3

Whenever I boot up a game that is directx dependent, my PC loses power and reboots automatically. No intermediate warning of a reboot/restart, either. I have played non directx based games fine, such as Minecraft, or web based unity games. However, games like Counter-Strike: Global Offensive, Call of Duty 6, or Diablo III all cause my computer to shut off. These are all directx based games.

I've been researching this issue for a while, consulting several specialists and IT gurus in the process.

My Specs:

Processor: i3-2120 @ 3.3 ghz (4 cpu) x64
Old GFX Card: Nvidia GT 620 (2gb vram)
New GFX Card: AMD R7 260x (1gb vram)
Old OS: Windows 8.1 64-bit
New OS: Windows 7 64-bit
Motherboard: Gigabyte Z77-D3H (Please note that i recently flashed my BIOS from F14 to F22)
Power Supply: Antec EA 650W Green-Series

My Steps Taken:

  1. Because i was already having -many- issues with an old GT 620, I replaced it with a brand new AMD R7 260x.

  2. After this I had issues installing the card due to an outdated BIOS. I flashed the BIOS to an updated version. The card works now.

  3. During my issues with the AMD card installation, I made absolutely sure that the 12V rail the GFX card is pulling current from was not the rail used by my drives, my motherboard, or my CPU. My power supply doesn't seem to be having issues with the amount of power my PC requires. 650 Watts seems to be more than enough. The same issues occur.

  4. After all of this I bought a brand new Corsair 60GB SSD, and installed a brand new copy of Windows 7 on it. The same issue with game crashes still occur on this copy of windows.

  5. My final tests were on the RAM. I recently replaced old ram with a pair of Corsair sticks. One of which had a chance of being corrupted, but worked anyways. The old sticks of RAM weren't broken, they were just only 2x2GB and i wanted 2x4GB for 8GB of total ram instead of 4GB. I've made -several- combinations of my RAM sticks to see if RAM was the culprit (the old RAM sticks has been Memtested, I am 100% certain they are not corrupt at all). My BIOS is also properly configured with the correct timings for the RAM. My problem still persists.

My Conclusion/Diagnosis:

And after some more minor research, I noted that other people who are having similar issues seem to also have the same motherboard as i do! How intriguing...

Noting this fact, I've come to my own diagnosis that my motherboard, after two and a half years of service, has gone bad. However, I am uncertain and am not excited for the idea of buying a new mobo - and possibly a new CPU because of how long my current i3 has been inside my current mobo.

Are there any possible things that could be bad in my PC besides the motherboard?

Thank you for actually reading all of this.

Edit 1:

As of now, with the PSU being the most practical culprit, my computer shuts off and reboots after just a few moments of being online. I will replace the PSU with the proper minimum wattage (around 450W, i'll probably grab another 500-650W supply)

8
  • 4
    Is there anything of interest in event viewer? Does anything happen with opengl games? Does a benchmark like furmark crash?
    – Journeyman Geek
    Sep 16, 2014 at 2:49
  • A motherboard just doesn't not work in this way, you don't mention, what version of direct x not your drivers
    – Ramhound
    Sep 16, 2014 at 2:50
  • @Ramhound My drivers are the latest versions provided by AMD Catalyst. It doesn't tell me the version. Also, this occurs with any version of directx.
    – Ashley
    Sep 16, 2014 at 3:39
  • 3
    @TristenHorton Can you replace you PSU? the fact that it seems fine doesn't mean it is. It might be too weak to hold under strain.
    – EliadTech
    Sep 16, 2014 at 6:00
  • 3
    I'd go down the PSU route. Seems like the mobo isn't getting the power it needs during high power draw. Try running other benchmarks that include DirectX as well as high mem, high cpu.
    – dubRun
    Sep 16, 2014 at 15:27

1 Answer 1

3

This looks like a power supply "problem". It seems your old card - the Nvidia GT 620 - has a power draw of up to 49W. Few PCs have a power draw of significantly more than 100W, without GPU. A standard 200W power supply covers this easily.

I can't manage to find the exact spec for your new card, the AMD R7 260x, but from what I can find it clearly has a much higher power draw and requires at least a 300W power supply, maybe even higher - see the box or the manual for the minimum requirements. In this test, a system with the AMD R7 260x had a power draw of 221W (with about 113W caused by the graphics card). A PC with a power supply that delivers no more than 200W should shut down if this happens. The linked test also explains that the power is only drawn when the graphics card is under load, i.e. when you start a game that makes use of the graphics card.

In your question you state that you are using a 650W power supply, so if everything works by design, this shouldn't happen. My guess is an issue with the power supply, or an issue with the power supplied to the power supply (are you using a faulty UPS? did you try a different wall socket?). It's also possible that one of your components draws excessive power far above spec, but if that is the case usually it's usually easy to smell the results.

4
  • 2
    Agreed this sounds like either an under rated psu or a cheap one.
    – jharrell
    Sep 16, 2014 at 17:13
  • 2
    Or a failing power supply. I've seen old ones deliver well below their rated power and turn off if asked for more. Sep 16, 2014 at 18:25
  • 1
    You will notice that i list my power supply and its wattage of 650W. It is not that my antec is underrated or cheap, it is most likely that it is old and degraded. Please read my newest edit to my post, it updates the current situation. The R7 260x does take approx. 300W.
    – Ashley
    Sep 17, 2014 at 1:55
  • @TristenHornton Sorry, reading is hard. Updated the reply.
    – Peter
    Sep 17, 2014 at 5:08

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .