors and scroll down to the power section where you will see what the volts are doing see my pic.
The original right hand window shows the machine running and is handy for that but for looking at the components in some detail close it and use the main left hand side panel
FOR OTHER COMPONENTS
Open each + down into the component itself and then click on it - in the right hand side will appear all sorts of details including brands speeds and other essential info that device. See pic for example.
Use both PARTS and I would be interested to see the results from the PSU please

Edit: HWiNFO, HWiNFO32/64 - Download go to the bottom of the page and download from the uS site

 

[iCsav]

Here are the test results.

Form Speccy:

CPU
AMD FX-8350 11 °C
Vishera 32nm Technology
RAM
8.00GB Single-Channel DDR3 @ 803MHz (10-10-10-27)
Motherboard
Gigabyte Technology Co. Ltd. 970A-DS3P (CPU 1) 27 °C
Graphics
L195 Wide (1440x900@60Hz)
2047MB NVIDIA GeForce GTX 950 (NVIDIA) 34 °C
Storage
931GB Seagate ST1000DM003-1SB10C ATA Device (SATA) 28 °C

From HW:

I'm not sure if I got you what you wanted from this one, so please tell me if you need anything else.

Also, another thing I noticed was that now, even playing offline videos in fullscreen results in the same thing happening, albeit without sound. This was never happening before 2 days ago. I use MPC-HC with KCP.

 

[ICIT2LOL]

Ok mate the rail voltage for the 3.3v rail is not good and I am not usre why the 12v and 5v rail volts are not being shown. I just wonder if you downloaded the right bit version of that software. Having said that it doesn't always support or rather the machines sensors are not supported by HW Info.

The best way to know for definite is to measure the volts on the 24 pin plug with a digital multimeter. They are very cheap these days and all you need to to is place the black lead on any black cabled pin and then check the other colours for the voltages see the codes in this for the colour / volts .
https://en.wikipedia.org/wiki/Power_supply_unit_(computer) scroll down that reference and you will fin the colour coding NB the Power Good should read 5v when the machine is running if not then it will shut down the machine or not let it start at all.

 

[iCsav]

Thank you for your help. I'm not sure that I can get the required multimeter anytime soon. I'll check if someone I know has it (iirc, someone does) but doing so may take a while - about 20 days. Is it alright to continue this thread after 20 days or perhaps PM you? I'm not sure how this forum's rules work. After how many days are we allowed to bump?

 

[ICIT2LOL]

Yep keep this thread going mate any response comes through as you move a along. If you start another it might well get lost amongst the others.

 

[Layback Bear]

Just for information.

I notice that the ram is in single channel on a dual channel motherboard.
Might also be a ram problem.

From post #3

8.00GB Single-Channel DDR3 @ 803MHz (10-10-10-27)
Motherboard
Gigabyte Technology Co. Ltd. 970A-DS3P (CPU 1) 27 °C

 

[iCsav]

@Layback Bear, could you expand on that? I don't know much about hardware. Do I need to check for something specifically?

 

[Layback Bear]

Sure I will. In layman terms.
Please note: You may have more than one problem.

If your ram is install as per the motherboard manual it should be in dual channel..
If you have a bad ram module or motherboard ram slot it could shift to single channel. You will need a ever number of ram modules install in the proper slots.
A bad ram module and/or ram slot can cause various problems. It's possible that ram problem could cause goofy voltage readings.

Sometime such a ram problems will not cause a BSOD.
To me the next step would be to verify you have the proper ram installed properly as per the motherboard manual.

Please answer these questions.

1. How many ram modules (sticks) installed?
2. Do you have your motherboard manual.
3. Is your ram installed as per the motherboard manual.
4. Exactly what ram do you have installed?
These might help.

 

[ICIT2LOL]

Is your ram installed as per the motherboard manual.

Yep good point as boards specify slots 1 and 3 as the primary slots and other 2 and 4 as I found out when using a Gigabyte board and then an Asus board.

If you still have issues then maybe the RAM should tested http://www.sevenforums.com/tutorials/105647-ram-test-memtest86.html if you do though and you get errors on just one stick then you will need to try it in the next primary slot because it might be bad and also use only this memtest and always make sure at least 8 passes anything less is useless - unless errors come up fairly quickly then you can stop it and move the stick and errors come up in the second slot I would be very suspicious of the stick.

It is a tedious process but it will eliminate memory problems as an issue.

 

[Mellon Head]

@iCsav:

Just so that we're all clear on the hardware angle of things, do you have the 6 pin power connector plugged into the GPU?

The problem that you're having could be explained by that connector not being plugged in.

 

[iCsav]

So. An update guys.
Some things I noticed recently:

1. The full screen issue (post #3) was solved by itself within a day or two of posting. I has come up again today. Same stuff as earlier. Only difference was that this time, when I pressed some keys, I heard the sound that is played when a key is pressed numerous times in succession or no command is assigned to the key.

2. In the Speccy screenshot I'm posting below, I saw that all of my PCIe slots were being used. I had a look at the motherboard recently, and iirc this was not the case. I'll check again to be sure.

3. In the second MSI Afterburner screenshot, I noticed that the temperature limit was set to 80 °C. Could this be a problem?

4. As I had mentioned before, Osu! was the only game I could play just fine. Today, I found that after testing another game for the same results and then the full screen video one, even osu doesn't last anymore. I can hardly try 1 round and when displaying results or going back to main menu, the BSOD happens. Somehow, now I can't even open GeForce Experience. BSOD as soon as I go to that window.

@Mellon Head: Sorry, I was not keeping track of the thread so I didn't notice your post. I check it as soon as possible.
I'll test for the RAM today, though I can't say about the PSU pin.
Anyways, here are the screenshots. I took as many as possible to capture all relevant information.